Polímeros: Ciência e Tecnologia (Polimeros) 1st. issue, vol 27, 2017 by Polímeros: Ciência e Tecnologia (Polimeros)

Substituindo Alumínio em dissipadores de calor CoolPoly® termoplásticos termicamente condutivos permitem o desenvolvimento de dissipadores de calor com arquiteturas e desempenhos únicos. Os termoplásticos termocondutivos CoolPoly® permitem a moldagem por injeção de estruturas que atendem ao desenho de equipamentos eletrônicos, realizando a transferência de calor com a mesma eficiência de trocadores de calor produzidos em alumínio. Quando necessário, os termoplásticos termicamente condutivos CoolPoly® podem incorporar funções dielétricas, reduzindo as tolerância dimensionais requeridas entre os componentes geradores de calor, reduzindo o calor e consequentemente a resistência térmica total. Os dissipadores de calor podem também ser moldados com CoolPoly® elastômero termoplástico, proporcionando flexibilidade suficiente para eliminar a necessidade de materiais de interface adicionais. Os dissipadores de calor produzidos com CoolPoly® tem vantagens significativas em relação aos dissipadores produzidos em alumínio, pois evitam a necessidade de processos adicionais relacionados aos processos de transformação de alumínio, além de evitar a necessidade de componentes adicionais para garantir o isolamento elétrico, uma vez que a base dielétrica CoolPoly® proporciona um bom isolamento elétrico e condução de calor.

Polímeros

OOLPOLY® THERMALLY CONDUCTIVE PLASTICS

Dissipadores de calor injetados com CoolPoly® oferecem: - Redução de custos

- Liberdade no projeto

- Isolantes elétricos

- Baixo fluxo de ar A maior resistência térmica em sistemas eletrônicos é muitas vezes a diferença entre o componente e o dissipador de calor. A diferença é especialmente significativa na concepção de soluções multi-chip. A base moldada tridimensional minimiza a diferença para cada fonte geradora de calor. Ao contrário de uma base plana bidimensional, uma base moldada pode ainda abranger os lados de componentes geradores de calor aumentando ainda mais a dissipação do calor. Estruturas aletadas são particularmente atraentes em ambientes de fluxo de ar limitado devido à sua espessura mínima e eficiência. Dissipadores de calor injetados a são soluções de gerenciamento térmico, ideal para módulos multi-chip, servidores, conversores de energia e outras soluções de gerenciamento térmico.

VOLUME XXVII - Issue I - JAN/MAR - 2017

- Melhor transferência de calor

Polímeros, now only in English Celanese, Alameda Ministro Rocha Azevedo, 38 conj. 102/604 – São Paulo/SP – Brasil CEP 01410-000 Telefone: (11) 31473360/3370, contato@celanese.com Para maiores informações visite: celanese.com/engineered-materials

I SEMINÁRIO INTERNACIONAL DE

CROSSLINKING VIA PERÓXIDOS Novembro 2017

Ÿ Seminário paralelo a XIV Jornadas

La noamericanas del Tecnología del Caucho

07/11 - Porto Alegre - RS 13/11 - São Paulo - SP

Ÿ Palestrantes Nacionais e Internacionais Ÿ Coquetel

Vagas Limitadas Inscreva-se!

Evento Comemora vo:

Apoio:

www.re loxseminario.vpeventos.com

SOLUÇÕES TECNOLÓGICAS

anos

Soluções Especiais para

Transformação de Plastômeros Surpreenda-se com as soluções tecnológicas Retilox! ROTOMOLDAGEM O RETI ROTOMOLD é um aditivo modificador reológico do PE, possui várias grades, atendendo as mais rígidas especificações na Rotomoldagem, seja para crosslinking total, ou como modificador da reologia do polímero, mantendo a reciclabilidade. Proporciona propriedades físicas excepcionais (abrasão, impermeabilidade, resistência ao impacto).

PVC

RETICROSS: trata-se de uma série de aditivos destinados a modificação de impacto de perfis e tubos. No segmento de calçados, aditivos especiais para emborrachamento, melhora das características físicas dos artefatos expandidos e redução do brilho.

SOPRO RETISOPRO é um aditivo modificador reológico do PE, em Masterbach, que melhora todas as características físico-químicas do PEAD virgem ou reciclado, mantendo a reciclabilidade do produto.

EXTRUSÃO DE TUBOS PE Aditivos para reticular (PEXa / PEXb PEXc) para melhor resistência a: temperatura, pressão, impacto e durabilidade. Aditivos especiais para melhoria das propriedades físicas, mantendo a reciclabilidade.

INJEÇÃO E RECICLAGEM DE PP

INJEÇÃO PE

Aos transformadores que reciclam ou injetam PP e necessitam corrigir o índice do melt index dos compostos, ou fazer modificação do impacto, a Retilox oferece grades exclusivas de excelente custo-benefício.

SOLUÇÕES TECNOLÓGICAS

anos

Retilox Química Especial Ltda Sistema de Gestão da Qualidade Certificado ISO 9001:2008

www.retilox.com.br contato@retilox.com.br +55 11 4705-9460

http://dx.doi.org/10.1590/0104-1428.2701

Polímeros, from now onwards published only in English….. As promised, from this volume (v27, number1, Jan-Mar/2017) onwards Polímeros is going to publish articles written exclusively in English. They will be an average of 48 articles per year, evenly distributed in four numbers. This was the last step towards becoming a really international scientific journal, which give to all of us a great pleasure in been part of the world community. In the beginning of this year Polímeros has published an extra issue containing the very last 16 accepted articles written in Portuguese. Recently one of our Associated Editor Prof. João B. P. Soares from University of Alberta, Edmonton, Canada, facing an increase in other duties, has asked us to resign from this position. We lose a great team leader, but we understand his reasons and sincerely acknowledge the work he has done to Polímeros. We still count on him to express his opinion as AdHoc Reviewer in his field of expertise. Thank you Prof. João, keep in touch. Towards the end of last year the members of the Council Board have approved and implemented the article-charge, to all accepted articles submitted to Polímeros. Up to now authors from 77% of all accepted articles agreed to help funding the journal. Next time you apply a project do not hesitate to budget funds to publish its results. Myself, in behalf of the Editorial Board, thanks all our Polymer Community Members for understanding the reason why and accepting the new financial police of Polímeros, this is the warranty it needs to sustain itself in the future ….

Sebastião V. Canevarolo Editor-in-Chief

Polímeros, 27(1), 2017

E E E E E E E E E E E E E E E E E E E E E E E E E

E E E E E E E E E E E E E E E E E E E E E E E E E E E E

ISSN 0104-1428 (printed) ISSN 1678-5169 (online)

P o l í m e r o s - I ss u e I - V o l u m e X X V I I - 2 0 1 7 I n d e x e d i n : “ C h e m ic a l A b s t r a c t s ” — “ RA P RA A b s t r a c t s ” — “A l l - R u s s i a n I n s t i t u t e o f S ci e n c e T e c h n ic a l I n f o r m a t i o n ” — “ R e d d e R e v i s t a s C i e n t i f ic a s d e A m e r ic a L a t i n a y e l C a r i b e ” — “ L a t i n d e x ” — “ W e b o f S ci e n c e ”

and

Polímeros E d i t o r i a l C o u nci l

Editorial Committee

Marco-Aurelio De Paoli (UNICAMP/IQ) - President

Sebastião V. Canevarolo Jr. – Editor-in-Chief

Members

A ss o ci at e E d i t o r s

Adhemar C. Ruvolo Filho (UFSCar/DQ) Ailton S. Gomes (UFRJ/IMA) Alain Dufresne (Grenoble INP/Pagora) Antonio Aprigio S. Curvelo (USP/IQSC) Bluma G. Soares (UFRJ/IMA) César Liberato Petzhold (UFRGS/IQ) Cristina T. Andrade (UFRJ/IMA) Edson R. Simielli (Simielli - Soluções em Polímeros) Elias Hage Jr. (UFSCar/DEMa) Eloisa B. Mano (UFRJ/IMA) João B. P. Soares (UAlberta/DCME) José Alexandrino de Sousa (UFSCar/DEMa) José António C. Gomes Covas (UMinho/IPC) José Carlos C. S. Pinto (UFRJ/COPPE) Júlio Harada (Harada Hajime Machado Consutoria Ltda) Laura H. de Carvalho (UFCG/DEMa) Luiz Antonio Pessan (UFSCar/DEMa) Luiz Henrique C. Mattoso (EMBRAPA) Osvaldo N. Oliveira Jr. (USP/IFSC) Raquel S. Mauler (UFRGS/IQ) Regina Célia R. Nunes (UFRJ/IMA) Richard G. Weiss (GU/DeptChemistry) Rodrigo Lambert Oréfice (UFMG/DEMET) Sebastião V. Canevarolo Jr. (UFSCar/DEMa) Silvio Manrich (UFSCar/DEMa)

Adhemar C. Ruvolo Filho Alain Dufresne Bluma G. Soares César Liberato Petzhold João B. P. Soares José António C. Gomes Covas José Carlos C. S. Pinto Regina Célia R. Nunes Richard G. Weiss Rodrigo Lambert Oréfice

D e s k t o p P u b l is h in g

www.editoracubo.com.br

“Polímeros” is a publication of the Associação Brasileira de Polímeros São Paulo 994 St. São Carlos, SP, Brazil, 13560-340 Phone: +55 16 3374-3949 emails: abpol@abpol.org.br / revista@abpol.org.br http://www.abpol.org.br Date of publication: March 2017

Financial support:

Polímeros / Associação Brasileira de Polímeros. vol. 1, nº 1 (1991) -.- São Carlos: ABPol, 1991Available online at: www.scielo.br

Quarterly v. 27, nº 1 (Jan./Fev./Mar. 2017) ISSN 0104-1428 ISSN 1678-5169 (electronic version)

Website of the “Polímeros”: www.revistapolimeros.org.br

1. Polímeros. l. Associação Brasileira de Polímeros. E2

Polímeros, 27(1), 2017

Editorial Section Editorial................................................................................................................................................................................................E1 News....................................................................................................................................................................................................E4 Agenda.................................................................................................................................................................................................E5 Funding Institutions.............................................................................................................................................................................E6

T e c h nic a l S e c t i o n Hyperbranched polyester polyol plasticized tapioca starch/low density polyethylene blends Manuel Guzmán, Diego Giraldo and Edwin Murillo.......................................................................................................................................... 1

Mechanical recycling of tags and labels residues using sugarcane bagasse ash Guilherme Augusto Mantovani, Jean Halison de Oliveira, Andressa dos Santos, Andrelson Wellington Rinaldi, Murilo Pereira Moisés, Eduardo Radovanovic and Silvia Luciana Fávaro........................................................................................................ 8

Removal of Remazol brilliant violet textile dye by adsorption using rice hulls Geyse Adriana Corrêa Ribeiro, Domingos Sérgio Araújo Silva, Clayane Carvalho dos Santos, Adriana Pires Vieira, Cícero Wellington Brito Bezerra, Auro Atsushi Tanaka and Sirlane Aparecida Abreu Santana....................................................................... 16

The effect of gelatin amount on the properties of PLA/TPS/gelatin extruded sheets Ana Paula de Oliveira Pizzoli, Fabio Yamashita, Odinei Hess Gonçalves, Marianne Ayumi Shirai and Fernanda Vitória Leimann................................................................................................................................................................................. 27

Use of pH-thermosensitive hydrogels for nickel ion removal and recovery Álvarez Casillas Cesar Andrés and Cortés Ortega Jorge Alberto..................................................................................................................... 35

Effect of shrimp shells milling on the molar mass of chitosan Helton José Alves, Maristela Furman, Cristie Luis Kugelmeier, Clayton Rodrigues de Oliveira, Vanessa Rossato Bach, Karine Natani Lupatini, Andressa Caroline Neves and Mabel Karina Arantes................................................................................................ 41

Characterization of low cost orally disintegrating film (ODF) Riana Jordao Barrozo Heinemann, Fernanda Maria Vanin, Rosemary Aparecida de Carvalho, Marco Antonio Trindade and Carmen Sílvia Fávaro-Trindade........................................................................................................................................................................ 48

Rice husk ash as filler in tread compounds to improve rolling resistance Mônica Romero Santos Fernandes, Ana Maria Furtado de Sousa and Cristina Russi Guimarães Furtado.................................................... 55

Characterization of biopolymers and soy protein isolate-high-methoxyl pectin complex Mírian Luisa Faria Freitas, Kivia Mislaine Albano and Vânia Regina Nicoletti Telis...................................................................................... 62

A quantitative relationship between Tgs and chain segment structures of polystyrenes

Xinliang Yu and Xianwei Huang........................................................................................................................................................................ 68

Nanocomposites films obtained from protein isolates of mechanically deboned chicken meat added with montmorillonite Bruna da Silva Menezes, William Renzo Cortez-Vega and Carlos Prentice...................................................................................................... 75

In vitro and in vivo cell tracking of PKH26-labeled osteoblasts cultured on PLDLA scaffolds Alice Rezende Duek, Gabriel Ciambelli Dias da Costa, Bruna Antunes Más, Maria Lourdes Peris Barbo, Adriana Cristina Motta and Eliana Aparecida de Rezende Duek..................................................................................................................... 83

Cover: Histological analysis of PLDLA scaffolds cultured with osteoblast cells after 4 weeks (A...). Arts by Editora Cubo.

Polímeros, 27(1), 2017

I I I I I I I I I I I I I I I I I

Global plastics market to register CAGR of 7.03% from 2017 to 2025

N N N N N N N N N N N N N

Global plastics market is projected to grow at a Compound Annual Growth Rate - CAGR of 7.03% from 2017 to 2025, as per Orian Research. Plastics are used in a variety of industries: construction, packaging, appliance, automobile, textile, transportation, and many others. A large number of manufacturers supply many different products to numerous end-users for a multitude of applications. The major drivers for plastics market would be its low cost, flexibility of use, easy manufacturing capabilities, growing construction in Asia-Pacific region among others. Thus, plastics can be regarded as synthetic or semi-synthetic organic solids which can be transformed into several useful products. Also, the paradigm shift of technology to recycle plastics for end-use applications provide significant opportunities to the entire supply chain of the plastics market. Polyethylene is derived from polymerization of ethylene and has the properties like chemical & thermal resistivity, flexibility, electrical & thermal insulation among others. Due to its light weight and easy manufacturing capability, polyethylene has found applications in various industries such as construction, electronics, and automotive. The polyethylene market is projected to grow at a considerable CAGR in the emerging economies. Also, the demand from different domains such as injection moulding, food & beverages and packaging has fuelled the demand. On the other hand, Polyethylene terephthalate (PET) is projected to grow at a significant CAGR during the forecast period (2017-2025). Asia Pacific plastics market is projected to account for a share of around 45% of the global plastics market by 2025. The growing automotive and construction sectors in countries such as India and China with the adoption of rapid technological advancement has significantly boosted Asia-Pacific’s plastics market. Presence of major automotive industries in Germany and France should drive European plastics market along with the growing demand for high performance and environmental friendly plastics materials such as bio‑based plastics and engineering thermoplastics. Also, the Central & South American region is projected to grow at a considerable CAGR during the forecast period. Source: Plastemart - www.plastemart.com

Celanese Completes Acquisition of Nylon Compounding Division of Nilit Celanese Corporation, a global technology and specialty materials company, today announced it has completed the acquisition of the nylon compounding division of Nilit, a major independent producer of high performance nylon polymers and compounds. Financial details of the transaction are not being disclosed at this time. This acquisition further extends Celanese’s leadership position in the engineered materials business to a global nylon solutions provider. The acquisition E4

includes Nilit Plastics’ nylon compounding product portfolio, customer agreements and manufacturing, technology and commercial facilities in Germany and China. In addition, the acquisition is complementary to the company’s capabilities and track record of innovation, quality and service. As previously announced, Nilit will retain ownership of its nylon fibers and nylon polymerization businesses worldwide, including facilities in Israel, the United States, China and Brazil. Nylon compounds continue to be a material of choice in automotive, E&E, consumer and industrial applications. This acquisition delivers on Celanese’s intention to complement and grow its broad portfolio by becoming a leading, global nylon compound supplier. “Nylon is increasing in applications and end uses in growth industries where Celanese is already focusing significant product, solution and customer development activities,” said Scott Richardson, senior vice president of the Celanese engineered materials business. “The addition of the Nilit nylon compounding product portfolio will extend Celanese’s engineered materials solutions offering, and when combined with the company’s world-class operating model, we are well positioned to be the first choice materials solutions provider for our customers.” Celanese will integrate Nilit Plastics’ nylon compounding product portfolio and production capabilities into its engineered materials business to include the following registered brands: • FRIANYL flame retardant grades for electrical and electronics industries, meeting most stringent industry requirements, and in almost all colors. • NILAMID technical grades for industrial and automotive applications. • NILAMID specialty portfolio for extended performance requirements in terms of thermal, electrical, mechanical and tribological properties in particular. • ECOMID grades containing high-quality polyamide fibers and textile recycled for a combination of quality, high lot-to-lot consistency and competitiveness. The most recent additions to the NILIT Plastics range, developed to cater to new market trends, include “XS” and “XT” types. NILAMID XS grades are compounds based on semi-aromatic polyamides, and FRIANYL XT and NILAMID XT, are compounds based on polyphthalamide (PPA) provide superior performance, particularly in terms of high heat resistance and mechanical strength. Richardson concluded: “The combined portfolios of Celanese, SO.F.TER. Group, and Nilit Plastics significantly increases the solution options available to our customers across thermoplastics and elastomers. Our combined polyamide portfolio will represent a broad range of PA6- and PA6,6-based solutions delivering a range of functional capabilities and modifications for customer needs globally.” Source: Business Wire - www.businesswire.com Polímeros, 27(1), 2017

October II Semana de Composites Avançados SAMPE Brasil Date: October 2-6, 2017 Location: São José dos Campos - Brazil Website: sampe.com.br/pt-br/ii-semana-de-compositesavancados-sampe-brasil-2017 Performance Polyamides USA Date: October 3-4, 2017 Location: Pittsburgh - USA Website: www.amiplastics.com/events/event?Code=C824 Polymer Failure & Defects: Problem Solving Case-Histories Date: October 3-5, 2017 Location: Marriott Marquis - USA Website: www.innoplastsolutions.com 25th Polychar - Annual World Forum on Advanced Materials Date: October 9-13, 2017 Location: Kuala Lumpur - Malaysia Website: www.25polychar.org.my Polyolefin Additives Date: October 10-12, 2017 Location: Cologne - Germany Website: www.amiplastics.com/events/event?Code=C820 IUPAC-FAPS 2017 Polymer Congress on Smart Materials for Emerging Technology Date: October 11-13, 2017 Location: Jeju - South Korea Website: iupac.org/event/iupac-faps-2017-polymer-congress Annual Meeting on Biopolymers Date: October 12-13, 2017 Location: Osaka - Japan Website: www.meetingsint.com/chemical-engineeringconferences/biopolymers 14° Congresso Brasileiro de Polímeros Date: October 22-26, 2017 Location: Águas de Lindóia - Brazil Website: www.cbpol.com.br 7th International Conference and Exhibition on Biopolymers and Bioplastics Date: October 19-21, 2017 Location: San Francisco - USA Website: biopolymers-bioplastics.conferenceseries.com

November Long-Fibre Thermoplastics 2017 Date: November 7-8, 2017 Location: Cologne - Germany Website: www.amiplastics.com/events/event?Code=C822 Plastimagen 2017 Date: November 7-10, 2017 Location: Mexico City - Mexico Website: www.plastimagen.com.mx PLASTEC Minneapolis Date: November 8-9, 2017 Location: Minneapolis - USA Website: plastecminn.plasticstoday.com 2nd International Conference and Exhibition on Polymer Chemistry Date: November 15-17, 2017 Location: San Antonio - USA Website: polymer.conferenceseries.com

30th International Plastics & Rubber Machinery, Processing & Materials Exhibition Date: November 15-18, 2017 Location: Jakarta - Indonesia Website: www.plasticsandrubberindonesia.com 9th International Exhibition for Plastics Industry Date: November 22-24, 2017 Location: Almaty - Kazakhstan Website: plastworld.kz

December Polymers in Flooring 2017 Date: December 5-6, 2017 Location: Berlin - Germany Website: www.amiplastics.com/events/event?Code=C849 Fire Resistance in Plastics 2017 Date: December 5-7, 2017 Location: Cologne - Germany Website: www.amiplastics.com/events/event?Code=C847 33rd Annual Meeting of the Polymer Processing Society (PPS-33) Date: December 10-14, 2017 Location: Cancun - Mexico Website: pps-33.com Polymers in Footwear Date: December 11-12, 2017 Location: Duesseldorf - Germany Website: www.amiplastics.com/events/event?Code=C867

January 21st Thermoplastic Concentrates Date: January 23-25, 2018 Location: Coral Springs - USA Website: www.amiplastics.com/events/event?Code=C852 21st International Trade Fair Plastics and Rubber (INTERPLASTICA 2018) Date: January 23-26, 2018 Location: Moscow - Russia Website: www.interplastica.de Polyethylene Films Date: January 30 - February 1, 2018 Location: Coral Springs - USA Website: www.amiplastics.com/events/event?Code=C853

February Salone SamuPlast Date: February 1–3, 2018 Location: Pordenone – Italy Website: www.samuexpo.com/samuplast PLASTEC West Date: February 6–8, 2018 Location: Anaheim – USA Website: plastecwest.plasticstoday.com 10th International Plastics Exhibition, Conference & Convention (PLASTINDIA 2018) Date: February 7–12, 2018 Location: Gujarat – India Website: www.plastindia.org/plastindia-2018/index.html

Polímeros, 27(1), 2017 E5

A A A A A A A A A A A A A A A A A A A A A

ABPol Associates Sponsoring Partners

Institutions UFSCar/ Departamento de Engenharia de Materiais, SP SENAI/ ServiĂ§o Nacional de Aprendizagem Industrial Mario Amato, SP UFRN/ Universidade Federal do Rio Grande do Norte, RN

PolĂmeros, 27(1), 2017

ABPol Associates Collective Members A. Schulman Plásticos do Brasil Ltda. Aditive Plásticos Ltda. Avamplas – Polímeros da Amazônia Ltda. CBE – Grupo Unigel Colorfix Itamaster Indústria de Masterbatches Ltda. Cromex S/A Cytec Comércio de Materiais Compostos e Produtos Químicos do Brasil Ltda. Formax Quimiplan Componentes para Calçados Ltda. Imp. e Export. de Medidores Polimate Ltda. Innova S/A Instituto de Aeronáutica e Espaço/AQI Jaguar Ind. e Com. de Plásticos Ltda Johnson & Johnson do Brasil Ind. Com. Prod. para Saúde Ltda. Master Polymers Ltda. Milliken do Brasil Comércio Ltda. MMS-SP Indústria e Comércio de Plásticos Ltda. Nexo International Ltda. Nitriflex S/A Ind. e Com. Politiplastic Politi-ME. Premix Brasil Resinas Ltda. QP - Químicos e Plásticos Ltda. Radici Plastics Ltda. Replas Comércio de Termoplásticos Ltda. Uniflon - Fluoromasters Polimeros Ind .Com. Imp. Export.Ltda

Polímeros, 27(1), 2017 E7

Palestrantes

Prof. Alessandro Gandini Grenoble Institute of Technology

Prof. Sadhan C. Jana, Ph.D. University of Akron

Prof. Osvaldo Novais de Oliveira Universidade de São Paulo

Patrocinadores

Expositores

Promoção

Organização

Apoio

http://dx.doi.org/10.1590/0104-1428.04816

Hyperbranched polyester polyol plasticized tapioca starch/low density polyethylene blends Manuel Guzmán1, Diego Giraldo1 and Edwin Murillo2* Grupo de Materiales Polímericos de Ingeniería, Universidad de Antioquia, Medellín, Colombia 2 Grupo de Investigación en Materiales Poliméricos – GIMAPOL, Departamento de Química, Universidad Francisco de Paula Santander, San José de Cúcuta, Colombia

*edwinalbertomurillo@gmail.com

Abstract In this work, low density polyethylene (LDPE)/plasticized starch (TPS) blends were prepared. The TPS employed in this study was obtained by plasticization of tapioca starch with a hyperbranched polyester polyol. Differential scanning calorimetry analysis showed that the melting temperature increased with the TPS content. The opposite effect was exhibited in the crystallization temperature and additional changes were not observed during the heating. X-ray diffraction analysis showed a reduction in intensity of the peak at Bragg’s angle 17.5°, proving a diminution on A type crystallinity with the increasing amount of LDPE. Micrographs obtained by scanning electron microscopy exhibited starch granules without destructure. TPS acted as a filler to LDPE, since the mechanical properties (Young’s modulus and tensile strength) improved ostensibly. The Young’ modulus and tensile strength decreased with the amount of LDPE, however, the elongation at break exhibited an opposite behavior. Keywords: LDPE, thermoplastic starch, blends, properties.

1. Introduction Starch is a natural carbohydrate accumulated by green plants in the form of granules. It is composed of linear polysaccharide molecules (amylose) and branched molecules (amylopectin)[1]. The starch plasticization has spurred considerable interest in the last years. The starch presents a great importance as substrate due to wide availability, its low cost and its renewable character[2]. The investigation of new plasticizers constitutes a major topic in this context. The main plasticizer employed in the plasticization process of starch is glycerol[3], but several other compounds such as urea[4], ethanolamine[4] and sorbitol[5] have also been employed. The hyperbranched polyester polyols (HBP) could be an alternative to plasticization process of starch due to the low viscosity it has in the molten and solution state, as well as its high number of OH groups and small hydrodynamic dimensions[6-9]. Low density polyethylene (LDPE) is a polymer of the largest tonnage of consumption in the world. LDPE has excellent mechanical properties and it is employed mainly in packaging industry[10,11]. This polymer is derived from petrochemical sources and it is not biodegradable. In order to reduce the dependence on the depleting petrochemical resources, this material was blended with biopolymers such as starch[12]. LDPE/starch blends are not compatible because LDPE has no polar groups that could interact with starch. However, TPS has been the target of research due to its low cost and ability to be modified or blended with other polymers in order to improve their properties and processing[13]. Furthermore, TPS blends allow for obtaining the biodegradable or partially biodegradable materials, which is the reason for the preparation of LDPE/starch blends.

Polímeros, 27(1), 1-7, 2017

Blends of recycled LDPE with starch would have two environmental advantages: the substitute of the virgin synthetic thermoplastic matrix by post-consumer materials, and the end products that would be biodegradable and cheap[14]. LDPE/starch blends have been prepared from a dispersion of starch granules in a LDPE matrix. Such blend, without a starch plasticizer, exhibits poor physico-chemical properties, even with only 10% of granulated starch[1]. St-Pierre et al.[12] analyzed the performance of mixtures of TPS (using glycerol as a plasticizer) in a LDPE matrix. Plasticization process was carried out in a single-screw extruder with gas removal, connected to a twin-screw extruder, where finally the components of the mixture were blended. Ning et al.[15] studied the effect of citric acid on the properties of glycerol/starch/linear low-density polyethylene blends. The presence of citric acid (CA) improved the dispersion and plasticization of starch. The rheological study proved that CA could decrease the viscosity and improve the fluidity of the blends. The blends exhibited poor interfacial adhesion and the mechanical properties of the blends without CA were very poor. The tensile strength and the elongation at break were greatly enhanced in the presence of CA. Pedroso and Rosa[14] prepared recycled LDPE/corn starch blends, the proportions of starch in the blends were 30, 40 and 50 wt%. The addition of starch to LDPE decreased melt flow index (MFI) values, tensile strength and elongation at break. This behavior was the same as showed by virgin LDPE/corn starch blends. The melting and crystallization temperatures of the blends were the same as those of pure polymers.

O O O O O O O O O O O O O O O O

Guzmán, M., Giraldo, D., & Murillo, E. Garg and Jana[16] evaluated the properties of LDPE/starch films using crosslinked and glycerol modified starch. The tensile strength, elongation and MFI of the films containing crosslinked starch were higher than those containing native starch, but the burst strength showed an opposite trend. For native starch modified with glycerol, the elongation and melt flow index of the films increased.

2.2 Preparation of the blends

Rodriguez-Gonzalez et al.[1] studied the effect of processing and glycerol content on the morphology and properties of LDPE/starch blends. Different types of commercial LDPE were used in the study. Micrographs showed that by using 27.5% or higher glycerol content and the high viscosity LDPE it was possible to break the granular structure of starch. It should be noted that granular structure of starch is preserved at low levels of glycerol and that granules act as filler[1].

2.3 Characterization of the blends

Guzmán and Murillo[17] prepared blends of maleic-anhydride (MA)-grafted polyethylene (LDPE-g-MA) and starch plasticized with a hyperbranched polyester - HBP (TPS) employing proportions of 20/80, 30/70, 40/60 and 50/50. They observed that HBP acted as a plasticizer for starch and the LDPE-g-MA/TPS blends, and that the A-type crystallinity of the starch was reduced. The thermal stability of the blends showed no significant changes regarding those of neat LDPE-g-MA and TPS. The viscosity of the blends was reduced with increasing shear rate and starch granules were observed by scanning electron microscopy. Furthermore, the tensile testing of the blends with increasing content of LDPE-g-MA showed a reduction in the tensile modulus and tensile strength, and an augment in the elongation at break. According to the reviewed literature, there has been no report on the studies of the blends of LDPE and starch plasticized with a HBP (TPS) with an exception of LDPE-g-MA/TPS blends[17]. Therefore, the goal of this study was to prepare LDPE/TPS blends and evaluate the influence of the proportions of LDPE and TPS on the structural, thermal, rheological, morphological, and mechanical properties of all materials prepared. Further goal was to compare the properties of the LDPE/TPS blends prepared with those of the LDPE-g-MA/TPS blends[17].

2. Experimental Section 2.1 Materials In a previous study, TPS was prepared from tapioca starch (60% wt) and HBP. The composition of starch was 17 wt% of amylose and 83 wt% of amylopectin. HBP was of the fourth generation, prepared from pentaerythritol and 2,2-bis(methylol)propionic acid[6-8]. The structural, thermal, rheological, morphological and mechanical properties of the TPS have been reported earlier[17]. LDPE 132I was supplied by Dow Plastics. This material has a density of 0.921 g/cm3 and melt flow index of 0.22 g/10 min, whereas the thermal properties determined by DSC and TGA are as follows: melting temperature (Tm) 114.1 °C, melting enthalpy (ΔHm) 91.5 J/g, crystallization temperature (Tc) 89.8 °C, and decomposition temperature (Td) 469.7 °C[11]. 2

LDPE and TPS were mixed before to be added to the torque rheometer. Mixtures of LDPE with TPS were prepared using a Thermo Scientific torque rheometer at a temperature of 150 °C at 50 rpm and a residence time of 6 minutes. The proportions of LDPE and TPS are given in Table 1.

For IR analysis, films of the blends were analyzed on a FTIR spectrometer Spectrum One (Perkin Elmer) in the spectral range between 500 and 4000 cm-1 performing 8 scans. The thermal transitions of the materials were determined by differential scanning calorimetry (DSC) using a TA Instruments Q-100 with a heating rate of 10 °C/min in the temperature range from 0 to 200 °C. Thermogravimetric analysis (TGA) was performed to analyze the thermal stability of the samples. These measurements were performed on a TA Instruments Q-500 at a heating rate of 10 °C/min in the temperature range from 50 to 550 °C. The X-ray diffraction (XRD) was performed with the purpose to identify crystalline forms present in the LDPE/TPS blends. Diffractograms were obtained on a PANalytical X’Pert PRO MPD diffractometer using Cu-K alpha radiation (λ=1.5406 A). The voltage and the operating current were 45 KV and 40 ms, respectively. The diffractograms were measured in the range of Bragg angle (2θ) from 10° to 30°. The scanning electronic microscopy (SEM) analysis of fractured surfaces of the compounds was executed on a scanning electron microscope JEOL JSM-6490LV. SEM was used with a beam acceleration voltage of 10 to 20 kV. The rheological analyzes were performed on a Malvern Kinexus rotational rheometer using a plate-plate geometry of 20 mm at a controlled temperature of 130 °C under static and dynamic conditions. The static analysis was performed in a static mode at shear rates from 0.01 to 1000 s–1. The rheological parameter recorded was the apparent viscosity (η). The dynamic mode was used for characterizing the viscoelastic behavior of the samples. Frequency sweeps were performed from 0.1 to 100 Hz to a value of 0.2 % strain. For tensile tests, the IV type test specimens were obtained with a piston-cylinder injecting machine at 150 °C and a residence time of 4 minutes, employing a pressure of 75 Psi. Tensile tests were conducted under the standard ASTM D 638 at a rate of 5 mm/min until the specimen failed.

3. Results and Discussion Figure 1 presents the behavior of the materials during torque rheometry. The torque behavior with respect to the time is shown in Figure 1a.

Table 1. Proportions of LDPE and TPS of the blends. LDPE/TPS Blend 20/80 30/70 40/60 50/50

LDPE (wt%) 20 30 40 50

TPS (wt%) 80 70 60 50

Polímeros, 27(1), 1-7, 2017

Hyperbranched polyester polyol plasticized tapioca starch/low density polyethylene blends

Figure 1. Torque rheometry of the blends: (a) torque vs time; and (b) temperatures vs time.

Figure 2. IR spectra of the LDPE/TPS blends.

Torque of all the mixtures increased to a maximum value of 80 Nm in the first minute. When the mixture was completely melted, the torque remained constant at 13 Nm (Figure 1a). The temperature of the mixtures was also recorded (Figure 1b), however, notable differences between the blends were not observed. Due to heat transfer effects (melting process), the temperature was reduced to about 110 °C and then increased until the sample was melted. The same behavior was observed for LDPE-g-MA/TPS blends[17]. Figure 2 shows the IR spectra of the LDPE/TPS blends. The signal that appears at 3439 cm-1 is assigned to stretching of OH groups (intramolecular and intermolecular bonding); this peak appears at lower frequency than that of the TPS (3355 cm-1)[17]. This was due to structural rearrangement or new interactions that occurred during the blend preparation. The same was observed for LDPE-g-MA/TPS blends[17]. The signal about 2900 cm-1 is due to -CH2 stretching and the signal at 1650 cm-1 is assigned to bending of OH groups of water. The signal around 1470 cm-1 is attributed to angular deformation of -CH2. A signal appears at 1309 cm-1, which corresponds to C-O bending of ester groups belonging to HBP in TPS. The signals that appear at 1125 and 1140 cm-1 are due to C-O and C-OH stretching, respectively. The signal at 1010 cm-1 is attributed to C-O bond stretching of C-O-C groups in the anhydroglucose ring. The signal at 720 cm-1 is due to methylene rocking[8] and is characteristic of LDPE[11]. The signals that belong to C-O and C-OH bonds also appear in the spectra of TPS and LDPE-g-MA/TPS blends[17]. Figure 3 shows the DSC (Figures 3a and 3b) and TGA (Figures 3c and 3d) thermograms of the samples. Despite the fact that TPS exhibits the glass transition temperature (Tg) at 30.8 °C[17], none of the blends exhibited Tg, possibly indicating a decreased interaction between starch and HBP in the TPS. The Tm of the blends (Table 2 and Figure 3a) are lower than that of LDPE (114.1 °C)[11], which is possibly due to the

Figure 3. Thermograms showing (a) Endothermic process (DSC); (b) Exothermic process (DSC); (c) Weight vs temperature (TGA); and (d) Deriv. weight vs temperature (TGA). Polímeros, 27(1), 1-7, 2017

Guzmán, M., Giraldo, D., & Murillo, E. plasticizing effect of HBP. ΔHm of the blends (Table 2) are lower than that of LDPE (91.5 J/g)[11]. This is an indication that the crystallization process of LDPE is seriously affected by the presence of TPS. The Tm values of the materials obtained in this study are similar to those found to LDPE-g-MA/TPS blends[17]. Therefore, the grafted MA in these materials did not affect the Tm. ΔHm increased with the LDPE amount, meaning that in the same sense the crystallinity degree of the blends was augmented. The same behavior was also observed for LDPE-g-MA/TPS blends prepared with the same proportions of TPS, except for that with the proportion of 50/50 with the Tm value of 25.70 °C, which was assigned to different packaging[17]. Despite that, the results of Tm obtained in this study are comparable with those reported for LDPE-g-MA/TPS blends (90.7-96.7)[17]. The Tc of the LDPE (89.8 °C)[11] was lower than those of the blends, which is an indication that the crystallization process occurs easier for the blends than for LDPE. This behavior is attributed to the presence of TPS. The LDPE/TPS (50/50) blend with the lowest Tc presented more difficulty to crystalize than other blends. This is related with the highest amount of TPS employed. The same behavior was also observed for LDPE-g-MA/TPS (50/50) blends[17]. The values of Tm obtained in this study are comparable with those obtained for LDPE-g-MA/TPS blends with the same proportions of TPS[17]. The crystallization enthalphy (ΔHc) follows the same trend as ΔHm. The reduction in Tm, ΔHm and ΔHc of the LDPE have also been observed for LDPE/corn starch blends[14]. The ΔHm values of the LDPE/TPS are similar to those reported for LDPE-g-MA/TPS blends[17]. The results obtained by DSC analysis allow us to conclude that the presence of TPS affects the crystallization process of LDPE. Another important aspect is that the TPS does not act as a nucleating agent for LDPE (it would increase its degree of crystallinity), since ΔHm of LDPE in the LDPE/TPS blends was lower than that of the proportional fraction of neat LDPE (it is an indication of reduction of crystallinity) and it did not increase with the TPS content. TGA thermograms of the blends (Figures 3c and 3d) show that the first visible change occurs around 100 °C, where a small amount of water was evaporated from TPS. The HBP and starch present in the TPS can absorb water; in case of HBP, it occurs during its synthesis and in case of starch it can happen during the storage. This weight loss appears in all blends. The thermal decomposition of the TPS (Td1) occurs at 302.3 °C[17]. This was observed for all blends and it was reduced with the TPS content. The second thermal decomposition (Td2) appears about 340 °C and is associated with HBP employed as a plasticizer for starch[7,17]. The decomposition temperature of LDPE is 469.7 °C[11]. Therefore, the third thermal decomposition of the blends above 470 °C is assigned to LDPE (Td3). The Td values of the blends

are summarized in Table 2. The Td of starch varied between 299.4 and 307.7 °C, while between 338.5 and 348.4 there are Td of HBP, and between 470.0 and 478.3 °C those for LDPE (Table 2). However, no characteristic pattern is present, since all the values are very close. This behavior is attributed to rearrangement that possibly occurred during the processing of the blends. Slightly enhanced thermal stability of the blends with increasing amount of LDPE was also observed. This means that LDPE elevates the thermal stability of the blends due to its high thermal stability. It has been demonstrated that the introduction of the high molecular weight material such as LDPE, induces a gradual increase in the initial decomposition temperature[18]. The same thermal behavior as exhibited for the blends in this study was seen for LDPE/plasticized starch blends, where a Td was observed for TPS (300 °C) and LDPE (around 400 °C)[1]. Furthermore, similar results were obtained by Sailaja et al. for blends of LDPE and tapioca starch[10]. In another study, the same results on thermal stability were observed for LDPE-g-MA/TPS blends, obtained with the same proportions of TPS (three weight losses assigned to TPS, HBP and LDPE-g-MA)[17]. XRD diffractograms are presented in Figure 4. The TPS exhibits three peaks, which appear at 2θ=15°, 17.5° and 23°. The presence of these peaks in the TPS is due to the A type crystallinity of the tapioca starch[17]. In the blends, the intensity of the peak at 2θ=17.5° increased slightly with increasing TPS content. This behavior indicates that during the processing of the blends the TPS crystallinity was modified and starch granules were restructured, possibly

Figure 4. XRD difractograms of the LDPE/TPS blends.

Table 2. Thermal properties of the LDPE/TPS blends. LDPE/TPS blends 20/80 30/70 40/60 50/50

Tm (°C) 108.9 109.7 110.7 112.0

ΔHm (J/g) 14.84 17.55 25.90 30.70

Tc (°C) 94.7 94.3 93.8 91.1

ΔHc (J/g) 17.04 23.13 29.62 30.02

Td1 (°C) 304.0 303.1 304.9 307.7

Td2 (°C) 338.5 345.6 344.9 348.4

Td3 (°C) 470.0 475.6 475.6 478.4

Polímeros, 27(1), 1-7, 2017

Hyperbranched polyester polyol plasticized tapioca starch/low density polyethylene blends due to rupture and formation of new interactions between starch and HBP. This confirms the rearrangement of TPS in the LDPE/TPS blends. The same reduction in intensity of this peak was observed for LDPE-g-MA/TPS blends[17]. Another peak is observed at 2θ=21.5°, which increases its intensity as the LDPE content is enhanced. This is attributed to the crystalline regions of LDPE, because this peak is characteristic of LDPE[19] and was not observed for TPS. It is associated with a orthorhombic unit cell[17,20]. These results are comparable with the results obtained by DSC analysis, where the LDPE degree of crystallinity was reduced with the TPS content. The difractograms obtained in this study are very similar to those of the LDPE-g-MA/TPS blends with the same proportions of TPS, since a reduction of peak at 2θ=17.5° with increasing TPS content and an enhanced intensity of peak at 21.5° was observed in correlation with the LDPE content. The apparent viscosity (η) of the blends is shown in Figure 5. It has been noted that as the shear rate increases, the samples exhibit a decrease in viscosity (shear thinning) caused by dissociation of hydrogen bonding or chain disentanglement[21]. At shear rate of 10-1 s-1, the LDPE/TPS (20/80), (30/70), and (40/60) blends reduced the viscosity with the TPS content. The LDPE/TPS (50/50), however, did not follow the same behavior and this could be attributed to the lowest amount of HBP present in the TPS or interactions between the macromolecules of HBP. The viscosity values of the LDPE/TPS (40/60) and (50/50) were very similar in the range of shear rate between 10-1 and 100 s-1. This behavior was possibly due to a rearrangement of starch granules, since these samples were prepared with higher LDPE amounts than other blends. The viscosity of the LDPE/TPS blends at a shear rate of 10-1 s-1 was higher than that exhibited by TPS (1328 Pa.s)[17], but lower than that of LDPE (94.85 Pa.s)[11]. Therefore the presence of the TPS reduces the viscosity of the LDPE. LDPE/TPS blends exhibited the same rheological behavior (pseudoplastic) and comparable viscosity values with those obtained to the LDPE-g-MA/TPS blends[17]. Furthermore, they did not show any trend with the proportion of the TPS employed for the preparation of the blends. Figure 6 shows SEM micrographs of the LDPE/TPS blends. Starch granules within two phases and a rough surface were observed on all micrographs. This is an indication that these blends are incompatible, which was expected, since starch is hydrophilic and LDPE is hydrophobic. The same was observed for LDPE-g-MA/TPS blends, which was associated with weak interactions between LDPE-g-MA and TPS[17]. The TPS also exhibited starch granules without disruption after the plastization process[17]. According to SEM micrographs, TPS was not equally dispersed in LDPE, since the 40/60 blend presented a smoother surface than other blends. Therefore, this particular blend exhibited the best dispersion of starch. This phenomenon has been ascribed to the weak interfacial adhesion between TPS and LDPE[21]. The mechanical properties of the blends were also investigated in correlation to the TPS content. Young’s modulus and tensile strength were reduced as the TPS content decreased, but the elongation at break increased (Table 3). As TPS is Polímeros, 27(1), 1-7, 2017

Figure 5. Viscosity vs shear rate of the LDPE/TPS blends. Table 3. Tensile properties of the LDPE/TPS blends. LDPE/TPS blends 20/80 30/70 40/60 50/50

Young’s modulus (MPa) 499.0 ± 11.8 469.6 ± 12.3 407.4 ± 55.7 321.4 ± 10.0

Tensile strenght (MPa) 99.18 ± 2.35 77.91 ± 2.12 70.46 ± 3.23 67.36 ± 4.35

Elongation at break (%) 3.12 ± 0.23 3.58 ± 0.14 3.96 ± 0.10 4.98 ± 0.10

a rigid and fragile material, by adding LDPE, it confers ductility reducing the Young’s modulus and tensile strength, but increasing the elongation at break (Table 3). All the blends showed a significantly higher Young’s modulus than neat LDPE (92 MPa)[11]. The same behavior was reported on for LDPE/starch plasticized with glycerol[1]. These are unusual results considering the high levels of immiscibility between LDPE and TPS. Some authors have found that compression during crystallization, exerted by a crystalline matrix, on an amorphous dispersed phase can result in good interfacial contact and a higher Young’s modulus[1,22]. In this study, Young’s modulus were higher than those obtained for the LDPE/starch blends plasticized with glycerol, whose values were between 44.0 and 66.2 MPa[1]. In another study, Young’s modulus values of LDPE/starch blends were between 240 and 290 MPa and increased with the starch content (30, 40 and 50 wt%)[14]. This behavior was ascribed to TPS acting as filler for LDPE. Another hypothetical reason is that HBP migrated from TPS during the processing of the blends and this possibly induced a rearrangement of starch granules, leaving the starch granules without plasticization. In the LDPE-g-MA/TPS blends, high values of Young’s modulus (between 224.1 and 468.1 MPa) were also observed; they enhanced with increasing the TPS content of the blends[17]. The materials obtained in this study exhibited higher Young’s modulus and tensile strength, but lower elongation at break than those obtained for the LDPE-g-MA/TPS blends with the same proportions of TPS[17]. It is important to note that the LDPE’s value of Young’s modulus is similar to LDPE-g-MA (93.58 MPa). Therefore, it was expected that 5

Guzmán, M., Giraldo, D., & Murillo, E.

Figure 6. SEM micrographs of the LDPE/TPS blends (a) 20/80; (b) 30/70; (c) 40/60; and (d) 50/50.

the Young’s modulus and tensile strength of these materials, obtained with the same proportion of TPS, would be almost equal. However, this was not the case, probably due to the fact that during the processing of these materials a structural rearrangement of TPS occurred, related to rupture and formation of the interaction between starch and HBP or interaction between macromolecules of HBP. The lack of adhesion between LDPE and starch, seen on the SEM micrographs of fractured surfaces, indicates a poor interfacial interaction and could explain the decrease in the mechanical properties following the addition of LDPE.

4. Conclusions In this study, LDPE/TPS blends were prepared. Using HBP as a plasticizer for tapioca starch makes an important contribution to the study of LDPE/TPS blends. According to the results, HBP induced a structural rearrangement of the LDPE/TPS blends. Furthermore, the presence of HBP in TPS reduced the viscosity of the LDPE/TPS blends, which was lower than that of the neat LDPE. The Tm and Tc values were practically the same, without significant changes, for all the LDPE/TPS blends. The LDPE/TPS blends crystalized easier than LDPE. The A type crystallinity in the blends was reduced. The thermal stability of the mixtures showed no significant changes and it did not exhibit dependence on the composition of the LDPE/TPS blends. All blends exhibited a shear thinning behavior. SEM analysis showed the presence of the granular structure of starch for all the 6

LDPE/TPS blends. All the blends exhibited a significantly higher Young’s modulus than neat LDPE. The thermal and rheological properties of the LDPE/TPS blends were comparable with those of LDPE-g-MA/TPS blends prepared with the same proportions of TPS[17].

