Hydrocarbon Engineering October 2021 by PalladianPublications

October 2021

A WORLD LEADER IN SULPHUR PROCESSING AND HANDLING End-to-end systems from receipt of molten sulphur to loading of solid material – single source supply by IPCO. Ƨǎǎbî§ØºăØǎeßùß±ßîØǎë òùºÌÌ ùºßÙư Ƨǎǎ ăëÌ§Ĕǎòù§§Ìǎ §Ìùòǎòë§ º±º ÌÌĕǎ ÌÌßĕ§£ǎùßǎî§òºòùǎ ßîîßòºßÙư Ƨǎǎ(º²·ǎ ë ºùĕǎ£îăØǎ²î ÙăÌ ùºßÙư Ƨǎǎ ßďÙòùî§ Øǎòùßî ²§ǎƿǎòºÌßǎ Ù£ǎßë§Ùƶ Ìßò§£ǎòùß ÊëºÌ§òư Ƨǎǎ ăòùßØǀ ăºÌùǎî§ Ì ºØ§îòǎ±ßîǎ ÙĕǎÌß ùºßÙư Ƨǎǎpîă Êƫǎî ºÌǎ Ù£ǎò·ºëǎÌß £ºÙ²ǎ Ù£ǎ ²²ºÙ²ǎòĕòù§Øòư ƧǎǎI î²§ǎò Ì§ǎ Ìß ÊǎëßăîºÙ²ư

VPP pǎ +p(ǎui iăÌë·ăîǎǘǎiăÌë·ăîº ǎ º£ǎŬŪŬū ºîùă Ìǎ ßÙ±§î§Ù §ǎȄǎ Ĕ·º ºùºßÙ ūǀŮǎPßĎ§Ø §îǎŬŪŬū

ipco.com/sulphur

CONTENTS October 2021 Volume 26 Number 10 ISSN 1468-9340

03 Comment

29 The key points Michael Gaura, AMETEK Process Instruments, USA, highlights the key points in a sulfur recovery unit tail gas treatment unit.

05 World news 08 A delicate return Contributing Editor, Nancy D. Yamaguchi, takes a look at the post-pandemic oil industry in the Middle East, and questions the extent of its recovery.

33 Contactless combustion analysis Dr Jonas Westberg, Dr Viacheslav Avetisov and Dr Peter Geiser, NEO Monitors, Norway, outline how an evolution in TDLAS signal processing is enabling new applications for the next generation of smart combustion analysers.

37 The critical measurements Dr Stephen Firth, Servomex, UK, examines the critical measurements for safety and control in PTA production processes.

42 Trading risk for reliability, flexibility and eﬃciency Sami Tabaza, Atlas Copco Gas and Process, USA, explains why integrally geared centrifugal compression technology is increasingly being used in NGL plants alongside oil-flooded gas screw compressors.

48 Simulation in a world of trouble: part two

19 Green SRU in biofuels Jan Klok, Desirée de Haan, Joost Timmerman and Rieks de Rink, Paqell B.V., the Netherlands, outline a number of differentiators that make biological desulfurisation suitable for the rising biofuels market.

25 Finding flexibility: part one In the first of two parts, Brandon Forbes and DJ Cipriano, Ametek CSI, and Marco van Son, Worley Comprimo, explore a more flexible way to degas sulfur.

In the second of two installments, Ralph H. Weiland and Nathan A. Hatcher, Optimized Gas Treating Inc., explore case studies to show how simulation is used in troubleshooting.

53 The problem with tramp air Bill Johnson and Chuck Baukal, John Zink Hamworthy Combustion, a Koch Engineered Solutions Company, USA, evaluate the impact of tramp air leakage on process heater operation, as well as corrective and preventative actions.

57 H2 ready

Eric Pratchard, Zeeco, USA, evaluates the challenges to consider when transitioning to firing hydrogen.

61 The last mile Stuart Morstead, Honeywell Connected Industrial, presents an overview of the main trends shaping the future of the industry, and outlines the need to connect processes, people and assets to meet new challenges.

THIS MONTH'S FRONT COVER JOIN THE CONVERSATION follow

@HydrocarbonEng

Hydrocarbon Engineering

join

Atlas Copco Gas and Process demonstrates how to increase reliability and flexibility in NGL plants using centrifugal compression technology.

2021 Member of ABC Audit Bureau of Circulations

Copyright© Palladian Publications Ltd 2021. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the copyright owner. All views expressed in this journal are those of the respective contributors and are not necessarily the opinions of the publisher, neither do the publishers endorse any of the claims made in the articles or the advertisements. Printed in the UK.

Hydrocarbon Engineering

CBP006075

COMMENT CONTACT INFO

CALLUM O'REILLY SENIOR EDITOR

MANAGING EDITOR James Little

james.little@palladianpublications.com SENIOR EDITOR Callum O'Reilly

callum.oreilly@palladianpublications.com EDITORIAL ASSISTANT Pippa Luck

pippa.luck@palladianpublications.com SALES DIRECTOR Rod Hardy

rod.hardy@palladianpublications.com SALES MANAGER Chris Atkin

chris.atkin@palladianpublications.com SALES EXECUTIVE Sophie Barrett

sophie.barrett@palladianpublications.com PRODUCTION Kyla Waller

kyla.waller@palladianpublications.com DIGITAL EDITORIAL ASSISTANT Bella Weetch

bella.weetch@palladianpublications.com DIGITAL EVENTS COORDINATOR Louise Cameron

louise.cameron@palladianpublications.com VIDEO CONTENT ASSISTANT Molly Bryant

molly.bryant@palladianpublications.com DIGITAL ADMINISTRATOR Lauren Fox

lauren.fox@palladianpublications.com ADMIN MANAGER Laura White

laura.white@palladianpublications.com CONTRIBUTING EDITOR Nancy Yamaguchi Gordon Cope

SUBSCRIPTION RATES Annual subscription £110 UK including postage /£125 overseas (postage airmail). Two year discounted rate £176 UK including postage/£200 overseas (postage airmail). SUBSCRIPTION CLAIMS Claims for non receipt of issues must be made within 3 months of publication of the issue or they will not be honoured without charge.

APPLICABLE ONLY TO USA & CANADA Hydrocarbon Engineering (ISSN No: 1468-9340, USPS No: 020-998) is published monthly by Palladian Publications Ltd GBR and distributed in the USA by Asendia USA, 17B S Middlesex Ave, Monroe NJ 08831. Periodicals postage paid New Brunswick, NJ and additional mailing offices. POSTMASTER: send address changes to HYDROCARBON ENGINEERING, 701C Ashland Ave, Folcroft PA 19032.

15 South Street, Farnham, Surrey GU9 7QU, UK Tel: +44 (0) 1252 718 999 Fax: +44 (0) 1252 718 992

ll eyes will soon be on Glasgow for the 2021 UN Climate Change Conference (COP26). Ahead of hosting the conference, the UK’s Prime Minister Boris Johnson has already claimed that it will represent the “turning point for humanity.” As the world’s political leaders decide on what actions should be taken to help decarbonise the planet, the global oil and gas sector faces continued uncertainty. In its recently published ‘2021 CEO Outlook’, KPMG notes that climate change risk continues to be seen as the number one threat by energy leaders. Of the 133 CEOs that were surveyed about their strategies and outlook over the next three years, 37% selected climate change as their top risk category as it relates to their organisation’s growth. Over half of respondents also report that they are under pressure from investors and regulators for increased reporting and transparency on environmental, social and governance (ESG) issues. Meanwhile, almost one-third of CEOs stated that attracting and retaining talent was their top priority over the next three years, with 86% of energy leaders planning to increase their workforce over the same period. However, attracting new talent to our sector at a time of heightened environmental concerns is challenging, as Regina Mayor, KPMG’s Global Head of Energy and Natural Resources, explains: “The industry has faced talent and skills shortages before, but there’s a sense that, this time, we may not be able to win the hearts and minds of the talented individuals who have gotten us to where we are today.” It is our responsibility, as an industry, to change this narrative in order to attract the next generation of talent, including passionate environmentalists. The AFPM recently published an interesting article in Politico explaining why the idea of an environmentalist working in the downstream sector need not be seen as the ultimate paradox.1 By seeking out careers in the refining and petrochemical sectors, these workers can advance positive outcomes and encourage change from within. In the article, the AFPM presents three case studies of environmentalists driving sustainability efforts at US refining and petrochemical companies. It highlights an environmental biologist supporting ecological and biodiversity efforts at Valero’s refineries and ethanol plants, a geologist who founded an environmental stewardship committee at CITGO, and a principal engineer and wildlife photographer aiding a wetland’s transformation at Westlake Chemical. Each of these workers entered the downstream sector with a vision to help improve the environment and aid sustainability efforts. As Bill Goulet – the nature photographer at Westlake Chemical – explains: “If you want to make the most impact and protect the environment, why would you not work for an oil, gas or a petrochemical plant? That is where the most good can, and is, being done. That is where you want the experts.” This is a message that we need to make loud and clear in order to ensure that our sector continues to attract (and retain) the next generation of talent, who will be essential to driving both growth and sustainability efforts within our industry. 1.

‘From the inside: how industry environmentalists are driving sustainability efforts within refining and petrochemical companies,’ Politico, (15 September 2021).

HYDROCARBON

ENGINEERING

October 2021

FFCs slurry oils sell for just a few hundred dollars/ton, but when converted into Carbon Black – a key component in high-end plastics, tire manufacturing, and other rubber-based components – they can reach a market price of $2,500 USD/ton. ET Black™: the industry technology of choice Eurotecnica’s modified furnace black process, ET Black™, efficiently converts slurry oils into a wide range of carbon black grades, giving you access to an exciting and fast-growing industry. With guaranteed operating flexibility and reliability, easy maintenance, and a low CAPEX, ET Black™ has become the technology of choice for industry leaders like ADNOC Refining. Contact us at www.igoforETBLACK.com to find out more.

WORLD NEWS China | Lummus

selected for DPC facility

ummus Technology has been awarded a technology contract by Zhangzhou CHIMEI Chemical Co., a subsidiary of CHIMEI Corp. of Taiwan, for a grassroots diphenyl carbonate (DPC) plant in Fujian Province, China. Zhangzhou CHIMEI will build a new polycarbonate (PC) production plant that leverages the Versalis DPC technology licensed by Lummus. The plant is expected to reach mass production in 4Q24. Zhangzhou CHIMEI’s new PC plant will implement eco-friendly features, introduce advanced

technologies to optimise energy efficiency, and produce intermediate products that can be recycled and reused in the circular economy. PC materials are some of the highest performing materials in the market because of their high-temperature resistance, high-impact strength and transparency. Lummus’ scope for this award includes the technology license, process design package services, operator training and technical services.

USA | EIA

expects increasing consumption of natural gas by US industry

ased on the US Energy Information Administration’s (EIA) September ‘Short-Term Energy Outlook’, industrial sector natural gas consumption is expected to rise throughout 2021 and to exceed pre-pandemic 2019 levels. The EIA forecasts the growth to continue into 2022, with natural gas delivered to industrial consumers averaging 23.8 billion ft3/d next year. If realised, this amount would be near the current record high for

annual industrial natural gas consumption set in the early 1970s. In its report, the EIA expects natural gas consumption in the US industrial sector to average 23.5 billion ft3/d in the second half of this year and 23.2 billion ft3/d for 2021. If realised, this amount of industrial natural gas consumption would exceed the 2019 average of 23.1 billion ft3/d and mark the most US industrial natural gas consumption since 1997.