5. References 1. Rodriguez-Gonzalez, F. J., Ramsay, B. A., & Favis, B. D. (2003). High performance LDPE/thermoplastic starch blends: a sustainable alternative to pure polyethylene. Polymer, 44(5), 1517-1526. http://dx.doi.org/10.1016/S0032-3861(02)00907-2. 2. Da Róz, A., Carvalho, A., Gandini, I. A., & Curvelo, A. (2006). The effect of plasticizers on thermoplastic starch compositions obtained by melt processing. Carbohydrate Polymers, 63(3), 417-424. http://dx.doi.org/10.1016/j.carbpol.2005.09.017. 3. Chillo, S., Flores, S., Mastromatteo, M., Conte, A., Gerschenson, L., & Del Nobile, M. A. (2008). Influence of glycerol and chitosan on tapioca starch-based edible film properties. Journal of Food Engineering, 88(2), 159-168. http://dx.doi. org/10.1016/j.jfoodeng.2008.02.002. 4. Ma, X. F., Yu, J. G., & Wan, J. J. (2006). Urea and ethanolamine as a mixed plasticizer for thermoplastic starch. Carbohydrate Polymers, 64(2), 267-273. http://dx.doi.org/10.1016/j. carbpol.2005.11.042. 5. Yang, J., Yu, J., & Ma, X. (2006). Study on the properties of ethylenebisformamide and sorbitol plasticized corn starch (ESPTPS). Carbohydrate Polymers, 66(1), 110-116. http:// dx.doi.org/10.1016/j.carbpol.2006.02.029. 6. Murillo, E. A., Cardona, A., & López, B. L. (2011). Rheological behavior in the molten state and solution of hyperbranched Polímeros, 27(1), 1-7, 2017

Hyperbranched polyester polyol plasticized tapioca starch/low density polyethylene blends polyester of fourth and fifth generation. Journal of Applied Polymer Science, 119(2), 929-935. http://dx.doi.org/10.1002/ app.32774. 7. Murillo, E. A., Vallejo, P. P., & López, B. L. (2010). Characterization of hydroxylated hyperbranched polyesters of fourth and fifth generation. E-Polymers, 120(1), 1-12. http:// dx.doi.org/10.1515/epoly.2010.10.1.1347. 8. Murillo, E. A., Vallejo, P. P., Sierra, L., & López, B. L. (2009). Characterization of hyperbranched polyol polyesters based on 2,2-bis (methylol propionic acid) and pentaerythritol. Journal of Applied Polymer Science, 112(1), 200-207. http://dx.doi. org/10.1002/app.29397. 9. Murillo, E. A., Vallejo, P. P., & López, B. L. (2010). Synthesis and characterization of hyperbranched alkyd resins based on tall oil fatty acids. Progress in Organic Coatings, 69(3), 235240. http://dx.doi.org/10.1016/j.porgcoat.2010.04.018. 10. Sailaja, R. R. N., & Seetharamu, S. (2008). Itaconic acid grafted - LDPE as compatibilizer for LDPE - plasticized Tapioca starch blends. Reactive & Functional Polymers, 68(4), 831-841. http://dx.doi.org/10.1016/j.reactfunctpolym.2007.12.003. 11. Guzmán, M., & Murillo, E. A. (2014). Funcionalización de polietileno de baja densidad con anhídrido maleico en estado fundido. Polímeros: Ciência e Tecnologia, 24(2), 162-169. http://dx.doi.org/10.4322/polimeros.2014.034. 12. St-Pierre, N., Favis, B. D., Ramsay, B. A., Ramsay, J. A., & Verhoogt, H. (1997). Processing and characterization of thermoplastic starch/polyethylene blends. Polymer, 38(3), 647-655. http://dx.doi.org/10.1016/S0032-3861(97)81176-7. 13. Halley, P. J., Truss, R. W., Markotsis, M. G., Chaleat, C., Russo, M., Sargent, A. L., Tan, I., & Sopade, P. A. (2007). Review of biodegradable thermoplastic starch polymers. ACS Symposium Series, 978, 287-300. http://dx.doi.org/10.1021/bk-2007-0978. ch024. 14. Pedroso, A., & Rosa, D. (2005). Mechanical, thermal and morphological characterization of recycled LDPE/corn starch blends. Carbohydrate Polymers, 59(1), 1-9. http://dx.doi. org/10.1016/j.carbpol.2004.08.018. 15. Ning, W., Jiugao, Y., Xiaofei, M., & Ying, W. (2007). The influence of citric acid on the properties of thermoplastic starch/

Polímeros, 27(1), 1-7, 2017

linear low-density polyethylene blends. Carbohydrate Polymers, 67(3), 446-453. http://dx.doi.org/10.1016/j.carbpol.2006.06.014. 16. Garg, S., & Jana, A. K. (2007). Studies on the properties and characteristics of starch-LDPE blend films using cross-linked, glycerol modified, cross-linked and glycerol modified starch. European Polymer Journal, 43(9), 3976-3987. http://dx.doi. org/10.1016/j.eurpolymj.2007.06.030. 17. Guzmán, M., & Murillo, E. A. (2015). The properties of blends of maleic - anhydride - grafted polyethylene and thermoplastic starch using hyperbranched polyester polyol as a plasticizer. Polymer Engineering and Science, 55(11), 2526-2533. http:// dx.doi.org/10.1002/pen.24143. 18. Zaky, M. T., & Mohamed, N. H. (2010). Influence of lowdensity polyethylene on the thermal characteristics and crystallinity of high melting point macro- and micro-crystalline waxes. Thermochimica Acta, 499(1-2), 79-84. http://dx.doi. org/10.1016/j.tca.2009.11.005. 19. Pushpadass, H. A., Bhandari, P., & Hanna, M. A. (2010). Effects of LDPE and glycerol contents and compounding on the microstructure and properties of starch composite films. Carbohydrate Polymers, 82(4), 1082-1089. http://dx.doi. org/10.1016/j.carbpol.2010.06.032. 20. Mandani, M. (2010). Structure, optical and thermal decomposition characters of LDPE graft copolymers synthesized by gamma irradiation. Materials Science, 33, 65-73. http://dx.doi. org/10.1016/j.cap.2010.06.021. 21. Ning, W., Jiugao, Y., Xiaofei, M., & Chunmei, H. (2007). High performance modified thermoplastic starch/linear lowdensity polyethylene blends in one-step extrusion. Polymer Composites, 28(1), 89-97. http://dx.doi.org/10.1002/pc.20266. 22. Leclair, A., & Favis, B. D. (1996). The role of interfacial contact in immiscible binary polymer blends and its influence on mechanical properties. Polymer, 37(21), 4723-4728. http:// dx.doi.org/10.1016/S0032-3861(96)00319-9. Received: Dec. 10, 2015 Revised: Apr. 12, 2016 Accepted: May 20, 2016

http://dx.doi.org/10.1590/0104-1428.2278

O O O O O O O O O O O O O O O O

Mechanical recycling of tags and labels residues using sugarcane bagasse ash Guilherme Augusto Mantovani1, Jean Halison de Oliveira2, Andressa dos Santos2, Andrelson Wellington Rinaldi2, Murilo Pereira Moisés3, Eduardo Radovanovic2 and Silvia Luciana Fávaro1* Laboratório de Química de Materiais e Sensores, Departamento de Engenharia Mecânica, Universidade Estadual de Maringá – UEM, Maringá, PR, Brazil 2 Laboratório de Química de Materiais e Sensores, Departamento de Química, Universidade Estadual de Maringá – UEM, Maringá, PR, Brazil 3 Universidade Tecnológica Federal do Paraná – UTFPR, Apucarana, PR, Brazil

*slfavaro@hotmail.com

Abstract In this study, an alternative method for recycling residues of labels and stickers (parings) containing biaxically oriented polypropylene (BOPP) and polyurethane-based glue was discussed. The recycling of this type of material is complicated, once the separation and the milling processes are difficult to be accomplished, due to the presence of a large amount of glue. In this study, sugarcane bagasse ash was used to enable the milling process of stickers residues. Composites were prepared with post-consumer polypropylene extrusion with different polypropylene/parings ash ratio. These materials were analyzed by tensile, three point flexural, hardness, density, water absorption, Izod impact tests, thermogravimetric analysis, environmental exposure and scanning electron microscopy. Addition of sticker residues/ash to the polypropylene matrix makes the material more rigid and does not affect significantly thermal and degradation properties. Thus, the recycling process proposed in this paper is environmentally and economically viable. Keywords: composites, parings, BOPP, recycling, mechanical properties.

1. Introduction Labels and stickers play the role of providing information on products, in order to attract the consumer, give information compulsory for the product distribution and show the producer’s brand. Many labels are printed on stickers, made of couche paper or biaxially oriented polypropylene (BOPP) film, once those materials allow a high-quality printing. When those labels and stickers are produced, many parts of them are cut out; the remaining material is called “parings” by the printing industry. Those adhesive parings are not recycled or reused; they are kept in the factories or are inconsequently thrown away on dumps. The companies that are interested in providing an adequate destination for the parings pay to dispose this residue in landfills. In Brazil, the cost for that disposal is around R$ 70,00 per ton of non-hazardous residue; in European countries, this cost can reach R$ 340,00 (2010)[1]. The amount of labels and stickers discarded is huge, for they are largely used. A small factory, for example, generates approximately 25 tons of residue per month. The mechanical recycling of that material is complex, once the separation and the milling processes are expensive and difficult to be done, due to the big amount of glue that the labels and stickers contain. Sugarcane bagasse ash is another solid industrial residue inappropriately managed. Brazil is one of the biggest producers of sugar and alcohol from sugarcane; it is one of the oldest and most important economic activity in the country, being Paraná State one of the biggest national producers. The main by-products from the sugar and ethanol industry

are: rinsing water, bagasse, leaves, sugarcane tips and ash, vinasse, filter cake and yeast. The bagasse is one of the most attractive by-products, due to its calorific power, making it the main energy source for producing sugar and alcohol. It is also used as energy source for producing electrical power to the national system, by means of cogeneration. After the sugarcane bagasse is burnt in boilers, residual ashes remain. The chemical composition of those ashes reveal the predominant presence of silica (SiO2). The silicon present in the ashes is absorbed from the soil by the sugarcane roots, as monosilicic acid (H4SiO4); after the transpiration, it gets stuck on the outer wall of the epidermal cells, as silica gel[2]. From the sugarcane mass that feeds the milling process are produced approximately 26% of bagasse (containing 50% of moisture) and from this material 0.62% of residual ash is generated[3]. In the harvest 2013/2014, 4 million tons of sugarcane bagasse ash were generated in Brazil. The ashes produced in the process of fuel burning in boilers are among the residues most generated in the country. According to ABNT NBR 10004:2004[4], because of the chemical and physical characteristics, the ashes are classified as a solid residue of rural category, belonging to nature of Class II (not inert). Nowadays, in Brazil, there is not legislation to deal specifically with sugarcane bagasse ash; it is frequently inappropriately discarded or takes space in landfills, what can generate environmental problems. The concern about sustainable development lead to the necessity of the realization of researches on the use of sugarcane bagasse

Polímeros, 27(1), 8-15, 2017

Mechanical recycling of tags and labels residues using sugarcane bagasse ash ashes; some researchers have already reported its use in Portland cements[5] and at the synthesis of porous materials[1], such as microporous zeolites. Commercially, a small amount of ashes is reused at the civil construction industry, as well as fertilizer. The most important effort should be in a way that minimizes the environmental impacts caused by the disposal of that residue in the environment, enlarging the possibilities for its use, for example, at the production of composite materials[6]. Therefore, this work presents an alternative for the reutilization of residues from stickers and labels, by means of mechanical recycling, using sugarcane bagasse ashes for milling the residues, inhibiting the glue action and evaluating the chemical and thermal properties of the composites formed by these residues and post-consumer polypropylene.

2. Materials and Methods 2.1 Materials The sugarcane bagasse ashes (CBCA) were obtained from a sugarcane and alcohol mill located in the region of Maringá (Paraná State, Brazil). The residues from labels and tags used, named parings in this study, were provided by the company INOVAFLEX – Labels and tags, located in Maringá (Paraná State, Brazil). The polypropylene used at the preparation of composites was acquired from a supplier of recycled material located in the region of Maringá.

2.2 Samples preparation For using CBCA in the milling process, they were previously calcinated. They stayed in a muffle at 600 °C for 4 hours, so all the organic matter was removed. The ashes resulting from this process are called in this study CBCA600[1]. The parings (resulting residue from the production of labels) were milled in a knife mill, together with CBCA600, at proportion parings/ashes equal to 1:3 (mass/mass). Afterwards, the milled material was sieved to remove the ashes excess, in a sieve with opening size n° 40 (ASTM). The excess ashes were removed by hand, almost 50% mass of milled mixture was retired by sieving. The milled and sieved residue was mixed with post-consumer polypropylene at different mass proportions: 10% residue with 90% polypropylene (PP10); 30% residue with 70% polypropylene (PP10); 50% residue with 50% polypropylene (PP50). Samples of pure post-consumer polypropylene (PP) were also prepared, as well as of the milled parings with ashes without adding polypropylene (PPR). The samples were processed in a twin-screw extruder Thermo Sientific MiniLab II HAAKE Rheomex CTW 5, using the temperature of 190 °C for the mold and 60 rpm of rotation. The test specimens for the mechanical trials were injected by means of an injection machine Thermo Sicentific HAAKE MiniJet II, with cannon temperature at 210 °C, mold temperature at 40 °C, injection pressure of 650 bar, injection time of 15 s, holding pressure of 300 bar and holding pressure time of 30 s. Polímeros, 27(1), 8-15, 2017

2.3 Characterization Calorimetric analysis (DSC) from the parings and from the post-consumer polypropylene were obtained. The tests were carried out in SHIMADZU DSC50 equipment, under nitrogen atmosphere (20 mL/min), at a heating rate of 10 °C/min. The quantity of ashes in each sample was determined by gravimetry. The composites samples were calcinated for the measurement of the quantity of sugarcane bagasse ashes introduced in each of the samples. The morphology of the composites was analyzed by means of scanning electron microscopy (SEM) in SHIMADZU SS-550 equipment. Images of the test specimens fractured region after the impact tests and of the exposed surface after the environmental exposure were obtained. The thermogravimetric analyses (TGA) of the composites were carried out in SHIMADZU TGA-50 equipment, operating at a heating rate of 10 °C/min from the room temperature through 700 °C, with 20 mL/min nitrogen flow. The density of the composites was verified at a pycnometer MICROMERITICS 1305, under helium pressure. The tensile tests were carried out at an universal testing machine EMIC DL 10000 with load cell of 5000 N, according to the ASTM D638 standard. A number of eight tests were carried out for each sample. The Izod impact tests were carried out in CEAST Resil Impactor Junior equipment, according to the ASTM D256 standard, which establishes the method for the determination of the resistance to the Izod impact for plastic materials. A number of five tests was carried out for each sample. The environment exposure tests were carried out according to the ASTM D1435 standard, which deals with the procedures for the exposure of plastic materials to the environment for evaluating degradation. Two degradation periods were analyzed, 60 and 120 days. At the end of these periods, the impact of the exposure on the composite materials properties was analyzed, with scanning electron microscopy techniques and mechanical strength trial: yield strength, ultimate tensile strength, elastic modulus, stretching of fracture and Izod impact strength.

3. Results and Discussions 3.1 Characterization of the raw material The sugarcane ashes have previously been characterized[1], showing that the CBCA is basically constituted of silica grains, with sizes varying between 10 and 400 μm, containing 14% of organic matter, referent to the remaining carbonized sugarcane fibers, and 2% of moisture. The thermal treatment to which CBCA600 ashes were subjected eliminates all the remaining organic matter. The x-ray difractograms obtained for the parings and for the post-consumer polypropylene used in this study are presented in Figure 1. The post-consumer polypropylene presented signals at the angles 2ϴ of 14°, 16.8°, 18.5° and 21.5°, typical of the isotactic polypropylene. The difractogram of the parings presented signals at the angles 14°, 17° and 18.8°, such as polypropylene. However, two other peaks 9

Mantovani, G. A., Oliveira, J. H., Santos, A., Rinaldi, A. W., Moisés, M. P., Radovanovic, E., & Fávaro, S. L. at 12.3° and 25.7° could also be observed, indicating the presence of syndiotactic polypropylene in the parings. The absence of a signal in the difractogram for the parings at 21°, when compared to polypropylene, may occur due to the material processing temperature and to the biaxial orientation process[7-9]. The polypropylene tacticity is an important parameter to be analyzed, for it influences directly on the physical and mechanical properties of material. The DSC results obtained for parings and post-consumer polypropylene used at the composite production are shown in Figure 2. The parings showed an endothermic signal close to 165 °C, characteristic of the polypropylene fusion. The post-consumer polypropylene showed two endothermic peaks, one of them close to 165 °C, characteristic of the polypropylene, and the other, close to 130 °C, characteristic of the fusion of high-density polyethylene. In general post-consumer polypropylene contain a certain amount of polyethylene as a contaminant. It occurs due to the process of density separation used at the recycling places, because polypropylene and polyethylene have very similar densities. The amount of polypropylene present in the mixture was determined, considering the fact that the pure polypropylene specific heat of fusion is 60 J/g[10] and considering the polypropylene area of fusion peak (56.3 J/g); then the value of 94% of polypropylene could be found.

Figure 1. X-rays difractogram (DRX) of the parings and of the polypropylene.

3.2 Characterization of the composites The mechanical recycling consists of separation, milling, washing, drying, agglutination and extrusion phases. The washing and drying phases may be skipped, depending on the residue condition; for industrial residues, for example, they are not carried out. However, the milling process is primordial to the process. As the labels and tags residues contain glue in their composition, the milling process becomes very arduous. Prior to this study, different milling techniques were tested, and the process using sugarcane bagasse ashes was the most effective. The ashes act inhibiting the glue, preventing it from sticking to the mill knifes. The results of the gravimetric analysis of the composites and of the post consumed PP are presented in Table 1. The non-volatile fraction of the samples is corresponding to the CBCA600 used for milling the parings, which is not eliminated in the sieving process, i.e., the ashes that stick to the parings surface. This way, it is possible to state, from the results of the gravimetry tests, that the CBCA600 mass is approximately 50% of the milled and sieved residue. The thermogravimetric analysis (TGA) of the samples is presented in Figure 3; the residual mass indicates the quantity of CBCA600. An increase on the thermal stability of the composites in relation of the pure matrix can be observed, by means of the temperature in which maximum mass loss rate occurs (dm/dT). The value presented by pure PP is 415 °C, while, for the composites, this temperature is around 450 °C to all the compositions. Another important aspect to be observed in Figure 3 is the quantity of remaining residue present in each sample, due to an increase of inorganic material (ashes) to the sample. With the addition of the lower percentage of parings and ashes, a low percentage of remaining CBCA600 for the composite PP10 could be observed; this quantity gradually increases when composites 10

Figure 2. DSC thermograms of the parings and of the polypropylene.

Table 1. Average composition of the samples mass. Sample PP PP10 PP30 PP50 PPR

Polypropylene (%) 100 90 70 50 --

Parings (%) -6 16 24 47

Ashes (%) -4 14 26 53

Figure 3. TGA curves of the composites and of the PP. Polímeros, 27(1), 8-15, 2017

Mechanical recycling of tags and labels residues using sugarcane bagasse ash

Figure 4. Micrograph of the composites fracture surface.

PP30 and PP50 are considered. In the sample prepared only with the mixture of parings and ashes (PPR), the quantity of CBCA600 reaches 60% (m/m). Those results are equivalent to the ones observed at the gravimetric analysis and prove a homogeneous reinforcement distribution in the matrix, once for the gravimetric analysis, some grams of material are used, and for the thermogravimetry, some milligrams are used. In Figure 4, micrographs of the PP10, PP30 and PP50 composites are presented. The silica distribution in the composites is homogeneous; there were not any problems for dispersing the ashes in the material during the process, and they do not tend to agglutinate. It is also possible to observe that, during the milling and extrusion processes, silica was broken into smaller grains. In the fracture area it is possible to observe that the places where the silica unglued from the matrix during the impact test are flat surfaces, indicating that the silica does not stick very well to the polymeric matrix. It is known that low interfacial resistances are generally identified on fracture surfaces due to the presence of empty spaces, associated to the particles yank, originated from the fissures spreading on the interfacial area. The fractographic analysis, in its turn, tends to associate high interfacial resistances to the presence of polymeric films on the particles[11]. The graph presented in Figure 5 presents the samples final density. The increase of the ashes amount in the material causes a proportional density increase in the composites, if compared to pure PP. The ashes density is close to 2.3 g/ cm3, while the polypropylene density is around 0.9 g/cm3. The highest density presented by the composite materials, enables its use for specific applications, in which materials with densities higher to the ones of the polypropylene and polyethylene are demanded. In Figures 6A to 6D are presented the results obtained from the tensile strength test for the composite materials and for the pure PP. In Figure 6A, the values obtained from the yield strength in the trials can be observed. It can be noticed that there is a decrease on the yield strength with the increase of the silica amount; this decrease is remarkable in the sample PPR. In Figure 6B are shown the values for the ultimate tensile strength. A decrease in these values occurs Polímeros, 27(1), 8-15, 2017

Figure 5. Density of the composites and of the PP.

with ashes addition. Nevertheless, for the composites with up to 50% of residue (around 25% of ashes), the tensile strength decrease is very low. The introduction of hard inorganic particulates in polymeric matrices is usually followed by a reduction in the polymer mechanical strength, leading to the production of composites mechanically poorer than the pure polymer. This fact is frequently observed, for example, when particulates, such as slate, calcium carbonate and talc are incorporated in polymers like polypropylene[12-15]. The fact that the ash addition has not influenced the polymer resistance so drastically highlights its capacity of not deteriorating the properties in relation to the pure polymer. That deterioration of properties would be linked to the action of the particulate matter as a generator of defects on the composite material[15]. Only for the sample PPR that decrease is highly remarkable, with a decrease of around 80% in its value. Once the silica adhesion to the polypropylene matrix is low, as the morphological analysis has shown, this result was already expected. As the matrix does not transfer adequately the effort to the reinforcement phase, the area of the specimen test section that supports the effort is smaller; consequently, the tension that the material can tolerate is also lower. Figure 6C presents the elastic module values of the composites. These values increase with the addition of 11

Mantovani, G. A., Oliveira, J. H., Santos, A., Rinaldi, A. W., Moisés, M. P., Radovanovic, E., & Fávaro, S. L.

Figure 6. Mechanical behavior of the composites and PP samples where (A) represents the yield strength; (B) ultimate tensile strength; (C) elastic modulus; and (D) rupture strength.

ashes to the composites. That increase is more effective for the sample PPR. As a consequence of this result, there is a decrease on the rupture stretching, as presented on the results of Figure 6D. The stretching values decrease with the increase of the silica quantity; a more drastic decrease was observed for the composites prepared without postconsumer PP addition, only with parings and ash. From the results of the impact strength tests, carried out with the Izod method, presented in Figure 7, it could be observed that the addition of ashes subtly decreased the material impact strength. As observed on the tensile strength tests, the composites PP10, PP30, PP50 and PPR are harder than the pure matrix. Harder materials absorb less impact energy during the failure. The impact strength decreases with the increase of ashes concentration. The addition of hard loads to the ductile matrix tends to weaken the material. The load acts as tensile concentration, restricting the mobility of the matrix, avoiding plastic deformation. Another aspect to be considered is the lack of adhesion between the matrix and the reinforcement. This result is in accordance with other studies that use inorganic loads as reinforcement in polypropylene matrix[16-18]. That small difference in the impact strength makes it possible to be used instead of pure polypropylene in applications in which the maximization of this characteristic is not preponderant.

3.3 Environment exposure test The term polymers degradation is associated to any destructive reaction that could be caused by chemical, physical, mechanical or biological agents, causing an 12

Figure 7. Impact strength of the samples.

irreversible change in the properties of polymeric materials. In many cases, the degradation reactions may be desirable, for example, for non-recyclable polymeric waste. In the composites, the degradation may occur in the matrix, in the reinforcement or in the system as a whole. In the sun light, this degradation considerably increases, because the oxidation rate of polymers is accelerated, and this effect is increased by the presence of atmospheric pollutants. The temperature and the moisture can also be mentioned as factors that contribute to the degradation process of polymers[19]. In order to evaluate the degradation process of the composites, the material was subjected to trials of natural aging, being exposed to environmental conditions very Polímeros, 27(1), 8-15, 2017

Mechanical recycling of tags and labels residues using sugarcane bagasse ash similar to its real conditions of utilization. Mechanical and scanning electron microscopy analyses were used to verify the occurrence of degradation of the material, as well as the reinforcement influence on this process. The surfaces of the samples exposed to degradation were evaluated by scanning electron microscopy, Figure 8. In all the samples, it is possible to notice the appearance of fissures with the degradation and the increase of cracking with longer exposure time, except for the sample PPR, which is already very porous. The cracking results from a phenomenon noticed at the photodegradation of the polypropylene, known as chemi-crystallization. The molecules of the amorph regions in the polypropylene are more susceptible to photodegradation, which causes molecules to split. The resulting smaller segments are more mobile and can rearrange in crystalline structures, generally on already existing crystals. This process increases the polypropylene

crystallity and causes the superficial spontaneous formation of fissures[20-22]. In Figures 9A to 9D are presented the results of the tensile tests after the exposure of the material to degradation, compared to the tests carried out in the material as soon as it was processed. It is possible to notice a tendency of decreasing resistance and maximum stretching; this decrease is mainly due to the photodegradation of the composites polymeric matrix, because the reinforcement phase silica is not subject to property loss in the short time evaluated. Rabello and White[20] also observed the polypropylene mechanical strength decrease due to the effect of the exposure to the ultraviolet radiation: chains splitting, fissuring and increase of crystallity. The elastic modulus presented an increase at almost all the samples, but did not present a direct relation between its value and the composites composition.

Figure 8. Micrograph of the composites surface after 120 days environmental exposure.

Figure 9. Mechanical behavior of the samples with the degradation where (A) represents the yield strength; (B) ultimate tensile strength; (C) elastic modulus; and (D) rupture strength. Polímeros, 27(1), 8-15, 2017

Mantovani, G. A., Oliveira, J. H., Santos, A., Rinaldi, A. W., Moisés, M. P., Radovanovic, E., & Fávaro, S. L. The composites can be applied to the production of different materials, including domestic-use tools, such as buckets, bowls, clothespins, and also products for civil construction, for manufacturing tiles, floorings, hoses and pipes. So those materials can be developed, more research must be carried out to evaluate the materials degradation in longer periods of environmental exposure.

5. References Figure 10. Behavior of the impact strength with the degradation.

The changes on the composites properties were not very different from the changes occurred at the samples of pure polypropylene. This way, the incorporation of parings/ashes to polypropylene allowed the production of materials with mechanical and composites thermal behavior properties that are not very different form the ones of the pure polypropylene, besides producing a low-cost material, which also contributes to the environment preservation as it favors the reuse of residues. Figure 10 presents the results of the impact strength after the exposure the material to environmental conditions compared to the results of the tests carried out before that. All the samples presented a small reduction on the impact strength when exposed to the weather conditions. Nevertheless, it is not possible to state that the environmental exposure time is directly related to the decrease of the impact strength, because for the samples PP, PP30 and PP50 the lowest impact strength was obtained after 60 days of exposure, while for the samples PP10 and PPR the lowest values occurred after 120 days of exposure to the environment.

4. Conclusions The use of sugarcane bagasse ash to inhibit the glue action and enable the milling of parings is promising. The mechanical characterization of the composites indicated that the addition of the parings residues milled with ashes to the polypropylene produces a harder material than the pure polypropylene. The composites also presented high thermal stability. The mechanical properties of the composites and of the pure PP behave in a very similar way, and the same fissuring phenomenon is noticed in all the samples. One of the biggest advantages of the composites preparation is the flexibility in relation to the composition polypropylene/parings/ashes, offering the possibility to produce materials with different final mechanical properties, related to their composition. The recycling method proposed stands a high chance to be successful, because of the low cost of the equipment used and because of the innovative and environment-friendly character. It potentially represents job creation at associations and cooperatives built to supply the recyclable material industry. 14

1. Moisés, M. P., Silva, C. T. P., Meneguin, J. G., Girotto, E. M., & Radovanovic, E. (2013). Synthesis of zeolite NaA from surgarcane bagasse ash. Materials Letters, 108(1), 243-246. http://dx.doi.org/10.1016/j.matlet.2013.06.086. 2. Mano, E. B., Pacheco, E. B. A. V., & Bonelli, C. M. C. (2010). Meio ambiente, poluição e reciclagem. São Paulo: Blucher. 3. Nunes, I. H. S., Vanderlei, R. D., Secchi, M., & Abe, M. A. P. (2008). Estudo das caracteristicas fisicas e químicas da cinza do bagaço de cana-de-açucar para uso na construção. Revista Tecnológica, 17, 39-48. http://dx.doi.org/10.4025/revtecnol. v17i1. 4. Associação Brasileira de Normas Técnicas – ABNT. (2004). ABNT NBR 10004: resíduos sólidos: classificação. Rio de Janeiro: ABNT. 5. Paula, M. O., Tinôco, I. F. F., Rodrigues, C. S., Silva, E. N., & Souza, C. F. (2009). Potencial da cinza do bagaço da cana-de-açúcar como material de substituição parcial de cimento Portland. Revista Brasileira de Engenharia Agrícola e Ambiental, 13(3), 353-357. http://dx.doi.org/10.1590/S141543662009000300019. 6. Meier, M. (2014). Sustainable polymers: reduced environmental impact, renewable raw materials and catalysis. Green Chemistry, 16(4), 1672. http://dx.doi.org/10.1039/c4gc90006e. 7. Türkçü, H. N. (2004). Investigation of the crystallinity and orientation of polypropylene with respect to temperature changes using FT-IF, XRD and raman techniques (Master’s dissertation). Bilkent University, Ankara. 8. Tartaglione, G., Tabuani, D., Camino, G., & Moisio, M. (2008). PP and PBT composites filled with sepiolite: morphology and thermal behaviour. Composites Science and Technology, 68(2), 451-460. http://dx.doi.org/10.1016/j.compscitech.2007.06.023. 9. Erp, T. B. V., Balzano, L. M., & Peters, G. W. (2012). Oriented gamma phase in isotatic polypropylene homopolymer. ACS Macro Letters, 1(5), 618-622. http://dx.doi.org/10.1021/ mz3000978. 10. Canevarolo, S. V. (2003). Técnicas de caracterização de polímeros. São Paulo: Artliber. 11. Sinien, L., Lin, Y., Xiaoguang, Z., & Zongneng, Q. (1992). Microdamage and interfacial adhesion in glass bead-filled high-density polyethylene. Journal of Materials Science, 27(17), 4633-4638. http://dx.doi.org/10.1007/BF01165998. 12. Premalal, H. G. B., Ismail, H., & Baharin, A. (2002). Comparison of the mechanical properties of rice husk powder filled polypropylene composites with talc filled polypropylene composites. Polymer Testing, 21(7), 833-839. http://dx.doi. org/10.1016/S0142-9418(02)00018-1. 13. Demjén, Z., Pukánszky, B., & Nagy, J. (1998). Evaluation of interfacial interaction in polypropylene/surface treated CaCO3 composites. Composites. Part A, Applied Science and Manufacturing, 29(3), 323-329. http://dx.doi.org/10.1016/ S1359-835X(97)00032-8. 14. Mareri, P., Bastide, S., Binda, N., & Crespy, A. (1998). Mechanical behaviour of polypropylene composites containing fine mineral Polímeros, 27(1), 8-15, 2017

Mechanical recycling of tags and labels residues using sugarcane bagasse ash filler: Effect of filler surface treatment. Composites Science and Technology, 58(5), 747-752. http://dx.doi.org/10.1016/ S0266-3538(97)00156-5. 15. Carvalho, G. M. X., Mansur, H. S., Vasconcelos, W. L., & Oréfice, R. L. (2007). Obtenção de compósitos de resíduos de ardósia e polipropileno. Polímeros: Ciência e Tecnologia, 17(2), 98-103. http://dx.doi.org/10.1590/S0104-14282007000200008. 16. Asuke, F., Aigbodion, V. S., Abdulwahab, M., Fayomi, O. S. I., Popoola, A. P. I., Nwoyi, C. I., & Garba, B. (2012). Effects of bone particle on the properties and microstructure os polypropylene/ bone ash particulate composites. Results in Physics, 2(1), 135141. http://dx.doi.org/10.1016/j.rinp.2012.09.001. 17. Fuad, M. Y. A., Ismail, Z., Ishak, Z. A. M., & Omar, A. K. M. (1995). Application of rice husk ash as fillers in polypropylene: effect of titanate, zirconate and silane coupling agents. European Polymer Journal, 31(9), 885-893. http://dx.doi. org/10.1016/0014-3057(95)00041-0. 18. Ramos, S. M. L. S., Carvalho, L. H., Spieth, E., & Rivadula, R. S. (1993). Efeitos da estabilização do Polipropileno nas propriedades térmicas, mecânicas e termo-mecânicas de

Polímeros, 27(1), 8-15, 2017

compósitos de Polipropileno/Atapulgita. Polímeros: Ciência e Tecnologia, 3(4), 26-31. 19. Thwe, M. M., & Liao, K. (2003). Durability of bamboo-glass fiber reinforced polymer matrix hybrid composites. Composites Science and Technology, 63(3-4), 375-387. http://dx.doi. org/10.1016/S0266-3538(02)00225-7. 20. Rabello, M. S., & White, J. R. (1997). Fotodegradação do polipropileno: um processo essencialmente heterogêneo. Polímeros: Ciência e Tecnologia, 7(2), 47-56. 21. Allen, N. S., Edge, M., Corrales, T., Childs, A., Liauw, C. M., Catalina, F., Peinado, C., Minihan, A., & Aldcroft, D. (1998). Ageing and stabilisation of filled polymers: an overview. Polymer Degradation & Stability, 61(2), 183-199. http://dx.doi. org/10.1016/S0141-3910(97)00114-6. 22. Paoli, M. A. (2009). Degradação e estabilização de polímeros. São Paulo: Artliber. Received: Nov. 09, 2015 Revised: Apr. 01, 2016 Accepted: Apr. 28, 2016

http://dx.doi.org/10.1590/0104-1428.2386

O O O O O O O O O O O O O O O O

Removal of Remazol brilliant violet textile dye by adsorption using rice hulls Geyse Adriana Corrêa Ribeiro1, Domingos Sérgio Araújo Silva2, Clayane Carvalho dos Santos1, Adriana Pires Vieira3, Cícero Wellington Brito Bezerra1*, Auro Atsushi Tanaka1 and Sirlane Aparecida Abreu Santana1 Inorganic Chemistry Laboratory & Analytics, Department of Chemistry, Centro de Ciências Exatas e Tecnologia – CCET, Universidade Federal do Maranhão – UFMA, Campus do Bacanga, São Luís, MA, Brazil 2 Department of Chemical Technology, Centro de Ciências Exatas e Tecnologia – CCET, Universidade Federal do Maranhão – UFMA, Campus do Bacanga, São Luís, MA, Brazil 3 Institute of Chemistry, Universidade Estadual de Campinas – UNICAMP, Campinas, SP, Brazil 1

*cwb.bezerra@ufma.br

Abstract The release of industrial effluents into the environment causes widespread contamination of aquatic systems. Adsorption is seen as one of the most promising treatment processes, and lignocellulosic materials have gained prominence as adsorbents. This study investigates the potential of rice hulls, either in natura or treated with nitric acid, as adsorbents for removal of the dye. The adsorbents were characterized by infrared spectroscopy, solid state 13C-NMR, thermogravimetric analysis, and pH at point of zero charge. The dye adsorption experiments were carried out in batch mode, using different experimental conditions. The kinetic adsorption data could be fitted using the model of Elovich. The Freundlich model provided the best fit to the isothermal data. The thermodynamic parameters confirmed the spontaneity of the adsorption process. These adsorbents offer an alternative for dye removal, with advantages including biomass availability and low cost. Keywords: adsorption, remazol, rice hull.

1. Introduction The discharge of industrial effluents into the environment without adequate treatment is frequently a major problem, due to the problems of contamination of water resources. Many industries, such as those concerned with leather processing, rubber, paper, plastics, fabrics, and paints produce large volumes of wastewater containing various chemicals harmful to aquatic ecosystems[1-3]. Among these, in the absence of adequate wastewater treatment, textile manufacturers produce (during the dyeing step) effluent with persistent coloration, which can contaminate natural waters, interfering in the passage of solar radiation and impairing processes of photosynthesis[4,5]. Many processes are used in the treatment of wastewater containing dyes, such as biological oxidation and chemical adsorption[5-7], photoelectrochemical processes[8], treatment with zero-valent iron[9] and zero-valent iron/H2O2[10], heterogeneous photocatalysis combined with ozone treatment[11], hydrogen peroxide combined with UV[12], and others. However, many of these require appropriate conditions and have high operational costs. Adsorption has been shown to be very promising for the removal of textile dyes, since adsorbents can offer industrial practicality, low cost, high removal rates, and (in some cases) recovery of the species involved without losing their chemical identity[13]. Adsorption is the removal of a chemical species from the fluid phase, with subsequent concentration on the surface of a substrate, usually a solid[14]. Adsorbents are

generally highly porous and have a large surface area[15]. One widely used material is activated carbon, which offers high efficiency due to its structural characteristics, high surface area, and chemical nature[3]. However, its use has some disadvantages, such as low selectivity and efficiency for certain disperse dyes[16], as well as losses during the recovery process[17], resulting in a need to identify less expensive and more effective adsorbents. There are a variety of alternative low-cost materials that can be used for the removal of textile dyes, and other contaminants. Agricultural residues are an abundant and inexpensive renewable source of potential adsorbents. Examples include babassu coconut shell[18], rice hull, rice straw[19,20], and banana pseudostem[21], amongst others. These agricultural residues can be used either in natura or modified with the insertion or release of active sites capable of improving the adsorption efficiency. Rice hull is an agricultural waste biomass produced by the rice processing industry. This bulky byproduct represents about 23% of the weight of the grain[22]. Few options for the use of this material have been reported in the literature, and it is still common to see deposits of the waste in the open and on the banks of rivers[23], where it takes about 5 years to decompose, while emitting a high volume of methane, a greenhouse gas that is also harmful to the ozone layer[24]. The aim of this study was therefore to analyze rice hull, in natura and treated with nitric acid, as an adsorbent for the removal of Remazol brilliant violet textile dye present in aqueous solutions.

Polímeros, 27(1), 16-26, 2017

Removal of Remazol brilliant violet textile dye by adsorption using rice hulls

2. Materials and Methods 2.1 Materials and reagents The rice hull used in this work was collected in São Bento city (Maranhão-BR). All reagents used were of analytical grade. The Remazol brilliant violet textile dye was produced by DyStar and provided by Toalhas São Carlos (São Paulo-BR). The purity of the dye was not specified. The chemical structure of Remazol brilliant violet dye is shown in Figure 1[18].

2.2 Preparation of the adsorbent The rice hull was ground and washed with distilled water in a ratio of 10:1 (v/w) for 30 min, with constant stirring, until the conductivity of the supernatant remained constant. The material was then dried at 333 K for 24 h and sieved to obtain the particle size range between 0.088 and 0.177 mm. This in natura rice hull material was denoted RHI. Then, it was processed with acid, as described previously[25], with small changes in contact time and temperature. RHI was placed in a solution of 2 mol L-1 nitric acid, at a ratio of 1:200 (w/v), at 323 ± 1 K, for 5 h under constant stirring. It was then washed with distilled water until the supernatant conductivity remained constant, dried at 333 K, and sieved to obtain the 0.088-0.177 mm particle size fraction. The treated rice hull was denoted RHT.

2.3 Characterization of the adsorbent The adsorbents were characterized by Fourier transform infrared spectroscopy (FTIR), 13C nuclear magnetic resonance (13C-NMR), thermogravimetric analysis (TGA), and pH at point of zero charge (pHpzc). The FTIR spectra in the 4000-400 cm-1 region were obtained using an MB series spectrometer Bomem-Hartmann & Braun, and resolution of 4 cm-1.The KBr tablets were prepared using a press, by diluting the sample of interest in KBr at a ratio of 1:100 (w/w). The 13C-NMR spectra were obtained using a Bruker AC300 spectrometer, with 7 mm rotor, cross polarization and magic angle spinning (CP-MAS), and a contact time of 3 min. The repetition time was 3 s and the frequency was 4 kHz. The TGA were performed with a DuPont instrument (model 9900), heating from room temperature to 1273.15 K at 10 K min-1, under an inert argon atmosphere. The pHpzc values of the adsorbents were estimated by placing 100 mg of the adsorbent (RHI or RHT) in contact with 25.0 mL of

KCl solution (0.1 mol L-1), with initial pH (pHi) ranging from 1 to 12, previously adjusted with HCl or NaOH. After 24 h of contact at room temperature and under constant agitation, the adsorbent was removed by filtration and the final pH (pHf) of the supernatant was measured.

2.4 pH study, adsorption kinetics and isotherms The study was carried out varying the initial pH of the Remazol brilliant violet dye solution. 100 mg of adsorbent (RHI or RHT) was placed in contact with 25 ml of dye solution (100 mg L-1) for 24 h at 298 ± 1 K, under continuous stirring. This test was performed at six pH values (initial pH between 1 and 6, adjusted with HCl and NaOH). To obtain the adsorption kinetics, 100 mg of adsorbent was placed in contact with 25 mL of dye solution at two concentrations (100 and 1000 mg L-1), using the optimum pH (the pH at which greatest adsorption was observed in the pH study), with constant stirring at room temperature. The contact time was varied between 5 and 600 min. Adsorption isotherms were obtained by placing 100 mg of adsorbent in contact with 25 mL of dye solution at various concentrations (100 to 1000 mg L-1) and four temperatures (283, 298, 313, and 328 ± 1 K), using the optimum pH and stirring constantly during the equilibribration time. At the end of the contact time, the adsorbent was separated by centrifugation and the new dye concentration in the supernatant was determined with a UV‑Vis spectrophotometer (model 2550, Shimadzu) fitted with a 1 cm optical path quartz cell, using the absorbance at 560 nm. The amount of adsorbed dye, q (mg g-1), was obtained using Equation 1[26], where Ci and Cf (mg L-1) are the initial and final dye concentrations, respectively, m (g) is the mass of adsorbent, and V (L) is the volume of dye solution.

( Ci − C f )V

(1)

3. Results and Discussion 3.1 Infrared spectroscopy Spectroscopy in the infrared region is a widely practiced characterization technique, because it allows qualitative evaluation of the presence of different functional groups.

Figure 1. Chemical structure of Remazol brilliant violet dye. Polímeros, 27(1), 16-26, 2017

Ribeiro, G. A. C., Silva, D. S. A., Santos, C. C., Vieira, A. P., Bezerra, C. W. B., Tanaka, A. A., & Santana, S. A. A. Figure 2 illustrates the vibrational spectra of all the materials studied. The large numbers of peaks found for the in natura and treated rice hulls indicate the heterogeneous nature of the adsorbents. The IR spectrum of the dye (see Figures 1 and 2a) presented many absorption peaks, including those corresponding to phenolic O–H, secondary amide N–H stretching, and aromatic C–H at 3671-3047 cm−1; aliphatic C–Hn at around 2925 cm−1; amide C=O stretching at 1639 cm−1; C=C stretching at 1532 cm−1; N=N stretching at 1436 cm−1; –SO (R–SO2–R) stretching at 1309 cm−1; –SO stretching of R–SO3- at 1207-989 cm−1; C–S stretching at 748 cm−1 (coupled with out-of-plane C–H and C=C bending) and at 626 cm−1[27,28]. The spectra for the adsorbents (Figures 2c and 2e) exhibited the characteristic bands of lignocellulosic materials reported in the literature[18,29-30], and did not differ in terms of the nature of the functional groups present. Broad stretching bands occurred between 3100 and 3600 cm-1, indicating the presence of -OH groups on the surfaces of the matrices. These were mainly due to glycosidic structures, the presence of silanol groups (Si-OH), and water adsorbed on the surfaces of the hulls[31]. A band between 2900-3000 cm-1 corresponded to CH stretching of the methyl groups that are common in the structures of lignocellulosic materials. A band located at 1730 cm-1 was characteristic of stretching of C=O of ketones, aldehydes, and carboxylic acids. A band at 1640 cm-1 could be attributed to adsorbed water, while a band at 1500 cm-1 corresponded to aromatic C=C vibrations. A band at 1370 cm-1 was due to CH plane deformation vibration, and a band at 1240 cm-1 was associated with CO stretching of the pyran ring. A band at 1060 cm-1 was due to the vibration of methoxyl (C-O-C) and β-1,4 bonds. Bands near 790 cm-1 were due to out-of-plane δ (CH) deformation of the aromatics present in lignin[32]. Comparison of the spectra for the in natura and treated adsorbents showed that the intensities of the bands at 3435, 1734, and 1067 cm-1 increased after treatment, as expected since these bands correspond to oxygen-containing groups (-OH, -C=O, and -COC-, respectively) that may have been formed during the treatment with nitric acid[33].

Figure 2. Infrared vibrational spectra of the RHI and RHT samples. 18

3.2 Solid state 13C nuclear magnetic resonance The solid state 13C-NMR spectra for the RHT and RHI matrices indicated the presence of structures of cellulose and its derivatives (Figure 3). A peak at 105 ppm corresponded to the C1 carbon, with a large displacement due to the bonding to two oxygen atoms (CH(OR)2). Signals at 90 and 83 ppm referred to C4, with a smaller displacement than C1 since the bond involves only one oxygen (CH-OR)[34,35]. A peak at 72 ppm was related to the C2, C3, and C5 secondary carbons attached to the hydroxyl (CH-OH). Peaks at 64 and 62 ppm were assigned to the C6 carbon, indicating regions of low crystallinity and amorphous character (see Figure 1A in the Appendix A)[36], confirming this characteristic for both matrices. The presence of methyl groups and carboxyl carbon-carbon double bonds were indicated at 21 and 172 ppm, respectively. A signal corresponding to lignin guaiacyl methoxy groups was found at 55 ppm, and signals corresponding to aromatic carbons could be seen between 120 and 150 ppm[37]. Comparison of the 13C-NMR spectra of the different matrices revealed increases in the peaks relating to C1, C2, C3, C4, and C5 after treatment (RHT sample) (see also Table 1A in the Appendix A). This could be attributed to the effectiveness of the treatment with nitric acid, forming active adsorption sites and increased intensities of peaks related to carbon atoms with oxygen groups. These results corroborated those obtained in the characterization by infrared spectroscopy, which were also indicative of an increase in oxygen-containing groups after treatment.

3.3 Thermogravimetric analysis Thermogravimetric analysis was used to determine the thermal stability of the adsorbents and provide quantitative data on mass loss according to temperature. The thermogravimetric (a) and derivative (b) curves obtained for the RHI and RHT adsorbents, under an inert argon atmosphere, are shown in Figure 4. The adsorbents showed similar behavior in terms of thermal decomposition, remaining stable up to about 473 K, as reported previously for other lignocellulosic materials[30]. Comparison of the percentage mass losses of the RHI and RHT adsorbents indicated that the in natura matrix showed greater weight loss corresponding to cellulose and hemicellulose, while the treated matrix showed greater loss of

Figure 3. The solid state 13C-NMR spectra for the RHT and RHI. Polímeros, 27(1), 16-26, 2017

Removal of Remazol brilliant violet textile dye by adsorption using rice hulls mass corresponding to lignin. The thermogravimetric curve showed three distinct stages of decomposition. The first, at around 373 K, should not be taken into consideration in terms of thermal stability, because it was only due to the release of physically adsorbed water present on the surfaces of the adsorbents[38]. The second stage, between 473 and 623 K, resulted in a greater loss of mass, due to the decomposition of hemicellulose and most of the cellulose. In the third stage (623-1043 K), which showed less mass loss than the previous stage, there was decomposition of lignin, which is structurally more stable than cellulose or hemicellulose[32].

3.4 pHpzc and effect of pH The pH at the point of zero charge (pHpzc) is a parameter that indicates the pH value at which a particular solid does not have excess net charges on its surface. This parameter is important because it enables prediction of the surface charge of the adsorbent as a function of pH, since at pH lower than the pHpzc, the surface charge of the solid is positive, while for pH values higher than the pHpzc, the surface charge is negative. Figure 5a shows a graph of the variation of ΔpH (pHi-pHf), as a function of the initial pH (pHi), for the in natura and treated adsorbents.

Figure 4. Thermogravimetric curves (a) and derived (b) of the RHI and RHT, obtained under an argon atmosphere, using a heating rate of 10 K min-1 in the temperature interval 298.15-1273.15 K.

The pHpzc values obtained for the RHI and RHT adsorbents were 5.3 and 3.8, respectively. Hence, below these pH values, the surfaces of the adsorbents would be positively charged, favoring anion adsorption, while at higher values, the surfaces would be negatively charged, favoring the adsorption of cations. The pHpzc value for RHI was very close to values reported in the literature for in natura rice hull (5.3)[19] and babassu coconut epicarp (6.7)[39,40]. In other work, it was found that in natura wood sawdust had a pHpzc of 5.3, which decreased to 4.0 after treatment with hydrochloric acid[41]. The decrease in the pHpzc value of the adsorbent after treatment is an indication that there were changes in the surface characteristics, with an increased number of protonated sites. This is supported by the increased prevalence of peaks related to oxygenated groups in the infrared and 13C-NMR spectra. The pH of the system is an important element in the adsorption process, because pH change alters the chemical balance of the ionic groups present in the adsorbent and the dye, influencing the electrostatic interaction. Figure 5b shows the removal percentages of Remazol brilliant violet dye by the RHT and RHI adsorbents. It can be seen that higher pH reduced adsorption of the dye, while at low pH there was greater protonation on the surfaces of the adsorbents, which favored anion removal. The highest dye removal percentages were achieved at pH 1.0 and 2.0. At pH 1.0, the maximum amounts of dye adsorbed were 46.2 and 73.2% for RHI and RHT, respectively. Higher removal percentages of Remazol dyes at pH 1.0 and 2.0 were observed previously using adsorbents such as babassu coconut mesocarp[42] and rice hull[19]. Importantly, use of the RHT adsorbent significantly increased the percentage removal, compared to RHI, showing the effectiveness of the acid treatment. The adsorption by RHT improved between pH 3.0 and pH 6.0, suggesting that in addition to electrostatic attraction, other interaction mechanisms were involved in the adsorption process, since a decreased removal capability was expected between these pH values, as found for the RHI adsorbent. Another interesting finding concerned the contribution of protonation to dye

Figure 5. (a) pHpzc of the RHI and RHT, (b) Influence of pH on adsorption of Remazol brilliant violet dye by the RHI and RHT adsorbents (m = 100 mg; Ci = 100 mg L-1, t = 24 h; T = 298 ± 1 K). Polímeros, 27(1), 16-26, 2017

Ribeiro, G. A. C., Silva, D. S. A., Santos, C. C., Vieira, A. P., Bezerra, C. W. B., Tanaka, A. A., & Santana, S. A. A. adsorption, with values of 66.3% and 18.4% obtained for RHI and RHT, respectively. These results provided further evidence of the effectiveness of the treatment, which resulted in dye removal being practically independent of the effect of protonation, due to the enhancement of other interactions that contributed to the adsorption.

= q

In the above equations, qe (mg g-1) is the amount adsorbed at equilibrium, k1 (min-1), k2 (g mg-1 min-1), and α (mg g-1 min-1) are the Lagergren, Ho, and Elovich adsorption rate constants, and β (g mg-1) is a parameter related to the activation energy. The equilibrium times were 180 and 240 min for dye concentrations of 100 and 1000 mg L-1, respectively. At the lower concentration, the equilibrium time was shorter due to the smaller number of interactions between the dye molecules and the active sites of the adsorbent. Most of the adsorption occurred within 60 min, reflecting fast adsorption kinetics. The high adsorption rates in the early stages of the process were due to greater availability of active sites. Similar equilibrium times have been reported previously for the removal of Remazol dyes using lignocellulosic materials. Tunç et al.[46] used cotton stem bark to remove Remazol B black dye and obtained an equilibrium time of 300 min. The values of the fitting parameters for the Lagergren, Elovich, and Ho kinetic models are shown in Table 1. The closeness of R2 (coefficient of determination) to unity indicates the high levels of explanation provided by the models, with the best results obtained using the Elovich model. This model has been successfully used previously for adsorption by materials with heterogeneous surface energy, and allows for a gradual decrease in removal rate with increasing concentration[47]. The adsorbents used in this study presented several functional groups that could act as adsorption sites. Furthermore, Remazol brilliant violet dye also has many groups capable of interacting with adsorbents, so that the adsorption process was energetically heterogeneous, in agreement with the Elovich model.