The Netherlands | Shell

to build biofuel facility

oyal Dutch Shell plc has announced a final investment decision to build an 820 000 tpy biofuels facility at the Shell Energy and Chemicals Park Rotterdam, the Netherlands, formerly known as the Pernis refinery. Once built, the facility will be among the biggest in Europe to produce sustainable aviation fuel (SAF) and renewable diesel made from waste. A facility of this size could produce enough renewable diesel to avoid 2 800 000 tpy of carbon dioxide (CO2) emissions a year, the equivalent of taking more than 1 million European cars off the roads. The new facility will help Europe to meet internationally binding emissions reduction targets. It will also help Shell to meet its own target of becoming a net-zero emissions energy business by 2050, in step with society’s progress towards achieving the climate goals of the Paris Agreement. Advanced production methods will be used to make the fuels. The facility is expected to use technology to capture carbon emissions from the manufacturing process and store them in an empty gas field beneath the North Sea through the Porthos project.

Sweden | Honeywell

and Preem conduct commercial co-processing trial to produce renewable fuel

oneywell has announced the completion of a commercial refinery trial with Preem AB for the co-processing of biomass-based pyrolysis oil in a fluidised catalytic cracking (FCC) unit. Utilising UOP’s proprietary bioliquid feed system with OptimixTM GF Feed Distributor, pyrolysis oil was successfully

co-processed in the FCC at Preem’s Lysekil refinery to produce partially renewable transportation fuel. This test marks the sixth commercial co-processing trial conducted by UOP worldwide of this technology in an FCC, and the first pyrolysis oil co-processing trial in Scandinavia using UOP’s Optimix GF Feed Distributor technology.

To meet Sweden’s long-term goals of greenhouse gas (GHG) emission reductions, fuel suppliers must reduce the carbon intensity of transportation fuels. Co-processing of biomass-based pyrolysis oil is one method to reduce the carbon intensity of transport fuels at the refinery compared to blending of biofuels downstream. HYDROCARBON

ENGINEERING

October 2021

WORLD NEWS DIARY DATES 12 - 13 October 2021 Gulf Coast Conference Galveston, Texas, USA www.gulfcoastconference.com

12 - 15 October 2021 & 21 - 22 October 2021 Downstream USA Online & Houston, Texas, USA www.reutersevents.com/events/downstream

13 - 14 October 2021 Valve World Expo Americas Houston, Texas, USA www.valveworldexpoamericas.com

25 - 27 October 2021

USA | MEGlobal

Americas plans 100% renewable electricity at petrochemical plant

EGlobal Americas Inc. has entered into an agreement with Calpine Energy Solutions LLC to purchase renewable energy to fulfil 100% of the expected power needs at MEGlobal’s Oyster Creek, Texas, petrochemical site, beginning in 2023. The deal highlights the strong commitment made to sustainability by the EQUATE Petrochemical Group, of which MEGlobal is a part.

Opportunity Crudes Conference Online www.opportunitycrudes.com

Russia | SIBUR

01 - 04 November 2021

Sulphur + Sulphuric Acid 2021 Online www.sulphurconference.com

15 - 18 November 2021 ADIPEC Abu Dhabi, UAE www.adipec.com

15 - 18 November 2021 ERTC Madrid, Spain www.worldrefiningassociation.com/ertc21

01 - 02 December 2021 14th Annual National Aboveground Storage Tank Conference & Trade Show The Woodlands, Texas, USA www.nistm.org

05 - 09 December 2021 23rd World Petroleum Congress Houston, Texas, USA 23wpchouston.com

14 - 16 December 2021 Turbomachinery & Pump Symposia Houston, Texas, USA tps.tamu.edu

To keep up with all the latest news on key industry events in light of the COVID-19 pandemic, visit hydrocarbonengineering.com/events

October 2021

HYDROCARBON

ENGINEERING

MEGlobal will purchase more than 1.5 million MWh of renewable electricity over the five-year term of the agreement, resulting in the displacement of more than 600 000 t of carbon dioxide. When compared to common carbon offsets, the amount of displaced CO2e from this purchase is equal to the amount of carbon sequestered by more than 1.8 million acres of forest in a year.

and TAIF finalise merger terms

IBUR and TAIF have entered into an agreement finalising the creation of a combined entity that will leverage the facilities of PJSC SIBUR Holding and create a petrochemical company in Russia. The new company will support Russia’s leadership in the chemical, petrochemical, and oil and gas sectors, while making the country’s products more globally competitive thanks to economies of scale, higher production efficiency, better sales processes, and improved customer

service in line with global best practices. Following the completion of all ongoing investment projects, the new combined company will be among the top five producers of polyolefin and rubber globally. The company’s investment programme will yield further growth of chemical non-commodity exports and import substitution, as well as unlock new ways of tackling sustainability and environmental challenges.

Europe | BP

and Brightmark explore plastics renewal plants

P and Brightmark have signed a Memorandum of Understanding to jointly evaluate opportunities for the development of the next generation of plastic waste renewal plants in Germany, the Netherlands, and Belgium. The companies will evaluate opportunities for projects that convert end-of-life waste plastics otherwise destined for incineration, landfill, or export, into valuable petrochemical feedstocks for plastics and other industrial applications.

Brightmark’s proprietary plastics renewal process recycles plastic waste that has reached the end of its useful life, including items not currently recyclable using conventional mechanical processes (types 3 – 7), such as plastic film, flexible packaging, styrofoam, and plastic beverage cups. As a first step, Brightmark and BP intend to work together to develop plans that could lead to the construction of an initial European plant.

MAKE EVERY MOLECULE MATTER Something as small as a series of chemical reactions can have a profound impact on the health of our planet and its people. This is why Shell Catalysts & Technologies’ mission is to Make Every Molecule Matter. We are focused on making investments in more and cleaner energy solutions – to enable industry to tackle global climate challenges starting at the molecular level. Our experienced team of scientists and engineers apply our diverse, unique owner-operator expertise to co-create solutions to your specific emissions or energy efficiency challenges. Learn more at catalysts.shell.com/MEMM

October 2021

HYDROCARBON

ENGINEERING

Contributing Editor, Nancy D. Yamaguchi, takes a look at the post-pandemic oil industry in the Middle East, and questions the extent of its recovery.

he COVID-19 pandemic caused untold suffering in 2020, until at last, in 2021 it seemed the worst was over. Economies around the globe began to reopen. Crude oil prices rose. In July 2021, the OPEC+ group announced that it would begin to phase out its production cut agreement. The group noted clear signs of improvement in demand and a reduction in crude oil stocks. The plan is to allow 400 000 bpd of additional supply each month, beginning in August. This halted the upward trend in oil prices, but it did not cause a massive reversal. It suggested that the market is ready to absorb additional supply, but not a massive volume and not all at once. Yet the questions remain: is the global economy truly on its way to recovery? Is the COVID-19 pandemic truly under control? The desire to believe this during the first half of 2021 caused a wave of exuberance, which opened the door for an increase in infections, particularly from the Delta variant. As this article is being completed in early September 2021, full economic recovery remains in jeopardy in key markets. COVID-19 has infected over 222 million people and caused over 4.6 million deaths. All Middle Eastern countries are affected by the virus – crude oil producers, refiners, traders, and marketers. Cases have trended back up noticeably in Iraq and Iran, the two most populous countries in the region and two of the largest oil producers and consumers. At the time of writing, Iran had over 5.2 million confirmed cases and over 112 000 deaths. Iraq had over 1.9 million cases and over 21 000 deaths. The Middle Eastern countries find themselves pondering the correct course in a world where the pandemic seemed to be coming under control, but the current resurgence shows that it has not yet

HYDROCARBON

ENGINEERING

October 2021

Table 1. COVID-19 in the Middle East COVID-19 cases

Deaths

Iran

3 603 527

87 837

Iraq

1 518 837

18 020

Israel

855 552

6454

Jordan

762 706

9922

UAE

665 533

1907

Lebanon

552 328

7888

Saudi Arabia

513 284

8115

Kuwait

388 881

2255

State of Palestine

315 876

3591

Oman

289 042

3498

Qatar

224 638

600

Bahrain

26 092

2381

Syria

25 849

1905

Yemen

6997

1371

Middle East

9 749 142

155 744

Source: Johns Hopkins, 21 July 2021

been defeated. Each country charts a delicate course, assessing the harm to their own populations and domestic economies, then rethinking developments in the global market and how they may change oil sector activity and investments. Now, achieving the delicate balance may grow more delicate. Internationally, even before the pandemic, global oil supplies from non-OPEC sources were growing, including light tight oil (LTO) production from US shale plays, and increases in Russian output. Oil demand growth was slowing in many countries, and demand was even starting a downward slide in many of the European countries that had been key customers for crude and product. Delegates from 195 countries have just approved a report authored by the UN-backed Intergovernmental Panel on Climate Change, which warned that the planet will warm by 1.5°C in the next two decades without drastic cuts to pollution. Some cuts in fossil energy use will be structural, and not related to the pandemic. Middle Eastern oil producers were already fighting to support prices without losing too much market share. During the worst of the pandemic, there was little to be done. But this year, as the world appeared to be emerging from the pandemic, Middle Eastern and allied producer countries began to consider when and how to relax their voluntary production ceilings. Oil prices in the vicinity of US$50 – 60/bbl seemed at first the essence of stability, then the first half of 2021 brought price recovery to US$70 – 75/bbl, causing exuberance. This article focuses on Middle Eastern oil in the lingering COVID-19 pandemic, which continues to grip the world. All through the region, the questions remain: Is the pandemic truly coming to an end? Will the delta variant delay or derail recovery? When shall we make the plans and investments to expand crude and refined product availability?

COVID-19 and the Middle East Figure 1. Middle Eastern oil producers recent rise in COVID-19 infections.

As of late July 2021, there were over 9.7 million confirmed cases of COVID-19 in the Middle East, and 155 744 deaths attributed to the disease. Iran had the highest number of cases, over 4.1 million, and the highest number of deaths, over 94 000, as of early August. By September, these numbers had risen to 5.2 million and 112 000, respectively. Table 1 provides country-level details on Middle Eastern COVID-19 cases and fatalities. Adding to concerns, there has been a recent uptick in infections. Figure 1 displays this trend for the five major oil producers: Iran, Iraq, Kuwait, Saudi Arabia, and the UAE. In order to compare the countries side by side, the data provides smoothed trend lines showing new infections per million population, illustrating a clear upward movement.

The pandemic and oil price volatility Figure 2. Brent crude spot prices.

October 2021 10 HYDROCARBON ENGINEERING

Figure 2 illustrates the COVID-19 pandemic’s profound impacts on Brent crude spot prices from the beginning of 2020 to early September 2021.

Decrease OPEX

Reduce emissions

Optimize capacity

Delivering sulphur solutions for a more sustainable world Our large range of technology solutions ensures our customers keep their costs low, reduce their carbon footprint and meet or exceed their sulphur recovery targets. Learn more with one of our experts email: comprimo@worley.com

Improve uptime

worley.com/comprimo

Figure 3. Middle East OPEC crude protection,

January 2020 – June 2021, ‘000 bpd (source: OPEC).

Saudi Arabia and Russia, agreed in principle to increase supply. The UAE objected to the initial baselines from which production would be allowed to rise, since the Emirates have invested in production capacity and are able to go well above the initial baseline. A compromise was reached that allowed OPEC+ to reach consensus once again. In mid-July 2021, the OPEC+ group announced that it would begin to increase output, adding 400 000 bpd of supply each month, beginning in August. Oil prices stuttered but did not collapse. The longer-term impact on prices will depend largely upon the health of demand as the months proceed and new supplies hit the market. With COVID-19 cases disturbingly on the upswing, demand recovery may be threatened. The US Energy Information Administration (EIA) noted that Brent spot prices averaged UA$73/bbl in June, but the agency forecast that global supply will grow faster than demand in the second half of 2021, resulting in modest decline in Brent spot prices to US$72/bbl on average. This forecast has not yet been revised to account for the current resurgence in COVID-19 cases, which has the potential to drag prices below US$70/bbl.

Pandemic impacts on Middle East oil production

Figure 4. Middle East shrinking share of global oil production, % (source: BP). In January 2020, Brent spot prices were approximately US$70/bbl. As the virus spread and economic activity was curtailed, prices went into freefall, bottoming out at US$9.12/bbl on 21 April 2021. The path back took over a year, with Brent spot prices not re-attaining the US$70/bbl level until June 2021. Prices strengthened to US$78.34/bbl on 5 July, the highest daily spot price in over three years. Prices have cycled up and down since then, however, as market caution returns alongside the rise of COVID-19 infections combined with the planned increase in OPEC+ crude supplies. Brent spot prices have been hovering around US$66 – 72/bbl. OPEC continues to monitor and support prices via its ‘Declaration of Cooperation’, which is nicknamed ‘OPEC+’. The group has also been nicknamed ‘ROPEC’ to emphasise the importance of Russian participation, which is not always in accord with Middle Eastern goals. Saudi Arabia remains the leader of the OPEC members while Russia is the leader of the non-OPEC members. The members often reported heated debates at their OPEC+ meetings, yet the group has demonstrated resilience and an ability to reach consensus time and time again. During the first half of 2021, prices rose strongly as markets reopened. The two top producers, October 2021 12 HYDROCARBON ENGINEERING

The COVID-19 pandemic caused a huge drop in Middle Eastern OPEC production. The OPEC+ production cut agreement kept output at a restrained level during the remainder of 2020 and into 2021. These cuts could not forestall a collapse in oil prices, but they helped stabilise prices over the ensuing year. Figure 3 shows the trend in Middle Eastern OPEC crude production from January 2020 through June 2021. Output hit a peak of 25.08 million bpd in April 2020, as producers bumped up production in anticipation of the cuts to come. Output then plunged to 17.62 million bpd in June 2020, a massive drop of 7.46 million bpd in a mere two months. Output crept back to the vicinity of 19.5 million bpd by the summer of 2020. OPEC reports that these key producers raised production by 0.97 million bpd in May and June 2021. Total OPEC production in June 2021 averaged 26.034 million bpd.

Middle East petroleum in the greater market The COVID-19 pandemic has been a tribulation across many economic sectors around the world, and it has had a huge impact on the Middle East. Millions of residents have suffered through the disease, thousands have lost their lives, and the disease continues to spread. In the petroleum sector, the region’s key oil producers had been serving as global swing producers, working together to support prices by keeping millions of barrels off the export market. Many of these producers are concerned about losing market share, and as Figure 4 illustrates, this concern is valid. The Middle East has seen its share of global oil production shrink. According to British Petroleum (BP,) Middle Eastern petroleum accounted for 35% of the global total in 2016, and this share fell to 31% in 2020.

OPEC anticipates that OECD demand will recover by 2.65 million bpd in 2021, and production will rise by 0.41 million bpd. The net import requirement will creep back up to 15.16 million bpd, still 4.53 million bpd below its 2018 level. The Middle East is also facing a drop in refinery utilisation rates, as shown in Figure 6. In the early 1990s, refinery utilisation rates were often 90% and higher. Utilisation rates trended gently down until the pandemic caused a sharp drop to 75% in 2020, according to data published by BP. Without a significant recovery in global demand, coupled perhaps with refinery closures abroad, low refinery utilisation will stifle enthusiasm for refinery investment plans.

Figure 5. OECD liquids supply and demand, million bpd.

Figure 6. Middle East refinery utilisation is falling, % (source: BP).

While it is true that producers outside the Middle East have also suffered from the pandemic, the Middle East has borne a relatively larger share. Figure 5 shows the liquids supply and demand balance for the countries of the Organisation for Economic Co-operation and Development (OECD). In 2018, OECD liquids supply was 28.3 million bpd, juxtaposed against demand of 47.99 million bpd, giving a supply deficit of approximately 19.69 million bpd – a rough estimate of the call on imports. In 2019, OECD liquids production rose to 30.01 million bpd, while demand declined to 47.69 million bpd, cutting net import requirements by 2 million bpd. In 2020, the pandemic caused OECD demand to constrict to 42.06 million bpd, a drop of 5.63 million bpd. Production, however, declined by only 0.86 million bpd, to an average of 29.15 million bpd. This cut OECD import requirements by another 4.77 million bpd. Therefore, between 2018 and 2020, OECD demand dropped by 5.93 million bpd, but its production rose 0.86 million bpd. The net import requirement fell from 19.69 million bpd in 2018 to 12.92 million bpd in 2020. Without the restraint shown by Middle Eastern oil producers, the oil price collapse would have been far worse. October 2021 14 HYDROCARBON ENGINEERING

Middle East recent market developments The recent swings in global oil prices and demand have had varied impacts within the Middle East. There is a divide between the net oil importers and net oil exporters, in terms of both crude oil and refined products. Looking first at the product importers, Israel has two refineries, but the larger one, Oil Refineries Ltd (ORL) at Haifa is under threat of closure by the government, which has proposed shutting down the heaviest polluting factories. ORL has proposed investments to become a cleaner energy company, but such a project could be years in coming, and the country would rely more on imported fuels. Jordan, Lebanon, Syria and Yemen rely heavily on product imports, though all four either have or have had refineries, and all appear to have plans to restore and/or expand their refinery presence. Yemen has continually proposed refinery construction plans, but nothing can be considered firm. The country has been embroiled in civil war since 2014. Jordan’s 90 000 bpd Zarqa refinery has run at low utilisation rates, but an expansion and upgrade project is planned. Syria has a nameplate capacity of 240 000 bpd, but refinery output has been less than half of this. Both of Syria’s refineries are slated for upgrades, and the Banias refinery is reportedly back in operation. These countries have modest-sized oil markets, with demand typically in the 100 000 – 200 000 bpd range. Their fuel needs are met with relative ease by regional refiners. The larger exporters have sophisticated refineries capable of producing EURO specification fuels, and they made investments based on the presence of these external markets. After a wave of optimism in mid-2021, export refiners face an uneasy prospect of another COVID surge derailing demand recovery. Most of the Middle Eastern countries have refining industries and goals to capture the value-added of processing and exporting petroleum and natural gas resources. For some, the presence of large domestic markets supported initial stages of refinery construction. As fuel quality standards tightened abroad and at home, billions of dollars were spent to upgrade and modernise refineries. At the present time, however, many export-led ambitions have been put on hold. As noted, refinery utilisation rates have fallen, and there is no guarantee of demand growth in the customary export markets. How are these refining centres faring, and do they

Sulfur recovery unit workers have a lot to worry about. Analyzers shouldn’t be one of them. Managing all the processes in a sulfur recovery unit (SRU) is arduous work—demanding skill, concentration, and dedication through every shift. Fortunately, the reliability, accuracy, robust design, and operating ease of AMETEK analyzers can make that tough work a little easier. AMETEK engineers have been designing industry-standard SRU analyzers for decades, and that shows in the products’ accuracy, reliability, and longevity. Because we make analyzers for every part of the process—from acid-feed gas to tail gas to emissions, including the gas treating unit, sulfur storage (pit) gas, and hot/wet stack gas—you get the convenience of one source for unparalleled engineering and support for all your analyzers, while your operators benefit from consistent interfaces and operating procedures. For decades, we’ve been dedicated to making your SRU operation the most efficient it can be for the long term. Learn more at www.ametekpi.com/SRU.

rose from 1 717 000 bpd in 2018 to 1 841 000 bpd in 2019, before retreating to 1 715 000 bpd in 2020. Upgrade programmes are planned at essentially all key refineries, but timing will be determined by availability of funds, materials, and contractors. The country is working to improve gasoline and diesel quality. Large grassroots projects are unlikely while sanctions remain, so the focus is on upgrading and expanding existing refineries, including the recently completed Persian Gulf Star plant. Lack of investment is expected to cut Iran’s crude production capability severely, so that even if sanctions are lifted, crude output may never regain historic levels. Iran placed its production capacity at 4.7 million bpd, but actual production in 2020 was less than 2 million bpd. As noted, the EIA reported that Iran’s crude oil export revenues collapsed from US$66 billion in 2018 to Figure 7. Drop in oil export revenues, Middle East OPEC, US$11 billion in the January – September period of billion US$ (source: US Energy Information Administration). 2020. A rebound in crude export revenues will be critical for many government initiatives. It is also seen as essential to tackle years of mismanagement have plans to expand capacity? Plans remain on the books, but of the agricultural sector and water supply. Energy and food ability to pay is a significant barrier. have been heavily subsidised, and Iran has been declared Figure 7 shows oil export revenues in the Middle Eastern ‘water bankrupt’ with respect to renewable water supply. OPEC countries Iran, Iraq, Kuwait, Saudi Arabia, and the UAE, Iraq according to the US Energy Information Administration (EIA.) In Iraqi crude production has declined significantly since 2018, these five countries earned US$528 billion in oil export COVID-19 hit. In 2019, production averaged 4.678 million bpd. revenues. This declined to US$448 billion in 2019, then plunged This fell by 0.627 million bpd in 2020 to an average of to US$194 billion in the January – September period of 2020 – a 4.051 million bpd. OPEC data for the first half of 2021 place drop of 63%. Iran suffered the largest drop; crude oil export Iraqi output at 3.915 million bpd, a drop of an additional revenues of US$66 billion in 2018 collapsed to US$11 billion in 0.136 million bpd. the January – September period of 2020. This was a drop of Iraq’s refining industry is organised mainly around three 83%. regional nodes: the North Refining Co.’s Baiji refinery, the The following sections provide a brief look at developments Midland Refining Co.’s Daura refinery, and the in the key refining countries. South Refining Co.’s Basra refinery. Smaller refineries, some Bahrain modular, have been installed to help meet local fuel needs. Bahrain has had a refinery expansion and modernisation project Total crude capacity is approximately 1.1 million bpd. underway at its 262 000 bpd Sitra refinery for several years. The Sophistication is low, however, and gasoline output lags project is largely complete, but full commissioning may be demand while fuel oil is in surplus. BP reports that Iraqi oil pushed back to 2022. The project is adding approximately demand was 919 000 bpd in 2020. The Basra and Baiji 100 000 bpd of crude capacity, plus hydrocracking, coking, and refineries are slated for upgrades, but these are not expected diesel hydrotreating. The domestic market is small, to be completed for several years. Work on the expansion at approximately 31 000 bpd in size, so the additional product Basra was halted during the pandemic. would be destined for export markets. Because the existing A government-funded grassroots refinery of 150 000 bpd output from the Sitra is already more than ample, there is no capacity is underway at Karbala. In recent years, there have rush to expand. been multiple proposals for new refinery projects, but the weak market quelled investor interest, and some projects were Iran dropped. Recently, the government announced that it had Weighed down by years of sanctions, Iranian crude oil exports awarded the contract for the long-planned Al-Fao (also have plunged, revenue has fallen, and outside investment has spelled Al-Faw) refinery to the China National Chemical faded. Despite limited revenue, the country forged ahead with Engineering Co. (CNCEC.) The government announced that the internal projects. The key refinery project was the Persian Gulf refinery will have a capacity of 300 000 bpd and will include a Star refinery, a three-phase gas condensate refinery at petrochemical plant. The Chinese government reportedly will Bandar Abbas. Production commenced in 2018, and in 2019, Iran provide funding. No schedule has been announced. at last became a net exporter of gasoline. Prior to this, Iran had Kuwait been heavily dependent on gasoline imports. Product exports According to OPEC, Kuwait cut its crude production from reportedly rose to 680 000 bpd in 2020. Iran has the largest 2.689 million bpd in 2019 to 2.431 million bpd in 2020, and to market in the Middle East. BP reports that Iranian oil demand October 2021 16 HYDROCARBON ENGINEERING

2.341 million bpd during the first half of 2021. This is a reduction totalling 348 000 bpd. Kuwait operates two large refineries, Mina Abdullah and Mina Al-Ahmadi, which have a combined capacity of 736 000 bpd. An older refinery, Shuiba, was decommissioned. Commissioning of the giant 615 000-bpd Al Zoor refinery project has been postponed because of the COVID-19 pandemic. This refinery will be the largest in the Middle East when complete. It was scheduled for 2020 startup, but contractors tried to invoke force majeure because of COVID-19 restrictions. A launch in 2021 remains possible, but only if the pandemic is under control and people and materials begin to travel more freely. Kuwaiti oil demand fell 468 000 bpd in 2018 to 446 000 bpd in 2019 and to 411 000 bpd in 2020, according to BP.