3.5 Adsorption kinetics Study of the adsorption kinetics is important, because it provides information about the mechanism of adsorption. This work considered the contact time between the adsorbent and the adsorbate required for the system to reach equilibrium. The adsorption kinetics depends on several factors, including the adsorbent mass, concentration of adsorbate, temperature, pressure, and stirring speed of the system. Figure 6 shows the adsorption kinetics for two concentrations of the Remazol brilliant violet dye (100 and 1000 mg L-1) at pH 2 (defined as optimal in the adsorption experiments), together with the fits obtained with the Lagergren (Equation 2)[43], Ho (Equation 3)[44], and Elovich (Equation 4)[45] kinetic models.

(

)

= q qe 1 − e− k1t (2)

1 ( ln ( αβ ) + ln t ) (4) β

( qe )2 k2t

1 + qe k2t (3)

3.6 Adsorption isotherms After establishing the optimal pH and the equilibrium time, isotherms were constructed to determine the maximum amounts of dye that could be removed by the adsorbents. Figure 7 shows the dye adsorption isotherms for the RHI and RHT adsorbents, at four temperatures (283, 298, 313, and 328 ± 1 K). The isotherms can be represented by mathematical equations used to provide information on the mechanism of adsorption, surface properties, and the affinity of the adsorbent for the adsorbate. Several models

Figure 6. Adsorption kinetics of Remazol brilliant violet dye using RHI and RHT, at pH 2.0 and using initial dye concentrations of 100 and 1000 mg L-1.

Table 1. Fitting parameters for the kinetic models. Models

Parameters

Elovich

RHT Ci=1000 mg L-1

Ci=100 mg L-1

Ci=1000 mg L-1

qe (mg g ) k1 (min-1) R2 qe (mg g-1)

10.17 0.033 0.9091 10.88

46.83 0.261 0.9208 48.35

15.12 0.079 0.7955 16.06

60.26 0.249 0.9031 62.40

k2 (g mg-1 min-1)

0.0049

0.0093

0.0074

0.0065

R2 α (mg g-1 min-1) β (g mg-1) R2

0.9684 3 0.628 0.9845

0.9695 253181 0.350 0.9967

0.9157 31 0.540 0.9900

0.9596 90545 0.250 0.9961

-1

Lagergren

RHI Ci=100 mg L-1

Polímeros, 27(1), 16-26, 2017

Removal of Remazol brilliant violet textile dye by adsorption using rice hulls have been reported in the literature, and in this work, the experimental data were tested using the Freundlich (Equation 5)[48], Langmuir (Equation 6)[49], and Temkin (Equation 7)[50] models. The parameters of the models are given in Table 2. 1/ n qe = K F Ceq (5)

qe = qe=

qmCeq K L 1 + K LCeq

(6)

B ⋅ ln(K T Ceq ) (7)

In the above equations, qm (mg g-1) is the maximum amount of dye adsorbed; KF (L mgn-1 g-n)1/n, KL (L mg-1), and KT (L mg-1) are constants of the Freundlich, Langmuir, and Temkin models, respectively; n is a parameter related to the energy of the surface sites; B (mg g-1) is a constant obtained from the expression B = RT/b, where b is a parameter related to the heat of adsorption, T is the absolute temperature (K), and R is the universal gas constant (8.314 J mol-1 K-1). The parameter, n describing the degree of heterogeneity of the system showed values between 1 and 5, which indicates favorable adsorption for both adsorption systems [51]. Application of the Freundlich model resulted in the highest value of the linear coefficient of determination (R2), with values close to unity for all temperatures studied Table 2.

Previous studies of lignocellulosic materials used for adsorption of Remazol brilliant violet have also found good fits using the Freundlich model. The materials tested were babassu coconut mesocarp[18], and coconut straw[52]. The amount of dye adsorbed increased progressively with increasing temperature, suggesting that the system was endothermic. This behavior could be explained by factors such as increased mobility of the molecules present in the solution (increased kinetic energy caused by the temperature rise), increased diffusion of adsorbate on the surface of the adsorbent, and dilation of the pores of the adsorbent[53].

3.7 Thermodynamics of adsorption Knowledge of the adsorption thermodynamic parameters is important because it enables characterization of the process as spontaneous, exothermic, or endothermic, and provides information on the affinity of the adsorbent for the adsorbate. Additionally, these parameters can provide information concerning the heterogeneity of the surface of the adsorbent and whether the process involves physical or chemical adsorption. The thermodynamic parameters adsorption enthalpy (ΔH°), entropy (ΔS°), and Gibbs energy (ΔGº) were obtained by linearization of the van’t Hoff equation, according to Equations 8, 9, and 10[54]. ln = K eq

ÄS° ÄH° (8) − R RT

Figure 7. Adsorption isotherms for Remazol brilliant violet dye using in natura (RHI) and treated (RHT) rice hulls as adsorbents. Table 2. Parameters of the isotherm models used to fit the adsorption equilibrium data for removal of Remazol brilliant violet dye by the RHI and RHT adsorbents. Temperature error = ±1 K. Isotherms Freundlich

Langmuir

Temkin

Parameters nF KF(L mgn-1 g-n)1/n R2 qm(mg g-1) KL(L mg-1) R2 B(mg g-1) KT(L mg-1) R2

Polímeros, 27(1), 16-26, 2017

RHI

RHT

283 K

298 K

313 K

328 K

283 K

298 K

313 K

328 K

1.663 0.751 0.9941 71.16 0.0017 0.9974 13.80 0.0214 0.9777

1.854 1.250 0.9965 68.63 0.0023 0.9943 13.74 0.0287 0.9772

1.990 1.709 0.9958 68.48 0.0028 0.9914 13.77 0.0354 0.9757

2.196 2.432 0.997 65.20 0.0037 0.9901 13.44 0.0448 0.9839

2.433 3.260 0.9972 61.41 0.0047 0.9774 12.13 0.0654 0.9771

2.519 3.924 0.9985 65.07 0.0053 0.9728 12.68 0.0781 0.9771

2.647 4.685 0.9983 66.49 0.0060 0.9606 12.40 0.1036 0.9708

3.211 8.156 0.9978 67.87 0.0102 0.9394 11.52 0.2568 0.9718

Ribeiro, G. A. C., Silva, D. S. A., Santos, C. C., Vieira, A. P., Bezerra, C. W. B., Tanaka, A. A., & Santana, S. A. A. ° ÄG =

ÄH° − TÄS° (9)

K eq =

qe 1 (10) q m − qe Ceq

In the above equations, qe is the amount adsorbed (mg g-1), qm is the maximum amount adsorbed at each temperature (mg g-1), Ceq is the equilibrium concentration of the adsorbate (mol L-1), T is the temperature of the adsorptive system (K), Keq are equilibrium constants, and R is the gas constant (8.314 J K-1 mol-1). Figure 8 illustrates the linearization of the experimental data according to the van’t Hoff equation, where linearity was observed between the points for the systems studied. The thermodynamic parameters for the adsorption of Remazol brilliant violet dye by the adsorbents are listed in Table 3. Negative values of the Gibbs energy (ΔGº) indicated that the interactions between the dyes and the RHT and RHI adsorbents were spontaneous and favorable, because the lower the value of ΔG°, the greater the driving force of the process, resulting in high adsorption[53]. This was supported by the more negative values of ΔG° for RHT, which also showed greater adsorption of the dye, compared to RHI. Positive enthalpy (ΔHº) values indicated the endothermic nature of all the adsorption processes, with the amount of dye adsorbed at equilibrium tending to increase with increasing temperature. The positive entropy (ΔSº) values indicated that entropy increased during the adsorption process, so the process tended towards a higher degree of disorder.

Figure 8. Van’t Hoff plots for the adsorption of Remazol brilliant violet dye by the in natura (RHI) and treated (RHT) rice hull adsorbents.

Bekçi et al.[55] reported that an enthalpy change of between 0 and 20 kJ mol-1 was characteristic of a physisorption process, while a change of between 80 and 400 kJ mol-1 reflected chemisorption. Alkan et al.[56] found that an enthalpy change of between 40 and 120 kJ mol-1 characterized a chemisorption process. However, Zhou et al.[57] considered that these ranges were not sufficient to confirm the type of adsorption. In fact, under favorable conditions, these processes can occur simultaneously or alternately[58], but the values obtained suggested that adsorption of the dye by the adsorbents occurred mainly due to physisorption.

3.8 Possible interactions between the dye and the adsorbents Elucidation of the mechanism of dye adsorption is a major challenge, as many interactions can occur during the adsorption process. Remazol Brilliant Violet is an anionic dye that contains a sulfonic group in its structure (which ionizes in aqueous solution, forming colored anions), together with aromatic rings. The RHT and RHI adsorbents were lignocellulosic materials composed primarily of cellulose, hemicellulose, and lignin. Characterization of the adsorbents revealed heterogeneous surfaces and the presence of hydroxyl and carbonyl groups. Comparison of the IR spectra for the dye and the rice hull surfaces (Figures 2a, 2c, and 2e) revealed that they were very similar, with the exception of bands at 1639 cm-1 (amide I) and 626 cm-1 (C–S stretching), which were only present in the dye spectrum. These bands were present in the spectra for the dye-saturated biomasses (Figures 2b and 2d), indicating that adsorption had occurred. Unfortunately, the characteristic N=N stretching band of the dye was overlapped with the C–H deformation vibration band, and could not be used for comparative analysis[27]. No other new bands were present in the spectra for the materials with adsorbed dye, suggesting similar predominant functional groups after adsorption. This was in agreement with the better definition and intensity of the IR bands for the materials with adsorbed dye, as well as with the estimated values of the thermodynamic parameters (which were indicative of physisorption)[58]. Some of the well-defined bands, such as those at 1100-1300 cm-1, could be attributed to protonated sulfonic groups[27,59] that probably resulted from interactions between the dye and the protonated surface, as illustrated in Figure 9. The pHpzc values showed that at pH below 5.3 (RHI) and 3.8 (RHT), the surfaces of the adsorbents were positively

Table 3. Thermodynamic parameters for the adsorption of Remazol brilliant violet dye by the RHI and RHT adsorbents. Adsorbent RHI

RHT

T (K)

Keq x10-4

ΔGº (kJ mol-1)

283 ± 1 298 ± 1 313 ± 1 328 ± 1 283 ± 1 298 ± 1 313 ± 1 328 ± 1

0.29 0.59 1.48 3.00 1.06 1.63 2.56 3.49

-18.60 -21.75 -24.91 -28.07 -21.81 -24.07 -26.32 -28.58

ΔHº (kJ mol-1)

ΔSº (J mol-1 K-1)

40.95

210.41

0.9978

20.71

150.27

0.9989

Polímeros, 27(1), 16-26, 2017

Removal of Remazol brilliant violet textile dye by adsorption using rice hulls here. The adsorption isotherms obtained for different temperatures showed that the removal efficiency increased with increasing temperature. The Freundlich isotherm model provided the best explanation of the experimental results for adsorption of the dye. The thermodynamic parameters (ΔG°, ΔH°, and ΔS°) obtained for these adsorption systems (at the temperatures studied) revealed that the reactions were spontaneous, energetically favorable, and endothermic, with a high degree of disorder.

5. Acknowledgments Figure 9. Possible adsorbent-dye interactions: (i) hydrogen bonds; (ii) electrostatic attraction; (iii) π electron resonance.

charged, with the presence of H+ ions favoring adsorption of anionic species such as the dyes studied here. This was supported by the results obtained in the study of the influence of pH on the adsorption, with greater removal achieved at pH 1 and 2. These findings indicated that the mechanism of adsorption of the dye by the adsorbent occurred due to electrostatic attraction, with the adsorbents being protonated in an acid medium, hence attracting the negatively charged sulfonic groups of the dye. The treated adsorbent showed greater dye removal efficiency, compared to the in natura material, especially at less acidic pH (pH 3-6), where the percentage removal remained essentially constant. However, a smaller influence of protonation was observed for the treated adsorbent. These findings confirmed that in addition to electrostatic attraction, other interactions were involved in the adsorption systems studied. The FTIR results showed that carboxyl, hydroxyl, and aromatic carbon functional groups could participate in dye adsorption. Considering all the information obtained in the experiments, three possible interaction mechanisms[50,60] can be proposed for the adsorption of Remazol Brilliant Violet dye by the adsorbents at pH 2: (i) hydrogen bonding, (ii) electrostatic attraction, and (iii) π electron resonance (Figure 9).

4. Conclusions The surface of rice hull was successfully modified by treatment with nitric acid, resulting in a decrease in the point of zero charge (pHpzc). Infrared and 13C-NMR analyses showed that oxygenated groups were introduced in the rice hull surface after treatment. Thermogravimetric analysis indicated that the matrix remained stable up to temperatures close to 473 K, and that the treated adsorbent showed greater weight loss due to the decomposition of lignin. The points of zero charge (pHpzc) were pH 5.3 and pH 3.8 for the in natura and treated materials, respectively. Below the pHpcz, the surface was positively charged, favoring adsorption of the anionic dye. Kinetic studies revealed that adsorption equilibrium was achieved within around 240 min. The Elovich kinetic model provided the best fit to the experimental data for adsorption of the dye, for the two concentrations employed Polímeros, 27(1), 16-26, 2017

The authors are grateful to FAPEMA, CNPq and CAPES for financial support.

6. References 1. Zeng, S., Duan, S., Tang, R., Li, L., Liu, C., & Sun, D. (2014). Magnetically separable Ni0.6Fe2.4O4 nanoparticles as an effective adsorbent for dye removal: Synthesis and study on the kinetic and thermodynamic behaviors for dye adsorption. Chemical Engineering Journal, 258, 218-228. http://dx.doi.org/10.1016/j. cej.2014.07.093. 2. Peng, X., Huang, D., Odoom-Wubah, T., Fu, D., Huang, J., & Qin, Q. (2014). Adsorption of anionic and cationic dyes on ferromagnetic ordered mesoporous carbon from aqueous solution: equilibrium, thermodynamic and kinetics. Journal of Colloid and Interface Science, 430, 272-282. PMid:24973701. http://dx.doi.org/10.1016/j.jcis.2014.05.035. 3. Crini, G. (2006). Non-conventional low-cost adsorbents for dye removal: A review. Bioresource Technology, 97(9), 1061-1085. PMid:15993052. http://dx.doi.org/10.1016/j. biortech.2005.05.001. 4. Durán-Jiménez, G., Hernández-Montoya, V., Montes-Morán, M. A., Bonilla-Petriciolet, A., & Rangel-Vázquez, N. A. (2014). Adsorption of dyes with different molecular properties on activated carbons prepared from lignocellulosic wastes by Taguchi method. Microporous and Mesoporous Materials, 199, 99-107. http://dx.doi.org/10.1016/j.micromeso.2014.08.013. 5. Guaratini, C. C. I., & Zanoni, M. V. B. (2000). Corantes Têxteis. Quimica Nova, 23(1), 71-78. http://dx.doi.org/10.1590/S010040422000000100013. 6. Sathishkumar, P., Mani, A., & Thayumanavan, P. (2012). Utilization of agro-industrial waste Jatropha curcas pods as an activated carbon for the adsorption of reactive dye Remazol Brilliant Blue R (RBBR). J. Journal of Cleaner Production, 22(1), 67-75. http://dx.doi.org/10.1016/j.jclepro.2011.09.017. 7. Zhong, Z.-Y., Yang, Q., Li, X.-M., Luo, K., Liu, Y., & Zeng, G.-M. (2012). Preparation of peanut hull-based activated carbon by microwave-induced phosphoric acid activation and its application in Remazol Brilliant Blue R adsorption. Industrial Crops and Products, 37(1), 178-185. http://dx.doi. org/10.1016/j.indcrop.2011.12.015. 8. Bertazzoli, R., & Pelegrini, R. (2002). Descoloração e degradação de poluentes orgânicos em soluções aquosas através do processo fotoeletroquímico. Quimica Nova, 25(3), 477-482. http://dx.doi.org/10.1590/S0100-40422002000300022. 9. Freire, R. S., & Pereira, W. S. (2005). Ferro zero: Uma nova abordagem para o tratamento de águas contaminadas com compostos orgânicos poluentes. Quimica Nova, 28(1), 130136. http://dx.doi.org/10.1590/S0100-40422005000100022. 10. Souza, C. R. L., & Peralta-Zamora, P. (2005). Degradação de corantes reativos pelo sistema ferro metálico/peróxido de hidrogênio. Quimica Nova, 28(2), 226-228. http://dx.doi. org/10.1590/S0100-40422005000200011. 23

Ribeiro, G. A. C., Silva, D. S. A., Santos, C. C., Vieira, A. P., Bezerra, C. W. B., Tanaka, A. A., & Santana, S. A. A. 11. Moraes, S. G., Freire, R. S., & Durán, N. (2000). Degradation and toxicity reduction of textile effluent by combined photocatalytic and ozonation processes. Chemosphere, 40(4), 369-373. PMid:10665401. http://dx.doi.org/10.1016/S00456535(99)00239-8. 12. Shu, H.-Y., & Chang, M.-C. (2005). Decolorization effects of six azo dyes by O3, UV/O3 and UV/H2O2 processes. Dyes and Pigments, 65(1), 25-31. http://dx.doi.org/10.1016/j. dyepig.2004.06.014. 13. Dallago, R. M., Smaniotto, A., & Oliveira, L. C. A. (2005). Resíduos sólidos de curtumes como adsorventes para a remoção de corantes em meio aquoso. Quimica Nova, 28(3), 433-437. http://dx.doi.org/10.1590/S0100-40422005000300013. 14. Adamson, A. W., & Gast, A. P. (1997). Physical chemistry of surfaces. New York: John Wiley & Sons. 15. Mak, S.-Y., & Chen, D.-H. (2004). Fast adsorption of methylene blue on polyacrylic acid-bond iron oxide magnetic nanoparticles. Dyes and Pigments, 61(1), 93-98. http://dx.doi.org/10.1016/j. dyepig.2003.10.008. 16. Babel, S., & Kurniawan, T. A. (2003). Low-cost adsorbents for heavy metals uptake from contaminated water: a review. Journal of Hazardous Materials, 97(1-3), 219-243. PMid:12573840. http://dx.doi.org/10.1016/S0304-3894(02)00263-7. 17. McKay, G. (1980). Colour removal by adsorption. American Dyestuff Reporter, 69, 38. http://dx.doi.org/10.1016/00431354(88)90165-0. 18. Vieira, A. P., Santana, S. A. A., Bezerra, C. W. B., Silva, H. A. S. S., Chaves, J. A. P., Melo, J. C. P., Silva, E. C., Fo., & Airoldi, C. (2009). Kinetics and thermodynamics of textile dye adsorption from aqueous solutions using babassu coconut mesocarp. Journal of Hazardous Materials, 166(23), 1272-1278. PMid:19150173. http://dx.doi.org/10.1016/j. jhazmat.2008.12.043. 19. Costa, E. P., Santana, S. A. A., Bezerra, C. W. B., Silva, H. A. S., & Schultz, M. S. (2009). Uso da casca de arroz como adsorvente na remoção do Corante têxtil vermelho remazol 5R. Caderno de Pesquisa, 16(2), 44-48. 20. Gong, R., Jin, Y., Chen, F., Chen, J., & Zhili, L. (2006). Enhanced malachite green removal from aqueous solution by citric acid modified rice straw. Journal of Hazardous Materials, 137(2), 865-870. PMid:16621265. http://dx.doi.org/10.1016/j. jhazmat.2006.03.010. 21. Rodrigues, N. F. M., Santana, S. A. A., Bezerra, C. W. B., Silva, H. A. S., Melo, J. C. P., Vieira, A. P., Airoldi, C., & Silva, E. C., Fo., (2013). New chemical organic anhydride immobilization process used on banana pseudostems: a biopolymer for cation removal. Journal of Industrial and Engineering Chemistry, 52(32), 11007-11015. http://dx.doi.org/10.1021/ie303409b. 22. Della, V. P., Kuhn, I., & Hotza, D. (2001). Caracterização de cinza de casca de arroz para uso como matéria-prima na fabricação de refratários de sílica. Quimica Nova, 24(6), 778782. http://dx.doi.org/10.1590/S0100-40422001000600013. 23. Della, V. P., Kuhn, I., & Hotza, D. (2006). Estudo comparativo entre sílica obtida por lixívia ácida da casca de arroz e sílica obtida por tratamento térmico da cinza de casca de arroz. Quimica Nova, 29(6), 1175-1179. http://dx.doi.org/10.1590/ S0100-40422006000600005. 24. Dyominov, I. G., & Zadorozhny, A. M. (2005). Greenhouse gases and recovery of the Earth’s ozone layer. Advances in Space Research, 35(8), 1369-1374. http://dx.doi.org/10.1016/j. asr.2005.04.090. 25. Ponnusami, V., Krithika, V., Madhuram, R., & Srivastava, S. N. (2007). Biosorption of reactive dye using acid-treated rice husk:Factorial design analysis. Journal of Hazardous Materials, 142(1-2), 397-403. PMid:17011118. http://dx.doi. org/10.1016/j.jhazmat.2006.08.040. 24

26. Chiou, M. S., Ho, P. Y., & Li, H. Y. (2004). Adsorption of anionic dyes in acid solutions using chemically cross-linked chitosan beads. Dyes and Pigments, 60(1), 69-84. http://dx.doi. org/10.1016/S0143-7208(03)00140-2. 27. Yuen, C. W. M., Ku, S. K. A., Choi, P. S. R., Kan, C. W., & Tsang, S. Y. (2005). Determining functional groups of commercially available ink-jet printing reactive dyes using infrared spectroscopy. Research Journal of Textile and Apparel, 9, 26-38. 28. Aboudan, M., & Kassab, R. (2015). Synthesis of new azo dyes derived from 2,7-dihydroxynaphthalene. International Journal of Academic Scientific Research, 3, 143-149. 29. Noda, I. (1993). Generalized two-dimensional correlation method applicable to infrared, raman, and other types of spectroscopy. Applied Spectroscopy, 47(9), 1329-1336. http:// dx.doi.org/10.1366/0003702934067694. 30. Gupta, V. K., Pathania, D., Sharma, S., Agarwal, S., & Singh, P. (2013). Remediation of noxious chromium (VI) utilizing acrylic acid grafted lignocellulosic adsorbent. Journal of Molecular Liquids, 177, 343-352. http://dx.doi.org/10.1016/j. molliq.2012.10.017. 31. Kamath, S. R., & Proctor, A. (1998). Sílica gel from rice hull ash: preparation and characterization. Cereal Chemistry, 75(4), 484-487. http://dx.doi.org/10.1094/CCHEM.1998.75.4.484. 32. Tserki, V., Matzinos, P., Kokkou, S., & Panayiotou, C. (2005). Novel biodegradable composites based on treated lignocellulosic waste flour as filler. Part I. Surface chemical modification and characterization of waste flour. Composites. Part A, Applied Science and Manufacturing, 36(7), 965-974. http://dx.doi. org/10.1016/j.compositesa.2004.11.010. 33. Fras, L., Johansson, L. S., Stenius, P., Laine, J., Stana-Kleinschek, K., & Ribitsch, V. (2005). Analysis of the oxidation of cellulose fibres by titration and XPS. Colloids and Surfaces, 260(1-3), 101-108. http://dx.doi.org/10.1016/j.colsurfa.2005.01.035. 34. Smits, J., & Grieken, R. V. (1981). Enrichment of trace anions from water with 2,2′-diaminodiethylamine cellulose filters. Analytica Chimica Acta, 123, 9-17. http://dx.doi.org/10.1016/ S0003-2670(01)83152-4. 35. Liu, C. F., Sun, R. C., Zhang, A. P., & Ren, J. L. (2007). Preparation of sugarcane bagasse cellulosic phthalate using an ionic liquid as reaction medium. Carbohydrate Polymers, 68(1), 17-25. http://dx.doi.org/10.1016/j.carbpol.2006.07.002. 36. Chang, S. T., & Chang, H. T. (2001). Comparisons of the photostability of esterified wood. Polymer Degradation & Stability, 71(2), 261-266. http://dx.doi.org/10.1016/S01413910(00)00171-3. 37. Tarley, C. R. T., & Arruda, M. A. Z. (2004). Biosorption of heavy metals using rice milling byproducts. Characterization and application for removal of metals from aqueous effluents. Chemosphere, 54(7), 987-995. PMid:14637356. http://dx.doi. org/10.1016/j.chemosphere.2003.09.001. 38. Han, R., Zhang, L., Song, C., Zhang, M., Zhu, H., & Zhang, L. J. (2010). Characterization of modified wheat straw, kinetic and equilibrium study about copper ion and methylene blue adsorption in batch mode. Carbohydrate Polymers, 79(4), 1140-1149. http://dx.doi.org/10.1016/j.carbpol.2009.10.054. 39. Holanda, C. A., Sousa, J. L., Santos, C. C., Santos, H. A. S., Santana, S. A. A., Costa, M. C. P., Schultz, M. S., & Bezerra, C. W. B. (2015). Remoção Do Corante Têxtil Turquesa De Remazol Empregando Aguapé (Eichhornia Crassipes) Como Adsorvente. Electronic Journal of Chemistry, 7(2), 141-154. 40. Vieira, A. P., Santana, S. A. A., Bezerra, C. W. B., Silva, H. A. S. S., Chaves, J. A. P., Melo, J. C. P., Silva, E. C., Fo., & Airoldi, C. (2011). Removal of textile dyes from aqueous solution by babassu coconut epicarp (Orbignya speciosa). Polímeros, 27(1), 16-26, 2017

Removal of Remazol brilliant violet textile dye by adsorption using rice hulls Chemical Engineering Journal, 173(2), 334-340. http://dx.doi. org/10.1016/j.cej.2011.07.043. 41. Janos, P., Coskun, S., Pilarová, V., & Rejnek, J. (2009). Removal of basic (Methylene Blue) and acid (Egacid Orange) dyes from waters by sorption on chemically treated wood shavings. Bioresource Technology, 100(3), 1450-1453. PMid:18848777. http://dx.doi.org/10.1016/j.biortech.2008.06.069. 42. Vieira, A. P., Santana, S. A. A., Bezerra, C. W. B., Silva, H. A. S. S., Chaves, J. A. P., Melo, J. C. P., Silva, E. C., Fo., & Airoldi, C. (2011). Epicarp and Mesocarp of Babassu (Orbignya speciosa): Characterization and Application in Copper Phtalocyanine Dye Removal. Journal of the Brazilian Chemical Society, 22(1), 21-29. http://dx.doi.org/10.1590/ S0103-50532011000100003. 43. Lagergren, S. (1898). About the theory of so-called adsorption of soluble substances. Handlingar, 24, 1-39. 44. Ho, Y., & McKay, G. (1998). The kinetics of sorption of basic dyes from aqueous solution by sphagnum moss peat. Canadian Society for Chemical Engineering, 76(4), 822-827. http://dx.doi.org/10.1002/cjce.5450760419. 45. Peers, A. M. (1965). Elovich adsorption kinetics and the heterogeneous surface. Journal of Catalysis, 4(4), 499-503. http://dx.doi.org/10.1016/0021-9517(65)90054-0. 46. Tunç, O., Tanaci, H., & Aksu, Z. (2009). Potential use of cotton plant wastes for the removal of Remazol Black B reactive dye. Journal of Hazardous Materials, 163(1), 187-198. PMid:18675510. http://dx.doi.org/10.1016/j.jhazmat.2008.06.078. 47. Ai, L., Zhang, C., & Meng, L. (2011). Adsorption of Methyl Orange from Aqueous Solution on Hydrothermal Synthesized Mg-Al Layered Double Hydroxide. Journal of Chemical & Engineering Data, 56(11), 4217-4225. http://dx.doi.org/10.1021/ je200743u. 48. Freundlich, H. M. F. (1906). Over the adsorption in solution. Journal of Physical Chemistry, 57, 385-470. 49. Langmuir, I. (1918). The adsorption of gases on plane surfaces of glass, mica and platinum. Journal of the American Chemical Society, 40(9), 1361-1403. http://dx.doi.org/10.1021/ja02242a004. 50. Temkin, M. I. (1972). Theoretical models of the kinetics of heterogeneous catalytic reactions. Kinetics and Catalysis, 13, 555-565. http://dx.doi.org/10.1021/cr00035a010. 51. Achak, M., Hafidi, A., Ouazzani, N., Sayadi, S., & Mandi, L. (2009). Low cost biosorbent “banana peel” for the removal of phenolic compounds from olive mill wastewater: Kinetic and equilibrium studies. Journal of Hazardous Materials, 166(1), 117-125. PMid:19144464. http://dx.doi.org/10.1016/j. jhazmat.2008.11.036.

Polímeros, 27(1), 16-26, 2017

52. Namasivayam, C., Kumar, M. D., Selvi, K., Begum, R. A., Vanathi, T., & Yamuna, R. T. (2001). Waste coir pith-a potential biomass for the treatment of dyeing wastewaters. Biomass and Bioenergy, 21(6), 477-483. http://dx.doi.org/10.1016/ S0961-9534(01)00052-6. 53. Saeed, A., Sharif, M., & Iqbal, M. (2010). Application potential of grapefruit peel as dye sorbent: kinetics, equilibrium and mechanism of crystal violet adsorption. Journal of Hazardous Materials, 179(1-3), 564-572. PMid:20381962. http://dx.doi. org/10.1016/j.jhazmat.2010.03.041. 54. Wang, L., & Li, J. (2013). Adsorption of C.I. Reactive Red 228 dye from aqueous solution by modified cellulose from flax shive: Kinetics, equilibrium, and thermodynamics. Industrial Crops and Products, 42, 153-158. http://dx.doi.org/10.1016/j. indcrop.2012.05.031. 55. Bekçi, Z., Seki, Y., & Cavas, L. (2009). Removal of malachite green by using an invasive marine alga Caulerpa racemosa var. Cylindracea. Journal of Hazardous Materials, 161(2-3), 1454-1460. PMid:18562093. http://dx.doi.org/10.1016/j. jhazmat.2008.04.125. 56. Alkan, M., Demirbas, O., Celikcapa, S., & Dogan, M. (2004). Sorption of acid red 57 from aqueous solution onto sepiolite. Journal of Hazardous Materials, 116(1-2), 135-145. PMid:15561372. http://dx.doi.org/10.1016/j.jhazmat.2004.08.003. 57. Zhou, Y. T., Nie, C. B.-W., & Zhu, J. C. (2009). Adsorption mechanism of Cu2+ from aqueous solution by chitosan-coated magnetic nanoparticles modified with α-ketoglutaric acid. Colloids and Surfaces. B, Biointerfaces, 74(1), 244-252. PMid:19683900. http://dx.doi.org/10.1016/j.colsurfb.2009.07.026. 58. Dabrowski, A. (2001). Adsorption - from theory to practice. Advances in Colloid and Interface Science, 93(1-3), 135-224. PMid:11591108. http://dx.doi.org/10.1016/S0001-8686(00)000828. 59. Silverstein, R. M., Bassler, G. C., & Morrill, T. C. (1991). Spectrometric identification of organic compounds. New York: John Wiley & Sons. 60. Chatterjee, S., Chatterjee, B. P., & Guha, A. K. (2007). Adsorptive removal of congo red, a carcinogenic textile dye by chitosan hydrobeads: Binding mechanism, equilibrium and kinetics. Colloids and Surfaces. A, Physicochemical and Engineering Aspects, 299(1-3), 146-152. http://dx.doi. org/10.1016/j.colsurfa.2006.11.036. Received: Nov. 09, 2015 Revised: Apr. 05, 2016 Accepted: Apr. 28, 2016

Ribeiro, G. A. C., Silva, D. S. A., Santos, C. C., Vieira, A. P., Bezerra, C. W. B., Tanaka, A. A., & Santana, S. A. A. Appendix A. Aditional materials. Table A1. Percentages of carbon (C), hydrogen (H), and nitrogen (N) in the in natura (RHI) and treated (RHT) adsorbents. Adsorbent RHI RHT Difference

%C 35.87 39.73 3.86

%H 5.04 5.37 0.33

%N 0.54 1.46 0.92

Figure A1. X-ray diffractograms of the in natura (RHI) and treated (RHT) rice hulls.

PolĂmeros, 27(1), 16-26, 2017

http://dx.doi.org/10.1590/0104-1428.2181

The effect of gelatin amount on the properties of PLA/TPS/gelatin extruded sheets Ana Paula de Oliveira Pizzoli1, Fabio Yamashita2, Odinei Hess Gonçalves1, Marianne Ayumi Shirai1 and Fernanda Vitória Leimann1* 1

Programa de Pós-graduação em Tecnologia de Alimentos, Universidade Tecnológica Federal do Paraná – UTFPR, Campo Mourão, PR, Brazil 2 Departamento de Ciência e Tecnologia de Alimentos, Centro de Ciências Agrárias, Universidade Estadual de Londrina – UEL, Londrina, PR, Brazil *fernandaleimann@utfpr.edu.br

Abstract Films and sheets composed by poly (lactic acid) (PLA)/thermoplastic starch (TPS) and TPS/gelatin blends have already been produced and characterized in the literature. However, materials produced with these three biopolymers have not been clearly investigated. In this work, extruded sheets were produced with PLA, TPS (glycerol as plasticizer) and different amounts of gelatin (0, 1, 3 and 5 wt%) in a pilot scale co-rotating twin-screw extruder coupled with a calender. The extruded sheets were characterized in regards to their water solubility, thickness, density, water vapor permeability (WVP), moisture sorption isotherms, mechanical properties and microstructure. The results showed an increase in solubility and WVP besides a decrease of about 30% in tensile strength, Young’s modulus and elongation at break. Extruded sheets microstructure revealed smother surfaces and homogeneous morphology with the addition of gelatin. The experiments demonstrated that extrusion and calendering process is a viable way to produce PLA/TPS/gelatin sheets with interesting properties. Keywords: biodegradable polymers, extrusion-calendering, hydrophilicity, mechanical characterization, microstructure.

1. Introduction Sustainable development policies tend to expand with the decreasing reserve of fossil fuel and the growing concern for the environment, consequently, biodegradable polymers have emerged as potential alternatives for petrochemical plastics[1]. Poly(lactic acid) (PLA) is becoming a popular biodegradable engineering plastic due to its mechanical properties and easy processability[2]. PLA is a versatile material with applications in the medical, textile, and packaging fields, but its brittleness and high cost in comparison to petroleum-based thermoplastics have limited its applications[3]. Polymer blends based on thermoplastic starch (TPS) have been used to lower the costs of other materials[4] and like starch, other polysaccharides, i.e. cellulose, chitin, chitosan and proteins i.e. gelatin, casein, pectin, etc. have found innumerable applications in biodegradable products[5]. Starch is a main storage energy source for higher plants and one of the primary sources of calories in human nutrition. Native starch occurs as water-insoluble granules whose sizes and shapes are dependent on its botanical source[4,6]. Starch granules can be gelatinized in the presence of a plasticizer and heat, during which the crystalline structure is disrupted (thermoplastic starch, TPS). This allows the starch to flow at high temperatures so it can be processed using conventional polymer processing equipment[7,8]. Gelatin is another low-cost and abundant raw material, and is well-known for its good film forming properties[9,10]. Many studies have been made on the combination of starch and gelatin[9-13] and also of starch and PLA[14,15]. However, the literature is lacking sufficient studies on the combination of these three biopolymers and how they

Polímeros, 27(1), 27-34, 2017

could eventually interact to form the final microstructure. For instance, a PLA-starch- gelatin blend (also containing calcium carbonate and glycerol) obtained by thermo pressing was proposed[16]. The resulting material showed to be easily degradable in a marine environment, which is highly desirable for most forms of applications. However, the authors focused mainly on the biodegradation behavior and on its amorphous/crystalline microstructure characterization under optical polarization microscopy. A more thorough evaluation should be performed to determine how blend composition could affect the mechanical and barrier properties of the material because they would give an insight on the material applicability and performance. In this context, the present study develops a novel biodegradable packaging blend based on PLA/TPS/gelatin. The influence of gelatin content on the mechanical properties, microstructure, water vapor barrier properties, humidity, density and thickness was investigated.

2. Materials and Methods 2.1 Materials PLA Ingeo 4043D (Natureworks LLC, Cargill, USA), cassava starch (Indemil, Brazil), glycerol (Dinâmica, Brazil), as plasticizing for the starch, and gelatin (Dinâmica, P.A., Brazil) were used to produce the extruded blend sheets. Magnesium nitrate and calcium chloride (Vetec, Brazil) were used to control the relative humidity during sheets conditioning.

O O O O O O O O O O O O O O O O

Pizzoli, A. P. O., Yamashita, F., Gonçalves, O. H., Shirai, M. A., & Leimann, F. V. 2.2 Production of pellets and sheets The procedure adopted for the production of pellets and sheets was described by Shirai et al.[17] with some modifications, and the sheets formulations are presented in Table 1. A control formulation (FC) was produced with 50:50 PLA:TPS with no added gelatin. In FG1, FG3 and FG5 formulations, gelatin percentage (wt%) is related to TPS mass (starch and glycerol). The proportion between starch and glycerol was kept constant in all formulations (33 g glycerol/100 g starch). Initially, gelatin was allowed to gelatinize in contact with glycerol during 24 hours at room temperature. After that, starch and PLA were added, manually mixed and extruded as cylindrical profiles in a single-screw extruder (BGM, EL-25 model, Brazil) using the following processing conditions: screw diameter of 25 mm, screw length of 28 D, screw speed of 30 rpm, and temperature profile of 90 / 180 / 180 / 180 °C at the four heating zones. The extruded cylindrical profiles were cooled at room temperature and, then, pelletized. The obtained pellets were extruded in a pilot co-rotating twin-screw extruder (BGM, D-20 model) coupled with a calender (AX-Plásticos, Brazil) for sheet production. The processing conditions employed were: screws diameter (D) of 20 mm, screws length of 35 D, temperature profile of 100 / 170 / 170 / 170 / 175 °C, screw speed of 100 rpm, and feed speed of 30 rpm. The distance between the calender rolls was kept at 0.8 mm and the rolls’ speed was adjusted depending on the formulation to maintain continuous processing.

2.3 Thickness, density and moisture content Sheet thickness was determined with the use of a digital micrometer (Starrett, Brazil) with 0.001 mm resolution. Ten random points were used from each sample. Two density determination samples (20 × 20 mm) were kept in a desiccator containing anhydrous calcium chloride (0% relative humidity) for two weeks and finally weighted[18]. Two moisture content determination sheets samples were weighted (mi1) and conditioned in a forced air oven by 24 h at 70 °C. After that, the sheets were weighted again (ms1) and the humidity (%) was calculated with the Equation 1. (mi1 − ms1 ) U= .100 mi1

(1)

Water solubility was defined as the dry mass content from the sheets that was solubilized after 24 h of immersion in water at 25 °C. The procedure adopted was described by Table 1. Composition of the PLA/TPS/gelatin extruded sheets FC (control sample), FG1 (0.5%gelatin), FG3 (1.5% gelatin) and FG5 (2.5% gelatin).

FC FG1 FG3 FG5

PLA (%) 50.0 50.0 50.0 50.0

SOL =

((mi2 − ma ) − ms2 ) .100 (mi2 − ma )

(2)

2.5 Water Vapor Permeability (WVP) The water vapor permeability of the sheets was determined in appropriate diffusion cells, using a relative humidity (RH) of 2% inside the cell and 75% outside the cell[19]. All tests were performed in triplicate.

2.6 Moisture sorption isotherm The moisture sorption isotherms of the sheets were obtained by static gravimetric method using saturated saline solutions (11.8%, 32.8%, 52.9%, 75% and 87%) to promote different relative humidity values. The sheets were previously dried in desiccator containing anhydrous calcium chloride (0% relative humidity) for two weeks, and then maintained in closed recipients with different saturated saline solution at 25 °C. The samples were weighed in regular intervals until three equal weight measurements were obtained (equilibrium condition). The absolute humidity (dry base) was determined by the oven method (105 °C for 24 h). The GAB (Guggenheim-Anderson-de Boer) model (Equation 3) was used to fit the experimental data. In this equation, the parameter “Xw” is the equilibrium moisture content (g water/g dry solid) at a known water activity (aw), mo is the monolayer water content, “C” is the Guggenheim constant (representing the sorption heat of the first layer), and “K” is the sorption heat of the multilayer. The parameters of the GAB model were determined using non-linear regression performed with the Statistica 7.0 software. The test was perfomed in triplicate. Xw =

m0 .C.K .aw (3) (1 − K .aw ).(1 − K .aw + C.K .aw )

2.7 Mechanical properties

2.4 Water solubility

Formulation

Soares et al.[14]. Sheet samples (2 × 2 cm) were weighted (mi2) and then immersed in water (200 mL, 25 +/- 2 °C) by 24 h. After that, the residual sheet was removed and dried at 70 °C in a forced air oven for 24 h. Finally the sheet was then weighed (ms2) and water solubility (%) was calculated with Equation 2; ma is the mass of water calculated from the humidity, and (mi2 - ma) the initial mass of the sheet on a dry basis.

Starch (%) 37.5 37.1 36.4 35.6

Glycerol (%) 12.5 12.4 12.1 11.9

Gelatin (%) 0 0.5 1.5 2.5

The tensile strength tests were performed with a texture analyser (Stable Micro Systems, TA XTplus model, England) based on the American Society for Testing and Material standards ASTM D-882-00[20]. The samples were previously conditioned at 23 ± 2 °C and 53 ± 2% of relative humidity for 48 h. The properties evaluated were tensile strength (MPa), elongation at break (%), and Young’s modulus (MPa). Ten samples were tested for each formulation.

2.8 Scanning electron microscopy (SEM) The microstructure of the surface and fractured sheets was analyzed with a scanning electron microscope (Philips, XL-30 model, Holland) with electron source of tungsten Polímeros, 27(1), 27-34, 2017

The effect of gelatin amount on the properties of PLA/TPS/gelatin extruded sheets and detectors of secondary and backscattered electrons at 20 kV. The sheet samples were immersed in liquid nitrogen and then fractured. After that, they were coated in gold using a sputter coater (BALTEC, SCD 005 model, Switzerland).

2.9 Statistical analysis The results were evaluated using analysis of variance (ANOVA), and treatment means were compared using Tukey’s test at the 5% significance level (p < 0.05) with Statistica 7.0 software (Stat-Soft, Tulsa, OK, USA).

3. Results and Discussions PLA/TPS/gelatin extruded sheets presented good processability and were successfully obtained. The results obtained for the formulations, taking into account the effect of the gelatin amount, are presented as follows.

3.1 Sheets appearance, thickness and density It was possible to observe visually, as shown at Figure 1, that the increase in the amount of gelatin (FG3 and FG5) resulted in a more yellowish appearance than formulation FG1, which was white. All sheets presented opaque appearance. Similar results were also observed by Fakhouri et al.[10,12]. Extruded TPS based materials are opaque and whitish due to the alignment of the chains and subsequent crystallization induced by extrusion which do not occurs in films produced by casting technique[10]. Thickness and density values obtained for PLA/TPS and PLA/TPS/gelatin sheets are presented in Table 2. No significant difference was detected (p > 0.05) between formulations with distinct gelatin amounts. Choi et al.[21] observed an increase in density of an artificial skin (composed

by sodium alginate and gelatin) as gelatin amount was increased. The authors used gelatin amounts between 50 and 90% from the polymeric content which is higher than mounts used here explaining the statistically equal results. Shirai et al.[17] also produced PLA/TPS blends by calendering-extrusion and observed lower density values (0.96 to 1.19 g.cm-3). They used a proportion between PLA and TPS of 1:3.33 while at the present study a proportion of about 2.5:1 was used, which could explain the difference. During the calendering-extrusion process, sheets thickness is controlled by the rolls speed, the distance between them and the stretching capacity of the formulation. According to Table 2, the thickness values obtained for all the sheets was closer (316 to 391 µm), suggesting that gelatin addition did not interfere in the processability of the PLA/TPS blends.

3.2 Sheets morphology The blend morphology was assessed by SEM observation of the sheets’ surface and fracture. In the micrographs presented in Figure 2 it is possible to observe that sheets fractures presented different characteristics. The control formulation (Figure 2, FC-A) presented a porous structure while the formulation with 1 wt% of gelatin showed fibrous characteristic that provides hollows in the structure. Table 2. Density and thickness values of the extruded PLA/TPS/ gelatin sheets. Formulation FC FG1 FG3 FG5

Density (g.cm-3) 1.25 ± 0.09 1.31 ± 0.09 1.22 ± 0.07 1.04 ± 0.10

Thickness (µm) 357 ± 47 316 ± 25 391 ± 40 320 ± 79

Figure 1. Sheets photographic images FC (control sample), FG1 (0.5% gelatin), FG3 (1.5% gelatin) and FG5 (2.5% gelatin). Polímeros, 27(1), 27-34, 2017

Pizzoli, A. P. O., Yamashita, F., Gonçalves, O. H., Shirai, M. A., & Leimann, F. V. The formulations FG3 and FG5 presented structures with low porosity and higher solidity. At the surface of all formulations, it is possible to observe that non-gelatinized starch granules are present. Between the formulations with gelatin, FG1 was the one that presented a more irregular surface. Similar results were obtained by Zhang et al.[22] that observed protrusions at the surface of gelatin/starch films, and that the density of the irregularities increased directly with the starch proportion. The authors suggest that the two phases present different shrinking rates during the drying period. In the case of the present study, it

Figure 2. Sheets micrographs where (A) fracture (magnification:

800 x) and (B) surface (magnification: 1,600 x): FC (control sample), FG1 (0.5% gelatin), FG3 (1.5% gelatin) and FG5 (2.5% gelatin).

is probable that the conformation rate of the polymers was different during the cooling step after lamination. Guzman-Sielicka et al.[16] produced PLLA, gelatin and TPS blends containing calcium carbonate by thermopressing. They observed an increase in the surface porosity when the amount of gelatin was increased in the formulations. This effect was undetectable in the images of Figure 2 probably due to the smallest proportions of gelatin used (up to 5 wt% in relationship to the total formulation) while the authors used from 10 to 40 wt%.

3.3 Water vapor permeability (WVP), moisture content and water solubility The water vapor permeability (WVP), moisture content and water solubility results are shown in Table 3. Permeability can be defined as the product of diffusivity and solubility when Fick and Henry laws fully apply. For most edible films the water vapor strongly interacts with polymer structure, which results in diffusion and solubility coefficients dependent on driving force[23-25].The addition of gelatin increased significantly the WVP of the sheets, conferring more hydrophilicity to the materials. The tightly bonds (hydrogen bonds and hydrophobic interactions) present in gelatin structure and the polar groups of amino acids resulted in brittle materials in dry state with high moisture absorption, as described by Karnnet et al.[26]. Gelatin is composed of repeated sequences of amino acids such as glycine, proline and hydroxyproline. These sequences are responsible for the triple helical structure of gelatin and its ability to form gels and immobilize water molecules. It is possible that gelatin have a greater water retention capacity than starch, explaining the higher values of WVP of gelatin added sheets[27]. Al-Hassan and Norziah[28] obtained WVP results ranging from 4 to 5 x 10-6 g.m-1.Pa-1.day-1 for sorgo and gelatin films (glycerol was used as plasticizer). Fakhouri et al.[11] reported WVP values of 3.38 g.m-1.Pa-1.day-1 for lipophilic corn starch and gelatin films plasticized with caprilic acid. Jamshidian et al.[29] characterized pure PLLA films produced by casting related lower WVP values (2.7 × 10-15 kg.mm-2.s 1 .Pa 1) than the ones obtained in this study, under relative humidity gradient of 0-90%. Soares et al.[14] obtained WVP values ranging from 21.6 to 43.68 x 10-6 g.m-1.Pa-1. day-1 in TPS/PLA (70/30) films produced by extrusion and thermopressing. Similar results of WVP were reported by Shirai et al.[17] in PLA/TPS (70/30) sheets plasticized with adipate and citrate esters, under the same gradient humidity. The results of moisture content (Table 3) of the sheets did not show significant difference (p > 0.05). However, it is possible to note that as the amount of gelatin in the

Table 3. Water vapor permeability (WVP), moisture content and water solubility values of PLA / TPS / gelatin sheets: FC (control sample), FG1 (0.5% gelatin), FG3 (1.5% gelatin) and FG5 (2.5% gelatin). Formulation FC FG1 FG3 FG5

WVP (g.m-1.Pa-1.day-1) (x 106) 2.83a ± 0.07 3.25a ± 0.003 3.45a ± 0.07 10.10b ± 1.14

Moisture (%) 4.51 ± 0.32 4.39 ± 1.78 5.73 ± 0.28 5.44 ± 0.18

Water Solubility (%) 8.67a ± 0.34 11.44b ± 0.65 20.34c ± 2.09 27.72d ± 2.79

Means followed by the same letters in the column did not show differences at 5% of significance level according Tukey test.