Oman According to BP, Oman’s oil demand fell from 240 000 bpd in 2019 to 209 000 bpd in 2020. The country has two refineries, Mina Al-Fahal and Sohar, with a combined capacity of approximately 222 000 bpd. The Omani government is working with Kuwait Petroleum Intl to build a joint venture refinery known by the abbreviation ‘OQ8’ at Duqm. This will be a 230 000 bpd deep conversion refinery with hydrocracker plus coker. The project is reportedly over 80% complete, though market conditions provide no urgency for full commissioning.

Qatar Qatar’s oil demand fell from 375 000 bpd in 2019 to 296 000 bpd in 2020, according to BP. The country has two refineries, Messaieed and Ras Laffan, with a combined capacity of 429 000 bpd. The Messaieed refinery, also known as MIC because of its location at Messaieed Industrial City, is undergoing an expansion and modernisation programme. The refinery completed its ultra-low sulfur diesel project in 2020. The refinery fully supplies the domestic market with a 10 ppm sulfur ULSD standard consistent with EURO 5 quality.

Saudi Arabia According to BP, Saudi Arabian oil demand has fallen for five consecutive years, declining from 3 879 000 bpd in 2015 to 3 544 000 bpd in 2020, a drop of 335 000 bpd. The country remains the region’s largest crude oil producer, refiner, and consumer, and it is the de facto head of the OPEC+ group. Saudi Arabian crude production rose from 9 198 000 bpd in 2008 to 10 317 000 bpd in 2018, a growth rate of 1.2% per year. The country has immense production capability, and it uses this capability to moderate global prices. Saudi Arabia went well below its quota under the production cut agreement, which was set at 10.508 million bpd, to ensure compliance. According to OPEC, Saudi Arabian crude output averaged 9.765 million bpd in 2019, 9.194 million bpd in 2020, and 8.471 million bpd during the first half of 2021. This was a drop of 1.294 million bpd. Saudi Arabia has a nameplate crude refining capacity of approximately 2.9 million bpd, centred around eight key refineries. These include the highly sophisticated, modern joint venture refineries known as SAMREF, YASREF, and SATORP. The 400 000 bpd Jizan refinery is largely complete, with some units operating, but full commissioning is not expected until 2022. Market conditions have not recovered, and the facility has also October 2021 18 HYDROCARBON ENGINEERING

been attacked by Yemeni missiles. Yet another complication stems from the fact that the refinery receives crude feedstock via tanker, raising the possibility of seaborne attacks. Saudi Arabia has been working for a solution that will extricate it from the Yemeni civil war. Saudi Arabia was the first Middle Eastern country to invest heavily in export refining, and it is also a pioneer in integrating refining with petrochemicals. For example, Saudi Aramco and Total are building a joint venture petrochemical plant next to the SATORP refinery.

UAE Oil demand in the UAE has continued to grow despite the COVID-19 pandemic. According to BP, the UAE consumed 1.229 million bpd of oil in 2018, 1.307 million bpd in 2019, and 1.331 million bpd in 2020. The UAE expanded its crude production capacity, though it cut production in line with the OPEC+ agreement and market demand. OPEC reports that UAE crude output fell from 3.08 million bpd in 2019 to 2.793 million bpd in 2020 and 2.627 million bpd during the first half of 2020. The UAE argued, with some success, that its OPEC+ production baseline should be increased, and its future allotment will grow from the new baseline. A compromise was reached that allowed the OPEC+ group to move ahead with its 2021 plan. The UAE then announced that it would increase export capability of its Murban crude. The UAE has expanded its refinery capacity, which now stands at approximately 1.14 million bpd. There are four main refinery complexes, the largest of which is Abu Dhabi National Oil Co.’s (ADNOC) Ruwais refinery. ADNOC reports that there has been significant progress made on its crude flexibility project, which will enable the plant to process up to 420 000 bpd of heavy sour crudes. Dubai’s ENOC refinery is planning to add a 70 000 bpd condensate refinery.

Conclusion: delicacy returns to a delicate balancing The Middle East has grown accustomed to taking a balancing position in the global oil market, though this has cost market share. The oil producers in the region serve as the anchor in the OPEC+ agreement, which has been far more successful than originally expected over the past years of structural oversupply, and currently during the COVID-19 pandemic. During the first half of 2021, it seemed that the coronavirus was finally coming under control. Oil demand and prices were heading up. Unfortunately, the highly contagious delta variant is contributing to another surge in infections, potentially jeopardising the economic recovery. The Middle Eastern countries are proceeding with a gradual increase in crude production, and they are resuming some of their work on downstream upgrading and expansions. All have been battered by the pandemic, however. All have lost oil export revenue. This will limit how many projects can move forward. Even in cases where governments and investors are able to proceed, some are pondering when (and whether) they should. Thus, the summer of 2021 has failed to bring a full recovery, and there is no guaranteed end to the COVID-19 pandemic. The Middle Eastern oil producers and refiners find a new delicacy returning to their constant battle to find a delicate balance in today’s global oil market.

Jan Klok, Desirée de Haan, Joost Timmerman and Rieks de Rink, Paqell B.V., the Netherlands, outline a number of differentiators that make biological desulfurisation suitable for the rising biofuels market.

n 2015, 196 countries signed the Paris Agreement, an international treaty with the goal of limiting global warming to well below 2°C. To achieve this, countries have set net-zero greenhouse gas emissions targets that involve both emissions abatement and reduction. Production of biofuels, e.g. fuel produced from biomass through contemporary processes, is undergoing a tremendous expansion with new biofuel projects being executed.1 When hydrogen sulfide (H2S) is formed during the production of biofuels, desulfurisation is required to control the emission of sulfur compounds. For projects with a high sulfur load, i.e. >100 tpd of sulfur, the combination of the amine and Claus process is the most cost-effective desulfurisation line-up. However, for gas streams with intermediate and low sulfur loads (e.g. up to 20 tpd of sulfur), several commercial technologies are available. The sulfur load of a typical biofuel project is in this range. In addition, the off-gases associated with biofuel production typically have a very unfavourable CO2/H2S ratio. Therefore, the

traditional Claus technology is not a suitable desulfurisation process for biofuel projects. Besides the economics (i.e. OPEX and CAPEX), other project specific requirements determine which technology is most suitable to apply. For example, flexibility, project timeline and CO2 footprint are often important considerations in the selection of the most adequate desulfurisation technology.2 The concept definition is known to be crucial for successful project development. Recently, several biofuel projects located in the US, Europe, and Asia, selected the Thiopaq O&G (TOG) biological desulfurisation process. This article discusses essential differentiators of this technology to highlight its suitability for sulfur recovery units (SRUs) in biofuel projects.

The technology Thiopaq is a biological desulfurisation process that uses naturally occurring bacteria to oxidise H2S to biological elemental sulfur. The process has its origins in biogas treatment and more than

HYDROCARBON 19

ENGINEERING

October 2021

Figure 1. A Thiopaq O&G facility treating high

pressure (60 barg) natural gas in four lines. Each line is designed to treat 5.4 tpd of sulfur. The TOG consists of three steps. 1: H2S removal in an absorber, 2: solution regeneration in a bioreactor, and 3: sulfur dewatering section using a settler for sulfur settling (3) and decanter-centrifuges, which are typically placed in a separate building (4).

to control the alkalinity, salinity and biomass concentration of the process solution. Based on lessons learned from operating plants, a new process line-up has been developed: the TOG Ultra. This process has a considerably higher conversion efficiency of sulfide to sulfur. It uses a more active biocatalyst, leading to a doubled bioreactor efficiency, and allowing for a smaller bioreactor. Furthermore, due to improvements in process conditions, the absorbed hydrogen sulfide is oxidised more effectively, leading to lower caustic consumption and make-up water. The optimised process conditions are the result of the implementation of an extra bioreactor between the absorber and the aerated bioreactor. The new line-up brings a further benefit of improved operational stability and improved settling of the produced bio-sulfur, resulting in easier separation in the sulfur recovery section.4 While the first TOG Ultra has been sold in Iraq in 2018, several others are currently under construction in Europe and Asia.

Process economics

Figure 2. Scanning electron microscopy picture of

produced biosulfur from a TOG. Biologically formed sulfur particles have a bipyramidal shape. Due to the relatively small particle size and hydrophilicity of the biologically produced elemental sulfur, the process solution does not require the addition of chemicals to stabilise the sulfur.

270 units have been installed worldwide.3 The process was introduced to the oil and gas industry in 2002 and is currently marketed under the name Thiopaq O&G (TOG). The traditional biotechnological desulfurisation process consists of three simple, integrated steps: absorption, solution regeneration, and sulfur recovery (see Figure 1). The process starts by absorbing H2S into the process solution, thereby forming dissolved sulfide which consumes the alkalinity of the process solution. The treated gas leaving the absorber column (containing <25 ppmv H2S) can be further processed. If needed, TOG can guarantee <4 ppmv H2S. The sulfide-rich process solution is mixed with air in the bioreactor where bacteria oxidise the dissolved sulfide to elemental sulfur (see Figure 2). This regenerates the process solution as alkalinity is restored. A slipstream of the process solution is directed to the sulfur recovery section. The sulfur is dewatered, yielding a sulfur cake of approximately 65 wt% solids. The filtrate is recycled back to the system. Caustic, make-up water and nutrients are supplied October 2021 20 HYDROCARBON ENGINEERING

Next to technical feasibility, a main driver for process selection is the process economics. As SRU selection is case specific, e.g. gas composition, flow, sulfur load and site requirements, extensive techno-economic studies have been performed.4, 5, 6, 7 As a first rule of thumb, TOG is economically attractive for gas streams with a sulfur load between 50 kg/d of sulfur up to 150 tpd of sulfur. At the lower end of the sulfur load spectrum (<50 kg/d of sulfur), non-regenerable scavengers are typically more cost-effective. At the higher end of the sulfur load, the cost-effectiveness depends on the nature of the feed gas. For example, the cost-effectiveness of TOG (Ultra) increases at higher and thus more challenging CO2/H2S ratios. Typically, the production of biofuels is associated with effluent gas streams containing these higher CO2/H2S ratios. TOG is also applied for direct treatment of natural gas streams.5, 8 For these gas streams, the general scheme of AGRU-SRU-TGT line-up can be replaced with TOG, resulting in: Less equipment. No requirement for burners and reboilers. Regeneration and sulfur recovery sections are operated at atmospheric pressure and ambient temperature. Examples of direct treatment of natural gas streams with TOG are shown in Figures 1 and 3. When the natural gas requires CO2 removal, a sweetening unit (AGRU) is required upstream of the TOG. In these cases, the technology is particularly cost-effective for lean acid gases (i.e. H2S concentrations <20%) up to 70 – 150 tpd of sulfur. To treat lean acid gas streams with the Claus process, one or more enrichment steps are required, which severely impacts the economics and operability.6 However, the technology might still be cost-effective for rich acid gas (i.e. without the need of an enrichment step).4 Other typical gas streams with challenging CO2/H2S concentrations are Claus tail gas streams. H2S removal with conventional tail gas treatment technologies is very energy consuming. While the Claus process produces valuable steam that can be reused in other parts of the plant, it has been shown that the energy used by a TGTU to achieve World Bank

DYNAMIC SIMULATION SERVICES FOR CENTRIFUGAL AND AXIAL FLOW COMPRESSION SYSTEMS

EXAMPLE: simulation of ESD events to GH¿ QH FKHFN KRW E\ SDVV YDOYH

Dynamic compressor maps

COMPSYS™ SCENARIOS 9 Normal and emergency VKXW GRZQV (ESD) 9 Machine VWDUW XS sequences (series and parallel) 9 Process FKDQJHV 9 Valves or other equipment failure events

92.6 % of MCS

BENEFITS

SLL

Cv - 92.1%

9 )LQHU VL]LQJ of hot by-pass valves

PHcorr (kJ/kg)

9 )DVWHU ¿ QHU G\QDPLFV FRPSDUHG WR RWKHU WRROV 9 7LPHO\ proper routing & sizing of control lines & valves

91.3 % (@ 0 s)

Valve size optimization

89.0 % (@ 0.3 s)

Cv - 9.5% -6.2% +28.2%

86.0% (@ 0.7 s)

+39.6%

9 %URDGHU RII GHVLJQ process & equipment checks

+13.3%

+83.5%

+3.2%

9 Realistic analysis of controllers VWDELOLW\ LQVWDELOLW\ 9 )LQHU SUH WXQLQJ of compressors antisurge, performance & load sharing controls 9 'H¿ QLWLRQ FKHFN RI FRPSOH[ operational procedures to reach target operating points ZZZ VDWH LWDO\ FRP 6DQWD &URFH D 9HQH]LD ,WDO\ LQIR#VDWH LWDO\ FRP SK

SEEING IS

BELIEVING

Optimized value 0%

Qcorr (m3/s) Selected size +25.4%

Consulting Simulation services & Software for process plants and machines

Take a look at our ABC Certiﬁcate. It shows our circulation has been independently veriﬁed to industry agreed standards. So our advertisers know they’re getting what they paid for.