Polímeros, 27(1), 27-34, 2017

The effect of gelatin amount on the properties of PLA/TPS/gelatin extruded sheets formulation increased there was a significant increase in solubility (p < 0.05). This result is probably due to the increased solubility of gelatin in water since plasticized gelatin films are completely soluble[30]. Water solubility of the control and FG5 (2.5% gelatin) sheets ranged from 8.67 to 27.72%, respectively. The dissolution of hydrophilic polymers is accounted to the penetration of water in the polymer bulk and subsequent swelling, followed by disruption of hydrogen and Van der Walls forces between polymer chains[31]. Water absorption and matrix solubilization of hydrophilic polymers was already investigated and the proposed mechanism takes into account the mechanical relaxation of the polymer chains as well as kinetic and thermodynamic factors[32]. Authors demonstrated that the predominant factor is kinetic in nature (water diffusion) although the thermodynamic factor and the physical state of water (liquid or vapor) may led to higher liquid water permeability than vapor water permeability. It is worth noting that polymer conformation was found to play an important role in water-matrix interactions in the case of water-sensitive films. The addition of a plasticizer (PEG400) to methylcellulose films was found to increase the dissolution rate in water due to the disruption of methylcellulose microstructure cause by the presence of PEG400[33]. A similar effect could explain the fact that water solubility increased more than 3 times with only 2.5% gelatin (FC5). Since starch and gelatin were both gelatinized in the extrusion processs, the new interactions formed during molecules rearrangement probably presented lower energy than those of the original materials.

of the packaging material necessary for the protection of a particular system. Furthermore, the sorption isotherm provides information about the hydrophilicity of TPS based materials. In the first part of the sorption curve (aw < 0.30), the amount of water adsorbed was similar for all the sheets. This first part corresponds to water field adsorbed as a monolayer, where polar groups of high binding energy to hydrophilic components (starch and proteins) are saturated with water molecules which are considered integral parts of the solid phase[25]. An increase in the moisture content was observed between aw from 0.3 to 0.9 mainly to PLA/TPS sheets with gelatin addition. According to other authors[34,35], the accumulation of water occurs at the polymeric matrix surface as well as at intermolecular free spaces resulting in partial swelling, which in turn may expose additional hydrophilic binding sites. It is possible that the gelatin addition to the sheets’ formulation promoted a higher exposure of binding sites increasing the interaction with water molecules. In this stage, it could be considered that water can bind either to starch primary or secondary hydroxyls in association with glycerol or gelatin in association with glycerol, or bind it directly to the free glycerol. The GAB model fairly represented the experimental data with correlation coefficient of 0.99, and similar behavior was observed in PLA/TPS sheets with addition of adipate and citrate esters and produced by calendaring-extrusion[17], TPS/PLA films produced by extrusion and thermopressing[36,37], and thermoplastic wheat flour and PLA blends[34].

Evaluating the solubility of gelatin and corn starch films plasticized with caprilic acid, Fakhouri et al.[11] observed values up to 44.82%. Soares et al.[14] determined the solubility of the sheets produced with PLA/TPS as equal to 35.2%. The higher solubility found by these authors in comparison with the present study was due to the difference between the proportion of PLA and TPS employed to produce the sheets, which was 30:70 (PLA/TPS). In the present research, the proportion of PLA/(TPS + gelatin) was 50:50. Due their hydrophobic character, PLA is not soluble in water and the higher concentration promoted solubility with different values.

3.4 Moisture sorption isotherms The moisture sorption isotherms of the PLA/TPS and PLA/TPS/gelatin sheets are presented in Figure 3 and the GAB model parameters are listed in Table 4. It is worth noting that the information derived from isotherms is helpful to determine the required water barrier properties

Figure 3. Moisture sorption isotherm for the PLA / TPS / gelatin sheets: FC (control sample), FG1 (0.5%gelatin), FG3 (1.5% gelatin) and FG5 (2.5% gelatin).

Table 4. GAB model parameters for moisture sorption isotherms for PLA / TPS / gelatin sheets: FC (control sample), FG1 (0.5% gelatin), FG3 (1.5% gelatin) and FG5 (2.5% gelatin). Formulation

GAB parameters mo (g water/g dry solids)

0.031 0.042 0.036 0.044

0.84 0.83 0.89 0.81

8.65 6.59 8.86 5.73

0.99 0.99 0.99 0.99

FC FG1 FG3 FG5

mo = monolayer water content; K = sorption heat of the multilayer; C = Guggenheim constant; R2 = correlation coefficient.

Polímeros, 27(1), 27-34, 2017

Pizzoli, A. P. O., Yamashita, F., Gonçalves, O. H., Shirai, M. A., & Leimann, F. V. Table 5. Mechanical properties of PLA/TPS/gelatin extruded sheets: FC (control sample), FG1 (0.5% gelatin), FG3 (1.5% gelatin) and FG5 (2.5% gelatin). Formulation FC FG1 FG3 FG5

Tensile strenght (MPa) 28.8a ± 2.1 19.4c ± 1.5 22.7b ± 0.9 19.8c ± 0.8

Elongation at rupture (%) 31.7a ± 6.1 37.2a ± 8.9 36.0a ± 9.7 20.5b ± 3.1

Young’s modulus (MPa) 614a ± 51 469c ± 27 520b ± 39 483b,c ± 25

Means followed by the same letters in the column did not show differences at 5% of significance level according Tukey test.

The monolayer moisture content (mo) indicates the maximum amount of water that can be adsorbed in a single layer per gram of dry film and it is a measure of number sorbing sites[38]. In general, it is associated with the material hygroscopicity and hydrofilicity. The inclusion of gelatin increased mo values, indicating greater water sorption capacity corroborating with WVP, moisture and solubility results discussed before. The mo values were lower than those reported by Abdillahi et al.[34] for wheat flour and PLA blends (0.05 to 0.09 g water/ g dry solids), and by Soares et al.[37] for TPS/PLA sheets (0.047 to 0.056 g water/ g dry solids). The difference was mainly due to the higher concentration of hydrophilic materials (starch and wheat flour) that was used to produce the materials. “K” parameter is related to the sorption heat of the multilayer, and when K = 1, it is assumed that there is no interaction of the water vapor in the multilayer or no variation in the energy of sorption in multilayer, which occur in homogeneous solids[39]. In our study, this parameter was not affected by gelatin addition. Finally, the parameter “C” is associated with the sorption heat of the monolayer, and no clear correlation between the samples was observed. Despite the fact that the “C” values are consistent with values found in other studies that worked with starch based materials[17,40].

observed, because of the presence of hydrophobic polymer like PLLA and their incompatibility between starch and gelatin, as described in other studies[15,41]. Although starch and gelatin are hydrophilic and compatible, they are immiscible and presented as two phases[13]. Several reports describe the production of PLLA/gelatin blend and TPS/gelatin blends by casting methods[10,12,13]. The main problem with films produced through this technique is the limitation on the quantity produced, and the use of solvents to disperse components. Although the gelatin addition affected the sheets’ mechanical properties, it was possible to produce them by extrusion and calendering process at pilot scale, which could be an alternative to offer biodegradable packaging for low moisture foods in commercial scale.

4. Conclusions PLA/TPS/gelatin sheets were successfully produced by calendaring-extrusion process at pilot scale. The inclusion of gelatin in PLA/TPS blends interfered significantly in the microstructural, mechanical and water barrier properties, while density and the water vapor permeability were unaffected. The PLA/TPS/gelatin sheets obtained have the potential to become biodegradable packaging for food.

3.5 Mechanical properties

5. Acknowledgements

The results associated with mechanical properties such as tensile strength, Young’s modulus and elongation at break are presented in Table 5. The gradual addition of gelatin in PLA/TPS sheets significantly affected the mechanical properties (p < 0.05). A decrease of about 30% was observed in tensile strength, Young’s modulus and elongation at break when 5% of gelatin was added, which can be explained by the incompatibility between the blend’s components. The mechanical properties of the sheets are directly linked to interfacial adhesion between the blend polymers. When the polymers are mixed, it is interesting that the dispersion and the distribution of the particles occur, forming a single polymeric phase. Poor dispersal could result in the formation of clusters from the entanglement of the polymer chains, which reduces the transmission of tension[37]. In the SEM images observed before, the presence voids and non-gelatinized starch granules complicate the load transfer under stress, explaining the results.

The authors would like to thank Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) and Fundação Araucária for their financial support and to Laboratório de Microscopia Eletrônica da Universidade Estadual de Londrina (UEL) for the SEM images.

Films made from blends of cassava starch and gelatin by casting technique presented increased tensile strength as gelatin was added, due to their reinforcement properties to the polymeric matrix[10,12]. In our study similar results were not 32

6. References 1. Bordes, P., Pollet, E., & Averous, L. (2009). Nano-biocomposites: biodegradable polyester/nanoclay systems. Progress in Polymer Science, 34(2), 125-155. http://dx.doi.org/10.1016/j. progpolymsci.2008.10.002. 2. Fortunati, E., Latterini, L., Rinaldi, S., Kenny, J. M., & Armentano, I. (2011). PLGA/Ag nanocomposites: in vitro degradation study and silver ion release. Journal of Materials Science. Materials in Medicine, 22(12), 2735-2744. PMid:22002470. http://dx.doi.org/10.1007/s10856-011-4450-0. 3. Liao, H.-T., & Wu, C.-S. (2009). Preparation and characterization of ternary blends composed of polylactide, poly(ɛ-caprolactone) and starch. Materials Science and Engineering A, 515(1-2), 207-214. http://dx.doi.org/10.1016/j.msea.2009.03.003. 4. Lescher, P., Jayaraman, K., & Bhattacharyya, D. (2012). Characterization of water-free thermoplastic starch blends for Polímeros, 27(1), 27-34, 2017

The effect of gelatin amount on the properties of PLA/TPS/gelatin extruded sheets manufacturing processes. Materials Science and Engineering A, 532(1), 178-189. http://dx.doi.org/10.1016/j.msea.2011.10.079. 5. Luckachan, G. E., & Pillai, C. K. S. (2011). Biodegradable polymers: a review on recent trends and emerging perspectives. Journal of Polymers and the Environment, 19(3), 637-676. http://dx.doi.org/10.1007/s10924-011-0317-1. 6. Genkina, N. K., Kozlov, S. S., Martirosyan, V. V., & Kiseleva, V. I. (2014). Thermal behavior of maize starches with different amylose / amylopectin ratio studied by DSC analysis. Starch, 66(7-8), 1-7. http://dx.doi.org/10.1002/star.201300220. 7. Taghizadeh, A., Sarazin, P., & Favis, B. D. (2013). High molecular weight plasticizers in thermoplastic starch/polyethylene blends. Journal of Materials Science, 48(4), 1799-1811. http://dx.doi. org/10.1007/s10853-012-6943-8. 8. Cavallaro, G., La Manna, G., Liveri, V. T., Aliotta, F., & Fontanella, M. E. (1995). Structural investigation of water/ lecithin/cyclohexane microemulsions by FT-IR spectroscopy. Journal of Colloid and Interface Science, 176(2), 281-285. http://dx.doi.org/10.1006/jcis.1995.9966. 9. Veiga-Santos, P., Oliveira, L. M., Cereda, M. P., & Scamparini, R. P. (2007). Sucrose and inverted sugar as plasticizer: effect on cassava starch-gelatin film mechanical properties, hydrophilicity and water activity. Food Chemistry, 103(2), 255-262. http:// dx.doi.org/10.1016/j.foodchem.2006.07.048. 10. Fakhouri, F. M., Costa, D., Yamashita, F., Martelli, S. M., Jesus, R. C., Alganer, K., Collares-Queiroz, F. P., & Innocentini-Mei, L. H. (2013). Comparative study of processing methods for starch/gelatin films. Carbohydrate Polymers, 95(2), 681-689. PMid:23648030. http://dx.doi.org/10.1016/j.carbpol.2013.03.027. 11. Fakhouri, F. M., Fontes, L. C. B., Innocentini-Mei, L. H., & Collares-Queiroz, F. P. (2009). Effect of fatty acid addition on the properties of biopolymer films based on lipophilic maize starch and gelatin. Stärke, 61(9), 528-536. http://dx.doi. org/10.1002/star.200800217. 12. Fakhouri, F. M., Maria Martelli, S., Canhadas Bertan, L., Yamashita, F., Innocentini-Mei, L. H., & Collares-Queiroz, F. P. (2012). Edible films made from blends of manioc starch and gelatin: influence of different types of plasticizer and different levels of macromolecules on their properties. LWT - Food Science and Technology, 49(1), 149-154. http://dx.doi. org/10.1016/j.lwt.2012.04.017. 13. Liu, X., Wang, Y., Zhang, N., Shanks, R. A., Liu, H., Tong, Z., Chen, L., & Yu, L. (2014). Morphology and phase composition of gelatin-starch blends. Chinese Journal of Polymer Science, 32(1), 108-114. http://dx.doi.org/10.1007/s10118-014-1377-1. 14. Soares, F. C., Yamashita, F., Müller, C. M. O., & Pires, A. T. N. (2013). Thermoplastic starch/poly(lactic acid) sheets coated with cross-linked chitosan. Polymer Testing, 32(1), 94-98. http://dx.doi.org/10.1016/j.polymertesting.2012.09.005. 15. Shirai, M. A., Grossmann, M. V. E., Mali, S., Yamashita, F., Garcia, P. S., & Müller, C. M. O. (2013). Development of biodegradable flexible films of starch and poly (lactic acid) plasticized with adipate or citrate esters. Carbohydrate Polymers, 92(1), 19-22. PMid:23218260. http://dx.doi.org/10.1016/j. carbpol.2012.09.038. 16. Guzman-Sielicka, A., Janik, H., & Sielicki, P. (2013). Proposal of new starch-blends composition quickly degradable in marine environment. Journal of Polymers and the Environment, 21(3), 802-806. http://dx.doi.org/10.1007/s10924-012-0558-7. 17. Shirai, M. A., Müller, C. M. O., Grossmann, M. V. E., & Yamashita, F. (2015). Adipate and citrate esters as plasticizers for poly(lactic acid)/thermoplastic starch sheets. Journal of Polymers and the Environment, 23(1), 54-61. http://dx.doi. org/10.1007/s10924-014-0680-9. 18. Müller, C. M. O., Laurindo, J. B., & Yamashita, F. (2011). Effect of nanoclay incorporation method on mechanical and Polímeros, 27(1), 27-34, 2017

water vapor barrier properties of starch-based films. Industrial Crops and Products, 33(3), 605-610. http://dx.doi.org/10.1016/j. indcrop.2010.12.021. 19. American Society for Testing and Material – ASTM. (1996). E96-00: standard test methods for water vapor transmission of materials. West Conshohocken: ASTM. 20. American Society for Testing and Material – ASTM. (2002). D882-02: standard test methods for tensile properties of thin plastic sheeting. West Conshohocken: ASTM. 21. Choi, Y. S., Hong, S. R., Lee, Y. M., Song, K. W., Park, M. H., & Nam, Y. S. (1999). Study on gelatin-containing artificial skin: I. Preparation and characteristics of novel gelatin-alginate sponge. Biomaterials, 20(5), 409-417. PMid:10204983. http:// dx.doi.org/10.1016/S0142-9612(98)00180-X. 22. Zhang, N., Liu, X., Yu, L., Shanks, R., Petinaks, E., & Liu, H. (2013). Phase composition and interface of starch-gelatin blends studied by synchrotron FTIR micro-spectroscopy. Carbohydrate Polymers, 95(2), 649-653. PMid:23648025. http://dx.doi.org/10.1016/j.carbpol.2013.03.045. 23. Bonilla, J., Atarés, L., Vargas, M., & Chiralt, A. (2013). Properties of wheat starch film-forming dispersions and films as affected by chitosan addition. Journal of Food Engineering, 114(3), 303-312. http://dx.doi.org/10.1016/j.jfoodeng.2012.08.005. 24. Chambi, H., & Grosso, C. (2006). Edible films produced with gelatin and casein cross-linked with transglutaminase. Food Research International, 39(4), 458-466. http://dx.doi. org/10.1016/j.foodres.2005.09.009. 25. Bertuzzi, M. A., Vidaurre, E. C., Armada, M., & Gottifredi, J. C. (2007). Water vapor permeability of edible starch based films. Journal of Food Engineering, 80(3), 972-978. http:// dx.doi.org/10.1016/j.jfoodeng.2006.07.016. 26. Karnnet, S., Potiyaraj, P., & Pimpan, V. (2005). Preparation and properties of biodegradable stearic acid-modified gelatin films. Polymer Degradation & Stability, 90(1), 106-110. http:// dx.doi.org/10.1016/j.polymdegradstab.2005.02.016. 27. Fiszman, S. M., Lluch, M. A., & Salvador, A. (1999). Effect of addition of gelatin on microstructure of acidic milk gels and yoghurt and on their rheological properties. International Dairy Journal, 9(12), 895-901. http://dx.doi.org/10.1016/ S0958-6946(00)00013-3. 28. Al-Hassan, A. A., & Norziah, M. H. (2012). Starch–gelatin edible films: Water vapor permeability and mechanical properties as affected by plasticizers. Food Hydrocolloids, 26(1), 108-117. http://dx.doi.org/10.1016/j.foodhyd.2011.04.015. 29. Jamshidian, M., Tehrany, E. A., Imran, M., Akhtar, M. J., Cleymand, F., & Desobry, S. (2012). Structural, mechanical and barrier properties of active PLA-antioxidant films. Journal of Food Engineering, 110(3), 380-389. http://dx.doi.org/10.1016/j. jfoodeng.2011.12.034. 30. Pereda, M., Ponce, A. G., Marcovich, R., Ruseckaite, A., & Martucci, J. F. (2011). Chitosan-gelatin composites and bi-layer films with potential antimicrobial activity. Food Hydrocolloids, 25(5), 1372-1381. http://dx.doi.org/10.1016/j. foodhyd.2011.01.001. 31. Turhan, K. N., & Şahbaz, F. (2004). Water vapor permeability, tensile properties and solubility of methylcellulose-based edible films. Journal of Food Engineering, 61(3), 459-466. http://dx.doi.org/10.1016/S0260-8774(03)00155-9. 32. Morillon, V., Debeaufort, F., Capelle, M., Blond, G., & Voilley, A. (2000). Influence of the physical state of water on the barrier properties of hydrophilic and hydrophobic films. Journal of Agricultural and Food Chemistry, 48(1), 11-16. PMid:10637042. http://dx.doi.org/10.1021/jf990809z. 33. Turhan, K. N., & Şahbaz, F. (2004). Water vapor permeability, tensile properties and solubility of methylcellulose-based 33

Pizzoli, A. P. O., Yamashita, F., Gonçalves, O. H., Shirai, M. A., & Leimann, F. V. edible films. Journal of Food Engineering, 61(3), 459-466. http://dx.doi.org/10.1016/S0260-8774(03)00155-9. 34. Abdillahi, H., Chabrat, E., Rouilly, A., & Rigal, L. (2013). Influence of citric acid on thermoplastic wheat flour/poly(lactic acid) blends. II. Barrier properties and water vapor sorption isotherms. Industrial Crops and Products, 50(1), 104-111. http://dx.doi.org/10.1016/j.indcrop.2013.06.028. 35. Godbillot, L., Dole, P., Joly, C., Roge, B., & Mathlouthi, M. (2006). Analysis of water binding in starch plasticized films. Food Chemistry, 96(3), 380-386. http://dx.doi.org/10.1016/j. foodchem.2005.02.054. 36. Müller, M. O. C., Pires, A. T. N., & Yamashita, F. (2012). Characterization of thermoplastic starch/poly(lactic acid) blends obtained by extrusion and thermopressing. Journal of the Brazilian Chemical Society, 23(3), 426-434. http://dx.doi. org/10.1590/S0103-50532012000300008. 37. Soares, F. C., Yamashita, F., Müller, C. M. O., & Pires, A. T. N. (2014). Effect of cooling and coating on thermoplastic starch/ poly(lactic acid) blend sheets. Polymer Testing, 33(1), 34-39. http://dx.doi.org/10.1016/j.polymertesting.2013.11.001. 38. Mali, S., Sakanaka, L. S., Yamashita, F., & Grossmann, M. V. E. (2005). Water sorption and mechanical properties of

cassava starch films and their relation to plasticizing effect. Carbohydrate Polymers, 60(3), 283-289. http://dx.doi. org/10.1016/j.carbpol.2005.01.003. 39. Brandelero, H., Grossmann, M. V., Yamashita, F. (2013). Hidrofilicidade de filmes de amido/poli(butileno adipato-cotereftalato) (pbat) adicionados de tween 80 e óleo de soja. Polímeros: Ciência e Tecnologia, 23(2), 270-275. http://dx.doi. org/10.1590/S0104-14282013005000011. 40. Müller, C. M. O., Laurindo, J. B., & Yamashita, F. (2012). Composites of thermoplastic starch and nanoclays produced by extrusion and thermopressing. Carbohydrate Polymers, 89(2), 504-510. PMid:24750751. http://dx.doi.org/10.1016/j. carbpol.2012.03.035. 41. Zhao, X., Liu, W., & Yao, K. (2006). Preparation and characterization of biocompatible poly (L -lactic acid)/ gelatin blend membrane. Journal of Applied Polymer Science, 101(1), 269-276. http://dx.doi.org/10.1002/app.23292. Received: May 19, 2015 Revised: Mar. 17, 2016 Accepted: Apr. 28, 2016

Polímeros, 27(1), 27-34, 2017

http://dx.doi.org/10.1590/0104-1428.2298

Use of pH-thermosensitive hydrogels for nickel ion removal and recovery Álvarez Casillas Cesar Andrés1 and Cortés Ortega Jorge Alberto1* 1

Departamento de Química, Universidad de Guadalajara, Guadalajara, Jalisco, México *jorcortes@hotmail.com

Abstract N-isopropylacrylamide / itaconic acid hydrogels were prepared in this study, varying the proportion of itaconic acid in the pregel mixture. The kinetics of the hydrogel swelling at 4 °C was determined, obtaining the kintetic patameters, in accordance with the second-order kinetic model. Similarly, the capacity to absorb water in terms of temperature was determined, along with the transition temperature of the samples. The influence of temperature on the capacity of the hidrogels to absorb nickel from aqueous solutions at 5% of NiCl2 and its subsequent recuperation was determined. Keywords: hydrogels, metals remover, recovery metals, pH-sensitive, thermosensitive.

1. Introduction Over recent years, the remediation of water contaminated with heavy metals has been carried out through the application of various methods[1-6], using both chemical and physical processes. Various studies have analyzed the possibility of using hydrogels for the retention and removal of heavy metals from contaminated water[7-11], reporting good results in terms of the removal process. However, there is no information available on either a process for recovering heavy metals from used polymer matrixes or the possible reuse of these matrixes. The costs of the remediation of contaminated water can be reduced through matrix reuse, provided that the material used for the removal of the metals is, as with the metals recovered during various inductrial processes. Poly(N-isopropylacrylamide) (NIPA) is a thermoresponsive polymer that has a lower critical solution temperature (LCST) of 32 °C[12] in water . Below the LCST[13,14], polyNIPA and water are completely miscible; above the LCST, phase separation occurs. As a consequence of this transition, crosslinked polyNIPA hydrogels exhibit a volume phase transition at around to this value. This temperature is known as the volume phase transition temperature (TTVP), which is characterized by a sudden decrease in the capacity to absorb water and occurs at around 32 °C. The addition of acid groups to the NIPA polymer chain has been found to increase both the capacity to absorb water and raise the transition temperature[15]. Therefore, this study aims to analyze the possibility of using thermosensitive and pH‑sensitive hydrogels for the removal and recovery of nickel ions from aqueous solutions, by placing the hydrogels in aqueous nickel solutions at below transition temperature, and, once equilibrium is attained, raising the temperature to above TTVP, in order to recover the nickel removed.

2. Materials and Methods 2.1 Materials N-isopropylacrylamide (NIPA) with a purity of 99%, from Aldrich. Itaconic acid (IA) with a purity of 99%, from Aldrich. The initiator used was potassium peroxydisulfate (K2S2O8),

Polímeros, 27(1), 35-40, 2017

at 99% purity, from Aldrich. N,N, methylenebisacrylamide (NMBA) was used as a cross-linking agent with a purity of 99% (TCI). The accelerator used was N, N, N, N, tetramethylenediamine (TMEDA) from Tokyo Kasei. Bidistilled water was used in all the experiments.

2.2 Synthesis of hydrogels Aqueous monomer solutions to 10 percent by mass (NIPA and IA, modifying the proportion of acid) were prepared in 20 mL glass containers in nitrogen atmosphera, to which was added 1% cross-linking agent in relation to the monomers, as well as NaOH solution for the complete neutralization (stoichiometric ratio) of the acid content of each sample. Once the solution had been formed, nitrogen was bubbling for one minute, 2% of initiator and 3% of accelerator was added in relation to the monomers (ensuring that all samples had a final monomer/water ratio of 10/90), the container was closed and sealed with parafilm, and left to react for 24 hours at 25 °C. The diagram of the polymerization reaction is shown in Figure 1

2.3 Cleaning Once the reaction had finished, the samples were left to dry, firstly for three days at room temperature and then in a vacuum oven at 40 °C. This was carried out to avoid rupturing the samples due to sudden drying. The dry samples are placed in distilled water at 25 °C and, once they have attained equilibrium swelling (at around 7 days)[11], they are placed in a temperature controlled bath at 40 °C for 2 days to remove from the samples the material that had not been added to the network. The water was then changed and the process repeated, until no signs were observed of turbitity in the wash water at 40 °C. The turbidity of wash water is an indicator of the presence of monomers and oligomers that contain NIPA. Once the process had been completed, the samples were left to dry, as described above, with their weight before and after the cleaning process used to determine the conversion level during the polimerization reaction.

O O O O O O O O O O O O O O O O

Andrés, A. C. C., & Alberto, C. O. J.

Figure 1. Scheme of the polymerization reaction of the synthesized hydrogels.

2.4 Swelling kinetics To determine the swelling kinetics, the dry samples are weighed and placed in water at 4 °C. The samples are weighed at different times, with their surface dried using absorbent paper. The swelling kinetics are determined using the difference between the weight of the dry and swollen sample. This process is repeated until the weight of the swollen sample does not change, which determines the equilibrium swelling (H∞), to the temperature of the experiment, with the swelling determined by means of the following equation[16]: = H

mt − m0 weight of hydrogel − weight of xerogel = m0 weight of xerogel

(1)

2.5 Equilibrium swelling in relation to temperature

placed in a controlled temperature bath at 30 °C, upon which the process described above is repeated, the temperature of the bath is raised to 40 °C, and the quantity of nickel removed is determined. This process is repeated for solutions with a pH of 3 (which is regulated by adding HCl) and a 5% NiCl2 weight. All experiments were carried out in systems without agitation.

3. Results and Discussions Swelling kinetics has been studied by various authors using the following equation[17-19]: dW 2 = K (W∞ − W ) dt

(2)

Where:

Once the samples attained the equilibrium swelling at mt − m0 weight of hydrogel − weight of xerogel (3) = W = 4 °C, the samples were placed in a controlled temperature mt weight od hydrogel bath at 30 °C and, once they had reached the equilibrium swelling (which was determined using the process described Which, on being integrated with W = 0 in t = 0, the following above), the temperature was raised. The temperatures used in was obtained: this study ranged from 4 to 46 °C. The transition temperature KW∞2t (TTVP) was determined using the equilibrium swelling curves W = (4) versus the temperature, measuring the change in the water 1 + KW∞t content in terms of equilibrium and temperature. On combining Equations 1-3, the following was obtained:

2.6 Nickel absorbing capacity

The xerogels are placed in 5% nickel (II) chloride aqueous solutions at 4 °C until the equilibrium swelling is reached (the time is determined by the swelling kinetics obtained previously) (a ratio of 300 grams of solution was used per gram of xerogel, for all experiments). Once this occurred, the samples were withdrawn from the medium and weighed, and a sample then taken from the residual solution to determine the concentration of Ni+2. Using the Genesys 10 U.V. spectrophotometer from the Thermoelectron Corporation, the samples are returned to the solution and 36

t ( H + 1) = H

( H ∞ + 1)2 + H ∞ + 1 t KH ∞2

H∞

(5)

Which is rearranged as: t ( H + 1) = mt + b H

(6)

With which H can be obtained in relation to time, with the following equation: Polímeros, 27(1), 35-40, 2017

Use of pH-thermosensitive hydrogels for nickel ion removal and recovery

t t ( m − 1) + b

(7)

With this expression, it is possible to determine the kinetic parameters with which the speed of water absorption and equilibrium swelling are obtained. Figure 2 shows the comparison between the experimental data and the theoretical prediction obtained with Equation 7, in which a good description of the experimental data is observed, thus confirming the standard deviation between the experimental data and the theoretical prediction, with said deviation presented in Table 1. Transition temperature is determined by calculating the change in equilibrium swelling in relation to the temperature, with the temperature at which this change acquires the absolute maximum value being determined as the transition temperature (Table s1 of Supplementary material). The kinetic parameters and the equilibrium swelling are presented in Table 1. It can be observed that, on increasing the concentration of itaconic acid, the capacity of the hydrogels to absorb water increases (Figure 3), The yield obtained from the polymerization reaction of the samples was greater than 98% in all cases, as does the transition temperature of the samples (Figure 4 and Figure s1 of Supplementary material). This is due to the increase of the carboxylic groups, which, in turn, increases the hydrophile groups, thus absorbing a greater amount of water[15,19,20]. Table 2 shows the decrease in the capacity of the hydrogels to absorb nickel solutions in relation to the capacity to absorb water, which is explained by the interactions of the nickel ions with the carboxylic groups, avoiding the association between the water molecules and these groups, reducing the capacity to absorb solution from the hydrogels. Similar results have been previously reported[21], from which it is possible to speculate on the benefits of adding hydrogels, instead of xerogels, to the nickel solutions in order to determine whether the capacity to absorb ions increases. In the hydrogels, the polymer chains are found to be extended, while the xerogel chains expand. The risk of elongating the chains to their maximum length could be reduced or avoided, depending on the quantity of solution that is being penetrated and the presence of nickel ions. On the other hand, it is possible to establish the nickel removal process depending on the prevailing climate in the area from where the water to be decontaminated is found, with the hydrogels (or xerogels) placed at low (lower than 10 °C) or medium temperatures (higher than 15 °C). For the recovery of the nickel removed, the loaded gels can be placed at greater temperatures for those hydrogels with a lower itaconic acid content in the matrix, while those hydrogels that have a greater proportion

of itaconic acid are placed in pH 3 solutions and greater temperatures. In this regard, an analysis must be undertaken of the minimum amount of solution with a pH of between 3 and 1 (taking into account of the possible degradation of material) that must be added to provoke the deswelling of the hydrogel and the amount of recovered nickel. It

Figure 2. Comparison between experimental data and the second order swelling kinetic model for the NIPA/IA 99/1 sample with the experimental data (●), and the theoretical prediction (-).

Figure 3. Swelling Kinetics of hydrogels depending on the proportion of NIPA/IA 99/1 (○), 98/2 (●) 97/3(□) 96/4 (+) y 95/5 (■).

Table 1. Equilibrium swelling and kinetic data of the hidrogeles synthized at 4 °C. NIPA/IA % weight Slope (m) H∞ (gwater /gxerogel) Dev std Intercept (b) K ghydrogel/(gwater-min) Transition temperature

Polímeros, 27(1), 35-40, 2017

99/1 1.017 59.140 0.582 38.374 0.027 38 °C

98/2 1.010 102.260 0.770 16.810 0.061 42 °C

97/3 1.007 142.644 2.248 9.426 0.108 44 °C

96/4 1.006 161.784 2.916 9.780 0.104 44 °C

95/5 1.007 150.042 2.837 7.331 0.138 44 °C

Andrés, A. C. C., & Alberto, C. O. J. has been observed that when the pH of the solution is regulated (adding HCl to obtain a pH of 3), the capacity of the hydrogels to absorb solution is seen to be reduced, owing to the fact that the acid groups are not ionized and, with this, the availability of spaces in the network

for the capture of water and metal ions is also reduced (Table 2). So it is known that increasing the temperature increases the solubility and therefore Ni2+ prefers to be in water and not in the hydrogel. Table 3 shows that on reducing the concentration of nickel ions in the solution, the capacity to absorb nickel solution is increased, as does the percentage of nickel ion removal, which corroborates that the presence of nickel ions reduces the active spaces for water absorbtion. With this, the space is opened to develop the process of removing metal ions, firstly by placing the hydrogels in the nickel ion solutions. Then, when the hydrogels become saturated with the metal, they are removed from the medium and placed in solutions with a pH of 3 or more, and/or at temperatures greater than 40 °C, at which the metal ions will be displaced from the hydrogel. The use of this procedure was intended to obtain metal saturated solutions and to explore the possibility of reusing the hydrogels. To determine the magnitude of the degradation of the hydrogels, all the samples were placed in the solutions at 5% of NiCl2 at 4ºC and, when they attained equilibrium swelling they were placed in a temperature controlled bath at 40 °C on four separate occasions. In all cases, similar swelling levels and Ni+2 ion absorbtion were obtained, with the loss of hydrogel mass determined in each experiment, which was less than 1% in all cases.

Figure 4. Equilibrium swelling in relation to temperature depending on the proportion of NIPA/IA 100/0 (○), 99/1 (●), 98/2 (□), 97/3 (■), 96/4 (+) and 95/5 (▲).

Table 2. Equilibrium swelling, mass of Ni+2 absorbed from the hydrogels when they are placed in 5% NiCl2 solutions depending on the proportion of NIPA/IA at different temperatures and pH levels. NIPA/IA % weight W∞ water 4 °C W∞ water 30 °C W∞ water 40 °C W∞ Ni 4 °C W∞ Ni 30 °C W∞ Ni 40 °C gNi/gxerogel 4 °C gNi/gxerogel 30 °C gNi/gxerogel 40 °C W∞ Ni pH 3 4 °C W∞ Ni pH 3 30 °C W∞ Ni pH 3 40 °C gNi/gxerogel 4 °C gNi/gxerogel 30 °C gNi/gxerogel 40 °C

100/0 25.235 13.165 0.3950 17.108 3.382 0.860 0.318 0.175 0.05 -

99/1 59.140 27.35 0.60 29.20 10.11 1.41 0.645 0.210 0.069 27.708 1.580 0.622 0.278 0.351 0.161

95/5 150.042 130.820 110.81 32.70 19.50 3.13 0.77 0.55 0.139 24.409 6.958 2.120 0.844 0.487 0.151

90/10 306.792 249.482 207.680 27.40 15.69 1.41 0.88 0.41 0.20 26.593 12.927 1.077 1.358 0.331 0.159

85/15 393.695 265.648 227.276 36.47 28.57 4.52 1.201 0.811 0.608 41.007 27.4334 2.566 0.981 0.677 0.142

Table 3. Equilibrium swelling of the hydrogels for the NIPA/IA 85/15 hydrogel, placed in NiCl2 solutions according to the salt concentration. Concentration NiCl2

gNi/gxerogel

3.000% 2.500% 1.000% 0.500% 0.350% 0.200% 0.100%

30.956 30.973 32.092 33.551 52.757 57.767 71.048

1.429 1.259 0.500 0.252 0.215 0.122 0.090

22.388 22.770 24.513 26.614 42.153 48.763 57.933

1.205 1.152 0.417 0.213 0.142 0.081 0.028

12.012 13.476 17.293 19.241 31.578 38.611 48.558

0.380 0.353 0.138 0.050 0.018 0.007 <0.005

4 °C

30 °C

40 °C

Polímeros, 27(1), 35-40, 2017

Use of pH-thermosensitive hydrogels for nickel ion removal and recovery

4. Conclusions It was found that the addition if itaconic IA increased the capacity to absorb water. The capacity to absorb Ni ions was determined using aqueous NiCl2 solutions with a composition of 5%. In all samples, the greatest capacity to absorb nickel was found at 4 °C. The possible use of these materials was determined for the recovery of nickel from industrial effluent and its recovery as a raw material, on removing the nickel from the solution at low temperatures and subsequently subjecting the saturated hydrogel to high temperatures. The low temperature depends on the proportion of itaconic acid in the sample. We can, therefore, establish that a low temperature would be 4 °C for the hydrogels with a 1 to 4% acid content. A temperature of 30 °C would be low for those hydrogels with a 5 to 15% acid content, these materials are left at this temperature to absorb the metal ions, with the loaded and saturated hydrogels then withdrawn from the medium and placed in the high temperatures, which would be 40 °C for the hydrogels with a low acid content and 46 °C for those hydrogels with a higher acid content. Small quantities of acid solution are added in order to obtain a pH value of between 1 and 3 in the medium.

5. References 1. Matlock, M. M., Howerton, B. S., & Atwood, D. A. (2002). Chemical precipitation of heavy metals from acid mine drainage. Water Research, 36(19), 4757-4764. PMid:12448518. http:// dx.doi.org/10.1016/S0043-1354(02)00149-5. 2. Ito, A. (2000). Removal of heavy metals from anaerobically digested sewage sludge by a new chemical method using ferric sulfate. Water Research, 34(3), 751-758. http://dx.doi. org/10.1016/S0043-1354(99)00215-8. 3. Lefers, J. B., van den Broeke, W. F., Venderbosch, H. W., de Niet, J., & Kettelarij, A. (1987). Heavy metal removal from waste water from wet lime(stone)–gypsum flue gas desulfurization plants. Water Research, 21(11), 1345-1354. http://dx.doi.org/10.1016/0043-1354(87)90008-X. 4. Cerón Neculpan, M. S. (2006). Remoción de cadmio presente en aguas de desecho de la industria metal-mecánica mediante membranas líquidas emulsificadas en extractores del tipo estanque agitado en proceso batch. (Tesis memoria para optar al titulo de químico). Universidad de Chile, Chile. 5. Huang, P., Ye, Z., Xie, W., Chen, Q., Li, J., Xu, Z., & Yao, M. (2013). Rapid magnetic removal of aqueous heavy metals and their relevant mechanisms using nanoscale zero valent iron (nZVI) particles. Water Research, 12(47), 4050-4058. PMid:23566331. http://dx.doi.org/10.1016/j.watres.2013.01.054. 6. Xiang, L., Chan, L. C., & Wong, J. W. (2000). Removal of heavy metals from anaerobically digested sewage sludge by isolated indigenous iron-oxidizing bacteria. Chemosphere, 41(1–2), 283-287. PMid:10819212. http://dx.doi.org/10.1016/ S0045-6535(99)00422-1. 7. Wu, Q., & Tian, P. (2008). Adsorption of Cu+2 Ions with Poly (N-isopropylacrylamide-co-methacrylic acid) Micro/ Nanoparticles. Journal of Applied Polymer Science, 109(6), 3470-3476. http://dx.doi.org/10.1002/app.28450. 8. Bekiari, V., Sotiropoulou, M., Bokias, G., & Lianos, P. (2008). Use of poly(N,N-dimethylacrylamide-co-sodium acrylate) hydrogel to extract cationic dyes and metals from water. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 312(2–3), 214-218. http://dx.doi.org/10.1016/j. colsurfa.2007.06.053. Polímeros, 27(1), 35-40, 2017

9. Ozay, O., Ekici, S., Baran, Y., Aktas, N., & Sahiner, N. (2009). Removal of toxic metal ions with magnetic hydrogels. Water Research, 43(17), 4403-4411. PMid:19625066. http://dx.doi. org/10.1016/j.watres.2009.06.058. 10. El Halah, A., Contreras, J., & López, F. (2011). Síntesis y aplicaciones de hidrogeles superabsorbentes de poli(acrilamidaco-monoitaconato de metoxietilo). Revista Latinoamericana de Metalurgia y Materiales, S3, 30-31. 11. Yildiz, U., Kemik, O. F., & Hazer, B. (2010). The removal of heavy metal ions from aqueous solutions by novel pHsensitive hydrogels. Journal of Hazardous Materials, 183(13), 521-532. PMid:20709450. http://dx.doi.org/10.1016/j. jhazmat.2010.07.055. 12. Shibayama, M., Motonaga, T., & Nakamoto, C. (2001). Preparation pressure dependence of structure Inhomogeneities and dynamic fluctuations in poly(N-isopropylacrylamide). Gels Macromolecules, 34(4), 911-911. http://dx.doi.org/10.1021/ ma001372v. 13. Biswas, C. S., Sulu, E., Hazer, B. (2014). Effect of the composition of methanol−water mixtures on tacticity of poly (N-ethylacrylamide) Gel Journal of Applied Polymer Science, 132, 41668-41678. http://dx.doi.org/10.1002/app.41668. 14. Biswas, C. S., & Hazer, B. (2015). Synthesis and characterization of stereoregular poly(N-ethylacrylamide) hydrogel by using Y(OTf)3 Lewis acid. Colloid & Polymer Science, 293(1), 143-152. http://dx.doi.org/10.1007/s00396-014-3399-1. 15. Cortés, J. A., Mendizábal, E., & Katime, I. (2008). Effect of comonomer type and concentration on the equilibrium swelling and volume phase transition temperature of N‐ isopropylacrylamide‐based hydrogels. Journal of Applied Polymer Science, 108(3), 1792-1796. http://dx.doi.org/10.1002/ app.27632. 16. Ortega, J. A. C. (2013). Synthesis of thermosensitive hydrogels of poly(N-isopropylacrylamide)-co-poly(N,N-dimethylacrylamide). Polímeros: Ciência e Tecnologia, 23(2), 189-195. http://dx.doi. org/10.4322/polimeros.2013.080. 17. Yıldız, U., & Hazer, B. (2000). Dispersion redox copolymerization of methyl methacrylate with macromonomeric azoinitiator as a macro crosslinker. Polymer, 41(2), 539-544. http://dx.doi. org/10.1016/S0032-3861(99)00217-7. 18. Fernandez, V. V. A., Tepale, N., Sánchez-Díaz, J. C., Mendizábal, E., Puig, J. E., & Soltero, J. F. A. (2006). Thermoresponsive nanostructured poly (N-isopropylacrylamide) hydrogels made via inverse microemulsion polymerization. Colloid & Polymer Science, 284(4), 387-395. http://dx.doi.org/10.1007/s00396005-1395-1. 19. Katime, I., Velada, J. L., Novoa, R., Díaz de Apodaca, E., Puig, J., & Mendizabal, E. (1996). Swelling kinetics of Poly(Acrilamide)/Poly(mono-n-alkyl Itaconates) hydrogels. Polymer International, 40(4), 281-286. http://dx.doi.org/10.1002/ (SICI)1097-0126(199608)40:4<281::AID-PI555>3.0.CO;2-H. 20. Bajpai, S. K., & Johnson, S. (2007). Removal of Ni+2 ions from aqueous solution by sorption into Poly(Acrylamide-co-Sodium Acrylate) hydrogels. Journal of Macromolecular Science, Part A: Pure and Applied Chemistry, 44(3), 291-297. http://dx.doi. org/10.1080/10601320601077328. 21. Rojas de Gáscue, B., Ramírez, M., Prin, J.L., Torres, C., Bejarano, L., Villarroel, H., Rojas L., Murillo, M., & Katime, I. (2010). Hidrogeles de acrilamida/ácido acrilico y de acrilamida/ poli(ácido acrilico): estudio de su capacidad de remediación en efluentes industriales. Revista Latinoamericana de Metalurgia y Materiales, 30(1), 28-39. Received: Aug. 21, 2015 Revised: Mar. 17, 2016 Accepted: May 06, 2016 39

Andrés, A. C. C., & Alberto, C. O. J.

Supplementary Material Supplementary material accompanies this paper. Figure s1. Fraction of water released from the hydrogels in relation to temperature as a function of the proportion of NIPA/IA 100/0 (○), 99/1 (●), 98/2 (□), 97/3 (■), 96/4 (+) and 95/5 (▲), taking the equilibrium swelling of 4ºC as a point of reference. Table s1. Change in the fraction released in relation to temperature, which was calculated using (F32ºC - F30ºC) / (32ºC -30ºC) 32ºC and in the same manner for the rest of the temperatures. This material is available as part of the online article from www.scielo.br/po

Polímeros, 27(1), 35-40, 2017

http://dx.doi.org/10.1590/0104-1428.2354

Effect of shrimp shells milling on the molar mass of chitosan Helton José Alves1*, Maristela Furman1, Cristie Luis Kugelmeier1, Clayton Rodrigues de Oliveira2, Vanessa Rossato Bach1, Karine Natani Lupatini1, Andressa Caroline Neves1 and Mabel Karina Arantes1 Laboratory of Catalysis and Biofuel Production – LabCatProBio, Universidade Federal do Paraná – UFPR, Palotina, PR, Brazil 2 Department of Chemistry – DQ, Universidade Federal de São Carlos – UFSCar, São Carlos, SP, Brazil

*helquimica@gmail.com

Abstract Shrimp shells are a raw material rich in chitin, a precursor of chitosan biopolymer. The variables of processing (demineralization, deproteination and deacetylation) can be manipulated to determine the main characteristics of chitosan, the degree of deacetylation (DD), and average molar mass. This study evaluated the influence of one of the unit operations of shrimp shell physical processing, the milling, on the final product characteristic, chitosan. After different milling conditions, the raw material was subjected to standard chemical processing for chitin extraction, followed by deacetylation to obtain chitosan, which is characterized by 1H NMR, SEM, XRD, N2 physisorption (BET) and viscometry. The results indicated that the milling time of the raw material can be manipulated to increase the material depolymerization, significantly influencing the molecular weight reduction of chitosan a desirable feature for many applications of this biopolymer, and usually obtained by complex chemical and enzymatic methods. Keywords: biopolymers, chitosan, degree of polymerization (DP), viscosity, particle size distribution.

1. Introduction When researching the reduction of the molecular weight of polymers or biopolymers, many studies seek to employ chemical and enzymatic methods, but less attention is given to physical methods. In a previous work, our group evaluated the influence of different drying routes over depolymerization and properties of the biopolymer chitosan (2-amino-2-deoxy-D-glucose and 2-acetamido-2-deoxyD-glucose copolymer)[1]. The results indicated that the physical drying process may decisively affect the molar mass of the biopolymer, and the drying of chitosan with supercritical CO2 (SAS) was reduced more than 10 times its molar mass (from 35.3 to 3.0 kDa). Although changes were not observed in the chemical properties of the material, drying with supercritical CO2 caused an increased degree of crystallinity and crystallite size, as well as a significant change in the textural properties, resulting in values of specific area and much higher pore volume. In the case of depolymerization of chitini and chitosan, the most recent studies described in the literature suggest the following methods: *

- Chitosan: oxidative degradation by catalysts with peroxide active sites (for example, W (O2)) leading to a depolymerization rate higher than 90% [2]; depolymerization by a plasma treatment solution applied to the metal‑chitosan complex allowed to obtain oligomers (in order of 103 Da); however, the technical i *

Chitin: a linear polymer predominantly made of (1-4)-ᵦ linked 2-acetamido2-deoxy-D-glucose, estimated to be the second most abundant natural polymer after cellulose. It is isolated from shrimp waste by demineralization and deproteination of shells.

Polímeros, 27(1), 41-47, 2017

efficiency varied with the complexed metal[3]. The use of enzymes such as lysozyme or celullase caused depolymerization, but the enzyme activity varies even with the deacetylation degree[4]. Ultraviolet-irradiated oxygen treatment applied to a chitosan solution (slightly acidic) gave the molar mass reduction up to seven times after 300 min of reaction[5]; use of gamma radiation, proposed as depolymerization chitosan mechanism; however, the darkening of the material is a problem inherent to this method, leading to studies that can inhibit an occurrence such as pH control and O2[6]; - Chitin: gamma radiation and combining enzymatic hydrolysis where the partial depolymerization of chitin through radiation promotes accessibility to enzymatic peroxide active sites so that the following processes lead to obtaining oligomers and even monomers[7]. Using high concentrations of inorganic acid (HCl 3-12M) under average temperatures allowed chitin degradation in short term experiments[8].

It is known that the average molar mass of polymers in general can vary according to the processing history. Besides drying, milling is an important unit operation to consider. The degradation of the polymer chain can be caused by the energy supplied by mechanically milling the material as the primary response of a polymeric material to an external mechanical force is the relaxation of the chain. Another important type of response to mechanical force of the milling is the polymer chain scission[9], which can also be obtained by means other than mechanical, such as contact with thermal energy, among others[10,11].

O O O O O O O O O O O O O O O O O

Alves, H. J., Furman, M., Kugelmeier, C. L., Oliveira, C. R., Bach, V. R., Lupatini, K. N., Neves, A. C., & Arantes, M. K. When a particulate solid material is milled in a ball mill, for example, there are mainly two types of combined actions that are responsible for the particles’ shearing of the material: impact and abrasion. In the first case, the fracture occurs when force is applied quickly and the intensity is greater than the particle strength in areas (area A in Figure 1) where the grinding elements (balls/beads) roll down by force of gravity during the rotation of the pitcher, forming a kind of cascade. The second mechanism is the attrition existing between the grinding elements and the particles of the material during the movement of the balls on the other in concentric layers (Zone B in Figure 1), resulting in small fractures. When reaching the critical speed the balls can achieve the highest point of the mill without detaching the wall. Above the critical velocity, it is possible to observe the cataract effect, as shown in Zone C of Figure 1[12]. Computer simulations that considered the effect of milling time on the split of the polymer chains, taking into account the bonds around it, were studied by Cook and Mercer[13]. By using experimental data from van der Hoff and Chopra in their deduced equations, the author found that the milling time is the most important parameter that influences the average molar mass of the polymer during this unit operation. Its kinetic study showed that the increase in milling time causes an increase in the number of disruptions in the polymer chains, which is proven by the reduction of the average molar mass[9]. In research performed by Delezuk et al.[14] with the β chitin, ultrasound irradiation was used to promote chitin deacetylation starting from the assumption that when macromolecule solutions are exposed to ultrasound irradiation, their viscosity decreases as a result of depolymerization processes. In the case of the production of chitosan, depolymerization increases the reactive sites accessibility on the polymer chain, increasing the reaction efficiency. In a later work with β chitin[15], a selection of particle size of the material was performed, and each fraction obtained was evaluated as to molar mass and average degree of chitin deacetylation (DD), leading to the understanding that these two parameters suffer changes during grinding, and to the observation that as the particle size gets smaller, the crystallinity and molar mass of chitin are smaller. In this work, we evaluated the effect of another physical stage of obtaining chitosan process, the milling, over their molar mass. However, unlike other studies assessing the influence of grinding on the molecular weight of a polymer, compared with the original, in this case, the polymer

is submitted to chemical transformation to become the product of interest and then the evaluation is carried out. It is to show how some milling parameters of shrimp shells, raw materials for the production of biopolymer chitosan, interfere with the viscosimetric molar mass of the final product of chemical processing—the chitosan. Besides the milling time, another variable selected for this initial study was the “grinding elements size” that is added into the ball mill, being fixed the other variables encountered during the material processing.

2. Materials and Methods 2.1 Cleaning of shrimp shells The shrimp shells used in this study were obtained after the harvest of marine shrimp from the Parana State coast (Brazil). The shells remained in the freezer at –10 °C until use. They were washed in running water to remove the excess organic material (meat, eggs, etc.) and dried at 60 °C for 24 h.

2.2 Milling After complete drying, the shells were ground for 60 seconds in a blender, and then fractionated to compose six samples (M1, M2, M3, M4, M5, M6) to be ground in different ways for this work. For milling, a laboratory ball mill (Solab) was used, with the rotation of 300 rpm, composed by a jar of 14 cm in diameter and 20 cm in length, with the ball charge fixed at approximately 1400 g and the sample mass to be milled around 30 g. There were three different milling times (30, 90 and 180 min), and two compositions of different sizes of porcelain balls (Composition A: 100% wt. of balls with 20 mm; Composition B: 70% wt. of balls with 20 mm + 30% wt. of balls with 15 mm) were chosen. Thus, each time, two compositions were tested in different balls, as can be seen in Table 1. After the grinding stage each powdery sample was sieved individually using a series of three stacked sieves with openings of 106, 63 and 43 µm, and then shaken in a sieve shaker for 10 min for the granulometry selection.

2.3 Chitin extraction The samples were submitted to a demineralization process of three washes (15 min each) with HCl solution of 0.55 mol.L-1, under agitation. After rinsing them to neutral pH, we proceeded to deproteination with three washes (20 min each) with NaOH solution 0,3 mol.L-1, under agitation and 80 °C, followed by washing to neutral pH.

2.4 Obtaining chitosan After deproteination, the chitin samples were still wet, subjected to solutions of NaOH 60% (w.V-1) under stirring of 125 rpm in a rotary incubator at 27-30 °C for 172 h. Then the material was washed with distilled water under vacuum filtration to neutral pH and dried in an electric oven at 60 °C for 24 h, finally obtaining the chitosan.

2.5 Chitosan characterization Figure 1. A schematic diagram of a tubular ball mill. The balls compose the milling elements and drive rollers help to rotate the milling chamber. Adapted from Loh et al.[12]. 42

Absolute density and particle size distribution (PSD): the absolute density of chitosan (sample M1) was determined by helium pycnometry (Quantachrome Ultrapycnometer Polímeros, 27(1), 41-47, 2017

Effect of shrimp shells milling on the molar mass of chitosan Table 1. Milling conditions and physicochemical properties of samples.

Sample

Time (min)

M1 M2 M3 M4 M5 M6

30 30 90 90 180 180

SHRIMP SHELL Milling Parameters Spheresb Spheresb 20 mm (% wt.) 100 70 100 70 100 70

15 mm (% wt.) 30 30 30

Particles

Crystallites

Ø > 106 µm (%)

D50 (µm)c

(Å)

78 81 79 50 54 34

2.50 2.40 2.20 2.00 1.75 1.70

40.1 41.4 41.8 42.3 43.2 42.8

Dap d f

CHITOSAN Textural Deacetylation

Viscositya

Specific area (m2.g-1)

DD (%)

[η] (mL.g-1)

(kDa)

6.8 9.1 11.4 11.6 11.9 11.2

79 80 80 79 78 77

1600 ± 43 1469 ± 31 1061 ± 28 1014 ± 22 589 ± 10 864 ± 14

502 436 281 268 131 217

MV e

K= 0.074 and α = 0.76 (solvent 0.3M HAc / 0.2M NaAc, at 25 °C)[16]; b The total charge of balls was 1390 ± 2.0 g; c The absolute density value determined by Helium pycnometry, and used in X-ray sedimentometry is 1.5454 g.cm-1; d The calculation was based on the data of the main peak at 2θ = 19.350; e Calculating MV was determined using the average value of [η]; f For M1, was chosen the crystalline peak of greatest intensity 2θ ≈ 100 to apply the Scherrer’s equation. a

1000), to be used in determining the PSD. Chitosan samples (M1 to M6) were subjected to the X-ray sedimentometry test (Sedigraph Model 5000 D, Micromeritics) to obtain the PSDs. Specific area (BET): The analyses of N2 adsorption (physisorption) were performed at a temperature of –196 °C (Quantachrome Co. Nova-2000). The samples were previously treated at 150 °C for 4 hours. The specific area was determined by BET equation (Brunauer, Emmett and Teller), using p/p0 ≤ 0.3[17]. X-ray diffractometry: X-ray diffractometry was performed at a range of 5° < 2Ө < 40°, with CuKα radiation (λ = 1.54 Å, 45 kV, 25 mA) and a speed of 1o.min-1, in a Rigaku Geingerflex diffractometer[18], allowing the evaluation of the crystallinity of the chitosan samples and also the diameter of the crystallites, by applying Scherrer’s equation (Equation 1)[19]: Dap =

Kλ (1) β0 cos θ

where, Dap is the average diameter of the crystallite (Å) in a perpendicular direction to the plane (110); β0 corresponds to the width of the peak referring to the main signal of the crystalline regions, and to the intensity of half-height (radians); K is a constant; θ corresponds to half of Bragg’s angle of the most intense signal (radians); and λ is the wavelength of radiation employed (Å). Molar mass viscosimetric: Molar mass viscosimetric was obtained as the intrinsic viscosity chitosan solutions (in 0.3 M HAc/0.2 M NaAc) using a Ubbelohde Dilution Viscosimeter (Cannon Instrument Co., USA) that has a capillary size of 0.44 mm in a water bath at 25 °C. The relation between the intrinsic viscosity, [η], and the average viscosimetric molecular weight of the polymer, MV, is established by the α equation of Mark-Houwink-Sakurada ([ η] =K M V ) where K and α are constants for a given polymer-solvent system, which in the case of chitosan, varies according to the degree of acetylation (DA)[16]. Chitin extraction was confirmed by determining the rates of demineralization and deproteination, and the average degree of acetylation of chitosan was obtained by 1H NMR, Polímeros, 27(1), 41-47, 2017

using a Bruker Avance III spectrometer and 9.4 Tesla, in the following experimental conditions: 400 MHz for hydrogen frequency (SWH), experimental temperature of 323 K, 16 scans (ns), wait time of 10 s (d1), acquisition time of 6.83 s (aq), and 65,536 data points (td). The %DA and %DD were calculated as described by Santos et al.[20].

3. Results and Discussions 3.1 Particle size distribution of shrimp shells after milling Figure 2 shows the particle size distribution for shrimp shells after milling under different times. Note that as the milling time increased, there was an overall significant reduction of the particle sizes. The increase in M3’s milling time (90 min) did not contribute to the reduction of particle size in relation to M1 or M2 (30 min). When comparing the compositions M1, M3 and M5 with M2, M4 and M6, it appears that the latter have a content of particles with an average diameter Ø reduced (< 106 µm) much higher, particularly for milling times of 90 and 180 min. Observing the result set, this trend is evident when comparing the M3 and M4 compositions (90 min) where M3 has Ø equal to 21%, while M4 has the value of 50%. This can be explained by the fact that the load balls and their diameter can decisively influence the material particle size distribution, and balls are typically larger and heavier, favoring the shear mainly by impact. The use of smaller balls reduces the empty spaces between the milling elements increasing the friction between the balls and the particles. It is worth noting in this case that the shrimp shells are not exclusively composed of chitin, but are typically also found minerals calcium, magnesium and phosphorus associated, which in turn are comminuted along with the biopolymer and protein residues, which may not have been eliminated during the washing stage. Note also that the common process of the preparation of chitin (kiln drying after deproteination, then disintegration in mortar and pestle for subsequent deacetylation) was not used in this work in order to avoid the interference of the physical processes of particle aggregation and disaggregation; this could compromise the results, and undermine the interpretation of them. Thus, it 43

Alves, H. J., Furman, M., Kugelmeier, C. L., Oliveira, C. R., Bach, V. R., Lupatini, K. N., Neves, A. C., & Arantes, M. K. can be said that the only variable that distinguishes samples M1 to M6 is the milling of shrimp shells.

3.2 Particle Size Distribution (PSD) The distribution curve of particle sizes (PSD) of samples M1 to M6, obtained by X-ray sedimentometry, is found in Figure 3. In general, curves with a similar inclination are observed, indicating that the PSD is similar between the samples. There were no significant differences when different compositions of the milling balls were used. However, the positioning of a given curve to the left indicates a more effective milling that result in finer particles, as in the case of M5 and M6. From these curves were extracted the D50 values, as shown in Table 1, which are a statistical parameter representing the average particle diameter when the cumulative mass is 50%. Therefore, larger D50 values indicate that the PSD curves have shifted more to the right in Figure 3, which is associated with a less effective milling of the material (M1 and M2), resulting in coarser particles. Overall, it can be said that the effect caused by using balls of different sizes compositions was higher on the milling of shrimp shells (Figure 2), than on the properties of chitosan particles (Figure 3).

Figure 2. Particle size distribution of shrimp shells milled into ball mill under different conditions (CA: composition A; CB: composition B).

3.3 Specific area (BET) The results of N2 physisorption corroborate those obtained by X-ray sedimentometry, since they indicate an increase in the specific area of the chitosan samples according to the milling time increase, which is associated with the reduction of particle size. With the increase of milling time from 30 min to 180 min, the specific area nearly doubled, from 6.8 m2.g-1 in M1 to 11.9 m2.g1 in M5. Significant differences were not observed for the use of different compositions of balls. However, after 90 minutes of grinding, the specific area of the particles of chitosan does not vary significantly. For milling times of 90 and 180 min were not observed differences in the specific area values b etween the compositions of spheres A and B employed. This may indicate that after the significant reduction of the particle size of shrimp shell up to 30 minutes milling, the use of smaller spheres (composition B) is not able to change significantly the surface characteristics of the particles so as to modify the specific area of chitosan.

3.4 Degree of deacetylation (DD) The method used in this work to promote the chitin deacetylation has been tested in our laboratory in order to obtain chitosan with higher deacetylation without heating or reflux and with experimental apparatuses that are not easily scalable. In an earlier work[1], when using reflux for 10 hours at 100 °C in 50% of NaOH solution, chitosan was obtained whose DD ranged between 95 and 97%. When performing the adaptation for this simplified method and stirring under room temperature, DD values obtained by 1 H NMR (Figure 4) located between 77 and 80% (Table 1) demonstrate the viability of this method when the DD range is suitable for the intended purpose of chitosan produced. Also noteworthy is that from the results shown, the variations employed in the shrimp shells milling process 44

Figure 3. Distribution of particle sizes of the chitosan samples.

did not cause significant differences in the deacetylation reactions, although they have been identified as able to reduce the viscosimetric molecular weight, indicating biopolymer degradation[14]. It was expected that the depolymerization would be accompanied by increasing deacetylation to increase the accessibility of reaction sites, but this was not observed. Another observation concerning the employed deacetylation method is that it can be used when minimizing the polymer degradation during the deacetylation is desired. While the molar mass of chitosan obtained by refluxing (10h, 100 °C) in our studies is very low (about 30 kDa), in this method, the largest value obtained was about 500 kDa.

3.5 X-ray diffraction and apparent crystallites diameter (Dap) It is noted that in Figure 5 the chitosan samples have typical diffractograms of a semi-crystalline material with crystalline major peaks at 2θ ≈ 10° and 20°. The peak at 10° is assigned to the reflections of the planes (010), derived from the α-chitin structure (orthorhombic crystals of nature), since the asymmetric peak at 20° is associated with the planes (010) and (020). Sample M1 has a diffraction pattern that is distinct from the others, because the crystalline peak at 10° is more intense than at 20°. As the milling time increases, Polímeros, 27(1), 41-47, 2017

Effect of shrimp shells milling on the molar mass of chitosan

Figure 4. 1H NMR spectra of samples M1, M2, M3, M4, M5 and M6. The solid lines in the regions of 3.30 to 3.10 ppm and of 2.08 to 2.00 ppm refer to the hydrogens of the amino and acetyl groups, respectively.

a significant decrease of peak intensity at 10° is observed for all samples, suggesting the preferential progressive destruction of the crystal planes (010). At the same time, there is a less pronounced increase in intensity at 20°. These changes are most marked for Composition A (M1, M3 and M5), which may indicate that the impact milling has a greater effect on the destruction of the planes (010), as can be seen in Figure 5a. It is worth mentioning that when analyzing the group of diffractograms of Figure 5, we can affirm that with increasing milling time, the crystallinity of the chitosan generally decreases, which is according to literature[21]. In Table 1, the apparent diameter values of the crystallites (Dap) can be verified. It is worth mentioning that given the partially amorphous character of chitosan, observed in the diffractograms of Figure 5, the determination of the crystalline peaks’ baseline used to determine the values of β0 in the equation of Scherrer (Equation 1) may lead to a small deviation of the estimated values, which in turn varied between 40.1 and 43.2 Å. However, some observations are pertinent: i) The lower Dap values for M1 and M3, when compared to M2 and M4, respectively, may be explained by the fact that when using Composition A (balls of the same size), the impact generated in the particles is more effective, which may result in crystallites with smaller diameters; the segments in molecular crystalline domains are broken by splits into a greater number of biopolymer chains. Polímeros, 27(1), 41-47, 2017

Figure 5. Difratograms of samples: (a) M1, M3 and M5 and (b) M2, M4 and M6. 45

Alves, H. J., Furman, M., Kugelmeier, C. L., Oliveira, C. R., Bach, V. R., Lupatini, K. N., Neves, A. C., & Arantes, M. K.

Figure 6. Variation Mv of the chitosan samples as a function of milling time (CA: composition A; CB: composition B).

ii) The M5 and M6 samples have similar values with each other for both D50 and for Dap. If compared with the other samples, the D50 values are lower due to the effective reduction of the shell particle size with the grinding time advance. In contrast, there was not such a significant change in Dap values of chitosan. This suggests that after 90 min the increase of Dap is less pronounced.

3.6 Viscosimetric molar mass (MV) The results in Figure 6 and Table 1 reveal that as the milling time of shrimp shells increases, the molar mass of the chitosan decreases significantly, independent of the composition of balls used. Note that between the milling time of 30 and 180 min, the viscosimetric molar mass decreases almost four times, from 502 kDa to 131 kDa. The polynomial equations of the second degree are those that best describe the observed behavior: Composition A y = 0.0134x2 – 5.2967x + 648.8 and Composition B: y = 0.0149x2 – 4.5867x + 560.2. Note that the quadratic equations are very similar, and therefore, they have the same trend. Although it was not done in this work the milling of chitosan to evaluate the reduction of its molar mass, but the milling of shrimp shells, it is known that chitin is chemically very similar to chitosan, and in this case, any changes made in the chitin structure can potentially be observed in the chitosan, so that the results of molar mass reduction of chitosan are agree with those found in previous studies[9,14]. Thus one can say that the processing stages of chitin to obtain chitosan, subsequent to milling, acted in the same manner over the material which shows que the only variable analyzed actually was the milling time.

4. Conclusions It may be concluded that milling is a stage of processing the shrimp shell, which significantly reduces the average molar mass of chitosan (nearly four times in the range of 30 to 180 min of milling). With the exception of reducing the particle size followed by an increase of the specific 46

area and decreased crystallinity, the other properties of the chitosan did not vary as significantly as the crystallites size and degree of deacetylation. The diameter of the balls used in the milling of shrimp shells can decisively influence the particle size distribution of the material, directly affecting the size distribution of particles of chitosan. When a composition containing 30% undersize milling balls was evaluated, the result was the reduction of particle size of shell and chitosan, due to reduced voids between the grinding elements and the increased friction between the balls and the particles. The milling time of shrimp shells can be used in a manner associated with the drying of chitosan to obtain a product with low molar mass. Both deal with physical processes (unit operations) already employed in the traditional route from chitosan production, and can possibly replace partially or totally other more sophisticated and expensive technologies (enzymes, irradiation, etc.), if they are well designed and controlled.

5. References 1. Arantes, M. K., Kugelmeier, C. L., Cardozo-Filho, L., Monteiro, M. R., Oliveira, C. R., & Alves, H. J. (2015). Influence of the drying route on the depolymerization and properties of chitosan. Polymer Engineering and Science, 55(9), 1969-1976. http:// dx.doi.org/10.1002/pen.24038. 2. Ma, Z., Wang, W., Wu, Y., He, Y., & Wu, T. (2014). Oxidative degradation of chitosan to the low molecular water-soluble chitosan over peroxotungstate as chemical scissors. PLoS One, 9(6), e100743. PMid:24971631. http://dx.doi.org/10.1371/ journal.pone.0100743. 3. Pornsunthorntawee, O., Katepetch, C., Vanichvattanadecha, C., Saito, N., & Rujiravanit, R. (2014). Depolymerization of chitosan-metal complexes via a solution plasma technique. Carbohydrate Polymers, 102, 504-512. PMid:24507312. http:// dx.doi.org/10.1016/j.carbpol.2013.11.025. 4. Jung, J., & Zhao, Y. (2011). Characteristics of deacetylation and depolymerization of ß-chitin from jumbo squid (Dosidicus gigas) pens. Carbohydrate Research, 346(13), 1876-1884. PMid:21700271. http://dx.doi.org/10.1016/j.carres.2011.05.021. 5. Yue, W., Yao, P., & Wei, Y. (2009). Influence of ultraviotelirradiated oxygen on depolymerization of chitosan. Polymer Degradation & Stability, 94(5), 851-858. http://dx.doi. org/10.1016/j.polymdegradstab.2009.01.023. 6. Yue, W. (2014). Prevention of browning of depolymerized chitosan obtained by gamma irradiation. Carbohydrate Polymers, 101, 857-863. PMid:24299848. http://dx.doi.org/10.1016/j. carbpol.2013.10.011. 7. Dziril, M., Grib, H., Laribi-Habchi, H., Drouiche, N., Abdi, A., Lounici, H., Pauss, A., & Mameri, N. (2015). Chitin oligomers and monomers production by coupling g radiation and enzymatic hydrolysis. Journal of Industrial and Engineering Chemistry, 26, 396-401. http://dx.doi.org/10.1016/j.jiec.2014.12.015. 8. Einbu, A., & Varum, K. M. (2007). Depolymerization and de-N-acetylation of chitin oligomers in hydrochloric acid. Biomacromolecules, 8(1), 309-314. PMid:17206822. http:// dx.doi.org/10.1021/bm0608535. 9. Brostow, W., & Corneliussen, R. D. (1986). Kinetics of milling of polymers. Materials Chemistry and Physics, 14(1), 1-8. http://dx.doi.org/10.1016/0254-0584(86)90013-1. 10. Sánchez-Jiménez, P. E., Pérez-Maqueda, L. A., Perejón, A., & Criado, J. M. (2010). A new model for the kinetic analysis of thermal degradation of polymers driven random scission. Polímeros, 27(1), 41-47, 2017

Effect of shrimp shells milling on the molar mass of chitosan Polymer Degradation & Stability, 95(5), 733-739. http://dx.doi. org/10.1016/j.polymdegradstab.2010.02.017. 11. Bressy, C., Ngo, V. G., & Margaillan, A. (2013). A first insight into the thermal degradation mechanism of silylated methacrylic homopolymers synthesized via the RAFT process. Polymer Degradation & Stability, 98(1), 115-121. http://dx.doi. org/10.1016/j.polymdegradstab.2012.10.023. 12. Loh, Z. H., Samanta, A. K., & Sia Heng, P. W. (2015). Overview of milling techniques for improving the solubility of poorly water-soluble drugs. Asian Journal of Pharmaceutical Sciences, 10(4), 255-274. http://dx.doi.org/10.1016/j.ajps.2014.12.006. 13. Cook, R., & Mercer, M. B. (1985). Dynamic overstresses in fibrous polymeric materials. Materials Chemistry and Physics, 12(6), 571-580. http://dx.doi.org/10.1016/0254-0584(85)900434. 14. Delezuk, J. A. M., Cardoso, M. B., Domard, A., & Campana-Filho, S. P. (2011). Ultrasound-assisted deacetylation of beta-chitin: influence of processing parameters. Polymer International, 60(6), 903-909. http://dx.doi.org/10.1002/pi.3037. 15. Delezuk, J. A. M. (2013). Chitosan production with controlled characteristics using high intensity ultrasound irradiation [Doctoral thesis]. University of São Paulo, São Carlos. 16. Kasaai, M. R. (2007). Calculation of Mark-Houwink-Sakurada (MHS) equation viscometric constants for chitosan in any solvent-temperature system using experimental reported

Polímeros, 27(1), 41-47, 2017

viscosimetric constants data. Carbohydrate Polymers, 68(3), 477-488. http://dx.doi.org/10.1016/j.carbpol.2006.11.006. 17. Brunauer, S., Emmett, P. H., & Teller, E. (1938). Adsorption of gases in multimolecular layers. Journal of the American Chemical Society, 60(2), 309-319. http://dx.doi.org/10.1021/ ja01269a023. 18. Boz, N., Degirmenbasi, N., & Kalyon, D. M. (2009). Conversion of biomass to fuel: transesterification of vegetable oil to biodiesel using KF loaded nano-g-Al2O3 as catalyst. Applied Catalysis B: Environmental, 89(3-4), 590-596. http://dx.doi. org/10.1016/j.apcatb.2009.01.026. 19. Muzzarelli, A. A. (1985). Chitin. In H. F. Mark, N. M. Bikales, C. G. Overberger & G. Menges (Eds.). Encyclopedia of polymers science engineering. New York: John Wiley. 430 p. 20. Santos, J. E., Soares, J. P., Dockal, E. R., & Campana-Filho, S. (2003). Caracterização de quitosanas comercias de diferentes origens. Polímeros: Ciência e Tecnologia, 13(4), 242-249. http://dx.doi.org/10.1590/S0104-14282003000400009. 21. Perrin-Sarazin, F., Sepehr, M., Bouaricha, S., & Denault, J. (2009). Potential of ball milling to improve clay dispersion in nanocomposites. Polymer Engineering and Science, 49(4), 651-665. http://dx.doi.org/10.1002/pen.21295. Received: Sept. 15, 2015 Revised: Mar. 29, 2016 Accepted: May 16, 2016

http://dx.doi.org/10.1590/0104-1428.2409

O Characterization of low cost orally disintegrating film (ODF) Riana Jordao Barrozo Heinemann , Fernanda Maria Vanin , Rosemary Aparecida de Carvalho , O Marco Antonio Trindade and Carmen Sílvia Fávaro-Trindade * O Department of Food Engineering, College of Animal Science and Food Engineering – FZEA, Universidade de São Paulo – USP, Pirassununga, SP, Brazil O O Abstract O Orally disintegrating films (ODF) produced with a hydrophilic polymers are a thin and flexible material, wich disintegrate in contact with saliva and can vehicule bioactive materials. The aim of this study was to develop and characterize ODF O formulation with potential to act as a carrier for different bioactives compounds prepared with low cost polymers. Gelatin (G), starch (S), carboxymethyl cellulose (CMC) and their blends (G:S, CMC:S, CMC:G, and CMC:S:G) were by casting technique with sorbitol as a plasticizer. The formulations were characterized in terms of visual O prepared aspects, FTIR, SEM, mechanical characteristics, hygroscopicity, dissolution (in vitro and in vivo) and swelling index. analysis revealed that no interaction between polymers in ODF was observed. By SEM, it was possible to observe O FTIR differences on surfaces by different polymers. ODF made with CMC and CMC:G presented higher water absorption and higher swelling index probably due to the higher water affinity by CMC. Formulations with G, CMC:G and O (P<0.05) CMC:S:G presented the highest values of tensile strength (P<0.05). ODF prepared with S alone presented the highest time, the others formulations showed in vitro dissolution ranging from 5.22 to 8.50 min, while in vivo O disintegration dissolution time ranged from 2.15 to 3.38 min. By the formulations made with G and blend of G:S and CMC:S:G it is possible to develop a ODF of low cost with desired characteristics being an alternative vehicle to deliver functional O compounds for continuous use. O Keywords: biofilm, dissolution, polymeric matrix, oral vehicle. O 1. Introduction Edible films produced by natural macromolecules have small amounts of saliva, passed through esophagus and O emerged as a potential product for the food industry to protect reached the stomach more quickly than a gelatin capsule . fruits and for its possible application as biodegradable packaging. Other researchers have used microcrystalline cellulose, O However, the pharmaceutical industry has used the same hydroxypropylmethyl cellulose and carboxymethyl cellulose 1

*carmenft@usp.br

[8]

technology to a different application. Orally disintegrating films (ODF) presented as strips derived from hydrophilic polymers that dissolve in oral cavity can delivery drugs and bioactive compounds such as caffeine, nicotine, drugs, refreshing compounds, vitamins, minerals and probiotics[1-4].

as base materials to prepare oral film[9,10]. The most common technique for film production is casting. It is based on the dispersion of a biopolymers in a solvent (water, ethanol and organic acids), addition of additives (plasticizers), yielding a solution that undergoes drying operation[3,5].

The advantage of using ODF is the high efficiency of absorption of some compounds by oral via without the need of water for swallow, being an alternative to bioactive administered in tablets and pills[1,3], and moreover the absorption through the buccal epithelium, without contact with gastrointestinal tract which could degrade some sensible compounds[5]. Moreover, some peoples have difficulty in take and dissolve tablets and capsules, once those materials are larger and necessity a strong buccal work to disintegrate and dissolving the drug or bioactive compost[6]. Therefore, the use of strips for oral dissolution presents some advantages, but development should take into consideration sensory characteristics of the product.

The ODF formulation should be hard enough to not be damaged during handling and transportation, and present suitably disintegration in the mouth, but these features depends on polymeric composition[3,11]. In order to support the stress in the mouth, an ideal oral film should be strong and yet, flexible, elastic, and soft[9,12]. In this way, in order to use those ODF as a material of controlled release it is firstly necessary to better understanding their properties as model, without bioactive compounds or drugs, for its future application.

A variety of biopolymers can be used for oral strips formation, alone or in blends. An alginate/gelatin blend film was developed and this considered a potentially useful in drug delivery systems[7]. Films produced with gelatin, polyvinyl alcohol and carboxyvinyl turn into a jelly in contact with

Therefore, the objective of this study was to develop ODF’s using low cost polymers, and characterize those materials in terms of some factors that may have inﬂuence on the drug release, as visual, mechanical and microstructure characteristics, higroscopicity, swelling and disintegration properties. The material developed could be a successful application for drug delivery or bioactive controlled release in vivo or different industry process.

Polímeros, 27(1), 48-54, 2017

Characterization of low cost orally disintegrating film (ODF)

2. Materials and Methods 2.1 Materials for preparation of ODF Formulations were prepared using cassava starch (ARGO CC3400, Corn Products, Brazil); type A gelatin (260 Bloom/ 30 mesh) (Gelita do Brasil, São Paulo-Brazil); carboxymethyl cellulose (CMC 30000, Plury Química, Brazil) and sorbitol (Synth, Brazil). All the listed ingredients are considered GRAS (Generally Recognized as Safe) for FDA (Food And Drug Administration) and allowed for food products preparation..

2.2 Preparation of polymer matrix For ODF production it was used different combinations of the macromolecules gelatin (G), starch (S), and carboxymethyl cellulose (CMC) according to Table 1. Sorbitol concentration, the plasticizer, was kept constant at 20 g/100 g of macromolecules. Macromolecules and plasticizer concentration were defined in according to preliminary tests. ODF’s were prepared by casting technique. For starch ODF production, 2 g of starch was dispersed in 100 mL of distilled water for 20 min. Then the solution was heated with constant magnetic stirring (Marconi, MA085 – Brazil) until reached 75 °C/ 20 min. Then sorbitol mass (20 g/ 100 g of macromolecules) was added. For gelatin formulation, the same protocol was followed. For CMC formulation (1 g of CMC/ 100 mL), the carbohydrate was solubilized in water and was stirred overnight in order to avoid macromolecule insolubilization. Then, sorbitol was added and the final concentration was corrected with the addition of water. For the development of binary blends CMC:S, CMC:G, G:S, and CMC:S:G, the corresponding solution of each polymer was prepared as previously described. Then, the solutions were mixed according to the concentrations described in Table 1 added by sorbitol mass. In order to avoid bubbles in ODF, all formulations were placed for 10 min in ultrasonic bath and then poured on acrylic plates (12 × 12 cm), according to a previously standardized mass to ensure uniform thickness after drying. Drying of different formulation was in air forced oven at 30 °C/ 24 to 48 h (Marconi, MA035/5 – Brazil). After drying, the films were carefully removed from plates

and placed in desiccators (NaBr saline saturated solution, relative humidity = 58%) at 25 °C for at least 48 h until analysis. The thickness of each strips was measured using digital micrometer (Mitutoyo), considering the average of at least nine values[13]. The overall aspects of the product were evaluated, considering its homogeneity, easiness of peeling from acrylic plate and presence of bubbles or grumps.

2.3 Scanning electron microscopy (SEM) For this analysis, 1 × 1 cm samples were cut from strips, cryofractured and mounted in copper stubs. Samples were gold coated. The internal structure of the sample was evaluated using a Jeol scanning electron microscope (JMS-T300, Tokyo, Japan) at 5kV in a climatized room[14]. Before this tests, ODFs were stored in desiccators containing silica gel for 10 days.

2.4 FTIR The infrared spectra (FTIR) of the ODF were recorded between 600 and 4000 cm-1 and at 4 cm-1 of resolution with a Spectrum One (Perkin Elmer, Shelton, CT, USA) spectrometer[15]. For each spectrum, 16 scans were co-added. Before this tests, ODFs were stored in desiccators containing silica gel for 10 days.

2.5 Mechanical properties The mechanical properties of ODF were evaluated by tensile test [TS = tensile strength (MPa) and E= elongation at break (%)] using a texturometer TA.XT2i (Stable Micro Systems, UK)[16]. Samples of the films (12.0 × 2.54 cm) were fixed on a specific probe (tensile grips), at an initial separation distance of 100 mm, and test speed was constant at 50 mm/ min.

2.6 Moisture and hygroscopicity evaluation The moisture evaluation was performed in a high-precision moisture analyzer (Ohaus MB35 - USA) by infrared radiation from a halogen source[17]. To determine the hygroscopicity, 1g samples of each formulation was placed on glass slides, where were conditioned by 25 °C/ 7 days in desiccators with saturated solution of Na2SO4 (RH 81%). The hygroscopicity was determined by the water mass absorbed by the sample[18].

2.7 Swelling index Table 1. Composition of orally disintegrating films (ODF) formulations produced with gelatin (G), starch (S) or carboxymethyl cellulose (CMC), or its blends. Composition (%) Starch CMC solution Formulation Gelatin solution solution (2 g/100mL) (1 g/100mL) (2 g/100mL) G 100 0 0 S 0 100 0 CMC 0 0 100 G:S 50 50 0 CMC:S 0 80 20 CMC:G 80 0 20 CMC:S:G 40 40 20

Polímeros, 27(1), 48-54, 2017

Samples of 2 × 2 cm of different formulations (140-160 mg) were placed in a metal sieve and immersed in 40 mL of saliva simulated solution[19] in water bath at 36 ± 2 °C, according to adapted procedures[9]. The samples were weighed every 30 seconds until the maximum absorption of water was reached. The swollen weight of the strips was performed. The swelling index was calculated as the ratio between the masses of the strips after and before immersion, respectively.

2.8 In vitro disintegration time The ODF’s disintegration time were evaluated according to previous developed method[20] with some modifications. For this, pieces of 2 × 3 cm of each formulation (0.17 to 0.20 mg) were placed in 50 mL of simulated saliva solution, in a water 49

Heinemann, R. J. B., Vanin, F. M., Carvalho, R. A., Trindade, M. A., & Fávaro-Trindade, C. S. bath at 36 ± 2 °C with mechanical stirring. The time for complete disintegration was visually evaluated. An average of triplicated analysis of each formulation was performed.

2.9 In vivo disintegration time An assay to evaluate differences in disintegration time of different formulations was conducted. For this evaluation, 17 panelists were recruited (5 males and 12 females; ages between 19 and 41) and trained on how to evaluate the desintegration time of samples. Panelists were conducted to individual cabinet where 3 × 2 cm samples were randomly served. A chronometer was used for panelists determine the time until the complete disintegration of the sample in the mouth. This study was conducted in accordance with ethical principles and approved by FZEA-USP ethical committee (Process 2010.1.1479.74.3). Prior to performing the test, panelists signed a free and informed consent term.

2.10 Statistical analysis All experiments were performed in triplicate. The data were statistically analyzed using SAS version 9.2, by ANOVA followed by Tukey test (5% of significance).

3. Results and Discussions All samples were homogenous, without bubbles and phase separation. It was possible to produce oral strips with the different tested formulations, but the formulation using only starch was difficult to be peeled of, breaking easily. ODF produced with G and CMC were brilliant and the ones with S in composition were opaque.

3.1 SEM analysis The structure characteristics of the different formulations evaluated are shown in Figure 1. Differences could be observed in the surface structure of different polymers. ODF produced with the blend carboxymethil cellulose: cassava starch (Figure 1F), and carboxymethil cellulose: cassava starch: gelatin (Figure 1G) presented roughness in the surface. All other blends (Figure 1D e 1E) and polymers alone (Figure 1A, 1B and 1C) presented very homogeneous surface. Roughness at surface in films made with starch blends has already observed, in general, the micrograph cross section of cassava starch-based films displayed an irregular and rough structure[21]. The authors suggested that this heterogeneous structure could be due to the retrogradation and partial crystallization of gelatinized starch before the formation of the film. The micrograph results suggested good compatibility among the polymers, without micro phase separation, however starch in combination with other polymers, presented surface modification in rugosity.

3.2 FTIR analysis FTIR spectroscopy was used to examine the interactions between macromolecules used to prepare the ODF. The infrared spectra of CMC, gelatin and starch, and their blends are represented in Figure 2. The spectrum of CMC, starch and gelatin alone were similar to previous reported 50

in the literature[15,22,23]. For CMC characteristic band were observed at 1587 cm-1 (stretching of C=O), 1415 cm-1 (CH2 carboxylic groups), 1322 cm-1 (absorption of CH2) e 1051 cm-1 (stretching of C-O), consistent with previous data reported[22]. For gelatin alone, band centered around 3300 cm-1 is mainly due to the extension of the group NH of amide A. In the range 3000-3500 cm-1 there is a absorption band due to hydroxyl groups (OH), in the films of S:G. Its intensity decreases, indicating that the polymer created linkages of hydrogen intra chains. The intense bands between 1700 and 1600 cm-1 and between 1600 and 1500 cm-1, are known, respectively, as bands of amide I and amide II. Amide III, with bands between 1200 and 1400 cm-1, represent components of the extension of C–N and N-H and absorptions resulting from the vibrations of groups C–H2 of the glycine and proline[23]. Starch oral film also presents similar FTIR spectra from the one previous presented[15]. When two or more substances are mixed, physical blends versus chemical interactions are reﬂected by changes in characteristic spectra peaks[24,25]. Apparently, no structural changes occurred since the peaks of the blends broads compared to the ones with single polymers were similar. The results suggest that the formation of new bonds between macromolecules for blends formulation of ODF do not occur.

3.3 Mechanical properties The Table 2 present the results of tensile strength and elongation obtained from ODF prepared G, S, CMC and their blends. It could be observed (Table 2) that the ODF with gelatin in the formulation presented higher tensile strength and elongation, compared to the others formulations (S, CMC and CMC:S). Probably, the improvements observed in ODF containing gelatin are a consequence of the high cohesivity of the polymeric matrix. Similar values of the tensile strength reported in this study were observed for ODF’s made with pullulan, sodium alginate and CMC blends[26]. Gelatin:alginate oral strips, produced for controlled drugs release, presented values of maximum tensile strength and elongation when the blend of gelatin and alginate was 50%[7]. The authors observed that increasing gelatin or alginate content rather than 50% decreased the tensile strength.

Table 2. Mechanical properties of orally disintegrating films (ODF) produced with gelatin (G), starch (S) or carboxymethyl celulose (CMC), or its blends (means ± standard deviation)*. OFD identification G S CMC G:S CMC:G CMC:S CMC:S:G

Tensile strength (MPa) 55.81 ± 5.15 a,b 22.47 ± 5.93 d 34.61 ± 9.35 c,d 47.49 ± 3.82 b,c 71.72 ± 9.08 a 32.03 ± 6.27 c,d 49.29 ± 2.53 b,c

Elongation (%) 7.40 ± 1.67 a,b 5.40 ± 1.43 a,b 5.78 ± 2.91 a,b 4.86 ± 1.56 a,b 8.95 ± 1.75 a 3.60 ± 1.11 b 7.49 ± 2.59 a,b

*Means followed by the same letter in each column are not different according to tuckey’s test (p ≤ 0.05).

Polímeros, 27(1), 48-54, 2017

Characterization of low cost orally disintegrating film (ODF)

Figure 1. Micrographs of orally disintegrating films (ODF): (A) gelatin (G); (B) cassava starch (S); (C) carboxymethil cellulose (CMC); (D) gelatin: cassava starch (G:S); (E) carboxymethil cellulose: gelatin (CMC:G); (F) carboxymethil cellulose: cassava starch (CMC:S); (G) carboxymethil cellulose: cassava starch: gelatin (CMC:S:G).

3.4 Hygroscopicity and swelling index Table 3 presented the values of hygroscopicity and swelling index from the different ODF formulations. The hygroscopicity is the amount of absorbed water under controlled conditions at high RH. From the results obtained it could be observed that the hygroscopic of ODF ranged from 12.6 g / 100g to 31.5 g / 100g (Table 3). All formulations absorbed water after one week storage, but the ODF composed of only CMC and CMC:G showed the Polímeros, 27(1), 48-54, 2017

highest values, ~31.4 g / 100 g (P< 0.05); it was also observed that these strips were sticky. The ODF produced with only G or S presented the lower hygroscopity. Their average value was only 14.8 g /100 g, significantly lower than the others ODF hygroscopic. The strips composed of the ternary blend CMC:S:G presented intermediate values of hygroscopicity, and, were not sticky. This diference in water adsorption could be related to the number of hydrophilic groups present in the structure of each agent[27], or to the number of active sites to linkage of water. There is no 51

Heinemann, R. J. B., Vanin, F. M., Carvalho, R. A., Trindade, M. A., & Fávaro-Trindade, C. S. information in the literature about hygroscopicity behavior recommended for storage of ODF. On other hand, the production of ODF with only carboxymethil celloluse, or in a blend with gelatin, could improve the final material hygroscopicity, which is not desirable. Therefore, these formulations were not evaluated in terms of dissolution in vivo. In relation to the swelling index (Table 3) it could be observed from the different ODF formulations a greatly variation depending on their composition, with water absorption values from 4.08 to 17.71 g of water/g. Strips composed of S, G and G:S showed the lowest capacity of swelling (P< 0.05). Strips composed by CMC presented the highest capacity of swelling (P< 0.05), differing from the others. When CMC was incorporated to the formulation, in relation to S or G formulation, the ODF water affinity was enhanced The CMC affinity with water was observed when reduced contents of carbopol (cross-linked polyacrylate polymer) and increased sodium carboxymethyl cellulose contents were applied for strips production, and the swelling index also increased[28]. This effect was also observed with the increase of sodium carboxymethyl cellulose contents[9]. Bajpai and

Figure 2. FTIR spectra of gelatin (G), starch (S), carboximethil celulose (CMC) and their blends in orally disintegrating films (ODF).

Shrivastava[29] observed the same effect in polymeric film made with crosslinked starch and carboxymethyl cellulose, the increase of CMC concentration increased the sweeling of films. CMC is a modified natural water-soluble polymer, wich contains hydroxy and carboxyl groups, and therefore improve hydrophilicity to the molecule[29], which could explain the minor levels of sweeling index of ODF’s produced without CMC. If the swelling of films exists, it should not be to extensive in order to prevent discomfort[9]. In this way, again the ODF contain only CMC and CMC:G were discarded, and therefore not considered for in vivo analyses.

3.5 In vitro disintegration time Some authors developed fast dissolving oral films[2,11,30], while others bioadhesive films[9,28] which can that take minutes or hours to release the active compound of interest. No official guidance time is referenced for oral fast disintegrating films/strips[1,3]. In the same way, no official time for disintegration of ODF was found. Therefore, in this paper, it was established the time of 4 min as the ideal for the in vitro disintegration of oral strips, as the time that allows gradual release of bioactive compounds to be added, without causing fatigue to the consumer. According to Table 3, starch ODF showed the highest disintegration time, > 100 min. It was observed that after 24 h/ 36 °C in the presence of simulated saliva it still remained intact, therefore without interest for the development of active compounds in an oral vehicle. Furthermore, even if a saliva with enzyme could be used, this probably would not reduce sufficient the in vitro disintegration time in order to be possible to select this ODF for in vivo analyses (see reduction time for ODF produced with starch blends). The others formulations showed disintegration time ranging from 5.22 to 8.50 min. Visually, it was observed that the formulation with only S and with CMC:S breakdown in small pieces and did not completely dissolve in the solution. The others formed a continue phase with water. The results are in the same magnitude order then ODF produced with gelatin and hydrolyzed collagen[31], which varied from ~ 6 to 9 minutes in function of hydrolyzed collagen concentration. Therefore, in accordance to in vitro dissolution time, ODF’s produced with only S, and the blend CMC:S, were not considered adequate. Thus, only G, G:S, CMC:G:S ODF’s were characterized in relation to in vivo disintegration time.

Table 3. Characterization parameters of orally disintegrating films (ODF) produced with gelatin (G), starch (S) or carboxymethyl celulose (CMC), or its blends (mean ± standard deviation)*. ODF identification G S CMC G:S CMC:G CMC:S CMC:S:G

Moisture

Hygroscopicity

Swelling index

(g H2O/g)

11.43 ± 0.74 b 9.95 ± 0.61 b 15.17 ± 1.46 a 11.49 ± 0.86 b 10.97 ± 0.41 b 9.95 ± 0.57 b 11.01 ± 0.09 b

16.90 ± 1.16 c,d 12.61 ± 1.89 d 31.25 ± 4.92 a 23.32 ± 0.49 b 31.49 ± 2.75 a 23.18 ± 0.91 b 18.75 ± 1.10 b,c

4.83 ± 0.78 d 4.08 ± 0.45 d 17.71 ± 0.35 a 4.37 ± 0.34 d 13.46 ± 2.86 b 8.96 ± 0.40 c 8.20 ± 1.06 c

In vitro disintegration time (min) 5.22 ± 0.20 d >100 a 6.07± 1.21d 7.96 ± 0.56 b,c 8.50 ± 0.60 b 8.37 ± 1.12 b,c 6.50± 0.67 c,d

*Means followed by the same letter in each column are not signicantly different according to tuckey’s test (P <0.05).

Polímeros, 27(1), 48-54, 2017

Characterization of low cost orally disintegrating film (ODF) uncomfortable sensation. Therefore, those ODF could be very useful for an innovative oral vehicle for controlled release of bioactives compounds.

5. Acknowledgements Authors are grateful for support provided by FAPESP (fellowship # 2009/ 09762-5) and to CNPq for grant # 479249/2010-5.

6. References Figure 3. In vivo disintegration time of different orally-disintegrating film (ODF) produced with gelatin (G), gelatin:cassava starch (G:S), and carboxymethyl cellulose:cassava starch:gelatin (CMC:S:G). Different lowercase letters indicate a significant difference (p <0.05) between the different formulations of ODF.

3.6 In vivo disintegration time In vivo test with three formulations was evaluated and results could be observed in Figure 3. The means of disintegration time varied between 2.15 and 3.38 min. The in vivo disintegration test showed that all panelists agreed that the dissolution of samples was comfort to them; however, as expected for in vivo tests, there was great variability in the results. It could be observed that the disintegration times of the test in vivo were lower than the in vitro test, probably due to a lack in enzymatic activity and mechanical action of the second. Few studies report the disintegration time of oral strips. Oral strips produced with maltodextrin presented in vivo disintegration of only 10 s[30]. For ODF based on hydroxypropyl methylcellulose, corn starch, polyethylene glycol, and lactose monohydrate with donepezil a disintegration mean time of 49 s[32]. For drug delivery by oral route, an ODF with hydroxy-propyl methyl cellulose and carbopol had a residence time of 23 min[9]. The disintegration time will depend on the characteristic of the bioactive or drug released[2]. Some compounds should be continued released, while others should be immediately absorbed. In this study, during the development was set the objective of gradual release of a hydrophilic bioactive as vitamins or minerals, but fast enough to do not cause stress to the consumer. This parameter was achieved with all tested formulations in vivo study.

4. Conclusions The study showed that the composition of ODF had important relevant inﬂuence on properties evaluated, and therefore probably on the ODF release properties. Gelatin, cassava starch, carboxymethyl cellulose, and its blends, were successful applied for ODF production by “casting” technique. However, ODF’s produced with only starch were not interesting due to difficulty in handling and slow time disintegration (> 100 min). ODF’s composed with only CMC, and blends of CMC:G were very sticky and therefore were discarded. On the other hand, formulations made by G, G:S and CMC:S:G were more suitable for ODF production, since it was homogeneous, presented average values of hygroscopicity, good performance in mechanical tests and an acepatble in vivo disintegration time, with no Polímeros, 27(1), 48-54, 2017

1. Dixit, R. P., & Puthli, S. P. (2009). Oral strip technology: overview and future potential. Journal of Controlled Release, 139(2), 94-107. PMid:19559740. http://dx.doi.org/10.1016/j. jconrel.2009.06.014. 2. Garsuch, V., & Breitkreutz, J. (2010). Comparative investigations on different polymers for the preparation of fast-dissolving oral films. The Journal of Pharmacy and Pharmacology, 62(4), 539-545. PMid:20604845. http://dx.doi.org/10.1211/ jpp.62.04.0018. 3. Nagaraju, T., Gowthami, R., Rajashekar, M., Sandeep, S., Mallesham, M., Sathish, D., & Kumar, Y. S. (2013). Comprehensive review on oral disintegrating films. Current Drug Delivery, 10(1), 96-108. PMid:22920576. http://dx.doi. org/10.2174/1567201811310010016. 4. Patel, V. F., Liu, F., & Brown, M. B. (2011). Advances in oral transmucosal drug delivery. Journal of Controlled Release, 153(2), 106-116. PMid:21300115. http://dx.doi.org/10.1016/j. jconrel.2011.01.027. 5. Morales, J. O., & McConville, J. T. (2011). Manufacture and characterization of mucoadhesive buccal ﬁlms. European Journal of Pharmaceutics and Biopharmaceutics, 77(2), 187-199. PMid:21130875. http://dx.doi.org/10.1016/j.ejpb.2010.11.023. 6. Palmer, J. B., Drennan, J. C., & Baba, M. (2000). Evaluation and treatment of swallowing impairments. American Family Physician, 61(8), 2453-2462. PMid:10794585. 7. Dong, Z., Wang, Q., & Du, Y. (2006). Alginate/gelatin blend films and their properties for drug controlled release. Journal of Membrane Science, 1(2), 37-44. http://dx.doi.org/10.1016/j. memsci.2006.01.002. 8. Okabe, H., Suzuki, E., Sugiura, Y., Yanagimoto, K., Takanashi, Y., Hoshi, M., Nogami, E., Nakahara, K., Sekiguchi, T., & Baba, M. (2008). Development of an easily swallowed film formulation. International Journal of Pharmaceutics, 355(1-2), 62-66. PMid:18191926. http://dx.doi.org/10.1016/j. ijpharm.2007.11.038. 9. Peh, K. K., & Wong, C. F. (1999). Polymeric films as vehicle for buccal delivery: swelling, mechanical, and bioadhesive properties. Journal of Pharmacy & Pharmaceutical Sciences, 2(2), 53-61. PMid:10952770. 10. Nishimura, M., Matsuura, K., Tsukioka, T., Yamashita, H., Inagaki, N., Sugiyama, T., & Itoh, Y. (2009). In vitro and in vivo characteristics of prochlorperazine oral disintegrating film. International Journal of Pharmaceutics, 368(1-2), 98-102. PMid:18992311. http://dx.doi.org/10.1016/j.ijpharm.2008.10.002. 11. Corniello, C. (2006). Quick dissolving strips: from concept to commercialization. Drug Delivery Technology, 6(2), 68-71. 12. Pathare, Y. S., Hastak, V. S., & Bajaj, A. N. (2013). Polymers used for fast disintegrating oral films: a review. International Journal of Pharmaceutical Sciences Review and Research, 21(1), 169-178. 13. Sobral, P. J. A., García, F. T., Habitante, A. M. Q. B., & Monterrey-Quintero, E. S. (2004). Propriedades de filmes comestíveis produzidos com diferentes concentrações de 53

Heinemann, R. J. B., Vanin, F. M., Carvalho, R. A., Trindade, M. A., & Fávaro-Trindade, C. S. plastificantes e de proteínas do músculo de tilápia-do-nilo. Pesquisa Agropecuária Brasileira, 39(3), 255-262. http:// dx.doi.org/10.1590/S0100-204X2004000300008. 14. Carvalho, R. A., & Grosso, C. R. F. (2004). Characterization of gelatin based films modified with transglutaminase glyoxal and formaldehyde. Food Hydrocolloids, 18(5), 717-726. http:// dx.doi.org/10.1016/j.foodhyd.2003.10.005. 15. Bergo, P. V. A., Carvalho, R. A., Sobral, P. J. A., Santos, R. M. C., Da Silva, F. B. R., Prison, J. M., Solorza-Feria, J., & Habitante, A. M. Q. B. (2008). Physical properties of edible films based on cassava starch as affected by the plasticizer concentration. Packaging Technology and Science, 21(2), 85-89. http://dx.doi.org/10.1002/pts.781. 16. Thomazine, M., Carvalho, R. A., & Sobral, P. J. (2005). Physical properties of gelatin films plasticized by blends of glycerol and sorbitol. Journal of Food Science, 70(3), E172-E176. http:// dx.doi.org/10.1111/j.1365-2621.2005.tb07132.x. 17. Favaro-Trindade, C. S., Santana, A. S., Monterrey-Quintero, E. S., Trindade, M. A., & Netto, F. M. (2010). The use of spray drying technology to reduce bitter taste of casein hydrolysate. Food Hydrocolloids, 24(4), 336-340. http://dx.doi.org/10.1016/j. foodhyd.2009.10.012. 18. Cai, Y. Z., & Corke, H. (2000). Production and properties of spray-dried Amaranthus betacyanin pigments. Journal of Food Science-Chicago, 65(7), 1248-1252. http://dx.doi. org/10.1111/j.1365-2621.2000.tb10273.x. 19. Queiroz, G. M. O. D., Silva, L. F., Ferreira, J. T. L., Gomes, J. A. D. C. P., & Sathler, L. (2007). Electrochemical behavior and pH stability of artificial salivas for corrosion tests. Brazilian Oral Research, 21(3), 209-215. PMid:17710285. http://dx.doi. org/10.1590/S1806-83242007000300004. 20. El-Tinay, A. H., & Ismail, I. A. (1995). Effect of some additives and processes on the characteristics of agglomerated and granulated spray-dried Roselle powder. Acta Alimentaria Hungaricae, 14, 283-295. 21. Phan, T. D., Debeaufort, F., Luu, D., & Voilley, A. (2005). Functional properties of edible agar-based and starch-based films for food quality preservation. Journal of Agricultural and Food Chemistry, 53(4), 973-981. PMid:15713008. http:// dx.doi.org/10.1021/jf040309s. 22. Oliveira, A. F. D., Soldi, V., Coelho, C. M. M., Miqueloto, A., & Coimbra, J. L. (2009). Preparation, characterization and properties of polymeric films with potential application in seed coatings. Química Nova, 32(7), 1845-1849. http:// dx.doi.org/10.1590/S0100-40422009000700030. 23. Andreuccetti, C., Carvalho, R. A., & Grosso, C. R. (2009). Effect of hydrophobic plasticizers on functional properties of gelatin-based films. Food Research International, 42(8), 1113-1121. http://dx.doi.org/10.1016/j.foodres.2009.05.010.