ABC. See it. Believe it. Trust it.

www.abc.org.uk

standard sulfur emission limits results in a net energy consumption.9 Alternatively, the THIOPAQ process always achieves <25 ppmv H2S, making it suitable to reach even the strictest sulfur emission specifications. As such, it is suitable in revamp cases where tight SO2 emissions standards are being introduced. In addition, it is also a suitable technology for greenfield applications.10

Process flexibility and fast track construction

Figure 3. The economical envelope based on sulfur load: the project sulfur loads where TOG economics are favourable compared to other technologies. At the lower end of the sulfur load spectrum (<50 kg/d of sulfur, indicated by the red line), technology A is typically more cost-effective, whilst for very high sulfur loads (>150 tpd of sulfur, indicated by the black line) technology B is most feasible. The area between the green and purple line marks the sulfur load that is representative for biofuel production facilities. At the higher end of the sulfur load, cost-effectiveness depends on the feed gas composition. The more challenging CO2/H2S feed gas ratio, the more favourable a TOG as SRU is.

Figure 4. A Thiopaq O&G facility in the US, designed for 1.5 tpd of sulfur. The facility treats associated gas, required to produce LNG. The overall facility pay-back time was in the order of 2.5 – 3 years.8

Figure 5. A: comparison of consumable costs between design criteria of consumptions per t of sulfur and actual data under turndown conditions collected over a year of operation.11 B: flexible design of the TOG facility with two absorbers (not visible in the picture), five road-transportable bioreactors, and pumps mounted on skids.

October 2021 22 HYDROCARBON ENGINEERING

A unique aspect of the technology is its catalyst – a mixture of different species of sulfide-oxidising bacteria. These organisms are naturally occurring and use the released energy from the oxidation of sulfide to elemental sulfur to multiply themselves. The use of a biocatalyst has several advantages over traditional catalysts, since traditional catalyst is subjected to degradation, and hence continuous replenishment is required, and the catalyst in the biological process is not affected by turndown, i.e. less sulfide entering the system will typically result in lower growth rates but still guarantees the complete oxidation of all captured hydrogen sulfide. Recently, turndown operation of a TOG unit, operated at 16% of the design capacity, has been studied and reported.11 H2S levels in the treated gas always remained within permit. In addition, based on its flexible, skid-built design, the facility is able to deal with turndown operation while securing design criteria for consumables per t of sulfur (see Figure 4). Another differentiator of this technology is the speed of construction. In 2017, an early production facility (EPF) was taken in operation, treating a gas from a gas field located in the Mediterranean Sea.12 TOG was selected to be used as a SRU in the EPF, as the technology can be designed and constructed in less than a year, has favourable CAPEX and OPEX, and is flexible in the amount and concentration of H2S it can handle. More TOG facilities have been constructed following fast track. For example, in the facility depicted in Figure 1, first gas was taken in 9 months after definition of the basis of design.

HAZID study Recently, a health, safety and the environment (HSE) comparison between TOG and other technologies has been performed, using the HAZard IDentification (HAZID) methodology.6 HAZID is used for early identification of potential hazards and issues. The benefit of this methodology is that it includes assessment of operational aspects of a process and is used to select the most suitable technology for a given project. In this HAZID study, three different SRU technologies including TOG were compared. The study indicated that the THIOPAQ O&G process has fewer hazards in comparison with the two other SRU technologies, because there is no free H2S once the feed gas enters the unit. Furthermore, it has a better operability because it requires less operator attention and maintenance, and it has an infinite turndown with respect to gas rate and H2S content.

Desulfurisation in the market of biofuels An important requirement for the SRU in biofuel projects is a strong flexibility towards fluctuations in sulfur load and gas composition because of frequent changes in feedstock.

Thiopaq O&G and Thiopaq O&G Ultra The proven gas desulphurisation technology.

stable by nature

How to reach lowest Opex and highest value when treating natural gas streams for sulphur? THIOPAQ O&G puts you in control of sulphur removal and sulphur recovery. Perform well on safety, sustainability, reliability, cost and operability. Oil & Gas companies worldwide rely on THIOPAQ O&G. See why on paqell.com/thiopaq. Paqell’s THIOPAQ O&G - exceptional achievements in H2S removal.

paqell.com

emissions, CCS will be of great support. Lastly, one of the end-products of TOG is the formation of a cake of biological sulfur. Due to the unique properties of the biosulfur, it can be used in the fertilizer and fungicide industry. The latter will not only reduce the carbon footprint but will also move the circular sulfur economy forward.

References 1.

Figure 6. HAZID comparison for Thiopaq O&G and two other technologies based on HSE, operations and maintenance.6

Furthermore, the SRU should be cost-effective in the sulfur load range from 400 kg/d of sulfur up to 20 tpd of sulfur and should be able to treat gas streams with challenging CO2/H2S ratios. While biofuel projects are long-term projects, return on investment (ROI) in biofuel projects using TOG is currently estimated to be less than three years. In addition to economics, TOG can also cope with fluctuations in feed gas. The recent studies on TOG discussed in this article show excellent performance under the required flexible turndown conditions. Another advantage of this technology as an SRU is that the treated gas is well suited for carbon capture and storage (CCS). Since biofuel production aims to reduce CO2

RODIONOV, M. V., et al. ‘Biofuel production: challenges and opportunities’, International Journal of Hydrogen Energy, 42.12 (2017): 8450-8461. 2. JACQUES, M. and KÖNING-ADOLPH, A., ‘Sulphur management for an unconventional gas-condensate field’, Sulphur, pp. 37 - 41, (2017). 3. O’CALLAGHAN, P., et al., ‘Assessing and anticipating the real world impact of innovative water technologies’, Journal of Cleaner Production, (2021): 128056. 4. KLOK, J.B.M., et al., ‘Techno-economic impact of THIOPAQ-SQ, the next generation of the THIOPAQ O&G process for sulfur removal’, SOGAT, Abu Dhabi, UAE, (2017). 5. VAN HEERINGEN, G, and KLOK, J., ‘A new generation’, Hydrocarbon Engineering, (February 2018), pp. 77 - 81. 6. LEE, K. et al., ‘Thiopaq O&G bio-desulphurisation; an alternative recovery Technology’, GPA, Edinburgh, UK, (2013). 7. CURTESCU, A. et al., ‘Why choose a one-size fits-all TGTU and risk leaving money on the table?’, CRU Middle East Sulphur, Abu Dhabi, UAE, (2017). 8. LANNING, A., et al., ‘Stranded No More – Turning Sour Casinghead Gas Into Profits Using Shell-Paques® Bio-Desulfurization’, Laurance Reid Gas Conditioning Conference, Norman, Oklahoma, USA, (2008). 9. SLAVENS, A., et al., ‘Sulphur Plants Produce Valuable Energy Use it Wisely!’, CRU Middle East Sulphur, Abu Dhabi, UAE, (2017). 10. CURTESCU, A. and VAN HEERINGEN, G., ‘Strike a balance’, Hydrocarbon Engineering, (October 2017), pp. 77 - 80. 11. KLOK, J. B. M., et al, ‘Desulfurization of amine acid gas under turn down: Performance of the biological desulfurization process’ Laurance Reid Gas Conditioning Conference, Norman, Oklahoma, USA, (2018). 12. KLOK, J. B. M., et al, ‘Sulfur recovery in Early Production Facility’, GPA, Kuwait, (2019).

3URGXFHG :DWHU 5H¿QHU\ :DVWHZDWHU 7UHDWPHQW ®

AntiMercure Systems ®

October 2021 24 HYDROCARBON ENGINEERING

• Proven 99% mercury removal from water • Continuous operation • Easy installation and start-up • Pre-engineered, skid mounted – 3 standard sizes

www.AntiMercure.com

In the first of two parts, Brandon Forbes and DJ Cipriano, Ametek CSI, and Marco van Son, Worley Comprimo, explore a more flexible way to degas sulfur.

rude oil and natural gas contain naturally occurring sulfur compounds. In order to avoid practical and environmental problems due to sulfur in the end products, the bulk of this sulfur must be removed. Various processes are used to convert and/or remove the sulfur compounds, the most common compound being hydrogen sulfide (H2S). One process, the modified Claus process, is almost universally used to convert H2S into elemental sulfur. An unintended by-product of the Claus process is incorporation of unreacted H2S into the elemental sulfur. Mixtures of H2S and air have two unfortunate properties; the mixture can be both poisonous and explosive. Control of H2S in the produced sulfur is therefore a high priority for

refineries and gas plants. Extended exposure to air-born H2S at concentrations as low as 5 ppmv is likely to cause respiratory irritation and other symptoms. Extended exposure to concentrations as low as 100 ppmv can cause death. H2S in sulfur is not stable and will gradually migrate into any available vapour space. This process is relatively slow, thus emission of H2S from the sulfur remains a concern for downstream sulfur storage, handling and transportation. H2S is explosive starting at concentrations of roughly 4% by volume. Thus, accumulation of H2S in the vapour space of storage vessels is also a significant concern. This combination of H2S/air mixtures being both poisonous and explosive leaves sulfur facilities with few good choices for HYDROCARBON 25

ENGINEERING

October 2021

Figure 1. Degassing chemical reaction.

refineries and gas plants have been adding degassing equipment to their facilities in greater number. Where degassing is used, the typical target is 10 ppmw H2S or less in the sulfur. At this level, the equilibrium concentration limit of the vapour space is less than 4%vol. H2S. Thus, it is not physically possible to achieve an explosive H2S/air mixture in a storage vessel. It should be noted, however, that fires are still possible as the sulfur itself is flammable. Also, the H2S concentration in the vapour space may still exceed lethal levels. The ICOnTM (in-situ Claus optimisation) degassing technology brings new degassing capabilities and options to refineries and gas plants. ICOn has the unique ability to be installed immediately after the Claus unit sulfur condensers – essentially eliminating the need to store un-degassed sulfur. ICOn also provides greater configuration flexibility to better accommodate the varying restrictions inherit in retrofit applications.