24. Guan, Y., Liu, X., Zhang, Y., & Yao, K. (1998). Study of phase behavior on chitosan/viscose rayon blend film. Journal of Applied Polymer Science, 67(12), 1965-1972. http://dx.doi. org/10.1002/(SICI)1097-4628(19980321)67:12<1965::AIDAPP2>3.0.CO;2-L. 25. Yin, Y. J., Yao, K. D., Cheng, G. X., & Ma, J. B. (1999). Properties of polyelectrolyte complex films of chitosan and gelatin. Polymer International, 48(6), 429-432. http://dx.doi. org/10.1002/(SICI)1097-0126(199906)48:6<429::AIDPI160>3.0.CO;2-1. 26. Tong, Q., Xiao, Q., & Lim, L. T. (2008). Preparation and properties of pullulan–alginate–carboxymethylcellulose blend films. Food Research International, 41(10), 1007-1014. http:// dx.doi.org/10.1016/j.foodres.2008.08.005. 27. Tonon, R. V., Brabet, C., Pallet, D., Brat, P., & Hubinger, M. D. (2009). Physicochemical and morphological characterisation of açai (Euterpe oleraceae Mart.) powder produced with different carrier agents. International Journal of Food Science & Technology, 44(10), 1950-1958. http://dx.doi.org/10.1111/ j.1365-2621.2009.02012.x. 28. Singh, S., Soni, R., Rawat, M. K., Jain, A., Deshpande, S. B., Singh, S. K., & Muthu, M. S. (2010). In vitro and in vivo evaluation of buccal bioadhesive films containing salbutamol sulphate. Chemical & Pharmaceutical Bulletin, 58(3), 307-311. PMid:20190433. http://dx.doi.org/10.1248/cpb.58.307. 29. Bajpai, A. K., & Shrivastava, J. (2005). In vitro enzymatic degradation kinetics of polymeric blends of crosslinked starch and carboxymethyl cellulose. Polymer International, 54(11), 1524-1536. http://dx.doi.org/10.1002/pi.1878. 30. Cilurzo, F., Cupone, I. E., Minghetti, P., Selmin, F., & Montanari, L. (2008). Fast dissolving films made of maltodextrins. European Journal of Pharmaceutics and Biopharmaceutics, 70(3), 895-900. PMid:18667164. http://dx.doi.org/10.1016/j. ejpb.2008.06.032. 31. Borges, J. G., Tagliamento, M., Silva, A. G., Sobral, P. J. D. A., & Carvalho, R. A. D. (2013). Development and characterization of orally-disintegrating films for propolis delivery. Food Science and Technology, 33, 28-33. http://dx.doi.org/10.1590/ S0101-20612013000500005. 32. Liew, K. B., Tan, Y. T., & Peh, K. K. (2012). Characterization of oral disintegrating film containing donepezil for Alzheimer disease. AAPS PharmSciTech, 13(1), 134-142. PMid:22167416. http://dx.doi.org/10.1208/s12249-011-9729-4. Received: Nov. 18, 2015 Revised: Apr. 23, 2016 Accepted: May 16, 2016

Polímeros, 27(1), 48-54, 2017

http://dx.doi.org/10.1590/0104-1428.2385

Rice husk ash as filler in tread compounds to improve rolling resistance Mônica Romero Santos Fernandes1,2*, Ana Maria Furtado de Sousa2 and Cristina Russi Guimarães Furtado2 Lanxess Elastômeros do Brasil S.A., Duque de Caxias, RJ, Brazil Instituto de Química, Universidade do Estado do Rio de Janeiro – UERJ, Rio de Janeiro, RJ, Brazil 1

*monicaromero.fernandes@gmail.com

Abstract In the tire industry carbon black is being replaced by silica as a filler in recent due to the development of “green tires”. Amorphous precipitated silica in combination with a silane coupling agent as a filler in tread compounds can result in fuel savings of 3% to 4% compared to a tire having treads made from compounds with carbon black. This means a 20% reduction of the rolling resistance and consequently lower greenhouse gas emissions. On the other hand, rice is one of the most important food crops generating around 22% in weight of husk during its milling, a material that is mainly used as fuel for energy generation, resulting in ash. Rice husk ash (RHA) contains over 70% of silica in amorphous form. In this paper we evaluated the effect of replacing carbon black with RHA in a basic tread formulation. Compounds mechanical, dynamic properties and morphology were analyzed. Keywords: dynamic mechanical thermal analysis, mechanical properties, rice husk ash, rolling resistance, scanning electron microscopy.

1. Introduction Carbon black is the most widely used filler in rubber compounds. However, in the tire industry it is being replaced by silica in recent decades due to the development of “green tires”, which have lower rolling resistance as well as better abrasion resistance (durability) and wet grip (safety)[1-4]. As the tire deforms under the vehicle’s weight during movement, a part of the mechanical energy available to turn the wheels is elastically stored while another part is dissipated as heat (hysteresis loss). This lost energy results in higher fuel consumption. Rolling resistance (RR) can be defined as the mechanical energy converted into heat by a tire moving over a unit of distance on the roadway. This lost energy is related to the viscoelastic properties of the rubber compounds, with a special role being played by the polymers and fillers used in the tread’s compound[4,5]. The use of amorphous precipitated silica in combination with a silane coupling agent as filler in tread compounds can result in fuel savings of approximately 3% to 4%, compared to a tire having treads made from compounds with only carbon black. This means a 20% reduction of the rolling resistance and consequently lower greenhouse gas emissions. This has a considerable positive impact because vehicles are responsible for about 20% of CO2 emissions into the atmosphere[6-8]. The retread industry is a technology follower of the tire industry, but even though aware of the importance of searching for new formulations, a significant portion of companies still mainly use carbon black in the tread recipe. This happens because the environmental and economic

Polímeros, 27(1), 55-61, 2017

incentives of silica technology do not outweigh the higher cost of the process and the raw material involved. Therefore, one possible solution being evaluated is the use of mixtures of carbon black, silica and alternative fillers. Rice is one of the most important food crops, with annual world production of approximately 670 million metric tons. During rice milling, around 22% by weight of rice husk is generated, a material that is mainly used as fuel for energy generation, resulting in ash. Rice husk ash (RHA) can contain over 70% silica in amorphous form, along with some amount of metallic impurities. Even though many applications are being studied for its use around the world, its disposal still represents an environmental issue[9-11]. One of these applications is its use as filler in rubber compounds[11-13]. RHA can also be used as a raw material for precipitated silica production, by, for example, a process based on the reaction of silica with aqueous sodium hydroxide (NaOH) in an autoclave, followed by precipitation with an acid, filtration, washings and drying stages[14,15]. There is already a tire company developing a similar process to prepare its own silica[16]. The aim of this paper is to evaluate the effect of replacing a part of carbon black with RHA in a basic tread formulation. Tread rubber compounds were prepared in lab scale and their mechanical and dynamic properties were analyzed. The morphology of the compounds was also compared. Finally, RHA was characterized for its particle size, specific area and carbon and moisture contents in order to allow interpretation of the data obtained.

O O O O O O O O O O O O O O O O

Fernandes, M. R. S., Sousa, A. M. F., & Furtado, C. R. G. at 80 °C and mixed for seven minutes. At the end of this period, the mixture was dumped.

2. Materials and Methods 2.1 Materials The RHA used was supplied by Frenzel Indústria de Borrachas e Plásticos Ltda. The grade used was “super fine” (SF). It comes from a factory in the Brazilian state of Rio Grande do Sul, obtained by a fluidized bed process. Emulsion styrene butadiene rubber (Buna SE 1712 TE), composed of 76.5% butadiene and 23.5% styrene and extended with 37.5 phr of a treated residual aromatic extract (TRAE); oil with low content of polycyclic aromatic hydrocarbons; and high cis-1.4 polybutadiene (Buna CB 24), obtained by solution polymerization using a Neodimiun based catalyst with cis content above 96%, were supplied by Lanxess Elastômeros do Brasil. Carbon black N234 was produced by Cabot Brasil e Comércio S.A.; amorphous precipitated silica (Zeosil 1165MP) was supplied by Rhodia; and TRAE oil (Fluibrax Euro 40) was acquired from Petróleo Brasileiro S.A. Bis(3-ethoxysilylpropyl) tetrasulfide (TESPT) was obtained from Evonik Degussa Brasil Ltda. The other materials used were commonly used materials in the rubber industry.

2.2 Experimental procedure A typical tread formulation with carbon black (CB) was used as the basic reference. In order to evaluate the effect of rice husk ash (RHA) on the tread properties, compounds were prepared in which several parts of carbon black were replaced by equal amounts of RHA. Additionally, a compound using only commercial high dispersion silica (SIL) was also prepared to help interpretation of the analytical data. The formulations are shown in Table 1. The compounds were prepared in a 2-liter tangential Banbury mixer with rotation of 80 rpm. Different mixture procedures were used, depending on the filler composition, as described below: a) The compound containing only carbon black (CB) was prepared in a single stage, in which all components but the acceleration system were fed into the Banbury mixer

b) Compounds with a mixture of carbon black and RHA were prepared in two stages. In the first one, all the ingredients, but the acceleration system, were added into the Banbury mixer at 80 °C. Then the temperature was increased to 150 °C and maintained at that level for three minutes for the silanization reaction to occur. After this period, the mixture was dumped. c) The compounds with 100% RHA and the other with 100% commercial silica were also prepared in two stages. In the first stage, all the ingredients, but the acceleration system, were added in the Banbury mixer at 80 °C. The temperature was raised to 150 °C and maintained at this level for 2.5 minutes. Then the mixture was dumped and milled for two minutes in an open mill. The mixture obtained was returned to the Banbury mixer and heated for 2.5 min, after which zinc oxide was added one minute before the material was dumped. After all mixtures were dumped from the mixer and milled for 2 minutes and then allowed to rest at room temperature (around 25 °C) for at least 16 hours before being accelerated in an open mill.

2.3 Characterization The carbon and moisture contents of the rice husk ash were determined by thermogravimetric analysis (TGA) (TA Q50 analyzer, from TA Instruments) with an initial temperature of 50 °C, heating rate 10 °C/min until 800 °C with N2 atmosphere, followed by introduction of air flow until the temperature reached 950 °C with a ramp of 10 °C/min. The particle size distribution of the rice husk ash was analyzed by laser diffraction (Mastersizer 2000, Malvern Instruments Co. Ltd.). The BET surface area was determined by using an accelerated surface area/porosimetry system (Micromeritics ASAPTM 2020).

Table 1. Tread rubber compounds in phr*. Component BUNA SE 1712 TE BUNA CB 24 Carbon Black (N 234) RHA (SF) Silane (Si-69) Silica (Zeosil 1165MP) Sulfur Zinc oxide Stearic acid TBBS DPG Vulkanox 4020/LG Vulkanox HS/LG Oil (Fluibrax Euro 40) Wax (Antilux 653) Process additive (Aflux 37)

CB 96.3 30 90 2.0 3.0 1.0 1.3 0.3 2.0 2.0 11.2 2.0 -

RHA 22.5 96.3 30 67.5 22.5 1.8 2.0 3.0 1.0 1.3 2.0 2.0 2.0 11.2 2.0 -

RHA 45 96.3 30 45.0 45.0 3.6 2.0 3.0 1.0 1.3 2.0 2.0 2.0 11.2 2.0 -

RHA 67.5 96.3 30 22.5 67.5 1.8 2.0 3.0 1.0 1.3 2.0 2.0 2.0 11.2 2.0 -

RHA 96.3 30 7.0 90.0 7.2 2.0 3.0 1.0 1.3 2.0 2.0 2.0 11.2 2.0 2.0

SIL 96.3 30 7.0 7.2 90.0 2.0 3.0 1.0 1.3 2.0 2.0 2.0 11.2 2.0 2.0

*Parts per hundred rubber.

Polímeros, 27(1), 55-61, 2017

Rice husk ash as filler in tread compounds to improve rolling resistance The Mooney viscosities of the crude rubber and compound (ML1+4) were determined following the ASTM D1646 method at 100 °C in an Alpha Technologies viscometer (MV 2000). Rheometry was determined in an MDR rheometer from Alpha Technologies (MDR 2000P), with 100 cpm frequency, 0.5°, temperature of 160 °C, during 30 minutes, for measurement of t’90. Analysis of the tensile properties followed the ASTM D412 method, using an Instron 5581 tensiometer. Shore A hardness was measured following the ASTM D2240 method in a Bareiss durometer while rebound was determined following the ASTM D1054 method in a Zwick & Co KG device. Abrasion resistance was measured following the DIN53516 method using a Maqtest device. Dynamic tests were conduct in a dynamical mechanical analyzer (DMA 50N 01 Db – METRAVIB) in tensile mode. The temperature dependence (temperature sweep) of the loss factor (tan delta) and the storage modulus (G’) were measured in the temperature range from -60 °C to 80 °C, at a heating rate of 3 °C/min, frequency of 10Hz and dynamic strain of 1%. The Payne effect was evaluated by strain dependence of the storage modulus (G’) using the DMA at 1Hz and 60°.

3. Results and Discussions 3.1 Characterization of the rice husk ash The RHA sample received from Frenzel presented a carbon content of 12.4% as determined by TGA (Figure 1). This carbon content was considered high since the material was produced with a fluidized process. It might be related to the presence of potassium in the husks, which acts at the melted surface, causing fixation of carbon in the ash. The RHA average particle size was determined by laser diffraction in a Martersizer 2000 with a mean value of 6.97 microns and a standard deviation of 0.137. The BET specific area was 81m2/g.

3.2 Properties of the compounds Table 2 shows the results of the properties determined for the formulations. A reduction can be observed in reinforcement as carbon black was replaced by the RHA in the tread formulation. This fact can also be observed through comparison between the RHA and SIL compounds. This behavior can be explained by the higher particle size of RHA and consequently the smaller specific surface area presented by this ash (81 m2/g) in relation to the other fillers used: carbon black (120 m2/g)

Figure 1. TGA analysis of SF RHA used in this paper. Table 2. Compounds properties. Property Mooney viscosity, MML (1+4) 300% Modulus (MPa) Tensile strenght (MPa) Elongation (%) Hardness (Shore A) Tear strenght, (kN/m)

Polímeros, 27(1), 55-61, 2017

CB 90 13.7 21.3 454 71 37

RHA 22.5 71 15.4 22.2 408 64 35

RHA 45 82 12.5 13.1 318 60 29

RHA 67.5 84 9.0 247 56 24

RHA 61 8.0 227 57 25

SIL 87 15.6 21.8 376 63 39

Fernandes, M. R. S., Sousa, A. M. F., & Furtado, C. R. G. and high dispersion silica (170 m2/g). According to the literature, the classification of fillers into non-reinforcing, semi-reinforcing or reinforcing is strongly dependent on the size of the particles, with 6 microns being the primary particle size at the limit of semi-reinforcing and non-reinforcing fillers[7]. In this way, this ash, with particle size near 7 microns, can be considered a non-reinforcing filler. In relation to the viscosity, the decrease in the compound Mooney viscosity with increasing RHA content was probably also related to this lower specific area, which is associated with lower bound rubber content, reducing the contribution of the in-rubber-structure’s effect to the elastic modulus of the compound[17].

3.3 Morphology of the compounds The SEM analysis of the tensile fracture surface is a technique which allows observing the filler dispersion in a rubber compound. Figures 2-4 show, respectively, the images of RHA, RHA 45 and SIL compounds. From Figures 2a and 3a, it is possible to verify the large size and broad distribution of the RHA particles,

Figure 2. (a) SEM photomicrograph of the tensile fracture surface of RHA; (b) EDS analysis of RHA. 58

which is confirmed by EDS analysis (Figures 2b and 3b). The particle size of RHA is in accordance with the results obtained by laser refraction. The presence of carbon black cannot be observed due to the fact that its particles have nano dimensions, which are not observable with SEM resolution. According to Figure 4, as expected, the silica aggregates are not visible from this technique. Only some small agglomerates are visible, which may be present even with the use of silane due to incomplete coverage of all particle surfaces. The identification of the small white points was carried out with EDS to identify the elements present in them, which confirmed the concentration of silicon.

3.4 Rebound and abrasion resistance evaluation Figure 5 shows the results of rebound and DIN abrasion of the compounds. It can be observed the rebound increases as the RHA content increased in the formulation. As the filler’s specific surface area is reduced, a reduction of the filler-filler interaction can be expected, with reduction of energy loss, resulting in a more elastic performance of the material[7]. Due to the same reason, a reduction of the bound rubber can also be expected, which also contributes to the elastic response, as it acts as an immobilized rubber. The abrasion resistance is related to the rupture of small

Figure 3. (a) SEM photomicrograph of the tensile fracture surface of RHA45; (b) EDS analysis of RHA45. Polímeros, 27(1), 55-61, 2017

Rice husk ash as filler in tread compounds to improve rolling resistance particles of the compound under the action of frictional forces, when sliding takes place between the compound surface and a substrate. It is normal to expect a compound containing fillers with large particle size to present a lower filler-polymer interaction, leading to lower cohesive forces and the removal of a significant amount of material. Moreover, in the case of silica-silane as filler, the formation of the covalent link between the filler and the polymer plays an important role in abrasion resistance, being reduced in this case due

to lower specific area and the presence of a relatively high quantity of carbon (12%) on the surface. Unfortunately, the surface activity of the RHA and commercial silica could not be compared by the available techniques used in these experiments.

3.5 Dynamic mechanical analysis of the compounds’ vulcanizates Dynamic mechanical analysis is usually used to evaluate tire performance. It is well recognized that high values of tan delta at 0 °C mean better wet grip whereas low values of tan delta at 60 °C are correlated with lower rolling resistance (lower fuel consumption)[4]. Figure 6 presents the temperature dependence of tan delta in the range of 50 to 70 °C for the prepared compounds. It can be noted that silica significantly reduces the value of tan delta at 60 °C in relation to carbon black in the compounds, as already stated in Michelin’s patent, indicating a reduction of hysteresis, resulting in low fuel consumption[3]. Similarly, the compounds with RHA also presented a reduction in tan delta values at 60 °C. This result is produced by a lower filler-filler interaction resulting from lower specific surface area of RHA, with formation of a weak filler network. In a rolling tire, this breakdown of the filler network during deformation would result in dissipation of energy, creating hysteresis and therefore increased rolling resistance. So, the weak filler network results in a low dissipation of energy, which provides more elastic performance of the compound, with a lower tan delta and lower fuel consumption. Lower values of tan delta for RHA in relation to commercial silica can be observed, which is also related to a weaker filler network. As Zeosil 1165 has a higher specific area, the interaction between filler particles is more intense, resulting in stronger filler-filler interaction.

Figure 4. (a) SEM photomicrographs of the tensile fracture surface of SIL; (b) Punctual EDS analysis of one of the white points identified in the image.

The Payne effect, the variation in modulus observed at small deformation amplitude during strain sweep test, is widely used to evaluate the filler network in rubber compounds, being related to filler dispersion as well as the filler-filler interaction. The small delta G’ values (difference between low strain G’ and high strain G’) are related to a weaker filler-filler interaction and consequently lower heat generation during deformation of the treads during rolling, and are expected as an indication of good dispersion (better polymer-filler interaction) and better performance in

Figure 5. Results of DIN abrasion and rebound from the studied recipes.

Figure 6. Tan delta versus temperature for the range of 50 to 70 °C.

Polímeros, 27(1), 55-61, 2017

Fernandes, M. R. S., Sousa, A. M. F., & Furtado, C. R. G.

6. References

Figure 7. G’ versus strain test at 60 °C – Payne Effect.

relation to fuel economy. A clear tendency of this behavior can be observed in Figure 7 as the content of RHA increases in the prepared compounds. The weaker filler-filler interactions observed with the use of RHA makes breakage of the network easier when the compound is submitted to low strains. It can also be observed that the delta G’ value for the compound prepared with commercial silica is higher than the one for the sample prepared with 100% RHA, which is related to the higher specific area and stronger filler-filler interaction in this compound.

4. Conclusions The results obtained allow concluding that the use of RHA in tread compounds significantly reduces the reinforcement in relation to commonly used fillers: carbon black and silica. This can be observed by reduction of tensile strength, elongation and tear resistance. It is well known that reinforcement is related to the surface area of the filler and it was verified that RHA sample used had a much lower surface area in comparison with the other fillers tested. Abrasion resistance is a critical property for tire and retread segments and it was also verified that RHA decreases this property. So, based on the data obtained, it can be concluded that this type of RHA acted as non-reinforced filler and should not be considered as a substitute for silica in a tread formulation, even with good indications of rolling resistance reduction (lower tan delta at 60 °C). The formulation with 25% replacement of carbon black by RHA showed satisfactory physical properties and abrasion resistance, also providing reduced rolling resistance (low tan delta at 60 °C), which can justify a more complete evaluation of this compound, including outside tests for abrasion evaluation.

5. Acknowledgements We would like to thank Lanxess Elastomeros do Brasil S.A. for making all the tests possible, all suppliers for donating samples and Instituto Nacional de Tecnologia – INT for the support in MEV analysis. 60

1. Eirich, F. R., Erman, B., & Mark, E. J. (2005). The science and technology of rubber. New York: Elsevier. 2. Li, Y., Han, B., Liu, L., Zhang, F., Zhang, L., Wen, S., Lu, Y., Yang, H., & Shen, J. (2013). Surface modification of silica by two-step method and properties of solution styrene rubber (SSBR) nanocomposites fillied with modified silica. Composites Science and Technology, 88, 69-73. http://dx.doi. org/10.1016/j.compscitech.2013.08.029. 3. Rauline, R. (1993). US Patent 5,227,425. Retrieved in 23 December 2015, from http://www.google.com/patents/ US5227425 4. Fernandes, M., & Santos, N. (2013). SBR em solución para neumaticos de alto rendimento. In XII Jornadas Latinoamericanas de Tecnologia Del Caucho. Buenos Aires: SLT Caucho. Retrieved in 23 December 2015, from http://www.sltcaucho. org/jornadas2013/trabajos-tecnicos/#inline_content-conferencias 5. Transportation Research Board. (2006). Tires and passenger vehicle fuel economy (Special Report, vol. 286). Washington: National Research Council of the National Academies. Retrieved in 23 December 2015, from http://onlinepubs.trb. org/onlinepubs/sr/sr286.pdf 6. Dierkes, W. (2005). Economic mixing of silica-rubber compounds Interaction between the chemistry of the silica-silane reaction and the physics of mixing (Doctoral thesis). University of Twente, Enschede. 7. Mihara, S. (2009). Reactive processing of silica-reinforced tire rubber, new insight into the time- and temperature-dependence of silica rubber interaction mixing (Doctoral thesis). University of Twente, Enschede. 8. Dierkes, W. K, Noordermeer, J. W. M. (2012). Annual showcase. Tire Technology International, 14-18. 9. Pouey, M. T. F. (2006). Beneficiamento da cinza da casca de arroz residual com vistas à produção de cimento composto e/ ou pozolânico (Master’s dissertation). Universidade Federal do Rio Grande do Sul, Porto Alegre. 10. Ismail, H., Nasaruddin, M. N., & Rozman, H. D. (1999). The effect of multifunctional additive in white rice husk ash filled natural rubber compounds. European Polymer Journal, 35(8), 1429-1437. http://dx.doi.org/10.1016/S0014-3057(98)00223-7. 11. Chandrasekhar, S., Pramada, P. N., & Praveen, L. (2005). Effect of organic acid treatment on the properties of rice husk silica. Journal of Materials Science, 40(24), 6535-6544. http:// dx.doi.org/10.1007/s10853-005-1816-z. 12. Costa, H. M., Furtado, C. R. G., Nunes, R. C. R., & Visconte, L. L. Y. (2001). Rice-husk-ash-filled natural rubber. ii. partial replacement of commercial fillers and the effect on the vulcanization process. Journal of Applied Polymer Science, 83, 2485-2493. http://dx.doi.org/10.1002/app.11514. 13. Ismail, H., Nasaruddin, M. N., & Ishiaku, U. S. (1999). White rice husk ash filled natural rubber compounds: the effect of multifunctional additive and silane coupling agents. Polymer Testing, 18(4), 287-298. http://dx.doi.org/10.1016/S01429418(98)00030-0. 14. Kalapathy, U., Proctor, A., & Shultz, J. (2000). A simple method for production of pure silica from rice hull ash. Bioresource Technology, 72(3), 257-262. http://dx.doi.org/10.1016/S09608524(99)00127-3. 15. Furtado, C., Mansur, C., Sousa, A. M., & Visconte, L. (2009). Silica sol obtaained from rice husk ash. Chemistry & Chemistry Technology, 3, 331-336. Retrieved in 23 December 2015, from ena.lp.edu.ua:8080/handle/ntb/3307 16. Pirelli. (2012, June 18). Pirelli apresenta na Rio + 20 projetos com o ministério do meio ambiente italiano e o Estado de São Paulo, visando reduzir o impacto ambiental na produção e uso Polímeros, 27(1), 55-61, 2017

Rice husk ash as filler in tread compounds to improve rolling resistance de pneus. Retrieved in 23 December 2015, from http://www. pirelli.com/tyre/br/pt/news/2012/06/18/pirelli-apresenta-nario-20-projetos-com-o-ministerio-do-meio-ambiente-italianoe-o-estado-de-sao-paulo-visando-reduzir-o-impacto-ambientalna-producao-e-uso-de-pneus 17. FrĂśhlich, J., Niedermeier, W., & Luginsland, H.-D. (2005). The effect of filler-filler and filler-elastomer interaction on

PolĂmeros, 27(1), 55-61, 2017

rubber reinforcement. Composites. Part A, Applied Science and Manufacturing, 36(4), 449-460. http://dx.doi.org/10.1016/j. compositesa.2004.10.004. Received: Dec. 23, 2015 Revised: Apr. 21, 2016 Accepted: May 17, 2016

http://dx.doi.org/10.1590/0104-1428.2404

O O O O O O O O O O O O O O O O

Characterization of biopolymers and soy protein isolate-high-methoxyl pectin complex Mírian Luisa Faria Freitas1*, Kivia Mislaine Albano1 and Vânia Regina Nicoletti Telis1 Laboratory of Physical Measures, Department of Food Engineering and Technology, Universidade Estadual Paulista “Júlio de Mesquita Filho” – UNESP, São José do Rio Preto, SP, Brazil

*mirianlfreitas@yahoo.com.br

Abstract This study aimed at characterizing the soy protein isolate and high-methoxyl pectin biopolymers individually, and the complexes formed by both at different proportions and pHs in order to find the most suitable pH and biopolymer ratios to food application as stabilizers. The biopolymers were evaluated through solubility, charges, turbidimetry, and optical microscopy analyses; the systems with the pair of biopolymers were analyzed through turbidimetry and optical microscopy. High-methoxyl pectin showed high solubility at all pHs investigated. The soy protein isolate showed low solubility at pH 4.5, which is close to its isoelectric point, and complete solubility at pH 11.0. The formation of complexes suggested an attractive interaction between the biopolymers, with high absorbance reading values and images of complexes from optical microscopy. These complexes were present in systems with pHs below the soy protein isolate’s isoelectric point, with positive charges; the high-methoxyl pectin, however, had negative ones. Keywords: attractive interaction, morphology, solubility, turbidimetry, zeta potential.

1. Introduction Soy has been the most used material for industrial production of protein concentrates and isolates due to its high amount of proteins and its products’ good technological performance. It also represents a vegetable alternative to lactic proteins[1]. The soy protein isolate (SPI) contains at least 90% protein; it is thus virtually free from lipids and carbohydrates[2]. Pectin, probably the most complex natural macromolecule, is the most common stabilizer used in protein-based acidic beverages, positively contributing to the final product’s taste, stability, and texture, even if it is added in small amounts[3-5]. Although pectins are part of the majority of plant tissues, the number of commercial sources used is very limited[4]. The mainly polysaccharide used is derived from citric fruit and classified regarding the degree of esterification in: high-methoxyl pectins, when a half or more carboxyl groups are esterified, and low-methoxyl pectins, when less than a half the carboxyl groups are esterified[6]. The formation of complexes between proteins and polysaccharides with opposite charges is a colloidal phenomenon involved in the structuring of several biological systems. There has been increasing interest in complexes formed by these biopolymers recently due to their potential applications in the food industry, being used as stabilizers in milk-based beverages, emulsifiers, foam stabilizers, fat replacers, besides being used in encapsulation, enzyme immobilization and recuperation, and protein separation processes, as explained by Dong et al.[7] in revision of the literature. Lam et al.[6] carried out a study on pectin stability in protein acidic solutions, in which they used soy protein isolate. It was found that high-methoxyl pectin showed higher stability

than low-methoxyl pectin. Jaramillo et al.[8] observed that, at a pH near the isoelectric point of the SPI (around pH 4.0), as pectin was added the protein solubility increased, which prevented its aggregation through an electrostatic interaction. They also verified that the resulting protein‑polysaccharide complexes could bear thermal treatment, although a few changes in their properties occurred. When the pair was used in emulsions, the polysaccharides increased the emulsion’s physical stability through electrostatic and/or steric effects, because they modify the rheological properties of the interface and increase the viscosity of the emulsion[9]. Values of pH at which there are solubilization or biopolymer complexes formation do not depend or depend very little on the total concentration of the biopolymers. On the other hand, they are strongly related to the isoelectric point of the protein, the ratio between the biopolymers, their ionic strength and molar mass[10-12]. Given this context, the aim of this study was to characterize the soy protein isolate and high-methoxyl pectin biopolymers through solubility, charges, turbidimetry, and optical microscopy analyses, besides characterizing the complex formed by the pair at different proportions and pHs, evaluating turbidimetry and optical microscopy.

2. Materials and Methods 2.1 Materials The biopolymers used were soy protein isolate (SPI) (Tovani Benzaquen IngredientesTM) and high-methoxyl pectin (HM) (CP KelcoTM). For preparation of solutions, sodium azide was also used (DinâmicaTM) and for pH

Polímeros, 27(1), 62-67, 2017

Characterization of biopolymers and soy protein isolate-high-methoxyl pectin complex

Biopolymers were characterized, individually, through solubility, zeta potential, turbidimetry, and optical microscopy analyses at different pHs. The complex high-methoxyl pectin-soy protein isolate was characterized through turbidimetry analyses and optical microscopy at different proportions and pHs.

Turbidimetry analysis for each system was carried out according to methodology proposed by Antonov and Zubova[15], and Marfil[12]. After obtaining the desired pH, it were measured the absorbance reading values of the aliquots in a spectrophotometer (Biospectro SP-220) at wavelength 590 nm. According to the authors, the time between the adjustment of desired pH and absorbance reading in the spectrophotometer cannot be longer than 10 s, because there may be precipitation of the formed complexes and interference in the readings after this time.

2.2.1 Solubility

2.2.5 Morphology

Biopolymers’ solubility was determined, in triplicate, at pHs between 3.0 and 11.0 (± 0.05), according to methodology proposed by Cano-Chauca et al.[13] with some modifications. It was weighed 0.4 g of the biopolymer and completed up to 40 mL of solution with deionized water. The solutions were then moved to the shaker (Marconi, MA 830/A) for agitation for 3 hours and were left to hydrate during one night at room temperature. The next day, the desired pH was adjusted, the solutions were centrifuged at 3000×g for 5 minutes (Hermle, Z 326 K) and 20 g supernatant were transferred to previously dried Petri dishes. These were placed in a vacuum drying oven (Marconi, MA 030) at 60 °C for 48 h. Solubility was calculated by difference in weight.

The morphology of individual solutions and elaborated systems with SPI:HM proportions of 1:1, 2:1, 3:1, 4:1, with a biopolymer concentration of 0.05%, was verified for pHs between 3.0 and 7.0 (± 0.05) using an optical microscope (Olympus CX31) with 40x magnifying lenses coupled to a digital camera (Olympus SC30).

adjustment, either 1N hydrochloric acid (DinâmicaTM) or sodium hydroxide (DinâmicaTM) solutions were used.

2.2 Methods

2.2.2 Preparation of stock solutions Stock solutions of the soy protein isolate were prepared by solubilizing the biopolymer in deionized water and adjusting its pH to 11.0 (± 0.05), for a complete solubilization, according to the solubility result at that pH and as suggested by Jaramillo et al.[8]. Then they were stirred for 3 h in a magnetic stirrer and allowed to hydrate during one night for a complete hydration. The solutions had their desired pHs adjusted as the analyses were carried out. Stock solutions of high-methoxyl pectin were prepared by solubilizing them in deionized water for 3 h in magnetic stirrer and hydration during one night at room temperature. The solutions had their desired pHs adjusted as the analyses were carried out.

3. Results and Discussion 3.1 Solubility As observed in Figure 1, high-methoxyl pectin is completely soluble, disregarding the pH of the solution. On the other hand, the soy protein isolate shows low solubility, around 10%, at its isoelectric point (between pH 4.0 and 5.0) and the solubility increases as it distances from this point, particularly in more alkaline conditions, reaching 100% at pH 11.0. Similar results were found by both Jaramillo et al. , and Renkema et al.[16] when studying the soy protein behavior at pHs from 3.0 to 7.0, and 2.0 to 8.0, respectively, confirming the pH between 4.0 and 5.0 as the isoelectric point of soy protein isolate, which is characterized by the lowest solubility due to charge neutralization. [8]

According to Malhotra and Coupland[17] and Jaramillo et al. , the poor solubility of soy protein isolate around its isoelectric point can limit its application in acid foods. The soy protein isolate presents better functionality in conditions of [8]

It was added 0.04% sodium azide in stock solutions to avoid microorganisms growth. 2.2.3 Zeta potential The charge analysis of soy protein isolate and high-methoxyl pectin biopolymers in 0.02% solution was carried out at pHs between 3.0 and 7.0 (± 0.05). It was used a zeta potential analyzer (ZetaPALS), according to methodology proposed by Perrechil and Cunha[14]. Measurements were obtained in triplicate. 2.2.4 Turbidimetry Solutions containing 0.05% soy protein isolate or high-methoxyl pectin were analyzed individually. Besides, different quantities from each biopolymer solutions were mixed in order to obtain systems with SPI:HM proportions of 1:1, 2:1, 3:1, and 4:1, with a final concentration of 0.05%. Individual solutions and systems were analyzed at pHs between 3.0 and 7.0 (± 0.05). Polímeros, 27(1), 62-67, 2017

Figure 1. Solubility (%) of soy protein isolate and high-methoxyl pectin vs pH. 63

Freitas, M. L. F., Albano, K. M., & Telis, V. R. N. higher solubility, since it can help emulsifying hydrophobic compounds as well as binding water in food systems. Jaramillo et al.[8] observed that the addition of pectin increased the solubility of soy protein isolate close to its isoelectric point and prevented the formation of very large aggregates. Moreover, thermal treatment (30 min, 90 °C) enhanced the solubility of the soy protein isolate-pectin complexes close to the isoelectric point of protein. Rocha et al.[18] studied biodegradable composite films based on cassava starch and soy protein and verified that the increase in pH of the filmogenic solution favored solubility, possibly due to the distance of the soy protein isoelectric point, where the maximum solubility of the film was observed at pHs between 10 and 12. Therefore, soy protein isolate solutions used in the other analyses were prepared at pH 11.0 and allowed to hydrate. Only after these steps they had their pH adjusted according to the necessity of the analysis.

3.2 Zeta potential As reported by other authors, such as Harnsilawat et al. , in the present study it was also observed that for polysaccharydes solutions such as high-methoxyl pectin when the pH increases, negative charges also increase, until this value reaches a plateau. To what proteins are concerned, below their isoelectric point, they acquire positive charges, whereas above this point, they acquire negative charges. As observed in Figure 2, for the soy protein isolate, the charges close to zero were observed between pHs 4.0 and 5.0, and the curve reaches zero at pH close to 4.6, which confirms its isoelectric point. Lam et al.[6] and Jaramillo et al.[8] reported similar results in their studies on charges of soy protein isolate; it was observed that the charges were zero at pH 4.5 and 4.4, respectively. It is possible to assume that at pHs below the protein’s isoelectric point, the interaction between the protein and the pectin is attractive, once the pectin is negatively charged and the protein is positively charged[6]. For the studied biopolymers, at pH 3.5 it is possible that there is an attractive interaction and formation of complexes due to the fact that they have opposite charges and at quite high values.

particulate matter. On the other hand, for pHs lower than 4.0 and higher than 5.0, the absorbance reading values for this biopolymer were lower, resulting along higher solubility values and, consequently, lower quantity of precitpitated matter. For the systems in which the biopolymers were present at different concentrations, as solutions became more alkaline, the absorbance reading values became lower, which suggests a lower attractive interaction between them and lower complex formation. This result confirms the one obtained from the biopolymers charges analysis, in which at pH higher than 4.5 the absorbance reading values were lower than 0.1 a.u. for all analyzed proportions. For the same pH, the increase in proportion of soy protein isolate was followed by an increase in the absorbance reading value, which leads to the conclusion that a higher complex formation occurred, mainly in more acid solutions. It is noteworthy that the systems at pH 3.0 generated absorbance reading values higher than the systems at pH 3.5. Even though they suggested a higher complex formation at pH 3.0, the systems were less stable than at pH 3.5, with a

[19]

Figure 2. Influence of pH on zeta potential (mV) of solutions of soy protein isolate and high-methoxyl pectin.

3.3 Turbidimetry The absorbance readings at different pHs for the biopolymer solutions and for the systems with different proportions of soy protein isolate:high-methoxyl pectin (SPI:HM) are shown in Figure 3. As expected, the solution containing only high-methoxyl pectin presented a low and constant absorbance reading value for all studied pHs. Once its solubility does not depend on the pH, there was no phase separation nor precipitation. This behavior was not observed for the soy protein isolate, which presented high absorbance values at pHs between 4.0 and 5.0. These higher absorbance values resulted along the lowest solubility values, which suggests that the solution’s turbidity is connected with the suspended 64

Figure 3. Influence of pH on absorbance reading, at wavelength 590 nm, for individual biopolymer solutions and the pair at different proportions. Polímeros, 27(1), 62-67, 2017

Characterization of biopolymers and soy protein isolate-high-methoxyl pectin complex

Figure 4. Images of systems at pH 3.0 with SPI:HM proportions of 1:1 (a), 2:1 (b), 3:1 (c), and 4:1 (d), obtained through optical microscopy with 40x magnifying lenses.

Figure 5. Images of systems at pH 3.5 with SPI:HM proportions of 1:1 (a), 2:1 (b), 3:1 (c), and 4:1 (d), obtained through optical microscopy with 40x magnifying lenses. PolĂmeros, 27(1), 62-67, 2017

Freitas, M. L. F., Albano, K. M., & Telis, V. R. N. precipitation and phase separation in few minutes, which is undesirable in food systems such as emulsions. Based on these remarks, solutions at pH 3.5 were considered ideal for an attractive interaction and complex formation between the studied biopolymers to happen, being suitable to elaborate stable food systems, such as acidic beverages based on protein or emulsions. Jasentuliyana et al.[20] studied the interaction between the soy protein isolate and citric pectin with the objective of enhancing the use of the soy protein isolate as a turbidity agent in acidic beverages (pH 3.7). The authors then separated the soy protein isolate in two fractions, a hydrophobic and a hydrophilic one and evaluated the interaction between these and pectin through turbidity studies. They did not observe significant differences between the two fractions and reported that the higher the protein proportion, the higher the solution’s absorbance reading value, confirming the use of the soy protein isolate as an opacity agent. Besides, the systems elaborated at protein:pectin proportion of 2:1 and 1:4 showed the highest stability values throughout 28 days. Another study on the pair formed by the soy protein isolate and high-methoxyl pectin was conducted by Albano and Telis[21], at pH 3.5. The authors confirmed the effect of stability by pectin in protein solutions close to the protein’s isoelectric point. Besides, they observed the formation of small complexes which, when submitted to rheological tests, showed a slightly pseudoplastic behavior, with a tendency to a Newtonian behavior. When the biopolymer solutions were submitted to ultrasound, it was observed that the complexes had their sizes reduced and suffered a consequent reduction of the phases separation after 24 h. In a study on the interaction between soy protein and gum Arabic conducted by Dong et al.[7] it was found that, at pH 3.0, the addition of gum arabic in a soy protein solution increased the system’s absorbance value. Moreover, the systems with a protein:polysaccharide proportion of 1.5:1 and 3:1 were instable.

3.4 Morphology Through images obtained from optical microscopy, it was possible to observe the morphology of both the biopolymer solutions individually and the systems at different proportions for the studied pHs, confirming the results obtained through solubility, charges, and turbidimetry tests. For the soy protein isolate solution, the formation of aggregates was observed at pHs between 4.0 and 5.0, at which lower solubility and higher absorbance values were noted. For the remaining pHs, there was no presence of complexes. The pectin solution was solubilized at all different pHs, and aggregates were not noted. For systems containing different biopolymers proportions, there was a higher formation of complexes at pHs 3.0 and 3.5, and those increased as the soy protein isolate concentration increased in the solution. These results are observed for pH 3.0 in Figure 4 and for pH 3.5 in Figure 5. Although the complexes were smaller in pH 3.5, as expected by the results of turbidimetry, they were more soluble and more stable being suitable to elaborate food systems. 66

The systems in higher pH solutions did not present formation of complex, thus confirming a lower attractive interaction between the biopolymers.

4. Conclusions Through tests for characterization of the biopolymers, it was observed that high-methoxyl pectin showed high solubility, disregarding pH, and that negative charges increased as pH increased until they reached a plateau. The soy protein isolate showed low solubility at its isoelectric point which increased in alkaline solutions, until it reached 100% at pH 11.0. Besides, positive charges below the isoelectric point and negative ones above this point were found. In solutions with a pH lower than the isoelectric point, an attractive interaction between the soy protein isolate and high-methoxyl pectin was verified by analyzing the formation of complexes. These complexes were bigger for the systems with a higher protein proportion. The complexes formed in pH 3.5, in the different ratios, have potential application in the food industry, for example, as emulsifiers, foam stabilizers, fat replacers or being used in encapsulation.

5. Acknowledgements The authors acknowledge Sao Paulo Research Foundation, FAPESP (Processes 2014/08520-6, 2013/10842‑9 and 2014/02910-7) and Coordination for the Improvement of Higher Level Personnel, CAPES.

6. References 1. Tömösközi, S., Lásztity, R., Haraszi, R., & Baticz, O. (2001). Isolation and study of the functional properties of pea proteins. Die Nahrung, 45(6), 399-401. PMid:11712241. http:// dx.doi.org/10.1002/1521-3803(20011001)45:6<399::AIDFOOD399>3.0.CO;2-0. 2. Lam, M., Paulsen, P., & Corredig, M. (2008). Interactions of soy protein fractions with high-methoxyl pectin. Journal of Agricultural and Food Chemistry, 56(12), 4726-4735. PMid:18517218. http://dx.doi.org/10.1021/jf073375d. 3. Canteri, M. H. H., Moreno, L., Wosiacki, G., & Scherr, A. P. (2012). Pectina: da matéria-prima ao produto final. Polímeros: Ciência e Tecnologia, 22(2), 149-157. http://dx.doi.org/10.1590/ S0104-14282012005000024. 4. Santos, M. S., Petkowicz, C. L. O., Haminiuk, C. W. I., & Cândido, L. M. B. (2010). Polissacarídeos extraídos da gabiroba (Campomanesia xanthocarpa Berg): propriedades químicas e perfil reológico. Polímeros: Ciência e Tecnologia, 20, 352-358. http://dx.doi.org/10.1590/S0104-14282010005000056. 5. Tromp, R. H., de Kruif, C. G., van Eijk, M., & Rolin, C. (2004). On the mechanism of stabilization of acidified milk drinks by pectin. Food Hydrocolloids, 18(1), 565-572. http:// dx.doi.org/10.1016/j.foodhyd.2003.09.005. 6. Lam, M., Shen, R., Paulsen, P., & Corredig, M. (2007). Pectin stabilization of soy protein isolates at low pH. Food Research International, 40(1), 101-110. http://dx.doi.org/10.1016/j. foodres.2006.08.004. 7. Dong, D., Li, X., Hua, Y., Chen, Y., Kong, X., Zhang, C., & Wang, Q. (2015). Mutual titration of soy proteins and gum arabic and the complexing behavior studied by isothermal titration calorimetry, turbidity and ternary phase boundaries. Polímeros, 27(1), 62-67, 2017

Characterization of biopolymers and soy protein isolate-high-methoxyl pectin complex Food Hydrocolloids, 46(1), 28-36. http://dx.doi.org/10.1016/j. foodhyd.2014.11.019. 8. Jaramillo, D. P., Roberts, R. F., & Coupland, J. N. (2011). Effect of pH on the properties of soy protein-pectin complexes. Food Research International, 44(1), 911-916. http://dx.doi. org/10.1016/j.foodres.2011.01.057. 9. Serfert, Y., Schroder, J., Mescher, A., Laackmann, J., Ratzke, K., Shaikh, M. Q., Gaukel, V., Moritz, H. U., Schuchmann, H. P., Walzel, P., Drusch, S., & Schwarz, K. (2013). Spray drying behavior and functionality of emulsions with β-lactoglobulin/ pectin interfacial complexes. Food Hydrocolloids, 31(1), 438445. http://dx.doi.org/10.1016/j.foodhyd.2012.11.037. 10. Mattison, K. W., Brittain, I. J., & Dubin, P. L. (1995). Proteinpolyelectrolyte phase boundaries. Biotechnology Progress, 11(1), 632-637. http://dx.doi.org/10.1021/bp00036a005. 11. Mattison, K. W., Wang, Y., Grymonpré, K., & Dubin, P. L. (1999). Micro and macro-phase behaviour in protein-polyelectrolytes systems. Macromolecular Symposia, 140(1), 53-76. http:// dx.doi.org/10.1002/masy.19991400107. 12. Marfil, P. H. M. (2014). Microencapsulação de óleo de palma por coacervação complexa em matrizes de gelatina/goma arábica e gelatina/alginato (Doctoral thesis). Universidade Estadual Paulista “Julio de Mesquita Filho”, São José do Rio Preto. 13. Cano-Chauca, M., Stringheta, P. C., Ramos, A. M., & CalVidal, J. (2005). Effect of the carriers on the microstructure of mango powder obtained by spray drying and its functional characterization. Innovative Food Science & Emerging Technologies, 6(1), 420-428. http://dx.doi.org/10.1016/j. ifset.2005.05.003. 14. Perrechil, F. A., & Cunha, R. L. (2013). Stabilization of multilayered emulsions by sodium caseinate and κ-carrageenan. Food Hydrocolloids, 30(1), 606-613. http://dx.doi.org/10.1016/j. foodhyd.2012.08.006. 15. Antonov, Y. A., & Zubova, O. M. (2001). Phase state of aqueous gelatin–polysaccharide (1)–polysaccharide (2) systems. International Journal of Biological Macromolecules, 29(2),

Polímeros, 27(1), 62-67, 2017

67-71. PMid:11518577. http://dx.doi.org/10.1016/S01418130(01)00140-4. 16. Renkema, J. M. S., Gruppen, H., & Van Vliet, T. (2002). Influence of pH and ionic strength on heat-induced formation and rheological properties of soy protein gels in relation to denaturation and their protein compositions. Journal of Agricultural and Food Chemistry, 50(21), 6064-6071. PMid:12358481. http://dx.doi.org/10.1021/jf020061b. 17. Malhotra, A., & Coupland, J. N. (2004). The effect of surfactants on the solubility, zeta potential, and viscosity of soy protein isolates. Food Hydrocolloids, 18(1), 101-108. http://dx.doi. org/10.1016/S0268-005X(03)00047-X. 18. Rocha, G. O., Farias, M. G., Carvalho, C. W. P., Ascheri, J. L. R., & Galdeano, M. C. (2014). Filmes compostos biodegradáveis a base de amido de mandioca e proteína de soja. Polímeros: Ciência e Tecnologia, 24(5), 587-595. http:// dx.doi.org/10.1590/0104-1428.1355. 19. Harnsilawat, T., Pongsawatmanit, R., & McClements, D. J. (2006). Characterization of β-lactoglobulin–sodium alginate interactions in aqueous solutions: A calorimetry, light scattering, electrophoretic mobility and solubility study. Food Hydrocolloids, 20(1), 577-585. http://dx.doi.org/10.1016/j. foodhyd.2005.05.005. 20. Jasentuliyana, N., Toma, R. B., Klavons, J. A., & Medora, N. (1998). Beverage cloud stability with isolated soy protein. Journal of the Science of Food and Agriculture, 78(1), 389-394. http:// dx.doi.org/10.1002/(SICI)1097-0010(199811)78:3<389::AIDJSFA130>3.0.CO;2-Z. 21. Albano, K. M., & Telis, V. R. N. (2015). Rheological investigation of ultrasound effect on interactions between soy protein isolate and pectin. In Proceedings of the VII Brazilian Conference on Rheology (pp. 22-25). Curitiba: Universidade Tecnológica Federal do Paraná. Received: Nov. 09, 2015 Revised: Apr. 26, 2016 Accepted: May 17, 2016

http://dx.doi.org/10.1590/0104-1428.00916

O O O O O O O O O O O O O O O O

A quantitative relationship between Tgs and chain segment structures of polystyrenes Xinliang Yu1 and Xianwei Huang1* 1

College of Chemistry and Chemical Engineering, Hunan Institute of Engineering, Xiangtan, Hunan, China *hxw1o3o@126.com

Abstract The glass transition temperature (Tg) is a fundamental characteristic of an amorphous polymer. A quantitative structure-property relationship (QSPR) based on error back-propagation artiﬁcial neural network (ANN) was constructed to predict Tgs of 107 polystyrenes. Stepwise multiple linear regression (MLR) analysis was adopted to select an optimal subset of molecular descriptors. The chain segments (or motion units) of polymer backbones with 20 carbons in length (10 repeating units) were used to calculate these molecular descriptors reflecting polymer structures. The relative optimal conditions of ANN were obtained by adjusting various network paramters by trial-and-error. Compared to the model already published in the literature, the optimal ANN model with [4-7-1] network structure in this paper is accurate and acceptable, although our model has more samples in the test set. The results demonstrate the feasibility and powerful ability of the chain segment structures as representative of polymers for developing Tg models of polystyrenes. Keywords: chain segments, glass transition temperature, polystyrenes, structure-property relationship.