Description of chemistry

Figure 2. ICOn degassing contactor cross-section. safe handling. Storing the sulfur in air-containing vessels, tanks, or concrete pits creates an explosion concern. Venting the storage vessels and transfer points to remove the evolved H2S creates an exposure concern. The general industry consensus at this time is: To aggressively sweep/vent sulfur storage containers to prevent H2S accumulation. To treat the vent gas via incineration or Claus processing to prevent exposure and environmental damage. But personnel exposure remains a concern throughout the sulfur handling process; fresh air breathing apparatus and aggressive venting are commonly used in areas with potential exposure to sulfur. Sulfur degassing to remove the H2S has long been used throughout the various sulfur industries. In the past, degassing was typically performed only as a necessary step for forming and transporting sulfur. But over time, degassing for safety reasons has become more common. Similarly, there is a trend to move the degassing process further upstream so that the safety benefits are provided earlier in the process. For this reason, October 2021 26 HYDROCARBON ENGINEERING

The H2S contained within the liquid sulfur exists in two chemical forms: dissolved H2S and chemically bound H2S. Dissolved H2S is simply H2S molecules mixed in with the liquid sulfur molecules. This form of H2S can be removed readily by agitation. The rate of removal is driven primarily by the sulfur surface area exposed to the vapour. Spraying the sulfur through a vapour space or sparging the sulfur with a vapour stream are both effective methods of removing dissolved H2S. All existing degassing technologies use some form of sparging, spraying, or other agitation with vapour. Chemically bound H2S is more challenging to remove. Above 159°C (318°F), sulfur rings will open up due to homolytic bond scission to form a chain of sulfur atoms with radicals at both ends. During the Claus process, these radicals react with the available H2S to form hydrogen polysulfide (H2SX). These polysulfides likely exist in a variety of sulfur chain lengths and quite possibly as HSX· radicals. Below 159°C (318°F), these chains are unstable and will slowly break down into elemental sulfur and H2S. But this process is very slow and is equilibrium-limited by the surrounding dissolved H2S (Figure 1 provides a graphical representation of the reactions). The ICOn degassing approach accelerates the breakdown of the hydrogen polysulfides by simultaneously using a catalyst and a sparge vapour. The catalyst accelerates the decomposition of the hydrogen polysulfides; the sparge vapour removes dissolved H2S. While these steps are distinct, they occur in parallel. The decomposition of the hydrogen polysulfides appears to be limited by the presence of dissolved H2S in the sulfur. Thus, continual removal of dissolved H2S is needed to facilitate the decomposition reaction. This ‘1-2-punch’ approach fully degasses the sulfur with less than five minutes’ residence time.

OGT | ProTreat® brings out

Impact of temperature

PTU 2

To achieve the above criteria, the ICOn contactor uses a packed-bed catalyst configuration. This is represented in Figure 2. The liquid sulfur moves through the catalyst in a horizontal direction. The ‘sides’ of the catalyst zone adjoin an inlet and an outlet plenum that provide even distribution of sulfur flow through the catalyst. The sulfur leaves the contactor by overflowing out of the outlet nozzle; the liquid level is set by the outlet nozzle elevation.

The degassing reaction described above occurs in the ICOn degassing contactor. To support the reaction, the contactor must be configured based on the following key criteria: The sulfur must be sparged in the presence of the catalyst so that the H2S is removed as it is formed. The sparge vapour rate must be high enough to provide adequate ‘mixing’ to promote the interaction of sulfur, vapour, and catalyst. The sparge vapour must contain a sufficiently low concentration of H2S to enable degassing to below 10 ppm H2S content in the sulfur. The sulfur must be provided at a temperature below 159°C, and preferably below 150°C. The sulfur must have adequate residence time in the contactor.

Design of the contactor

THE EXPERT IN YOU

NU MB

Sulfur freezes at 120°C (248°F). Liquid sulfur above this temperature nominally takes on a molecular form analogous to the shape of an octagon comprised of eight sulfur atoms. But at temperatures above 159°C (318°F), these octagons open up into diradical chains, which combine to form polymer chains. The ‘friction’ between these polymer chains accounts for the sharp rise in viscosity that takes place at 159°C. Liquid sulfur begins to form in the sulfur condenser following either the thermal or catalytic conversion stages. Most of the sulfur condenses at a temperature well above the transition point described above. Thus, diradical sulfur chains will form as condensed sulfur rings open up. Hydrogen sulfide reacts with these open diradical polymer chains forming hydrogen polysulfides. This is significant for a couple of reasons. First, it creates the need for degassing as discussed above. Second, the ‘capped’ chains are prevented from combining into longer chains; this effectively dampens the viscosity spike observed in elemental sulfur. The magnitude of the spike is reduced, and the peak of the spike occurs at a higher temperature. This is significant for SRU design as sulfur from the first several condensers in a typical Claus unit is hotter than 159°C, and yet readily flows through the run-down lines to the pit. This is also significant for degassing as hydrogen polysulfides above 159°C will not readily break down into elemental sulfur and H2S. At these higher temperatures, the sulfur ring scissions cause diradical chain formation to occur readily. Removing H2S from the sulfur under these conditions will result in immediate joining (polymerisation) of the newly formed diradical sulfur chains into highly viscous sulfur. Any rings that remain ‘open’ will also eagerly re-join with an H2S molecule. Only at temperatures below 159°C will the sulfur resume its stable ring structure, eliminating the reactive diradicals. Thus, degassing is best accomplished at temperatures below 159°C.

SIM

OF N O U L AT I

You need the right answers! We can help provide them.

Optimized Gas Treating, Inc., Buda, TX 78610 +1 512. 312 .9424, www.ogtrt.com

Figure 3. ICOn unit installed in rundown line.

Figure 4. ICOn unit installed with rundown header. The sparge vapour moves up through the catalyst in the vertical direction. The bottom of the catalyst zone uses a sparge plate to create a uniform vapour flux through the catalyst. The sparge vapour rate and the sparge plate must be configured such that the upward flow of vapour through the orifices prevents sulfur from leaking into the vapour pocket below the plate. Any sulfur that entered the vapour pocket would not be fully degassed and may interfere with the vapour distribution. Vapour leaves the contactor through an outlet nozzle at the top of the vessel above the catalyst.

Example configurations A myriad of different configurations are possible. The following process flow diagrams illustrate some of the more unique configurations available. Many other configurations beyond these are possible. Figure 3 shows a configuration in which the degassing contactor is installed in-line with the sulfur run-down lines. No pump is required, and the sulfur is cooled and degassed before draining into the pit. With this arrangement, the cooler and contactor must be installed near the condensers, and the available elevation drop must accommodate the equipment. October 2021 28 HYDROCARBON ENGINEERING

This configuration is most suited for grassroots units where the required hydraulics can be designed into the system. Note that the sulfur from the fourth condenser is not being degassed. This stream typically represents approximately 5% of the total sulfur flow and contains around 20 ppmw H2S. The contribution of this stream is small enough that it can be skipped. The H2S concentration of the aggregate sulfur in the pit will still be well below the 10 ppmw threshold. On the vapour side, process gas is being pulled from the Claus tail gas line and is returned to the same. An ejector is used to create the necessary vapour movement through the contactor. The ejector is placed after the contactor to lower the contactor operating pressure; this ensures that sulfur will flow from the condensers to the contactor at all times. A demister is used to capture any sulfur droplets prior to the ejector. Figure 4 is another pre-pit configuration, but this one uses a collection header and a pump. The collection header is sealed and shares its vapour space with condenser number 4; it therefore requires no vapour sweep. Sulfur is pumped from the header, through a cooler, and into the contactor. The pump flow rate is controlled based on the level in the header. Using a pump allows the contactor to be operated at nearly any pressure. In this example, the contactor is operated at a pressure higher than the Claus thermal reactor pressure. The contactor is sparged with air, and the sparge vapour is sent to the Claus thermal reactor – supplementing the existing air supply to the thermal reactor. The vapour rate required for degassing is considerably smaller than that required for the Claus unit; it is expected that the addition of ICOn will not affect the size or control scheme of the combustion air blower. When using a pump for the sulfur supply, it is possible to combine the sulfur streams from multiple Claus units into a single degassing contactor. In this configuration, connecting the sparge gas to multiple return points would keep the degassing contactor on-line even if one of the Claus units is down.

Conclusion Through the development of ICOn, there is now a sulfur degassing technology that can be integrated seamlessly into a sulfur recovery unit without requiring a recycle of vent gas to the thermal reactor. Ametek CSI and Worley Comprimo have joined forces to bring this new flexible technology to the industry to reduce the risks associated with the production of sulfur from both an operational perspective as well as sulfur handling and transport. In part two of this article, the design and three years of operational results of the first commercial installation will be discussed.

Reference 1.

KELLER, A., GROVES, B., FORBES. B., and WILLINGHAM, T., ‘New Pre-Pit Sulfur Degassing Technology’, presented at Laurance Reid Gas Conditioning Conference, Norman, Oklahoma, US, (2016).

Michael Gaura, AMETEK Process Instruments, USA, highlights the key measurement points in a sulfur recovery unit tail gas treatment unit.

ulfur recovery units (SRUs) have been utilised to remove elemental sulfur from hydrogen sulfide (H2S) rich acid gas streams in hydrocarbon processing, refining, and steel production plants for over 70 years. Releasing emissions containing 50, 60, or 70+% H2S has never been an option, as this would be lethal to humans. Burning these gas streams will reduce the H2S levels, but sulfur dioxide (SO2) emissions would be a result, meaning this is not an acceptable option either. Modern SRUs have been designed to use the Claus process and have historically achieved 90+% recovery of elemental sulfur, significantly reducing the amount of H2S and SO2 released

to the environment. However, <99% recovery is no longer acceptable to many global regulatory agencies, so various plant designs and additions have been developed to achieve sulfur recovery rates of 99.8% or even 99.9%. This article focuses on the continuous, analytical gas measurements that can be performed in tail gas treatment units (TGTUs) or tail gas cleaning units (TGCU) to ensure reliable, efficient, and safe operations in the effort to reduce sulfur emissions from the SRU. Figure 1 provides an inlet to outlet overview of an SRU that contains a TGTU. A TGTU is designed to receive the gas output from the Claus unit, referred to as the tail gas, and further capture, and therefore reduce, sulfur compounds that will be released from HYDROCARBON 29

ENGINEERING

October 2021

Figure 1. Typical SRU, with TGTU. Claus effluent – SO2, carbonyl sulfide (COS), carbon disulfide (CS2), elemental sulfur (SX) – to H2S.

Figure 2. Typical TGTU layout (regenerator not shown).

Figure 3. Recommended analyser sample points. the plant. Compared to the feed gas at the start of the Claus unit, a typical tail gas stream will contain much lower levels (<5%) of sulfur compounds – such as H2S and SO2 – at the TGTU inlet, and these will ultimately be lowered to ppm levels at the emissions stack outlet. The sulfur compounds are converted to H2S and removed through a series of processes that incorporate thermal and chemical elements as shown in Figure 2. TGTUs almost always include an inline heater or burner, a reduction reactor including a catalyst (such as cobalt molybdenum [CoMo] or other proprietary active material), a quench tower, an absorber, and a regenerator.