1. Introduction The glass transition temperature (Tg) is known as the glass temperature or the transition temperature between glass and rubber states of amorphous materials. Tg is a fundamental characteristic and is taken as the most crucial property of amorphous polymeric materials[1]. The nature of the theory in the glass and glass transition is unsolved, however, is taken as the deepest and most interesting problem in solid stated theory. Though Tg can be determined experimentally, the discrepancies in reported Tg values in the literature may be quite large, because (1) the transition happens over a comparatively wide temperature range, and (2) many factors affect Tg values, which include the structural, constitutional and conformational features of polymers, molecular weight, and experimental conditions such as the measuring method, duration of the experiment, and pressure during the measurement[2]. In addition, experimental determination of Tgs cannot apply to those polymers that are not yet synthesized. Hence, it is necessary to develop theoretical methods for the prediction of Tgs.

Quantitative structure-property relationship (QSPR) models can be used to predict Tg values of polymers. This approach is based on the assumption that the variation of physicochemical properties of the compounds is dependent on changes of molecular structure, which can be characterized with descriptors. A major goal of QSPR approach is to develop a mathematical relationship between the property of interest and structural features[3]. Some researchers have predicted Tgs of polymers with QSPR models. Van Krevelen[4] predicted Tgs by using the group additive property theory. This method is only applicable to polymers whose contribution values are known. Bicerano[2] developed a more universally QSPR model with R2 (the square of the correlation coefficient R) being 0.95 and standard error (s) being 24.65 K for a data set of

320 polymers. The Tg model was based on the solubility parameter and the weighted sum of 13 topological bond connectivity parameters of the monomer structures. But the model is not validated with the test set. Joyce et al.[5] built models for Tg prediction based on the monomer structures of 360 polymers. The model predicted the Tg values for a test set of polymers with a root mean square (rms) error of 35 K. Katritzky et al.[6] introduced a four-parameter model with R2 0.928 for 21 medium molecular weight polymers and copolymers based on their repeat units. On a larger data set, Katritzky et al.[7] developed a QSPR for the molar glass transition temperature (Tg/M) of 88 uncross-linked linear homopolymers. The model has five molecular descriptors and the s for Tg is 32.9 K. On the same data, Cao and Lin[8] developed a QSPR model (R2 = 0.9056) by using ﬁve molecular descriptors that focus on the inﬂuence of chain stiffness and intermolecular forces. Yu et al.[9] developed stepwise multiple linear regression (MLR) for 107 polystyrenes and generated a QSPR model (R = 0.959 and s = 15.20 K) from the training set of 96 polystyrenes. The MLR model produced a rms error of 20.5 K for the test set comprising 11 polystyrenes. Recently, some quantum chemical descriptors calculated from repeating units or monomers were used to develop QSPR models for Tgs of polymers[10-12]. Due to the large and variable size of polymer molecules, the QSPR models stated above, together with QSPR models of other polymer properties, are modeled by extrapolation from monomer structures or repeating units[1]. These methods fail to account for the influences from neighboring repeating units. Especially for the Tg, the glass transition is resulted from Brownian motion of chain segments subjected to freezing or thawing. In this work, the chain segments (localized units or motion units) with 20 carbons (10 repeated units) in length were used to calculate descriptors for their corresponding polystyrenes and to develop QSPR models for their Tgs.

Polímeros, 27(1), 68-74, 2017

A quantitative relationship between Tgs and chain segment structures of polystyrenes

2. Materials and Methods 2.1 Data set Table 1 shows the experimental Tg data for 107 polystyrenes, which are taken from Brandrup et al.[13]. The entire set contains a Tg value range of 208-490 K.

The pendant groups presented in the benzene ring include halides, carbonyls, ethers, hydrocarbon chains, hydroxyl, hydroxyimino, aromatic rings, and other functional groups. These polystyrenes were randomly divided into a training set (70 polystyrenes) and a test set (37 polystyrenes). The training set was used to build a QSPR model, and the test set was adopted to evaluate the model.

Table 1. Molecular descriptors and Tg data of 107 polystyrenes. No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 a

Polystyrene Training set poly(4-dodecylstyrene) poly(4-(octyloxymethyl)styrene) poly[4-(1-ethlhexyloxymethyl)styrene] poly[4-(4-hydroxybutoxymethyl)styrene] poly(2-hexyloxycarbonylstyrene) poly(5-bromo-2-butoxystyrene) poly(2-pentyloxymethlstyrene) poly(4-octanoylstyrene) poly[4-(1-hydroxy-3-piperidinopropyl)styrene] poly(4-butyrylstyrene) poly(2-methoxystyrene) poly(4-methoxymethylstyrene) poly(5-bromo-2-methoxystyrene) poly(2-methoxymethylstyrene) poly(4-propoxycarbonylstyrene) poly(4-ethoxycarbonylstyrene) poly(4-isopropoxycarbonylstyrene) poly(4-phenoxystyrene) poly(4-diethylcarbamoylstyrene) poly(2-ethylstyrene) poly(3, 5-dimethylstyrene) poly(2, 5- dichlorostyrene) poly(4-methylstyrene) poly(3, 4-dimethylstyrene) poly(4-[(1-hydroxyimino)-2phenethyl]styrene) poly(4-methoxycarbonylstyrene) poly(4-acetylstyrene) poly(2-ethoxycarbonylstyrene) poly(4-cyanostyrene) poly(3-hydroxymethystyrene) poly(2, 4-dichlorostyrene) poly(2,4,5-trimethylstyrene) Poly[4-(bis(trimethylstanny)methyl)styrene] poly(2, 5-dimethylstyrene) poly(4-tert-butylstyrene) poly(2,4,6-trimethylstyrene) poly(2-carboxystyrene) Poly(4-benzoylstyrene) Poly(4-phenylacetylstyrene) Poly(2-phenylaminocarbonylstyrene) Poly(4-phenylstyrene) Poly(4-piperidinocarbonylstyrene) Poly[4-(3-piperidinopropionyl)styrene] Poly(4-propoxysulfonylstyrene) Poly(4-p-toluoylstyrene) Poly{3-[bis(trimethylsiloxy)boryl]styrene} Poly (4- [bis(trimethylstannyl)methyl] styrene)

ChiA_B(e)

SpMax_EA(bo)

H7s

DLS_01

Tg(K)a

Tg(K)b

0.288 0.286 0.282 0.281 0.277 0.275 0.283 0.278 0.275 0.270 0.271 0.274 0.267 0.274 0.271 0.268 0.265 0.261 0.270 0.273 0.265 0.262 0.268 0.265 0.260 0.265 0.263 0.268 0.262 0.271 0.262 0.263 0.261 0.265 0.263 0.263 0.261 0.257 0.259 0.258 0.260 0.269 0.272 0.261 0.255 0.264 0.261

3.734 3.734 3.768 3.734 4.010 3.982 3.903 3.820 3.765 3.820 3.901 3.734 3.980 3.903 3.820 3.820 3.822 3.752 3.841 3.901 3.879 3.963 3.720 3.919 3.857 3.820 3.812 4.010 3.767 3.767 3.958 4.085 3.867 3.963 3.871 4.135 4.003 3.934 3.829 4.016 3.914 3.841 3.821 4.697 3.951 3.811 3.867

8.818 8.417 9.048 10.155 10.629 8.121 6.016 9.644 10.082 8.314 5.874 6.070 6.733 6.404 9.101 8.878 9.488 9.224 8.503 5.356 5.377 6.754 5.042 5.207 11.994 7.214 5.501 8.940 6.441 6.319 7.169 6.131 5.232 5.744 4.717 6.020 7.687 8.698 10.145 8.500 6.531 8.483 11.064 9.650 10.707 15.282 5.136

0.50 0.50 0.50 0.00 0.25 0.50 0.50 0.25 0.00 0.50 0.50 0.50 0.50 0.50 0.25 0.25 0.25 0.50 0.25 0.50 0.50 0.50 0.50 0.50 0.00 0.25 0.50 0.25 0.50 0.25 0.50 0.50 0.50 0.50 0.50 0.50 0.00 0.50 0.50 0.00 0.50 0.25 0.25 0.25 0.50 0.25 0.50

221 231 250 293 318 320 320 323 327 347 348 350 359 362 365 367 368 373 375 376 377 379 382 384 384 386 389 391 393 398 406 409 413 416 422 435 450 371 351 464 434 387 311 490 372 308 413

234 239 283 305 335 335 335 329 356 348 371 334 376 356 364 375 373 363 376 372 398 395 392 402 398 414 403 379 394 400 388 421 414 398 415 432 466 385 357 465 401 379 336 484 370 314 415

Tg data were taken from Brandrup et al.[13]; bTg data were calculated with the ANN model.

Polímeros, 27(1), 68-74, 2017

Yu, X., & Huang, X. Table 1. Continued... No. 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 a

Polystyrene poly(4-hexylstyrene) poly[4-(1-hydroxy-1-methylbutyl)styrene] poly[4-(1-hydroxy-1-methylhexyl)styrene] poly[4-(1-hydroxy-1-methylpropyl)styrene] poly(4-ethylstyrene) poly(4-nonylstyrene) poly(4-tetradecylstyrene) poly[4-(2-hydroxybutoxymethyl)styrene] poly(5-bromo-2-pentyoxystyrene) poly(4-isopentyloxystyrene) poly(2-butoxymethylstyrene) poly(4-valerylstyrene) poly(4-butoxycarbonylstyrene) poly(4-methoxy-2-methylstyrene) poly(2-isopropoxymethylstyrene) poly(4-isobutoxycarbonylstyrene) poly(4-fluorostyrene) poly(styrene) poly(4-propionylstyrene) poly(2,3,4,5,6,-pentafluorostyrene) poly(4-chlorostyrene) poly(2, 4-dimethylstyrene) poly(4-bromostyrene) Test set poly(4-chloro-3-fluorostyrene) poly(2-isobutoxycarbonylstyrene) poly(4-hydroxymethystyrene) poly[4-(1-hydroxy-1-methylethyl)styrene] poly(4-hexyloxymethystyrene) poly(4-propoxymethylstyrene) Poly[4-(sec-butoxymethyl)styrene] poly(5-bromo-2-propoxystyrene) poly(2-butoxycarbonylstyrene) poly[2-(2-dimethylaminoethoxycarbonyl)styrene] poly(2-ethoxymethylstyrene) poly(2-isopentyloxymethylstyrene) poly(4-sec-butylstyrene) poly(4-methoxystyrene) poly(2-pentyloxycarbonylstyrene) poly(3-methylstyrene) poly(2,5-difluorostyrene) poly(2-propoxycarbonylstyrene) poly(2-fluoro-5-methylstyrene) poly(4-chloro-3-methylstyrene) poly(2-chlorostyrene) poly(4-dimethylaminocarbonylstyrene) poly(2-methoxycarbonylstyrene) poly(2-methylstyrene) poly(4-hydroxystyrene) poly(4-decylstyrene) poly(2-isopentyloxycarbonylstyrene) poly(5-bromo-2-ethoxystyrene) poly(2-hydroxymethystyrene) poly(2-octyloxystyrene) poly(4-octylstyrene) poly(3, 4-dichlorostyrene)

ChiA_B(e)

SpMax_EA(bo)

H7s

DLS_01

Tg(K)a

Tg(K)b

0.283 0.269 0.273 0.266 0.273 0.287 0.289 0.276 0.277 0.275 0.281 0.273 0.273 0.268 0.273 0.267 0.265 0.271 0.267 0.249 0.266 0.265 0.267

3.734 3.885 3.885 3.884 3.733 3.734 3.734 3.734 3.982 3.734 3.903 3.820 3.820 3.965 3.903 3.821 3.720 3.622 3.820 4.413 3.720 3.958 3.720

7.830 6.338 10.191 5.442 4.802 8.706 9.289 11.064 7.429 8.632 6.267 8.735 10.176 6.787 7.998 9.820 7.357 4.565 6.429 17.543 5.821 5.417 5.249

0.50 0.25 0.25 0.25 0.50 0.50 0.50 0.00 0.50 0.50 0.50 0.50 0.25 0.50 0.50 0.25 0.50 0.50 0.50 0.50 0.50 0.50 0.50

246 403 364 459 350 220 237 319 322 330 340 343 349 358 361 363 368 364 375 378 383 385 391

252 416 344 441 355 236 232 309 337 316 339 337 344 371 341 363 376 356 377 377 392 402 395

0.261 0.268 0.271 0.261 0.284 0.279 0.276 0.273 0.273 0.269 0.277 0.277 0.273 0.270 0.275 0.268 0.260 0.271 0.263 0.264 0.267 0.263 0.265 0.268 0.266 0.287 0.270 0.270 0.271 0.285 0.286 0.262

3.919 4.010 3.733 3.871 3.734 3.734 3.734 3.982 4.010 4.010 3.903 3.903 3.764 3.733 4.010 3.752 3.963 4.010 3.963 3.919 3.880 3.837 4.010 3.880 3.720 3.734 4.010 3.982 3.901 3.903 3.734 3.919

7.155 7.650 5.854 5.408 8.839 8.429 8.869 7.680 10.353 10.534 7.002 6.615 5.284 5.556 9.383 5.205 9.602 8.667 8.028 6.218 5.996 6.944 7.907 5.080 6.714 8.641 10.984 7.632 6.795 6.698 8.727 6.395

0.50 0.25 0.25 0.25 0.50 0.50 0.50 0.50 0.25 0.25 0.50 0.50 0.50 0.50 0.25 0.50 0.50 0.25 0.50 0.50 0.50 0.25 0.25 0.50 0.25 0.50 0.25 0.50 0.25 0.50 0.50 0.50

395 400 413 438 253 295 313 327 339 342 347 351 359 362 365 370 374 381 384 387 392 398 403 409 433 208 341 353 433 286 228 401

388 401 394 454 249 284 309 345 345 352 342 346 365 374 354 394 365 374 373 390 382 423 408 392 416 236 344 355 402 322 239 395

Tg data were taken from Brandrup et al.[13]; bTg data were calculated with the ANN model.

Polímeros, 27(1), 68-74, 2017

A quantitative relationship between Tgs and chain segment structures of polystyrenes Table 1. Continued... No. 103 104 105 106 107 a

Polystyrene poly(4-hexanoylstyrene) poly(2-phenoxycarbonylstyrene) poly(2-methaminocarbonylstyrene) poly(4-p-anisoylstyrene) Poly(2-phenethyloxymethylstyrene)

ChiA_B(e)

SpMax_EA(bo)

H7s

DLS_01

Tg(K)a

Tg(K)b

0.275 0.258 0.266 0.257 0.267

3.820 4.016 4.010 3.953 3.903

10.423 8.596 7.803 10.577 6.696

0.50 0.25 0.00 0.25 0.50

339 397 462 376 336

321 427 456 387 372

Tg data were taken from Brandrup et al.[13]; bTg data were calculated with the ANN model.

2.2 Descriptor computation A polymeric material consists of a mixture of giant molecules. Therefore, it is impossible to calculate descriptors directly from molecular structures of the polymeric material. Two approaches have been adopted to resolve this problem. One is using the repeating unit to calculate descriptors for the corresponding polymer. The other is using the monomer as representative of the corresponding polymer[1]. Tg is a temperature point used to express transition region, where polymer chain segments can move from frozen to movement (or vice versa). Below the glass transition region, 1-4 chain atoms are involved in motion. Further, these motions are largely restricted to vibrations and short-range rotational motions. During the glass transition region, 10-50 chain atoms attain sufficient thermal energy to move in a coordinated manner. In Tg region, these chain atoms (motion units) are first mobilized before the whole molecule starts moving. On further heating, the increased energy allotted to the chains permits them reptate out through entanglements rapidly and flow as individual molecules[14,15]. The structures of polymer chain segment have an effect on its glass transition and are correlated to Tgs. According to above theory of glass transition, descriptors calculated from the chain segments are more accurate in describing structures affecting polymer Tgs than that from repeating units and monomers. From a theoretical point of view, the chain segment used to calculate descriptors is longer, the descriptors are more accurate in characterizing polymers. The motion units related to glass transition of polymers usually contain 10-50 carbons in length. In addition, a too long segment taken into account may cause difficulty in calculating descriptors, and a too short segment cannot sufficiently represent the structure of motion units. Thus, chain segments with 20 carbons in chain length were used to calculate molecular descriptors for the corresponding polymers. Polymeric chain segments containing 20 main chain carbons of polystyrenes were first sketched using ChemBioDraw Ultra 11.0 in ChemBioOffice 2008 program. For example, the structure model consisting of 10 repeating units end-capped by two hydrogens (see Figure 1) was adopted as the representative structure of poly(styrene) (No. 65 in Table 1) to calculate the descriptors. Subsequently, the sketched 2D molecular structures were converted to 3D structures and optimized using a molecular mechanics (MM2 force field) in ChemBio3D Ultra 11.0 with the convergence criterion of minimum rms of gradient value being 0.01 kcal/molÅ. The optimized molecules were saved in Sybyl mol2 (.mol2) format as the input files for Dragon software[16]. Lastly, 4885 descriptors Polímeros, 27(1), 68-74, 2017

Figure 1. The calculated models of poly(styrene).

were calculated for each energy-minimized motion unit with Dragon software. Descriptors with constant or near constant values and with pair correlation greater than or equal to 0.90 were removed in order to reduce redundant and non-useful information. After excluding redundant and non-useful variables, 551 descriptors were remained to undergo descriptor selection. A relative optimal subset of descriptors was obtained by applying MLR analysis in IBM SPSS Statistics 19.

2.3 Artificial neural network The optimal descriptors subset was fed to artificial neural network (ANN) as input vectors. ANNs are computational models, which simulate the human brain behavior. The common networks consist of an input layer, some number of hidden layers (intermediate layers) and an output layer. Each layer includes a number of processing nodes, called neurons or units. Each node in the network is influenced by those nodes to which it is connected in a highly complex and parallel way. The degree of influence is dictated by the values of the links or connections. Through a training algorithm, the overall behavior of ANNs can be modified by adjusting the weights (or the values of the links or connections). After learning from the input dataset, ANNs acquire knowledge and can be applied on test set data not present in the training set. The output layer produces the prediction values of properties interested. One of the most popular algorithms applied in the training phase is the error back-propagation (BP) algorithm. The number of neurons in the hidden layer shouled be optimized by trial and validation until no obvious improvement was seen for that model[17].

3. Results and Discussions By analyzing the correlation between the 551 descriptors and Tgs of 70 polystyrenes in the training set with stepwise MLR analysis in IBM SPSS Statistics 19, Equation 1 and the corresponding statistical results were obtained. 71

Yu, X., & Huang, X. Tg = 1346.200 − 4546.871ChiA _ B(e) + 100.47 SpMax _ EA(bo) − 12.653H 7 s − 122.231DLS _ 01

(1)

n = 70, R = 0.955, R2= 0.912, s = 16.717, F = 167.652 where n is the number of samples from the training set; s is the standard error of estimate; R is the correlation coefficient; F is the Fischer ratio. The four molecular descriptors, ChiA_B(e), SpMax_EA(bo), H7s and DLS_01 appearing in above MLR model and the corresponding descriptor values are shown in Table 1. Their descriptor characteristics are listed in Table 2; and their definitions[18] are shown in Table 3. Calculated results with Equation 1 are depicted in Figure 2A. The rms errors of Tgs of the training and test sets are 16.1 and 22.4 K, respectively.

Moreover, all variance inflation factor (VIF) values are less than 2, far less than the default value of 10. Thus these descriptors are “pure” without “mixing” or contamination from other descriptors, and each descriptor reﬂects some particular molecular structures affecting Tgs.

According to the t-test, the most significant descriptor in the MLR model is ChiA_B(e) (2D matrix-based descriptors)[16]. ChiA_B(e) denotes the average randic-like index from burden matrix weighted by Sanderson electronegativity and is defined as follow:

The four descriptors are then fed to ANN as input vectors. The optimal condition of the neural network was obtained by adjusting various parameters by trial-and-error. The architecture of the ﬁnal optimum BP neural network is [4-7-1], with the number of hidden layer being 1, the nodes in hidden layer being 7, the permission error being 0.00001, the momentum being 0.6, and the sigmoid parameter being 0.9. The results from ANN method are listed in Table 1 and depicted in Figure 2B, which indicate that the predicted Tg values are close to the experimental ones. The rms error of training set is 13.6 K (R = 0.939). The test set rms error is 17.1 K (R = 0.902) which is less than the errors from the test set in previous model (20.5 K)[9]. The mean relative error for the 107 polystyrenes in Table 1 is 3.4%, less than that from the model of Yu et al.[9] (3.7%). Furthermore, it should be noted that the test set in this paper possesses 37 polystyrenes, more than the number of samples (11 polystyrenes)[9]. And it is much easier to obtain better results on small test set of polymers. In comparison to previous model on Tgs of polystyrenes[9], the statistic qualities of our model is accurate and acceptable. Therefore, it is feasible to calculate molecular descriptors from the chain segments of polymer backbones comprising 10 repeating units for developing Tg model of polystyrenes. Table 2 shows that each descriptor in Equation 1 has a Sig.-value near to 0, and less than the default level of 0.05, which suggest that these descriptors are significant for Tgs.

Figure 2. Plots of calculated vs. experimental T g values of polystyrenes: (A) for MLR model; (B) for ANN model.

Table 2. Characteristics of descriptors appearing in MLR model. Descriptors Constant ChiA_B(e) SpMax_EA(bo) H7s DLS_01

Unstandardized Coefficients 1346.200 -4546.871 100.047 -12.653 -122.231

Std. Error 110.153 273.430 14.650 0.959 13.833

Standardized Coefficients -0.685 0.288 -0.556 -0.364

Sig.

VIF

12.221 -16.629 6.829 -13.200 -8.836

0.000 0.000 0.000 0.000

1.249 1.309 1.307 1.247

Table 3. The symbol, class and definition for descriptors appearing in MLR models. Symbol ChiA_B(e) SpMax_EA(bo) H7s DLS_01

Class 2D matrix-based descriptors Edge adjacency indices GETAWAY descriptors Drug-like indices

Definition Average randic-like index from burden matrix weighted by Sanderson electronegativity Leading eigenvalue from edge adjacency matrix weighted by bond order H autocorrelation of lag 7 / weighted by I-state Modified drug-like score from Lipinski (4 rules)

Polímeros, 27(1), 68-74, 2017

A quantitative relationship between Tgs and chain segment structures of polystyrenes

ChiA_B(e) =

Chi_M(e) nBO

(2)

Where nBO is the number of graph edges. Chi_M(e) is the Randic-like index calculated by applying Sanderson electronegativity as the vertex weighting scheme and a H-depleted molecular graph as a square matrix: Chi_M(e) =

nSK −1 nSK

∑

i= 1

∑ αij ⋅ VSi ( M ; e) ⋅ VS j ( M ; e) 

j = i +1

−1/ 2

(3)

Here nSK means the number of graph vertices; VSi(M) is the ith matrix row sum; αij are the elements of the adjacency matrix, which are equal to one for pairs of adjacent vertices, and zero otherwise. ChiA_B(e) reflects information about interatomic distances, bond distances, ring types, planar and non-planar systems and atom types[16]. A small ChiA_B(e) indicates a small interatomic distances, which results in a low degree of freedom for rotation and leads to high Tg.

The second significant descriptor is the GETAWAY (GEometry, Topology, and Atom-Weights AssemblY) descriptor, H7s (H autocorrelation of lag 7 / weighted by I-state). The descriptor H7s encodes information on structural fragments, such as the effective position of substituents and fragments in the molecular space, and accounts information on molecular size and shape as well as for specific atomic properties[16]. A large H7s suggests that a polymer has a large side group, which decreases the volume ratio of phenyl ring to other substituent groups. While the aromatic or cyclic structure in bulky side groups increases rotational barrier for backbone chain and leads to high Tg. Therefore, a polymer with large H7s may have a low Tg. The next significant descriptor is the Drug-like indice DLS_01. The descriptor DLS_01, being modified drug-like score from Lipinski (4 rules), is calculated as 1 minus Lipinski Alert Index (LAI), while LAI is defined as the ratio between the number of satisfied conditions over the total number of conditions, i.e., (1) there are more than 5 H-bond donors; (2) there are more than 10 H-bond acceptors (N and O atoms); (3) molecular weight (MW) is over 500; and (4) Moriguchi’s logP (MLogP) is over 4.15[16]. DLS_01 is related to the number of intermolecular hydrogen bonds, which increase intermolecular force and determine the magnitude of molecular aggregates. Polymer molecules with small DLS_01 hold together more strongly due to intermolecular hydrogen bonds and are unable to mover that easily, and possess high Tgs. According to the t-test, the last significant descriptor in the MLR model is SpMax_EA(bo). Edge adjacency index, SpMax_EA(bo), is derived from the H-depleted molecular graph and encodes the connectivity between graph edges. It is leading eigenvalue from edge adjacency matrix weighted by bond order. SpMax_EA(bo) reflects molecular shape and implies the substituent position in the phenyl ring for styrenes[16]. Compared to styrenes with substituents lying in p-or m-positions of the phenyl ring, a styrene with a substituent lying in o-positions usually has a larger SpMax_EA(bo), which can be seen from Table 1. The substituents in o-positions will enhance rotational barrier Polímeros, 27(1), 68-74, 2017

Figure 3. Williams plot for polystyrenes with a warning leverage of 0.214.

for backbone chain, increase rigidity of polymer chains and result in higher Tgs[9]. Despite a variety of factors affecting the Tg values of polymeric materials, intermolecular forces and molecular flexibility (or rigidity) are two important factors related to Tgs. The descriptor DLS_01 reflects the intermolecular forces, while descriptors ChiA_B(e), SpMax_EA(bo) and H7s indicate the stiffness of polymer. Therefore, the four descriptors can predict Tgs sufficiently. Figure 3 (Williams plot) was obtained to visualize the applicability domain of the ANN model in this paper. According to Williams plot based on standardized residuals vs. leverages, predictions for only those samples that fall into this domain may be considered reliable[19,20]. Figure 3 shows that only the two samples No. 44, poly(4-propoxysulfonylstyrene) and No. 67, poly(2,3,4,5,6,‑pentafluorostyrene) in the training set have larger leverage h values (0.448 and 0.456, respectively), greater than the warning leverage h* (= 0.214). But their standardized residual values (0.430 and 0.075, respectively) are less than 3. Thus the two samples, poly(4-propoxysulfonylstyrene) and poly(2,3,4,5,6,‑pentafluorostyrene), can stabilize the ANN model of polystyrenes and make it more accurate.

4. Conclusions Four molecular descriptors calculated from the chain segments of main chains comprising 10 repeating units were adopted for developing QSPR model of Tgs for polystyrenes. MLR analysis was used to select the optimal subset of descriptors after molecular descriptor generation for each chain segment. The developed ANN model was proved to be accurate and acceptable, with the absolute mean errors for the whole data set is 3.4%, which is less than that of the model published in the literature, although our model possesses more samples for the test set. Therefore, it is feasible calculating molecular descriptors from the chain segments comprising 10 repeating units in length to develop ANN model of Tgs for polystyrenes. 73

Yu, X., & Huang, X.

5. Acknowledgements The project was supported by the National Natural Science Foundation of China (No. 21472040) and Scientific Research Fund of the Hunan Provincial Education Department (No. 16A047).

6. References 1. Katritzky, A. R., Kuanar, M., Slavov, S., Hall, C. D., Karelson, M., Kahn, I., & Dobchev, D. A. (2010). Quantitative correlation of physical and chemical properties with chemical structure: utility for prediction. Chemical Reviews, 110(10), 5714-5789. PMid:20731377. http://dx.doi.org/10.1021/cr900238d. 2. Bicerano, J. (1996). Prediction of polymer properties. 2nd ed. New York: Marcel Dekker. 3. Karelson, M., Lobanov, V. S., & Katritzky, A. R. (1996). Quantum-chemical descriptors in QSAR/QSPR studies. Chemical Reviews, 96(3), 1027-1044. PMid:11848779. http:// dx.doi.org/10.1021/cr950202r. 4. Van Krevelen, D. W. (1976). Properties of polymers their estimation and correlation with chemical structure. 2nd ed. New York: Elsevier. 5. Joyce, S. J., Osguthorpe, D. J., Padgett, J. A., & Price, G. J. (1995). Neural network prediction of glass-transition temperatures from monomer structure. Journal of the Chemical Society, Faraday Transactions, 91(16), 2491-2496. http:// dx.doi.org/10.1039/ft9959102491. 6. Katritzky, A. R., Rachwal, P., Law, K. W., Karelson, M., & Lobanov, V. S. (1996). Prediction of polymer glass transition temperatures using a general quantitative structure-property relationship treatment. Journal of Chemical Information and Computer Sciences, 36(4), 879-884. http://dx.doi.org/10.1021/ ci950156w. 7. Katritzky, A. R., Sild, S., Lobanov, V., & Karelson, M. (1998). Quantitative Structure-Property Relationship (QSPR) correlation of glass transition temperatures of high molecular weight polymers. Journal of Chemical Information and Computer Sciences, 38(2), 300-304. http://dx.doi.org/10.1021/ci9700687. 8. Cao, C., & Lin, Y. (2003). Correlation between the glass transition temperatures and repeating unit structure for high molecular weight polymers. Journal of Chemical Information and Computer Sciences, 43(2), 643-650. PMid:12653533. http://dx.doi.org/10.1021/ci0202990. 9. Yu, X. L., Wang, X. Y., Li, X. B., Gao, J. W., & Wang, H. L. (2006). Prediction of glass transition temperatures for polystyrenes by a four-descriptors QSPR model. Macromolecular Theory

and Simulations, 15(1), 94-99. http://dx.doi.org/10.1002/ mats.200500057. 10. Liu, A. H., Wang, X. Y., Wang, L., Wang, H. L., & Wang, H. L. (2007). Prediction of dielectric constants and glass transition temperatures of polymers by quantitative structure property relationships. European Polymer Journal, 43(3), 989-995. http://dx.doi.org/10.1016/j.eurpolymj.2006.12.029. 11. Liu, W. Q., Yi, P. G., & Tang, Z. L. (2006). QSPR models for various properties of polymethacrylates based on quantum chemical descriptors. QSAR & Combinatorial Science, 25(10), 936-943. http://dx.doi.org/10.1002/qsar.200510177. 12. Yu, X. L., Yi, B., Wang, X. Y., & Xie, Z. M. (2007). Correlation between the glass transition temperatures and multipole moments for polymers. Chemical Physics, 332(1), 115-118. http://dx.doi.org/10.1016/j.chemphys.2006.11.029. 13. Brandrup, J., Immergut, E. H., & Grulke, E. A. (1999). Polymer handbook. 4th ed. New York: Wiley-Interscience. 14. Sperling, L. H. (1992). Introduction to physical polymer science. New York: John Wiley & Sons. 15. Gedde, U. W. (1995). Polymer physics. London: Chapman & Hall. 16. Talete SRL. (2012). Dragon: software for molecular descriptor calculation. Version 6.0. Milano: Talete SRL. Retrieved in 20 February 2016, from http://www.talete.mi.it/ 17. Yu, X. L., Yi, B., Liu, F., & Wang, X. Y. (2008). Prediction of the dielectric dissipation factor tan Î´ of polymers with an ANN model based on the DFT calculation. Reactive & Functional Polymers, 68(11), 1557-1562. http://dx.doi.org/10.1016/j. reactfunctpolym.2008.08.009. 18. Todeschini, R., & Consonni, V. (2009). Molecular descriptors for chemoinformatics. Weinheim: Wiley-VCH. 2 v. 19. Tropsha, A., Gramatica, P., & Gombar, V. K. (2003). The importance of being earnest: validation is the absolute essential for successful application and interpretation of QSPR Models. QSAR & Combinatorial Science, 22(1), 69-77. http://dx.doi. org/10.1002/qsar.200390007. 20. Wang, Y. N., Chen, J. W., Li, X. H., Wang, B., Cai, X. Y., & Huang, L. P. (2009). Predicting rate constants of hydroxyl radical reactions with organic pollutants: algorithm, validation, applicability domain, and mechanistic interpretation. Atmospheric Environment, 43(5), 1131-1135. http://dx.doi.org/10.1016/j. atmosenv.2008.11.012. Received: Feb. 20, 2016 Revised: May 02, 2016 Accepted: May 24, 2016

PolĂmeros, 27(1), 68-74, 2017

http://dx.doi.org/10.1590/0104-1428.2253

Nanocomposites films obtained from protein isolates of mechanically deboned chicken meat added with montmorillonite Bruna da Silva Menezes1, William Renzo Cortez-Vega2* and Carlos Prentice1 Laboratory of Food Technology, School of Chemistry and Food, Universidade Federal do Rio Grande, Rio Grande, RS, Brazil 2 Laboratory of Bioengineering, Faculty of Engineering, Universidade Federal da Grande Dourados, Dourados, MS, Brazil

*carlos.prentice@gmail.com

Abstract The aim of this study was to evaluate the properties of nanocomposite films of protein isolates from mechanically deboned chicken meat with organoclay (montmorillonite). For the film development, a 23 experimental design was performed with three levels, protein isolate (2, 3.5, 5 g.100 mL-1 of solution), montmorillonite (0.3, 0.5, 0.7 g.100mL-1 of solution) and glycerol (25, 30, 35 g.100 mL-1 CPI). The tensile strength varied between 6.7 and 9.1 MPa, elongation to break from 26-66%, opacity of 13.1 to 35.7 and solubility from 38.5% to 81.8%. Assessing the structural properties, interleaving of the isolate and montmorillonite can be noted. The results obtained in the experimental design indicate that 2.0 g of CPI.100 g-1 of solution, 0.8 g of MMT.100 g-1of solution and 0.2 g of glicerol.100 g-1CPI are the ideal parameters for preparing nanocomposite films. Keywords: by-product, chicken, protein isolate, film.

1. Introduction Concentrated and isolated proteins are produced in large scale to serve as functional ingredients in a wide and increasing range of food applications. When replacing conventional proteins, protein concentrates and isolates maintain or improve the quality and acceptability of the products that have been incorporated[1]. The interest of maintaining or improving the quality of packaged goods and reducing the waste of packaging at the same time has encouraged the exploration of new packaging materials such as biodegradable films formulated with raw materials derived from renewable resources[2]. Among the studied raw materials, natural biopolymers, such as polysaccharides and proteins appear most promising for developing films, because they are abundant, renewable, cheap and capable of forming a continuous matrix[3]. The choice of material for use in the formulation of films and coatings is very important, as this will depend on the interactions between the components of the material, which may interfere with barrier, mechanical and sensory properties of films[4]. This association between biopolymers and nanoparticles aims to obtain synergistic effects, it is one of the most innovative ways to enhance the properties of these matrices. Depending on the geometry and nature of the nanoparticle, new properties such as gas barrier, mechanical strength, transparency and thermal stability are improved[5]. In this field, special attention has been given to montmorillonite (MMT) because of its small particles, extremely large surface area and good interleaving properties. Montmorillonite is composed of silicate layers with a thickness

Polímeros, 27(1), 75-82, 2017

of approximately 1 nm in the planar structure, and a lateral dimension between 200 and 300 nm. Its typical chemical structure consists of two tetrahedral silica layers surrounding a layer of octahedral aluminum hydroxide or magnesium[6]. With the addition of between 2 and 10% of montmorillonite clay, nanocomposites can introduce significant improvements in their mechanical, optical and barrier properties. This advantage of adding low clay content, compared to the traditional composites, means the production of lighter components, a desirable factor in many applications. Another interesting properties usually presented by polymer/clay nanocomposites is higher microbial stability[7]. The main reason for the differences in performance between composite and nanocomposite materials is related to the high surface area of t he latter, resulting in high interaction between the matrix and the nanoparticles[8]. Thus, the aim of this study is to take advantage of a by-product from the chicken processing industry for the production of chicken protein isolate films and evaluate their properties by varying the concentration of the protein isolate, the plasticizer, and montmorillonite clay.

2. Materials and Methods 2.1 Development of films For the development of active protein isolate films from mechanically deboned chicken meat (MDCM), a 23 experimental design was carried out, with three central points and axial points. Response surface methodology was applied to study the simultaneous effects of the independent variables

O O O O O O O O O O O O O O O O

Menezes, B. S., Cortez-Vega, W. R., & Prentice, C. (concentration of MDCM protein isolate (CPI = 2, 3.5 and 5 g); montmorillonite concentration (MMT = 0.3, 0.5 and 0.7 g) and glycerol plasticizer (G = 25, 30 and 35% / g CPI) on the tensile strength (MPa), elongation to break (%), water vapor permeability (WVP), solubility and opacity responses. The films were developed by the casting technique, where each solution of the film was prepared according to the experimental design data shown in Table 1.

2.2 Thickness After the packaging period, the thickness of the films was obtained using a digital micrometer (INSIZE, IP 54) Resolution 0.0100 ± 0.0005 mm. The thickness was determined as the arithmetic mean of five random measurements of the area of the film.

2.3 Mechanical properties The testing of tensile strength (TS) and elongation to break (E) was performed according to the standard method of the American Society for Testing and Materials, ASTM D-882[9]. A section of the film with a width of 2.5 cm and a length of 9 cm was inserted into a texturometer (Stable Micro System, TA.XT plus) and subjected to a stretching force until the film ruptured. The tensile strength was expressed in MPa and elongation to break in%.

2.4 Water vapor Permeability (WVP) Water vapor permeability (WVP) assays were done gravimetrically at 25 °C according to the E96-95 method[10]. The films were coupled with CaCl solution, previously dried at 105 °C for 2h, and stored in a desiccator with a relative humidity of 75% for 8 days, being weighed every 24 hours. The WVP calculation was performed according to Equation 1. P=

M L t A. ∆P

(1)

where M is the mass of absorbed moisture (g); t is time in days; L is the thickness in mm; A is the area in m2 and ΔP is moisture variation on in KPa.

2.5 Solubility Water solubility of the films was determined in triplicate by the Fakhouri et al.[11] method. The films were cut into squares of 2 cm, and the initial percentage of dry matter of the sample was determined in an oven (A1-SED, De Leo) at 105 °C for 24 hours. After weighing, the samples were immersed in containers with 25 mL of distilled water, and stirred slowly in an orbital shaker (TE-420, Tecnal) for Table 1. Variables used in the experimental design to develop active films. Real variables CPI (g)* Glycerol (g)* MMT (g)*

-1.68 1.0 0.2 0.2

Coded variables -1 0 1 2.0 3.5 5.0 0.5 1.0 1.5 0.3 0.5 0.7

1.68 6.0 1.8 0.8

*CPI: Chicken protein isolate of MDCM; MMT: Montmorillonite.

24 hours. After this period, samples were removed and dried (105 °C for 24 hours) to determine the mass of dry matter not dissolved in water.

2.6 Opacity The opacity of the films was obtained using a visible UV spectrophotometer (Biospectro, SP-22) and calculated by the formula: opacity =

A 450 X

(2)

where A450 is the absorbance at 450nm, given by the equation A450 = -logT and X is film thickness (mm).

2.7 Scanning electron microscopy (SEM) Sample films were analyzed to determine the surface characteristics using a scanning electron microscope (JSM 6060, JEOL, Japan) operating at 15kV. Five samples were mounted in a brass tube and coated with a gold layer before obtaining the images. The photographs were taken at 10000x magnification. This analysis was performed at the Center of Electron Microscopy of South Zone (CEME‑SUL) at the Federal University of Rio Grande.

2.8 X-ray diffraction (XRD) The analysis of X-ray diffraction (XRD) were obtained using a diffractometer (Siemens D500 of Bragg-Brentanogeometry) with Cu radiation, operating at 40kV and 17.5mA, graphite monochromator for diffracted X-ray beams. The measurements were obtained in steps of 0.05 degrees (2Ɵ), counting time of 5 seconds/step and measurement intervals in 2Ɵ from 2 to 60 degrees. The analysis was performed at the Electron Microscopy Center (CME) of the Federal University of Rio Grande do Sul – UFRGS.

2.9 Infrared spectroscopy (FTIR) The films were analyzed according to Attenuated Reflectance FTIR-ATR method; using spectrophotometric equipment on infrared (Prestige, 210,045-Japan) with 40 scans in the Laboratory of Synthesis and Inorganic Catalysis in the Federal University of Rio Grande.

2.10 Statistical analysis To statistically determine significant differences with 95% confidence between means (p = 0.05), analysis of variance and Tukey’s test were used and for the development of the films a response surface methodology was applied to study the simultaneous effects of the variables independent on responses on Statistica 7.0 software (Statsoft, USA).

3. Results and Discussion 3.1 Film properties 3.1.1 Physical thickness Table 2 shows the thickness values obtained, with the amount of filmogenic solution placed on the plate and thickness. Polímeros, 27(1), 75-82, 2017

Nanocomposites films obtained from protein isolates of mechanically deboned chicken meat added with montmorillonite The thickness was determined as the arithmetic average of five random measurements of the area of the film, it was found that with an increase in solids, thickness increased too, same behavior found by Brandelero et al.[12] who determined thicknesses of starch films in the same way and obtained a variation in thickness from 0.11 mm to 0.33 mm. On the other hand, Cortez-Vega et al.[13] also determined the film thickness of croaker protein isolate by the mean of 10 thickness measurements. Therefore, the value of the average thickness (0.2 mm) was maintained for the planning, according to the amount of solution placed in the plate (15.5 g), since these values showed better tensile strength and elongation to break characteristics from preliminary tests. 3.1.2 Physicochemical properties Responses surfaces were obtained to evaluate the effect of independent variables, chicken protein isolate (CPI), montmorillonite (MMT) and glycerol (G) on the dependent variables (tensile strength, elongation, opacity, solubility and water vapor permeability). Table 3 shows the matrix of the experimental design and the results obtained for the assays. The central points for the answers showed little variation, indicating a good repetitiveness of the process. Table 2. Determination of the thickness of chicken isolate and nanoclay films.

Mean

Amount of filmogenic solution (g) 14.5 15.0 15.5 16.0 17.0 15.6

Thickness (mm) 0.18 ± 0.11 0.19 ± 0.07 0.20 ± 0.10 0.21 ± 0.08 0.22 ± 0.12 0.20 ± 0.09

Table 3 shows the experimental design matrix and results obtained for the dependent variables tensile strength, elongation, opacity, solubility and water vapor permeability (WVP). However, only the tensile strength, opacity and solubility were influenced by the process variables. For the variables elongation to break and water vapor permeability (WVP), Fcalculated was less than Ftabulated, thus the model was not predictive, not being possible to generate the response surface[14]. Therefore, the elongation to break and WVP were not influenced by the process variables. According to the process conditions, the tensile strength of the films of chicken protein isolate with montmorillonite ranged from 6.7 to 9.1MPa and elongation to break of 26‑66%. On the other hand, Zavareze et al.[15] found tensile strength ranging from 4.0 to 5.7MPa and elongation to break of 102-193% in myofibrillar fish protein films, in croaker protein isolate films with the addition of montmorillonite found TS from 7.2 to 10.7MPa and elongation to break from 39.6 to 45.8%[13].It can be seen then that the films obtained in this work showed reasonable resistance compared to other films derived from meat protein isolates, possibly due to the addition of montmorillonite, obtaining values c lose to those of Cortez-Vega et al.[13] who also used nanoclay. As for the elongation, variability greater than that of other authors was obtained. Elongation to break depends principally on the plasticizer values, and increasing the concentration of glycerol and sorbitol improves the elongation to break of films[16]. Therefore, one can view the tensile strength conditions as a function of the protein isolate concentrations, montmorillonite and glycerol in Figure 1a and 1b, and conclude that the tensile strength becomes higher with maximum (0.8 g) and minimum amounts (0.2 g) of MMT and less glycerol, regardless of the amount of protein isolate. According to the process conditions, the opacity ranged from 13.1 to 35.7 and the solubility from 38.5% to 81.8%[17],

Table 3. Results and experimental design matrix used to evaluate the mechanical properties of chicken protein isolate and nanoclay films. Assay 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

CPI (g/100mL) -1(2.0) +1 (5.0) -1 (2.0) +1 (5.0) -1 (2.0) +1 (5.0) -1 (2.0) +1 (5.0) -1.68 (1.0) +1.68 (6.0) 0 (3.5) 0 (3.5) 0 (3.5) 0 (3.5) 0 (3.5) 0 (3.5) 0 (3.5)

MMT (g/100mL) -1 (0.3) -1 (0.3) -1 (0.3) -1 (0.3) +1 (0.7) +1 (0.7) +1 (0.7) +1 (0.7) 0 (0.5) 0 (0.5) 0 (0.5) 0 (0.5) -1.68 (0.2) +1.68 (0.8) 0 (0.5) 0 (0.5) 0 (0.5)

Glycerol (g/100mL) -1 (0.5) -1 (0.5) +1 (1.5) +1 (1.5) -1 (0.5) -1 (0.5) +1 (1.5) +1 (1.5) 0 (1.0) 0 (1.0) -1.68 (0.2) +1.68 (1.8) 0 (1.0) 0 (1.0) 0 (1.0) 0 (1.0) 0 (1.0)

Tensile Strength* (MPa) 8.6 ± 0.11 8.6 ± 0.15 7.6 ± 0.20 7.6 ± 0.13 7.5 ± 0.18 6.7 ± 0.23 6.8 ± 0.26 8.4 ± 0.11 8.4 ± 0.22 7.4 ± 0.24 7.9 ±0.19 7.0 ± 0.13 9.1 ± 0.12 8.9 ± 0.11 7.8 ± 0.22 8.0 ± 0.02 8.0 ± 0.09

Elongation to break* (%) 41 ± 0.06 44 ± 0.11 26 ± 0.16 29 ± 0.13 36 ±0.17 66 ± 0.22 28 ± 0.04 36 ± 0.12 44 ± 0.05 46 ± 0.13 56 ± 0.16 35 ± 0.26 52 ± 0.11 40 ± 0.10 39 ± 0.05 39 ± 0.01 39 ± 0.07

Opacity*

Solubility* (%)

WVP* (g.mm/m2.d.kPa)

13.7 ± 0.32 18.6 ± 0.53 13.1 ± 0.52 16.0 ± 0.18 14.8 ± 0.82 35.7 ± 0.07 15.3 ± 1.06 16.5 ± 0.49 15.2 ± 0.50 20.0 ± 0.36 17.6 ± 0.83 14.7 ± 0.41 13.5 ± 0.59 17.8 ± 0.89 14.7 ± 0.59 16.5 ± 0.29 16.0 ± 0.33

46.5 ± 0.74 39.7 ± 0.60 53.1 ± 0.54 53.6 ± 0.25 38.5 ± 0.09 71.8 ± 0.34 48.4 ± 0.05 54.0 ± 0.24 81.8 ± 0.78 46.5 ± 0.09 38.6 ± 0.13 43.6 ± 0.22 42.2 ± 0.20 40.7 ± 1.10 40.9 ± 0.37 45.7 ± 0.06 45.8 ± 0.26

1.74 ± 0.68 1.23 ± 0.18 1.21 ± 0.01 1.30 ± 0.15 1.23 ± 0.40 1.64 ± 0.01 2.83 ± 0.11 1.04 ± 0.25 2.01 ± 0.31 1.51 ± 0.09 1.33 ± 0.09 1.74 ± 0.44 1.00 ± 0.01 1.76 ± 0.33 1.09 ± 0.03 1.08 ± 0.01 1.07 ± 0.03

*Mean values of 3 repetitions expressed as mean ± standard deviation; CPI: Chicken protein isolate; MMT: Montmorillonite; WVP: water vapor Permeability.