TGTU operational overview As shown in Figure 2, the Claus unit tail gas is mixed with fuel gas in the heater/burner to raise its temperature as required for proper reaction in the reduction reactor. As hydrogen will be required to convert the sulfur components to H2S, some TGTU burners are designed to generate hydrogen in the burner, while other designs include introduction of hydrogen prior to the reaction vessel, or rely on excess H2 being present in the Claus tail gas. The combination of heat and a catalyst present in the reduction reactor converts any sulfur components in the October 2021 30 HYDROCARBON ENGINEERING

SO2 + 3H2 --> H2S + 2H2O S + H2 --> H2S H2O + CO --> H2 + CO2 COS + H2O --> CO2 + H2S CS2 + 2H2O --> CO2 + 2H2S

(1) (2) (3) (4) (5)

The quench tower cools the gas stream, necessary for proper absorber operation. An amine solution is frequently present in the absorber. This amine selectively captures H2S present in the tail gas treatment stream. Eventually, the absorber becomes saturated with H2S and the saturated media is then referred to as a ‘rich’ or ‘sour’ amine. When the absorbing media can no longer capture H2S as expected, it must be regenerated or ‘cleaned’. The regenerator uses temperature, pressure, and a residency time to clean the amine. H2S that has been absorbed by the amine is released and routed back to the front end of the Claus unit for reprocessing. The regenerated amine is returned to the absorber, where it is again utilised in the capture of H2S. The TGTU effluent gas will ideally contain very low levels (<300 ppm) of H2S after treatment and will be sent to a thermal oxidiser to convert the H2S to SO2, prior to being released to the environment. Measuring key gas stream components, at various points in the TGTU process, has been performed for decades. Based on trial and error, the most valuable, reliable, and accepted analytical measurement points in a TGTU are located at two locations – one before and one after the absorber, referred to as AT1 and AT2 in Figure 3 At the first measurement point, the gas stream to be sampled has progressed through the reduction reactor and quench tower. Recall that all of the unreduced sulfur compounds present at the inlet of the TGTU should be converted to H2S after hydrolysis and hydrogenation in the reduction reactor, so measurement of H2S after the quench tower is considered representative of the total amount of sulfur that has entered the TGTU. A process gas analyser is now commonly installed between the outlet of the quench tower and the inlet of the absorber tower to measure what is anticipated to be no more than 5% H2S. Process engineers can utilise this measurement value to account for differences

Hydrocarbon Engineering Online

The home for the latest hydrocarbon processing news, analysis, comment and events Visit our website today: www.hydrocarbonengineering.com

PROCESS TECHNOLOGY MADE IN GERMANY

ϥ MEMBRANE

TECHNOLOGY

FOR EMISSION CONTROL AND PRODUCT RECOVERY

ϥ COMPRESSORS FOR PROCESS GASES

ϥ BALL VALVES

between the amount of elemental sulfur that was captured in the Claus unit, with the measured sulfur components that were delivered to the SRU inlet. Effectively, this is a recovery efficiency measurement. Another measurement is also being made in the analysers installed at this location. The H2S is typically measured with an ultraviolet (UV)-based optical bench, but a thermoconductivity detector (TCD) can also be added to some analysers to measure residual hydrogen (H2). Experts in the design and operation of TGTUs rely on excess hydrogen to ensure the unreduced sulfur compounds are converted to H2S in the reduction reactor. H2 should not be removed in the quench tower, so consistent measurements of percent level H2 after the quench column is a good indicator that enough H2 was present in the reduction reactor to convert the SO2, COS, CS2 and SX to H2S. The second location of common gas analysis sampling and analysis in the TGTU is after the absorber. H2S is again the primary measurement, and the concentrations are expected to be much lower than they were prior to the absorber. This makes sense, as the primary purpose of the absorber is to strip out the H2S and send it back to the Claus inlet for further processing and sulfur removal. Concentrations of H2S at this measurement point are expected to be <500 ppm. If concentrations begin to trend up, and H2S concentrations prior to the absorber inlet are fairly constant, a reasonable assumption is that the amine (or absorber) has become saturated and should be regenerated. Note that the amine concentration may also have been ‘poisoned’ with SO2 when the Claus unit experiences operational upsets. The air demand analyser located at the outlet of the Claus unit will report elevated SO2 levels entering the TGTU, but operators or systems may not react quick enough to prevent a negative impact on the scrubber media. If the H2S levels at the absorber outlet remain elevated after regeneration of the absorber, there is a good chance the scrubber media needs to be disposed of and replaced. This is an expensive decision, and is never made without detailed analysis and review. With an increased focus on reducing the carbon footprint associated with energy and other materials production, frequent regeneration of the absorber material is undesirable. This process requires considerable energy, and many users are evaluating the best way to balance their carbon consumption and their sulfur emissions. As such, H2S levels of ‘0’ at the absorber outlet will not be applicable for most users, as accomplishing 100% sulfur recovery will have other negative environmental consequences. As it is known that thermal oxidiser emissions stacks require SO2 measurement, knowing the H2S concentration at the absorber outlet can also assist with identifying issues with thermal oxidiser operation. If H2S levels are increasing after the absorber, but SO2 concentrations in the emissions measurement are not, the thermal oxidiser operation should be evaluated. Many users are also including H2 analysis at this absorber outlet, as it is a relatively inexpensive addition to the H2S analyser and H2 concentrations are not impacted by the absorber media. Therefore, redundancy of the critical H2 measurement is added to the system, with minimum additional engineering or installation requirements. COS should also be measured at the absorber outlet for two reasons: COS levels should not exceed a few hundred ppm at this measurement point, unless there are issues with their reduction in the reduction reactor. If COS levels are rising, there could be an issue with the catalyst in the reduction reactor ageing. A secondary reason to measure COS at this location is to account for differences in readings at the SO2 stack emission analyser when compared to the H2S reading at the absorber outlet. If both H2S and COS are measured prior to sample gas movement to the thermal oxidiser, they should be approximately equal to the concentration of SO2 being emitted at the flare.

Conclusion

www.borsig.de BORSIG GmbH Phone: + 49 (0)30 4301-01 Fax: + 49 (0)30 4301-2236 ( PDLO LQIR‫ך‬ERUVLJ GH (JHOOVVWUDVVH %HUOLQ October 2021 *HUPDQ\

32 HYDROCARBON

ENGINEERING

Not every sulfur plant will have a TGTU, but when a TGTU is present, it can be operated in a more efficient manner and with less environmental impact, with proper installation and utilisation of a few continuous process gas analysers. Experience has shown that the gas analysers can pay for themselves by reducing environmental incidents (that result in regulatory fines), increasing plant process uptime and assisting in the prevention of damage to expensive catalysts and scrubbing agents. Continued improvements from analytical solution providers have reduced the amount of scheduled and unplanned maintenance required on the systems, further lowering the total cost of ownership of the analytical measurement points.

Dr Jonas Westberg, Dr Viacheslav Avetisov and Dr Peter Geiser, NEO Monitors, Norway, outline how an evolution in TDLAS signal processing is enabling new applications for the next generation of smart combustion analysers.

ombustion is one of the most important industrial processes providing heat and power for operation. Gas analysers based on tunable diode laser absorption spectroscopy (TDLAS) are ideal for highly selective and sensitive real-time and in-situ combustion monitoring, resulting in optimised fuel consumption and thus lower costs while minimising pollutant generation. TDLAS is now considered a proven technology and is used in many industrial plants around the globe. Nonetheless, the technology has advanced significantly over the years and now enables smart combustion analysis1, an evolution from traditional ‘one-laser-one-gas’ analysers to more sophisticated ‘one-laser-multi-gas’ analysers. The new signal processing principles and analysis capabilities described have subsequently created a desire in the industry to use the technology not only for combustion optimisation, but also for combustion safety. This desire presents entirely new challenges for the TDLAS technology, as shown in Figure 1. On the one hand, real-time combustion control requires carbon monoxide (CO) concentration measurements in the ppm range with high sensitivity to provide low-noise feedback to fuel and air control for optimal combustion with the lowest pollutant emissions. On the other hand, accurate measurements of much higher CO concentrations up to percentages are needed for safety purposes, while sensitivity is no longer as important. Combined, a multi-purpose gas analyser designed for in-situ combustion control and safety should be capable of performing gas measurements with both high precision and high dynamic range. This article will discuss how this new challenge is leading to a further evolution in TDLAS signal processing, enabling new applications for the next generation of smart combustion analysers.

Tunable diode laser absorption spectroscopy TDLAS analysers excel at trace gas analysis and routinely demonstrate lower detection limits (LDLs) of ppm or even below, but while they provide excellent precision for combustion control applications, they often lack the dynamic range needed for simultaneous hazardous gas level monitoring. In addition, the high selectivity associated with laser systems often comes at the cost of susceptibility HYDROCARBON 33

ENGINEERING

October 2021

to interfering gas components, which is usually a major concern when choosing a suitable wavelength region for the instrument. In practice, a strategy of minimising spectroscopic interferences rather than avoiding them is often employed. A third issue is that of accurately assessing the instrument baseline, which is an often overlooked topic in TDLAS. Looking more closely at the basic principles of TDLAS can help to understand this problem. In TDLAS, the wavelength of a laser is scanned across a narrow spectral region where the gas of interest absorbs light, see Figure 2a. As the laser light propagates through the gas, a fraction will be absorbed, which causes a dip in the transmission that can be quantified by collecting the laser light on a photodetector and monitoring its response. As the name implies, laser absorption spectroscopy measures the characteristic absorption profile (or transmission profile) of a gas from which the concentration can be calculated given knowledge of the measurement conditions (optical path length, temperature, pressure, etc.). Since absorption (A), or transmission (T=1-A), are relative entities, a natural question becomes: what is the reference level? Where is the baseline (the transmission profile without any gas absorption)? This is crucial for accurate concentration measurements, since a 2% error in measuring an absorption of 5% requires an accuracy on the baseline of 0.1%. Figure 2 highlights this problem. It is impossible to know how much light is absorbed without an accurate estimation of the baseline. Most TDLAS instruments are therefore designed to not only

probe the absorption line, but also a wavelength region that is not affected by the gas absorption. This allows the instrument to distinguish absorption associated with the molecules in the gas, from transmission losses due to particles entering the beam path or mechanical misalignment. From the example in Figure 2a it is easy to define such regions using the ‘wings’ of the dip. Traditionally, the baseline is estimated by fitting a low-order polynomial to these regions (other methods can also be employed). But what happens if there are no regions free from gas absorption? This is obviously a problem since it prevents an accurate estimation of the baseline, which leaves the absorption undetermined. Such a scenario is shown in Figure 2b. How could the baseline be estimated in this case? When looking for a suitable wavelength region for a combustion application, this situation is frequently encountered and sometimes impossible to avoid. Partly because of this, traditional TDLAS instruments often target a single gas using a carefully selected wavelength region where an isolated absorption line can be found, like in Figure 2a. It is clear that a different approach to TDLAS is required to tackle high levels of absorption and complex spectra, such as shown in Figure 2b. In summary, a modern multi-gas TDLAS analyser for combustion control and safety should be able to meet the following challenges: Large dynamic ranges (low-to-high concentrations). Congested spectra (many closely spaced lines). Gas interferences.

Smart baseline-insensitive TDLAS

Figure 1. Combustion diagram.

Figure 2. (a) Traditional TDLAS spectrum with a single absorption line.

The regions on the sides of the absorption line can be used to estimate the baseline. (b) Congested TDLAS spectrum with strong absorption. The baseline cannot be estimated.

October 2021 34 HYDROCARBON ENGINEERING

To fulfil these requirements and to address current limitations in the TDLAS technology, efforts have been focused here on the development of a new baseline-insensitive approach, specifically designed to face the future challenges of real-time combustion control and safety. This is made possible by merging modern digital signal processing principles with well-established laser absorption spectroscopy technology, resulting in analysers capable of background signal suppression reminiscent of that encountered in modulated systems – all without sacrificing the linearity and dynamic range associated with modulation-free systems. The following example can be considered in order to better understand why this is a significant leap forward: A TDLAS analyser is tasked with monitoring CO and methane (CH4) in a large industrial furnace. As previously mentioned, low concentration measurements are used to optimise the combustion process and high concentrations measurements are used for safety requirements (Figure 1). Ideally, the analyser should be able to measure both CO and CH4 simultaneously, but laser systems quickly run into difficulties in such applications since it is very challenging to find suitable isolated absorption lines for these

two gases, especially at high temperatures. Often, a situation similar to that in Figure 2b will be observed, i.e. access to the baseline cannot be assumed. In addition, the concentration ranges required to control combustion and detect potentially unsafe operation can be very large, e.g. up to 1% for CO and up to 5% for CH4. For 1 m of optical pathlength, the corresponding absorption is not very strong. However, for a 20 m heater, the effective path-integrated concentration becomes 20%·m for CO and to 100%·m for CH4. This requires an analyser with exceptional dynamic range, capable of measuring absorptions from the very low up to the point where almost all laser light is absorbed by the gas. The upper panel in Figure 3a shows an example of extreme absorption levels of CO that can be present in this type of application. Note the almost complete attenuation of light at the peak of the line compared to Figure 2b. Isolated absorption lines like these are ideally suited for TDLAS analysers and many systems can achieve impressive dynamic ranges if the baseline is accessible. NEO Monitors’ instrument is no exception, Figure 3b shows a linearity validation check measuring CO. With a single point calibration, the analyser is linear up to ~99.9% of relative CO absorption – a dynamic range of more than 3 × 105. A more difficult case is shown in Figure 4. Here, the CH4 concentration is gradually increased, but unlike the isolated absorption line of CO in Figure 3, CH4 absorbs over the entire laser scan. It is clear that an instrument capable of handling spectra similar to these cannot rely on traditional baseline fitting since the strong gas absorption blocks all access to the baseline. Enabling measurements even under such conditions

Figure 3. Demonstration of the dynamic range capabilities. CO concentrations are varied from low ppm·m up to 17.5%·m with excellent linearity.