Polímeros, 27(1), 75-82, 2017

Menezes, B. S., Cortez-Vega, W. R., & Prentice, C. in chitosan films added with clay, found opacity from 14.0 to 31.9, results within the range of values found in this study. In croaker protein isolate films with the addition of montmorillonite found opacity ranging from 12 to 13.5 and a solubility ranging from 18.1% to 27.6%[13]. Thus it can be concluded that the films of this study showed higher opacity and solubility and greater variability from the other studies. State that it is important that the films exhibit low opacity, i.e, greater transparency for viewing the packaged product and also stating that it is best for films to have a lower solubility to protect the food and ensure quality during storage[15]. From Figures 2a and 2b, one can see the response surfaces of the opacity against chicken protein isolate concentrations, glycerol and montmorillonite. It was possible to visualize the opacity conditions as a function of the concentrations of protein isolate, montmorillonite and glycerol in Figures 2a and 2b, concluding that the opacity (13.7) decreases with a greater amount of MMT (0.8g), least amount of glycerol (0.2 g), and higher values of protein isolate (6 g). This behavior is different to that found by Nascimento et al., because according to the author, increase in opacity occurs with increasing clay[17].

In this case, this may be due to dilution of montmorillonite along with the other components of the film. In Figures 3a and 3b the response surfaces of the solubility against the concentrations of chicken protein isolate, glycerol and montmorillonite. It was possible to visualize the solubility conditions as a function of the concentration of protein isolate, montmorillonite and glycerol in Figure 3a and 3b, concluding that the solubility (13.7%) decreases with greater amounts of MMT (0.8 g) and smaller amount of protein isolate (1 g), and is independent of the amount of glycerol. This behavior was similar to that showed by Nascimento et al.[17], where the solubility was lower when the chitosan concentrations were lower and with the addition of clay. From the response surface, the best formulation for the preparation of the films was chosen for further analysis. Interpreting the results of the response surface, the films with better properties were those with a maximum quantity of MMT (0.8 g /100 mL), smaller amount of glycerol (2 g/100 mL), since their quantity did not significantly influence the properties and intermediate quantity of the isolate (2 g/100 mL) so as not to make the film opaque.

Figure 1. Response surface of tensile strength as a function of the concentration of (a) protein isolate and MMT (b) protein isolate and glycerol.

Figure 2. Response Surface of opacity as a function of the concentration of (a) protein isolate and MMT (b) protein isolate and glycerol. 78

Polímeros, 27(1), 75-82, 2017

Nanocomposites films obtained from protein isolates of mechanically deboned chicken meat added with montmorillonite 3.1.3 Structural properties The film with the best formulation, determined from the experimental design, was characterized for its structural properties. 3.1.3.1 Infrared spectroscopy (FTIR) Figures 4a and 4b shows the infrared spectra of films without and with the addition of montmorillonite (MMT). The spectra show an increase in the bands of 50 cm-1 with the incorporation of MMT, one can see the peaks in the same wavelength on both films, but the number of waves with MMT seems to be slightly lower due to enlargement in some cases, as is in the case of the 3000-3600 cm-1 region. This is possible due to the interaction of the chain of the film components with MMT through hydrogen bonds[18]. Peaks around 2800 to 2900 cm-1 can be associated to asymmetric and symmetric stretching of the C-H bonds (from CH3 and CH2 groups). It can be verified that they are expanded further in films with addition of montmorillonite, Santos[19] studied the effects of various clays and found that they all had this functional group, because they presented peaks around this 2800 cm-1 band. In this study, the wider band shown approximately at the region of 2800 cm-1, in

Figure 4b, represents the functional group of montmorillonite nanoclay. Figure 4b shows bands in the 3600-3400 cm-1 region, corresponding to stretching vibration (axial) of the hydroxyl groups related to the water adsorbed between the MMT plates. The characteristic bands of Si-O-Si bonds were observed in the region between 1150-1020 cm-1 and in the range of 945-810 cm-1, corresponding to octahedral layers of the alumino silicate Si-O-Al[20]. The peak located around 1000 cm-1 may be related to the interaction between the plasticizer (glycerol OH group), and the structure of the film[21]. Therefore, by spectroscopic analysis, the main structural characteristics of the prepared films and the interactions between the protein isolate and montmorillonite could be observed. These results suggest that there was an interaction between CPI and MMT, the amplitudes of the peaks of MMT films were higher, although both films exhibit similar peaks. 3.1.3.2 Scanning electron microscopy (SEM) Figures 5a and 5b shows the scanning electron microscopy for the chicken protein isolate films with and without incorporation of montmorillonite, respectively,

Figure 3. Response Surface of solubility as a function of the concentration of (a) protein isolate and MMT (b) protein isolate and glycerol.

Figure 4. Infrared spectroscopy of CPI films (a) without the addition montmorillonite (b) with the addition montmorillonite. Polímeros, 27(1), 75-82, 2017

Menezes, B. S., Cortez-Vega, W. R., & Prentice, C.

Figure 5. SEM of protein isolate films from mechanically deboned chicken meat (a) without the addition of montmorillonite (b) with the addition of montmorillonite.

while Figure 6 shows the microscopy of natural MMT[22]. Scanning electron microscopy can be used to evaluate the homogeneity of the films, the structure of the layer, pores, cracks and surface smoothness[23]. In the scanning electron microscopy with a 10,000x magnification, shown in Figure 6, the difference in the morphology of the films can be seen. In Figure 5a, a compact and homogeneous structure was observed, showing the interaction of the protein with the plasticizer, unlike Figure 5b, which, in the presence of MMT, shows granules in their structure, representing clusters of MMT nanoclay. Micrograph (Figure 7) shows non-uniform granule‑shaped small particles well separated into different parts, however these are not presented in form of beads which is the natural form of MMT (Figure 6)[22]. These changes are directly related to interleaving and adsorption of the proteins and plasticizers with montmorillonite.

Figure 6. SEM of natural montmorillonite (MMT-Na), at a magnification of 2000x.Source: Sarier et al.[22].

These discontinuities in the film structure can provide the formation of preferential pathways for diffusion of water vapor, which would explain the high water vapor permeability (WVP) of films Bodini, however in this case there was no effect of MMT on the WVP of the films[24]. 3.1.3.3 X-Ray Diffraction (XRD) The XRD patterns of the protein isolate-based films from MDCM are shown in Figure 8. It can be seen that the spectra are different, suggesting the presence of distinct structures in the different layers of the film. It can be noted that the nanocomposite film with the presence of MMT shows peaks, representing crystalline phases not presented in films without MMT[25]. This is due to the fact that when the radiation is incident on the surface layer of a film with MMT, it has an incident irradiation reflex, representing an amorphous film, probably due to its composition, this potential depends strongly on the substrate. In study analyzed XRD in thin chalcogenide films and found that the films are mostly amorphous, with oxidation evident[26]. This oxidation is not advantageous, because if the film is oxidized, there is great chance of oxidizing the product packaged as well. 80

Figure 7. SEM of protein isolate films from mechanically deboned chicken meat with addition of montmorillonite at a 3000x magnification.

The small peak near 20° in the active nanocomposite film with MMT is characteristic of the crystalline phase of polymers[27]. And the other peaks between 20 and 40°; 40° and 60° correspond to the reflection of mineral structures, and Polímeros, 27(1), 75-82, 2017

Nanocomposites films obtained from protein isolates of mechanically deboned chicken meat added with montmorillonite

Figure 8. XRD of films of protein isolate from MDCM without montmorillonite (blank) and with montmorillonite (active film).

may be zinc or iron, which are more evident in nanocomposite films because of the composition of the montmorillonite[28]. In their sample without MMT, obtained no peaks either, indicating that no crystalline structure was evident[18].

4. Conclusion It was possible to obtain nanocomposite films from chicken protein isolate added with montmorillonite. On assessing the structural properties (SEM, FTIR and XRD), interleaving between the isolate and montmorillonite can be noted. It was found that the elongation to break and water vapor permeability of the films were not affected by variables in the experiment (chicken protein isolate; glycerol; montmorillonite), while the other properties were. The low solubility and opacity and high tensile strength of the films occurred at a high montmorillonite concentration (0.8 g.100 mL-1), low glycerol (0.2 g.100 mL-1), intermediate chicken protein isolate (2 g.100 mL-1) and a heat treatment of 70 °C.

5. Acknowledgements The authors acknowledge the financial support by the National Council for Scientific and Technological Development (CNPq) and the grant provided by the Coordination for the Improvement of Higher Education Personnel (CAPES) of Brazil.

6. References 1. Hua, Y., Huang, Y., Qiu, A., &Liu, X. (2005). Properties of soy protein isolate prepared from aqueous alcohol washed soy flakes.Food Research International, 38(3), 273-279. http:// dx.doi.org/10.1016/j.foodres.2004.05.010. 2. Souza, C. O., Silva, L. T., &Druzian, J. I. (2012). Comparative studies on the characterization of biodegradable cassava starch films containing mango and acerola pulps.Química Nova, 35(2), 262-267. http://dx.doi.org/10.1590/S0100-40422012000200006. Polímeros, 27(1), 75-82, 2017

3. Quintero, E. S. M., &Sobral, P. J. A. (2000). Preparo e caracterização de proteínas miofibrilares de tilápia-do-nilo para elaboração de biofilmes.Pesquisa Agropecuária Brasileira, 35(1), 179-189. http://dx.doi.org/10.1590/S0100-204X2000000100020. 4. Mali, S., Grossmann, M. V. E., &Yamashita, F. (2010). Filmes de amido: produção, propriedades e potencial de utilização. Ciências Agrárias, 31, 137-156. Retrieved in 22 July 2015, from http://www.uel.br/revistas/uel/index.php/semagrarias/ article/view/4898/4363 5. Chivrac, F., Pollet, E., Dole, P., &Avérous, L. (2010). Starch-based nano-biocomposites: plasticizer impact on the montmorillonite exfoliation process.Carbohydrate Polymers, 79(4), 941-947. http://dx.doi.org/10.1016/j.carbpol.2009.10.018. 6. Schlemmer, D., Angélica, R. M., Gomes, A. C. M. M., &Sales, M. J. A. (2009). Morfologia de filmes de amido termoplástico e montmorilonita (TPS/MMT) usando óleos vegetais do cerrado como plastificantes. In Anais do 10º Congresso Brasileiro de Polímeros (pp. 1-8). São Carlos: ABPol. 7. Souza, M. A., Pessan, L. A., &Rodolfo, A., Jr.(2006). Nanocompósitos de poli (cloreto de vinila) (PVC) / Argilas organofílicas.Polímeros: Ciência e Tecnologia, 16(4), 257-262. Retrieved in 22 July 2015, from http://www.scielo.br/pdf/po/ v16n4/01.pdf 8. Ray, S. S., &Okamoto, M. (2003). Polymer/layered silicate nanocomposites: a review from preparation to processing. Progress in Polymer Science, 28(11), 1539-1641. http://dx.doi. org/10.1016/j.progpolymsci.2003.08.002. 9. American Society for Testing and Materials – ASTM. (2001). E96-95: standard test methods for water vapor transmission of thin plastic sheeting. Philadelphia: ASTM. 10. American Society for Testing and Materials – ASTM. (1995). E96-95: standard test methods for water vapor transmission of thin plastic sheeting. Philadelphia: ASTM. 11. Fakhouri, F. M., Fontes, L. C. B., Gonçalves, P. V. M., Milanez, C. R., Steel, C. J., &Collares-Queiroz, F. P. (2007). Films and edible coatings based on native starches and gelatin in the conservation and sensory acceptance of Crimson grapes. Ciência e Tecnologia de Alimentos, 27(2), 391-393. http:// dx.doi.org/10.1590/S0101-20612007000200027. 12. Brandelero, R. P. H., Grossmann, M. V., &Yamashita, F. (2013). Hidrofilicidade de filmes de amido/poli (butileno 81

Menezes, B. S., Cortez-Vega, W. R., & Prentice, C. adipato co-tereftalato) (Pbat) adicionados de tween 80 e óleo de soja.Polímeros: Ciência e Tecnologia, 23, 270-275. http:// dx.doi.org/10.4322/S0104-14282013005000011. 13. Cortez-Vega, W. R., Bagatini, D. C., Souza, J. T. A., &Prentice, C. (2013). Nanocomposite biofilms obtained from Whitemouth croaker (Micropogonias furnieri) protein isolate and Monmorilonite: evaluation of the physical, mechanical and barrier properties.Brazilian Journal of Food Technology, 16(2), 90-98. http://dx.doi.org/10.1590/S1981-67232013005000011. 14. Rodrigues, M. I., &Lemma, A. F. (2009). Planejamento de experimentos e otimização de processos. São Paulo: Cárita. 15. Zavareze, E. R., Halal, S. L. M., Telles, A. C., &PrenticeHernández, C. (2012). Biodegradable films based on myofibrillar proteins of fish.Brazilian Journal of Food Technology, 4, 53-57. http://dx.doi.org/10.1590/S1981-67232012005000038. 16. Laohakunjit, N., &Noomhorm, A. (2004). Effect of plasticizers on mechanical and barrier properties of rice starch film.Stärke, 56(8), 348-356. http://dx.doi.org/10.1002/star.200300249. 17. Nascimento, S. D., Oliveira, T. A., Santos, F. K. G., Aroucha, E. M. M., &Leite, R. H. L. (2013). Effect of clay addition on the properties of chitosan biofilm.Revista Verde de Agroecologia e Desenvolvimento Sustentável, 8(1), 306-312. Retrieved in 22 July 2015, from http://www.researchgate.net/profile/ Suliene_Nascimento/publication/264274306_Efeito_da_ adio_de_argila_nas_propriedades_de_biofilme_de_quitosana/ links/5435a9a90cf2bf1f1f2b360e.pdf 18. Slavutsky, A. M., Bertuzzi, M. A., Armada, M., García, M. G., &Ochoa, N. A. (2014). Preparation ans characterization of montmorillonite/brea gum nanocomposites films.Food Hydrocolloids, 35, 270-278. http://dx.doi.org/10.1016/j. foodhyd.2013.06.008. 19. Santos, S. M. (2011). Influência da adição de montmorilonita nas propriedades térmicas e mecânicas de nanocompósitos com matriz de epóxi (Master’s dissertation). Instituto Alberto Luiz de Coimbra, Rio de Janeiro. 20. Delpech, M. C., Miranda, G. S., &Santo, W. L. E. (2011). Dispersões aquosas a base de nanocompósitos de poliuteranos e argilas hidrofílicas brasileiras: síntese e caracterização. Polímeros: Ciência e Tecnologia, 21(4), 315-320. http://dx.doi. org/10.1590/S0104-14282011005000054.

21. Bergo, P., &Sobral, P. J. A. (2007). Effects of plasticizer on physical properties of pig skin gelatin films.Food Hydrocolloids, 21(8), 1285-1289. http://dx.doi.org/10.1016/j.foodhyd.2006.09.014. 22. Sarier, N., Onder, E., &Ersoy, S. (2010). The modification of Na-montmorillonite by salts of fatty acids: An easy intercalation process.Colloids and Surfaces: Physico Chemical and Engineering Aspects, 371(1-3), 40-49. http://dx.doi. org/10.1016/j.colsurfa.2010.08.061. 23. Bilbao-Sainz, C., Avena-Bustillos, R. J., Wood, D. F., Williams, T. G., &Mchugh, T. H. (2010). Nanoemulsions prepared by a low-energy emulsification method applied to edible films. Journal of Agricultural and Food Chemistry, 58(22), 1193211938. PMid:20977191.http://dx.doi.org/10.1021/jf102341r. 24. Bodini, R. B. (2011). Desenvolvimento de materiais poliméricos bioativos à base de gelatina e própolis (Master’s dissertation). Faculdade de Zootecnia e Engenharia de Alimentos, Universidade de São Paulo, Pirassununga. 25. Jesus, F. A. A., &Macedo, Z. S. (2010). Síntese de filmes finos de germanato de bismuto.Scientia Plena, 6(3), 1-6. Retrieved in 22 July 2015, from http://www.scientiaplena.org.br/sp/article/ view/257/49 26. Moura, P. R., Almeida, D. P., Lima, J. C., Ponciano, C. R., &Campos, C. E. M. (2009). Propriedades estruturais de ligas e filmes finos calcogênicos submetidos à luz síncrotron.Revista Brasileira de Aplicações de Vácuo, 28, 1-6. http://dx.doi. org/10.17563/rbav.v28i1-2.481. 27. Pattabi, M., Amma, B. S., &Manzoor, K. (2007). Photoluminescence study of PVP capped CdS nanoparticles embedded in PVA matrix.Materials Research Bulletin, 42(5), 28-35. http://dx.doi. org/10.1016/j.materresbull.2006.08.029. 28. Fernandes, D. M., Silva, R., Hechenleitner, A. A. W., Radovanovic, E., Melo, M. A. C., &Pineda, E. A. G. (2009). Synthesis and characterization of ZnO, CuO and mixed Zn and Cu oxide.Materials Research Bulletin, 115(1), 5-10. http:// dx.doi.org/10.1016/j.matchemphys.2008.11.038. Received: July 22, 2015 Revised: May 23, 2016 Accepted: June 03, 2016

Polímeros, 27(1), 75-82, 2017

http://dx.doi.org/10.1590/0104-1428.2372

In vitro and in vivo cell tracking of PKH26-labeled osteoblasts cultured on PLDLA scaffolds Alice Rezende Duek1, Gabriel Ciambelli Dias da Costa1, Bruna Antunes Más1,2, Maria Lourdes Peris Barbo1, Adriana Cristina Motta2* and Eliana Aparecida de Rezende Duek1,2 Laboratório de Biomateriais, Pontifícia Universidade Católica de São Paulo – PUCSP, Sorocaba, SP, Brazil 2 Faculdade de Engenharia Mecânica, Universidade Estadual de Campinas – UNICAMP, Campinas, SP, Brazil 1

*motta.adrianam@gmail.com

Abstract The importance of monitoring in vivo interaction that occurs between cells /bio/tissue recipient in the understanding of tissue regeneration processes becomes ever greater. This study aims to monitor and evaluate the influence of scaffold implants of poly (L-co-D, L lactic acid) - PLDLA synthesized in the laboratory, previously cultured with primary osteoblastic cells heterologously stained with the fluorescent vital dye, PKH26, on the tissue regeneration process in 8 mm central critical defects of the Wistar rat calvaria. The results obtained by MTT assay and monitoring of cells stained with PKH26 dye over 14 days of culture showed that the dye was cytocompatible with osteoblastic cells and did not exert a negative influence on the growth of unstained cells. In the in vivo study, macroscopic observations made during deployment times corroborate the results in vitro, as no apparent signs of toxicity were observed in the implanted bone defect area. The use of mobile monitoring with the dye, PKH26 in vivo is an effective strategy for the understanding of cell behaviour in the presence of PLDLA polymer. Keywords: PKH26, PLDLA, scaffolds, tissue engineering.

1. Introduction The growth of tissue engineering and regenerative medicine in recent years reflects the efforts of researchers in the search for new strategies in an attempt to compensate for, as one example, the issue of limited availability of autologous tissue in procedures involving bone grafts. It is estimated that over 2.2 million grafts are performed annually worldwide for the repair of bone defects[1,2], and the growth forecast that surrounds this market reflects the significant volume of this industry, which was US $2.1 billion in 2013 and is projected to increase to $2.7 billion in 2020, according to Global Data. Among the materials with the greatest potential to be used in tissue engineering, bioresorbable polymers belonging to the family of poly (α-hydroxy acids), such as poly (L-co-D, L lactic acid) (PLDLA), stand out for their properties such as biocompatibility and versatility in terms of degradation over time. Regarding the most efficient way to promote the organization, growth and differentiation of cells in the tissue formation process in injured tissue, the most appropriate type of devices are the scaffolds, which must meet a series of requirements, such as excellent biocompatibility, good mechanical properties and adequate porosity[3,4]. The degradation rate of the copolymer is intermediate between poly (L- lactic acid) and poly (DL- lactic acid), thus proving very interesting for most applications[5]. Furthermore the lactic DL- acid units in the copolymer chain sequence hinder the crystallization of lactic L- acid, while the strength of the material is maintained for the period necessary for recovering the treated tissue[6].

Polímeros, 27(1), 83-91, 2017

Understanding the wide range of cellular and molecular responses involved in an in vivo system is quite complex; this has often led to misunderstanding and erroneous or contradictory interpretation of the actual biological events that guide a biomaterial implantation success and therefore, which properties and features should be developed in the biomedical implantable devices[7]. The use of so-called “constructs”, which are scaffolds grown onto autogenous or allogenic bone cells employed in tissue engineering and aimed at providing a higher rate of regeneration in various tissues, but little is known about the biological function and behaviour of the transplanted cells in the in vivo environment, and unsatisfactory results are reported frequently, for example, low viability of the transplanted cells and immune response of the organ / tissue implant[8,9]. The monitoring of live biological interactions has been shown extremely important and the use of fluorescence microscopy techniques stands out in this sense, due to the ease of handling and low cost of application, both in relation to previous procedures for cell labelling, as does the cost of acquisition and maintenance of the equipment itself. As a rule, to investigate the biological behaviour of cells seeded in vivo, it is essential to select such cells. Accordingly, the fluorescent dye, PKH26 is an efficient marker that can be used to track and trace cells through its ability to bind to membrane lipid regions of several cell types[10,11].

O O O O O O O O O O O O O O O O

Duek, A. R., Costa, G. C. D., Más, B. A., Barbo, M. L. P., Motta, A. C., & Duek, E. A. R. The use of PKH26 is extending to stem cell transplants for applications including diseases of the retina, myocardium and brain and bone regeneration[12,13]. In this work, the application of PKH26 dye, subjected to the method of evaluation and monitoring of polymeric constructs containing primary osteoblastic halogen cells, aims to contribute to the knowledge needed to develop new biomaterials and the preparation of strategies that provide a greater clinical success of the application of biomaterials focused on orthopaedics and bone regeneration processes.

2. Materials and Methods 2.1 Porous scaffolds PLDLA 70/30 Solutions Lab synthesized in Biomaterials - PUC/SP for Motta and Duek[14] were prepared by dilution of the copolymer in methylene chloride (Merck) (10% w/v) at room temperature. To obtain scaffolds, PLDLA was added to the sucrose solution (Synth) (75% w/v) with controlled particle size between 200 and 500 μm. The solution was poured into a cylindrical mold of silicone with a diameter of 8 mm. After evaporation of the solvent and sucrose removal in a 1% solution of polyvinyl alcohol (PVA), scaffolds were dried under vacuum and stored in a desiccator.

2.2 Isolation and collection of osteoblastic cells Osteoblastic cells were obtained by explantation of the calvarial bone fragments of three Wistar rats (250-300 g) at 20 days of age, according to the method of Declercq et al.[15].

2.3 Cell labelling Distribution and cellular behaviour of cultured osteoblasts in the scaffolds in in vitro assays and in vivo were assessed by fluorescence microscopy. Prior to cell culture in the scaffolds, the cells were labelled with the fluorescent vital dye, PKH26 (General PKH26-GL cell linker kit, Sigma  ) according to the manufacturer’s instructions. The cell membrane of 2 x 105 cells per mL were stained with the fluorescent marker and cultured in 48-well polystyrene culture plates containing pre-sterilized glass coverslips. After 24 and 14 days of culture, the coverslips were washed in PBS and fixed in 4% paraformaldehyde (PFA-Merck)[16]. Image acquisition and evaluation was performed using a NIKON ECLIPSE E800 fluorescence microscope (NIKON Instruments) with an excitation wavelength of 543 nm and an emission wavelength.

2.4 Growing cell scaffolds The scaffolds of cylindrical PLDLA were cut into 0.5 mm thick slices, disinfected in 70% ethanol by 1 hour and rinsed in ultrapure water. Following cell staining with PKH26 dye, the stained cells were reseeded onto 96-well plates at a density of 2 x 105 cells per mL and cultured on scaffolds for either 24 h or 14 days. 84

2.5 In vitro cell viability 2.5.1 MTT assay Cell viability of osteoblasts marked by PKH26 dye was investigated by the metabolic MTT assay[17]. A concentration of 2 x 105 cells per mL was seeded on supports and controls in standard DMEM, supplemented with 10% fetal bovine serum (FBS), and further incubated for 24 h in an incubator (5% CO2 at 37 °C). Then the medium was removed, the wells rinsed with 0.1 M PBS buffer and to each well was added 100 ul of DMEM medium containing 10 uL of 3‑(4,5-dimethylthiazol‑2-yl) -2,5 diphenyl tetrazolium bromide, MTT (5 mg/mL), followed by an incubation period of 4 h at 37 °C in the dark. After this time the solution containing MTT was replaced by 200 μL of dimethylsulfoxide (DMSO) solution, and 25 ul Glycine/Sorensen buffer. Afterwards, 100 ul of the solution contained in the wells were transferred to a new plate and the absorbance of MTT was determined at 570 nm by microplate reader Elx-800-UV (Bio-Tek Instruments, USA).

2.6 Preparation of implants Cell viability and distribution of osteoblastic cells labelled by fluorescent dye, PKH26 was also investigated by fluorescence microscopy. After 24 h or 14 days of cultivation, the cured scaffolds were fixed in 13% gelatin solution, diluted with distilled water, plus 30% sucrose. The material was immersed in a solution containing 30% sucrose for a 2-3 day period and then frozen at -70 °C for 24 h and subjected to the criotomia technique in a cryostat (Shandon Cryotome E). Cryosections were obtained from the series surface of the scaffold, allowing the monitoring of the distribution of cells marked with PKH26 for the process of proliferation and cell migration into the scaffold. The slides were photographed on a fluorescence inverted optical microscope (NIKON - E 800).

2.7 In vivo experiments Thirty-four Wistar rats of both sexes (250-300 g) were used. Rats were divided into groups according to implant time (4, 8 and 12 weeks) and subdivided by treatment used, and only the 4-week time point (n = 7) was monitored for cells labelled by fluorescent dye, PKH26. The treatment groups were as follows: A control group was given pure PLDLA implants. The PLDLA group was given implants cultured with osteoblast cells marked by PKH26. 2.7.1 Implantation The main steps in the surgical procedure are represented in Figure 1A-F. Animals were subjected to general anaesthesia administered intramuscularly with 10% ketamine hydrochloride solution (40 mg/kg) and xylazine 2% (5 mg/kg). A 30 mm incision was made in the parietal region of the skull, following the path of the sagittal suture. The musculature and periosteum were also displaced to expose the parietal bone, as shown in Figure 1A. A central critical defect was made in the skull of the animal with the aid of a trephine drill of 8 mm diameter (Figure 1B). At the defect site, a tight scaffold was formed by a cylindrical mold of silicone (Figure 1E), developed into a similar shape to that of the Polímeros, 27(1), 83-91, 2017

In vitro and in vivo cell tracking of PKH26-labeled osteoblasts cultured on PLDLA scaffolds hematoxylin and eosin (HE). The slides were photographed with a fluorescence optical microscope (NIKON - E 800). The central region of each critical defect was subjected to histological analysis. To this end, the caps were divided into six sections following the coronal plane and covering the length of the defective region. With the aid of a calibrated examiner using an optical microscope fitted with incandescent and fluorescent light filters, observations and analysis of new bone formation, ACS waste, waste clot elements and tissue inflammation were carried out.

2.9 Fluorescent cell monitoring and analysis The monitoring of fluorescent cells in vivo was carried out with the help of NIS -Elements Advanced Research software image analysis installed on an optical microscope (NIKON E800). Areas and fluorescent dots (corresponding to labelled cells in culture) of the implant and new bone formation that exhibited fluorescence were detected and evaluated for distribution and cell migration inside the implants. Figure 1. Steps of the surgical procedure performed for the implantation of scaffolds: (A) incision to expose the parietal bone; (B) critical defect; (C) extraction of bone fragment; (D) defect area; (E) implanted scaffold; (F) suture.

extracted bone fragment (Figure 1C). Figure 1D illustrates the defect area and intact blood vessels, suggesting that there was no damage to the animal’s brain. The skin on the defect region was repositioned and sutured followed by local antisepsis (Figure 1F). All experimental procedures involving the use of animals were approved by the Ethics Committee on the use of animals in the Centre of Medical and Health Sciences at PUC/SP Sorocaba Campus (n.2013/13).

2.8 Processing of material for histological analysis After fixation, the material was subjected to decalcification in EDTA solution (4.13%) for 21 days. Samples were prepared for histological analysis in accordance with techniques used for light microscopy, using wax as a means of inclusion. Bone fragments containing the implants were subjected to the following processes: dehydration in a sequence of three tanks containing ethanol solutions of increasing concentration (80, 90 and 100%) where they remained for 30 min; diafanization I and II in xylene for 30 min followed by soaking in liquid paraffin I and II at 60 °C for 30 min. The material was embedded in cubic shapes measuring approximately 2.5 cm and filled with paraffin at 60 °C forming blocks that remained at rest for 24 h at room temperature to cure. We then trimmed the blocks for the following histological cutting process. For in vivo monitoring of fluorescent cells, samples were fixed in a 13% hard gelatin solution diluted with distilled water plus 30% sucrose. The material was then immersed in a solution containing 30% sucrose for 2-3 days. The material was then frozen at -70 °C for 24 h and immediately subjected to freezing in a cryostat (Shandon Cryotome E). Histological sections of 3 μm thickness were obtained using a Leica type RM2245 microtome, and stained with Polímeros, 27(1), 83-91, 2017

3. Results The manufacture of polymeric scaffolds obtained by the porogeno leaching method is widely used for both in vitro assays and in vivo, in order to assess the cytocompatibility, biocompatibility and influence of new materials on the biological behaviour of cells and tissues[18]. Some practical advantages of this method lie in its low manufacturing cost and pore size control, which are compatible with the biological specificity of each cell and tissue type[19].

3.1 In vitro study 3.1.1 Fluorescent labelling In order to monitor the manner of operation and colouring behaviour of PKH 26, primary osteoblast cells previously stained with PKH26 were grown on glass coverslips and monitored by fluorescence microscopy after 24 h and 14 days in culture (Figure 2). Cells were positive for the dye, PKH26, showing a fluorescence signal intensity sufficient for monitoring, which indicated PKH26 was suitable for use in this study. After 24 h of cultivation (Figure 2A), PKH26-labelled cells were isolated to show the projections of well-defined cytoplasmic membrane and slightly rounded morphology or Splay typical of cellular adhesion to the substrate in the early stages of cultivation. In addition, there were no apparent signs of dye cytotoxicity, such as the appearance of cellular debris or altered cellular morphology. After 14 days in culture (Figure 2B), the surfaces of all glass coverslips analysed (n = 3) were almost entirely covered with a confluent layer of cells exhibiting fluorescent signals with very similar intensity to that found after 24 h of cultivation. Due to the high density of overlapping cells adhered on the coverslip and the affinity of the dye for extracellular matrix lipophilic groups synthesized by the cells during the cultivation period, it was not possible to observe the division of the cytoplasmic membrane of osteoblasts or their morphology[20]. However, the presence of juxtaposed 85

Duek, A. R., Costa, G. C. D., Más, B. A., Barbo, M. L. P., Motta, A. C., & Duek, E. A. R.

Figure 2. Confirmation of cell labelling with the fluorescent dye, PKH26. (A) Osteoblastic cells grown on glass coverslips after 24 h; (B) osteoblast cell culture after 14 days.

cell groups inserted into the extracellular matrix were clearly distinguishable in that they exhibited a sharp and strong bright red fluorescence intensity relative to the array[21].

3.2 Cell viability According to the results presented in Figure 3 comparing the growth curves of stained and unstained osteoblast cells cultured in culture wells (n = 6), cell labelling with the dye did not cause any cytotoxic effects on osteoblast cells. From the growth curves of unstained (control) cells and osteoblasts stained with PKH26, it can be seen that the adhesion rate and cell proliferation displayed no significant difference over any of the cultivation times analysed (p > 0.01), confirming thus the cytocompatibility of the dye. Because it is a cytocompatible, nonspecific cell membrane dye, PKH 26 is widely used in monitoring studies applying the practices of tissue engineering and cell therapy[8,22,23]. However, several researchers have shown that the long functional life of the dye, or fluorescence emission in vitro and in vivo, and the degree of dye cytocompatibility may be related to the relationship between the dye concentration and cell concentration used in the experiment as well as the greater or lesser affinity of the dye for a particular cell type or tissue[24,25]. Thus, the results obtained by in vitro monitoring analyses and cell viability assay of osteoblast cells in culture stained with PKH 26 allow us to conclude that the concentrations and procedures used in this experimental protocol have proven effective for cell labelling of osteoblasts, suggesting that the deployment of cultivated PLDLA scaffolds with PKH26-stained osteoblasts, proposed in the second stage of this project, will not present risks or negative influences on the process of bone repair in vivo.

3.3 In vivo experiments After the implantation step, animals were observed according to the conditions resulting from the surgical procedure, as the clinical evolution of the neurological point of view, motor response, sensitivity, and power and water consumption. Such analysis showed that the process of implantation of scaffolds seeded with osteoblastic cells 86

Figure 3. Comparison of the growth of unstained osteoblast cells (no fluorescent labelling) and osteoblast cells stained with PKH26 after 24 h, 5 days and 12 days of cultivation.

and the surgical procedure did not exert negative influences on the recovery of the animals. After an implantation period of 4, 8 or 12 weeks, the rats were sacrificed by an overdose of halothane[26]. The calvaria containing the implant region were removed and immediately placed in Bouin’s fixative solution 10% for a period of 24 h. 3.3.1 Extraction and material processing Macroscopic evaluation of the implanted material and appearance of the implantation site was performed throughout all of the bone regeneration times studied. An improvement in the appearance of the implant site with the gradual absorption of the material over 4, 8, and 12 weeks (Figures 4A-C) as well as improved biological interaction and integration of bulk tissue and scaffold PLDLA suggested that the implanted device served its function of temporarily synthetic matrix . No acute inflammatory response signal or rejection of the implanted material was found, proving the biocompatibility of PLDLA synthesized in the laboratory, which is in agreement with a study by Más et al.[27] on osteoblast interaction with a copolymer surface. Polímeros, 27(1), 83-91, 2017

In vitro and in vivo cell tracking of PKH26-labeled osteoblasts cultured on PLDLA scaffolds 3.3.2 Histological analysis and fluorescent labelling One of the points discussed in relation to the implantation of previously cultured cells is their identification; on way to distinguish newly formed tissue and to monitor the way it behaves in the new environment is by the labelling cells with specific dyes. Another issue is whether the defect is filled with surrounding cells or with implanted cells. To evaluate the effect of implantation of scaffolds seeded with osteoblast cells marked with PKH26, the devices were grown for 14 days and implanted into the central bony defects in the

skull in Wistar rats. The first group of cells was labelled with fluorescent scaffolds for a duration of 4 weeks. The dye was clearly seen, which allowed monitoring of the cells. Figure 5 shows the micrographs of histological sections of the middle region of the implanted scaffold PLDLA area cultivated with osteoblast cells stained by PKH26. The population of cells located at the edge of the bone defect (Figure 6A) in contact with the original bone suggests that, during cell culture, there was a preference for the region as a result of further diffusion of nutrients from the culture medium in peripheral regions of the scaffolds of PLDLA[28].

Figure 4. Calvarias removed from animals at days 4 (A), 8 (B) and 12 (C) weeks, showing the evolution of regeneration in the defect area.

Figure 5. Cell monitoring PLDLA scaffolds cultured with osteoblast cells after 4 weeks of implantation. (A) Photomicrograph illustrating the population of osteoblastic cells (O) displaying red fluorescence, located on the edge of the bone defect (DB - “bone defect”); (B) Population of cells located in the central region of the implant with visible presence of the polymer scaffold (P).

Figure 6. Histological and fluorescent PLDLA of scaffolds cultured with osteoblast cells after 12 weeks of implantation. BD - bone defect and NB - neoformed bone. Polímeros, 27(1), 83-91, 2017

Duek, A. R., Costa, G. C. D., Más, B. A., Barbo, M. L. P., Motta, A. C., & Duek, E. A. R. Figure 5B illustrates the central area of the critical defects implanted with PLDLA scaffold where, although some regions show small areas colonized by pre-cultured osteoblasts and labelled with PKH26, the physical structure of the scaffold was shown to promote migration/cellular distribution within the implanted material. Figure 7 shows histological images of the bone defect area after 4 weeks (A and B) 8 weeks (C and D) and 12 weeks (E and F). On all edges illustrated in Figure 7, irregular formation and growth of bone tissue was observed. The resulting defects at 4 weeks (Figures 7A and 7B), appear to be smaller than the primary defect and, especially in areas where the external diploe is located in proximity to skeletal muscle growth. At the 4-week time point assessed, no cases of bone formation were observed without direct contact with the defect edges. At 8 weeks post implantation, the defect was apparently less severe than at the previous time point, showing continuous and asymmetric bone formation. Moreover, sharp bone growth can be noted in the closing trend from the edges, as evidenced by the greater amount of newly formed bone (Figures 7C and 7D). In no case were

bone formation focuses forming calluses for the internal diploe observed, similar to that within the limits of the brain itself. At 12 weeks, the reduction of the defect was greater with irregular growth, but throughout the surrounding area. Secondary bone formation occurs preferentially in parallel beams between them (Figures 7E and 7F), simulating the original aspect diploe. At that time, it became a clear case of bone formation without direct contact with the edges (Figure 6). Figure 8 shows the histological analysis of fluorescent and cultivated PLDLA scaffolds with osteoblast cells after 4 days (A and B), 8 weeks (C and D) and 12 weeks (E and F) of implantation. The images in Figure 8 refer to the same area of t he bone defect with the left panel representing bone stained with hematoxylin and eosin and the right panel, the corresponding fluorescent image, indicating the osteoblast cells labelled with the dye, PKH26. At 4 weeks, the distribution of fluorescent cells throughout the defect area is clearly apparent, suggesting that they contributed to the process of tissue repair in vivo

Figure 7. Histological analysis of PLDLA scaffolds cultured with osteoblast cells after 4 weeks (A and B) 8 weeks (C and D) and 12 weeks (E and F) of implantation. 88

Polímeros, 27(1), 83-91, 2017

In vitro and in vivo cell tracking of PKH26-labeled osteoblasts cultured on PLDLA scaffolds

Figure 8. Histological analysis of PLDLA and fluorescent scaffolds cultured with osteoblast cells after 4 weeks (A and B) 8 weeks (C and D) and 12 weeks (E and F) of implantation.

as shown in the highlighted areas of Figure 8B. At 8 weeks, similarly, the image suggests the migration of dye-labelled cells, evidenced both in the defect area and at the edges of the newly formed bone. Already at 12 weeks, the image reveals the formation of bone spurs, irregular, primary and surrounded by osteoblast cells. Figure 6 shows, independent of defect edge (BD), an area of newly formed bone (TON) significant after 12 weeks of implantation. In Figure 6B, the arrows point to cells precultured and implanted with fluorescent dye, both in the bone defect area linked to the edge and the unrelated surrounding area, suggesting a contribution of these active cells to the formation of this fragment. Among the important requirements of a dye of choice stands out the following: the ability to remain detectable after a long time, ease of handling and non-interference in immunocytochemical responses of cells implanted in the Polímeros, 27(1), 83-91, 2017

tissue. In this work, we verified the cytocompatibility the PKH26 dye in the cultivation of osteoblast cells for 14 days and also by MTT assay.

4. Conclusions In this work, the cytocompatibility the PKH26 dye was verified in osteoblasts in cell culture for 14 days on PLDLA scaffolds by MTT assay. The dye did not exert negative influences on cell growth of osteoblasts in relation to unstained cells. In the in vivo study, macroscopic observations made during deployment times corroborated the in vitro results, as no apparent signs of toxicity were observed in the implanted bone defect area. The use of mobile monitoring with the dye, PKH26 in vivo is an effective strategy for the understanding of cell behaviour in the presence of PLDLA polymer. 89

Duek, A. R., Costa, G. C. D., Más, B. A., Barbo, M. L. P., Motta, A. C., & Duek, E. A. R.

5. References 1. Dantas, T. S., Lelis, E. R., Naves, L. Z., Fernandes-Neto, A. J., & Magalhães, D. (2011). Bone graft materials and their application in dentistry. UNOPAR Científica Ciências Biológicas e da Saúde, 13(2), 131-135. Retrieved in 9 November 2015, from pgsskroton.com.br/seer/index.php/biologicas/article/ viewFile/1248/1198 2. Kunz, F., Bergemann, C., Klinkenberg, E. D., Weidmann, A., Lange, R., Beck, U., & Nebe, J. B. (2010). A novel modular device for 3-D bone cell culture and nondestructive cell analysis. Acta Biomaterialia, 6(9), 3798-3807. PMid:20227531. http:// dx.doi.org/10.1016/j.actbio.2010.03.015. 3. Ikeda, T., Ikeda, K., Yamamoto, K., Ishizaki, H., Yoshizawa, Y., Yanagiguchi, K., Yamada, S., & Hayashi, Y. (2014). Fabrication and characteristics of chitosan sponge as a tissue engineering scaffold. BioMed Research International, 6, 1-8. PMid:24804246. http://dx.doi.org/10.1155/2014/786892. 4. Kroeze, R. J., Helder, M. N., Govaert, L. E., & Smit, T. H. (2009). Biodegradable Polymers in Bone, Tissue Engineering. Materials (Basel), 2(3), 833-856. http://dx.doi.org/10.3390/ ma2030833. 5. Alexander, J. T., Branch, C. L., Jr, Subach, B. R., & Haid, R. W., Jr. (2002). Applications of a resorbable interbody spacer via a posterior lumbar interbody fusion technique. Orthopedics, 25(10), S1185-S1189. PMid:12401030. 6. Moser, R. C., Mcmanus, A., Riley, S., & Thomas, K. (2005). Strength retention of 70:30 Poly(Llactide-co- D, L lactide) following real-time aging. Journal of Biomedical Materials Research. Part B, Applied Biomaterials, 75(1), 56-63. PMid:16001395. http://dx.doi.org/10.1002/jbm.b.30238. 7. Brazelton, T. R., & Blau, H. M. (2005). Optimizing techniques for tracking transplanted stem cells in vivo. Stem Cells, 23(9), 1251-1265. PMid:16109764. http://dx.doi.org/10.1634/ stemcells.2005-0149. 8. Yoo, J. J., Bichara, D. A., Zhao, X., Randolph, M. A., & Gill, T. J. (2011). Implant-assisted meniscal repair in vivo using a chondrocyte-seeded flexible PLGA scaffold. Journal of Biomedical Materials Research. Part A, 99(1), 102-108. PMid:21800420. http://dx.doi.org/10.1002/jbm.a.33168. 9. Polzer, H., Volkmer, E., Saller, M. M., Prall, W. C., Haasters, F., Drosse, I., Anz, D., Mutschler, W., & Schieker, M. (2012). Long-term detection of fluorescently labeled human mesenchymal stem cell in vitro and in vivo by semi-automated microscopy. Tissue Engineering. Part C, Methods, 18(2), 156-165. PMid:21951128. http://dx.doi.org/10.1089/ten. tec.2011.0275. 10. Matz, R. L., Erickson, B., Vaidyanathan, S., KukowskaLatallo, J. F., Baker, J. R., Jr., Orr, B. G., & Banaszak Holl, M. M. (2013). Polyplex exposure inhibits cell cycle, increase inflammatory response, and can cause protein expression without cell division. Molecular Pharmaceutics, 10(4), 13061317. PMid:23458572. http://dx.doi.org/10.1021/mp300470d. 11. Wang, W.-J., Wu, S.-P., Liu, J.-B., Shi, Y.-S., Huang, X., Zhang, Q.-B., & Yao, K.-T. (2013). MYC regulation of CHK1 and CHK2 promotes radioresistance in a stem cell-like population of nasopharyngeal carcinoma cells. Cancer Research, 73(3), 1219-1231. PMid:23269272. http://dx.doi.org/10.1158/00085472.CAN-12-1408. 12. Canola, K., Angenieux, B., Tekaya, M., Quiambao, A., Naash, M. I., Munier, F. L., Schorderet, D. F., & Arsenijevic, Y. (2007). Retinal stem cells transplanted into models of late stages of retinitis pigmentosa preferentially adopt a glial or a retinal ganglion cell fate. Investigative Ophthalmology & Visual Science, 48(1), 446-454. PMid:17197566. http://dx.doi. org/10.1167/iovs.06-0190. 90

13. Boomsma, R. A., Swaminathan, P. D., & Geenen, D. L. (2007). Intravenously injected mesenchymal stem cells home to viable myocardium after coronary occlusion and preserve systolic function without altering infarct size. International Journal of Cardiology, 122(1), 17-28. PMid:17187879. http://dx.doi. org/10.1016/j.ijcard.2006.11.022. 14. Motta, A. C., & Duek, E. A. R. (2007). Synthesis and characterization of poly (L-co-D,L acid lactic). Polímeros: Ciência e Tecnologia, 17(2), 123-129. http://dx.doi.org/10.1590/ S0104-14282007000200011. 15. Declercq, H. A., Verbeeck, R. M. H., Ridder, L. I. F. J. M., Schacht, E. H., & Cornelissen, M. J. (2005). Calcification as an indicator of osteocoindutive capacity of biomaterials in osteoblastic cells cultures. Biomaterials, 26(24), 4964-4974. PMid:15769532. http://dx.doi.org/10.1016/j.biomaterials.2005.01.025. 16. Duffy, G. P., Mcfadden, T. M., Byrne, E. M., Gill, S. L., Farrell, E., & O’Brien, F. J. (2011). Towards in vitro vascularisation of collagen-GAG scaffolds. European Cells & Materials, 21(12), 15-30. PMid:21225592. http://dx.doi.org/10.22203/ eCM.v021a02. 17. Ignatius, A. A., & Claes, L. E. (1996). In vitro biocompatibility of bioresorbable polymers: poly (L, DL-lactide) and poly(L-lactideco-glycolide). Biomaterials, 17(8), 831-839. PMid:8730968. http://dx.doi.org/10.1016/0142-9612(96)81421-9. 18. Thadavirul, N., Pavasant, P., & Supaphol, P. (2014). Development of polycaprolactone porous scaffolds by combining solvent casting, particulate leaching, and polymer leaching techniques for bone tissue engineering. Journal of Biomedical Materials Research. Part A, 102(10), 3379-3392. PMid:24132871. http:// dx.doi.org/10.1002/jbm.a.35010. 19. Sabir, M. I., Xu, E. X., & Li, L. (2009). A review on biodegradable polymeric materials for bone tissue engineering applications. Journal of Materials Science, 44(21), 5713-5724. http://dx.doi. org/10.1007/s10853-009-3770-7. 20. Chen, J., Wang, C., Lu, S., Wu, J., Guo, X., Duan, C., Dong, L., Song, Y., Zhang, J., Jing, D., Wu, L., Ding, J., & Li, D. (2005). In vivo chondrogenesis of adult bone-marrow-derived autologous mesenchymal stem cells. Cell and Tissue Research, 319(3), 429-438. PMid:15672263. http://dx.doi.org/10.1007/ s00441-004-1025-0. 21. Christian, W., Johnson, T. S., & Gill, T. J. (2008). In vitro and in vivo cell tracking of chondrocytes of different origin by fluorescent PKH 26 and CMFDA. Journal of Biomedical Science and Engineering, 1(3), 163-169. http://dx.doi.org/10.4236/ jbise.2008.13027. 22. Kang, E. J., Byun, J. H., Choi, Y. J., Maeng, G. H., Lee, S. L., Kang, D. H., Lee, J. S., Rho, G. J., & Park, B. W. (2010). In vitro and in vivo osteogenesis of porcine skin-derived mesenchymal stem cell–like cells with a demineralized bone and fibrin glue scaffold. Tissue Engineering. Part A, 16(3), 815-827. PMid:19778183. http://dx.doi.org/10.1089/ten. tea.2009.0439. 23. Lee, J. Y., Choi, M. H., Shin, E. Y., & Kang, Y. K. (2011). Autologous mesenchymal stem cells loaded in Gelfoam for structural bone allograft healing in rabbits. Cell and Tissue Banking, 12(4), 299-309. PMid:20652421. http://dx.doi. org/10.1007/s10561-010-9194-4. 24. Li, P., Zhang, R., Sun, H., Chen, L., Liu, F., Yao, C., Du, M., & Jiang, X. (2013). PKH26 can transfer to host cells in vitro and vivo. Stem Cells and Development, 22(2), 340-344. PMid:22913652. http://dx.doi.org/10.1089/scd.2012.0357. 25. Park, B. W., Kang, E. J., Byun, J. H., Son, M. G., Kim, H. J., Hah, Y. S., Kim, T. H., Mohana Kumar, B., Ock, S. A., & Rho, G. J. (2012). In vitro and in vivo osteogenesis of human mesenchymal stem cells derived from skin, bone marrow and dental follicle tissues. Differentiation, 83(5), 249-259. PMid:22469856. http://dx.doi.org/10.1016/j.diff.2012.02.008. Polímeros, 27(1), 83-91, 2017

In vitro and in vivo cell tracking of PKH26-labeled osteoblasts cultured on PLDLA scaffolds 26. Lemos, M. M. (2006). Experimental study on the effect of induced tendinitis in the gastrocnemius muscle: histopathology and raman spectroscopy. Franca: Universidade de Franca. 27. Más, B. A. (2011). Imobilização de colágeno em arcabouços de poli (L-co-D,L ácido lático) (Master’s dissertation). Faculdade de Engenharia Mecânica, Universidade Estadual de Campinas, Campinas. 28. Tavassol, F., Schumann, P., Lindhorst, D., Sinikovic, B., Voss, A., von See, C., Kampmann, A., Bormann, K. H., Carvalho, C., Mülhaupt, R., Harder, Y., Laschke, M. W., Menger, M. D.,

Polímeros, 27(1), 83-91, 2017

Gellrich, N. C., & Rücker, M. (2010). Accelerated angiogenic host tissue response to poly (L-lactide-co-glycolide) scaffolds by vitalization with osteoblast-like cells. Tissue Engineering. Part A, 16(7), 2265-2279. PMid:20184434. http://dx.doi. org/10.1089/ten.tea.2008.0457. Received: Nov. 09, 2015 Revised: May 06, 2016 Accepted: June 06, 2016

Polímeros

OOLPOLY® THERMALLY CONDUCTIVE PLASTICS

Dissipadores de calor injetados com CoolPoly® oferecem: - Redução de custos

- Liberdade no projeto

- Isolantes elétricos

VOLUME XXVII - Issue I - JAN/MAR - 2017

- Melhor transferência de calor