THE DAWN OF A NEW ERA FOR COMBUSTION ANALYSIS

T-Flange

LaserGas™ III Ultra

Say hello to our latest combustion analysis solution: Combine our LaserGas™ III O2 and the new LaserGas™ III CO Ultra analyzers with the innovative T-Flange - Process control and safety with low CAPEX and OPEX - Independent measurements with high precision and high dynamic range for large stacks -www.neomonitors.com Hot-swap analyzers mounted on the T-Flange

Performance You Can Trust www.neomonitors.com

Figure 4. Demonstration of the advantages of the baseline-insensitive approach by measurements of CH4. Despite an unknown baseline, the instrument is linear up to 25%·m.

was the main motivation for developing a new baseline-insensitive approach for TDLAS. The principle is based on a novel and proprietary combination of filters for digital signal processing. In this application example, it enables concentration measurements of CH4 up to 25%·m with excellent linearity and precision without access to the baseline and at absorption levels where most TDLAS systems would struggle (see Figure 4b). Finally, Figure 5 shows a measurement of a gas mixture containing 21%·m CH4 and 8.5%·m CO. This type of measurement presents yet another difficulty, namely gas interferences. The progress in processing power and advanced statistical algorithms over the last decades has opened up new avenues for on-instrument, real-time gas interference suppression. NEO Monitors implement these modern tools in the proprietary framework called IROSS (in-situ real-time overlapping spectra separation).1 Recent improvements to this framework allow an accurate and robust decomposition of a measured spectrum into its individual components. Proven-in-use TDLAS signal processing of the individual spectra can then be performed to extract the concentration information. An added benefit of the IROSS framework is its capability to predict process conditions from the measured spectra, which gives a secondary pathway to parameters such as gas temperature, pressure, and concentration. Benefits of the new baseline-insensitive approach compared to traditional TDLAS are outlined below:

Traditional TDLAS Normalisation required. Baseline fit. Limited to spectral regions and absorption levels where the baseline can be found. ~104 dynamic range.

Baseline-insensitive TDLAS No normalisation needed. Baseline-insensitive. Linear response up to 100% absorption, even if the baseline is not accessible. Up to 106 dynamic range. Gas interference handling by real-time source separation.

Conclusion As the TDLAS technology evolves from traditional ‘one-laser-one-gas’ analysers to more sophisticated ‘one-laser-multi-gas’ analysers, advanced signal processing plays a critical role. The new baseline-insensitive signal processing approach eliminates the need for traditional baseline matching and enables applications with high absorption levels over the entire tuning range of the laser. In addition, source separation algorithms have been further improved to increase the ability to separate signals from multiple gas components on the instrument in real time. Taken together, these two developments enable modern TDLAS analysers to cover applications with multiple gas components and high dynamic range requirements for combined combustion control and combustion safety.

Figure 5. Demonstration of the handling of gas interferences. CH4 is kept at a constant level of 21%·m while varying the CO concentration from 0 to 8.5%·m. October 2021 36 HYDROCARBON ENGINEERING

Reference 1.

GEISER, P., AVETISOV, V., WANG, J., and SIEKER, L., ‘Smart combustion analysis,’ Hydrocarbon Engineering, (December 2019), pp. 54 - 58.

Dr Stephen Firth, Servomex, UK, examines the critical measurements for safety and control in PTA production processes.

urified terephthalic acid (PTA) – sometimes referred to as polymer-grade terephthalic acid – is a white crystalline substance that plays an important role as a precursor for plastic production. Its high melting point of 427˚C (801˚F) makes it particularly valuable in ensuring the thermal stability of the plastics created from it. It is most commonly used as the key component in the manufacture of polyethylene terephthalate (PET), a recyclable thermoplastic resin with US Food and Drug Administration (FDA) approval for use as food and drink containers and bottles. PET is also the most widely used of the polyester type of man-made fibres. Approximately 80 – 100 million tpy of PTA is produced, with the vast majority manufactured in Asia – eight of the ten largest PTA plants in the world are currently located in the region. There is a continuing and growing demand for PTA throughout the world, particularly in the fast-expanding Asian economies, which also make up the largest demand for PTA; China alone uses 60% of the PTA produced globally.

PTA production processes There are a number of licensed methods for the production of PTA. These all follow roughly the same basic process. PTA is manufactured from p-xylene by careful and specific air oxidation in a reactor at high pressure and elevated temperature. Liquid acetic acid, which is highly flammable, is used as a solvent for this reaction. The crystalline PTA product is separated from the reaction liquor in separate crystalliser vessels and then recovered and purified. Gas analysis plays a vital role in the PTA production process, delivering the measurements that support product quality, process efficiency, and safety. The two critical measurements within the process are made at the oxidation reactors and the crystallisers.

Measurements at the oxidiser Gas analysis at the oxidiser stage is focused on oxygen (O2), carbon monoxide (CO), and carbon dioxide (CO2). These measurements are primarily for safety, but they also focus on process control. HYDROCARBON 37

ENGINEERING

October 2021

Figure 1. Purified terephthalic acid (PTA) application. Air is passed into the oxidation reactors, oxidising the p-xylene to terephthalic acid and generating CO2 and carbon monoxide CO. Some O2 remains unreacted, so the most critical gas analysis measurement is to monitor this residual O2 level in the off-gas (Figure 1, point 1), which should be around 4 – 5% O2. If the level rises too high, it means a dangerous situation could be developing in the reactor; sudden, runaway oxidation of the flammable materials could occur, resulting in an explosion. However, if the O2 level is too low, then insufficient oxidation occurs, leading to poor efficiency and a low product yield. To achieve the optimum results, the O2 level must be monitored with the best possible accuracy and the fastest response time – below 60 seconds, at least. Additionally, many plants require a measurement of the CO2 level – and sometimes the CO level – in the off-gas (Figure 1, point 2), as this reveals more information about the progress of the oxidation reaction. An infrared gas analyser can be used for this measurement, ideally configured to deliver simultaneous measurements for both CO2 and CO. Because of the importance of the O2 measurement, the system must operate with minimal errors. Therefore, to ensure reactor safety, a voting system is used to monitor the O2 concentration (Figure 1, point 3 [multiple measurements]). Voting systems use multiple analysers, and the process relies on the measurement agreed upon by the majority of analysers. Nearly all PTA plants use a three-analyser voting system, where if one analyser detects a significant change, it is outvoted by the other two and no action is taken. However, if two of the three analysers (or all of them) detect a change, this reading is held as correct, and action may be taken, ranging from informing the operator to automatically halting the process. Voting systems provide an extra layer of reliability in safety applications, and also allow any problem with an analyser to October 2021 38 HYDROCARBON ENGINEERING

be detected at an early stage without endangering the process. If two analysers agree on a measurement and the third differs, it indicates a potential problem that can be investigated and corrected before the process is affected.

Measurements at the crystalliser Off-gas applications in the crystalliser are broadly similar, and use identical analytical solutions for the measurement of O2 and CO2. The crystalliser measurements are all about safety. At this stage, acetic acid vapour is driven off as the PTA product crystallises out of the solvent liquor. Acetic acid is not just corrosive but highly flammable, so a measurement of the residual O2 in the crystalliser vapour is vital to provide a warning of any explosion risk. Monitoring the presence of CO2 in this vapour (Figure 1, point 4) provides an indication of any post-oxidation that may be occurring, and so is a useful measurement. Dewpoint control is also very important to stop any sample condensing.

Gas analyser technologies The off-gas from the process is generally at a pressure of >10 barg and a temperatures of 50˚C, so the area classification required by gas analysis equipment may be ATEX Category 2 or 3 (Zone 1 or 2), depending on the plant conditions. Gas analysis measurements are normally specified on a dry basis with required ranges between 0 – 10% for O2, 0 – 5% CO2, and 0 – 1% CO. Speed of response is critical for gas analysis in this application, particularly for the O2 measurement. Typically, 30 – 45 sec. for overall T90 of the complete system is required. A paramagnetic analyser is recommended for the O2 measurements, as this sensing technology is highly specific to oxygen, and so delivers high levels of accuracy in the reaction conditions. It also offers a fast response to changing O2 concentrations in the reactor.

Need a reprint?

We can tailor to your requirements, produce 1 - 12 page formats, print colour or mono and more +44 (0)1252 718999 reprints@hydrocarbonengineering.com

Paramagnetic cells each consist of two nitrogen-filled glass spheres, mounted on a rotating suspension within a magnetic field. Light shines on a centrally-located mirror, and is reflected onto a pair of photocells. Because O2 is naturally paramagnetic, it is drawn into the magnetic field, and so displaces the glass spheres, causing the suspension to rotate. This motion is detected by the photocells, which generate a signal to a feedback system. This, in turn, sends a current through a wire mounted on the suspension, creating a motor effect. The current produced is directly proportional to the concentration of O2 within the gas mixture, allowing an accurate and linear percentage reading to be made. As this technology is non-depleting, paramagnetic cells never need changing, and the performance does not deteriorate over time, with significant benefits to ongoing maintenance costs and sensor lifespan. For the other measurement needs, a single infrared analyser is usually sufficient. Infrared sensing technology depends upon the property of some gases to absorb unique light wavelengths. By focusing an infrared light source through a sample cell holding a continuously flowing sample and onto a detector, after wavelength selection, it is possible to determine the concentration of a selected gas in a mixture. Infrared sensing delivers a real-time measurement response and, like paramagnetic technology, is non-depleting, so has very low maintenance requirements. Gas filter correlation (Gfx) sensing is an enhanced version of infrared sensing that offers a higher degree of accuracy where low-level measurements are needed or background gases may interfere with the measurement. It is often used to measure ppm CO in the reactor off-gas. A well-designed sample conditioning system is also required to ensure the analyser is able to cope with the high-pressure, high-temperature off-gas, which will contain trace p-xylene and significant levels of corrosive acetic acid vapour. For both reactor and crystalliser measurements, the sample conditioning system must be well designed to ensure a fast overall response time, and to handle the significant levels of condensibles in the sample, removing them prior to the measurement. Sampling in the reactor off-gas stream also needs to correctly handle the high pressures involved. A water-washing ‘lute’ system may be used, and highly corrosion-resistant materials (e.g. titanium, Hastelloy) may be specified for construction, because of the presence of acetic acid and possible catalyst traces.

The analyser complies with Safety Integrity Level (SIL) 2 and is fully certified for ATEX Category 2 or 3 (Zone 1 or 2) hazardous areas, including use with flammable samples. It operates across a wide ambient temperature range, with a good speed of response, typically delivering a T90 response time of 4 sec. Up to six O2 transmitters can be linked to a single control unit, making the technology a good choice for voting systems. For single-component CO2 analysis, the SERVOTOUGH SpectraExact 2500 analyser series is an option, while a variant, the SpectraExact 2550, is used for dual-component CO2 and CO measurements. Using infrared sensing technology, the SpectraExact offers a stable, sensitive performance with minimal cross-interference. The robust measurement cell is separated from the electronics, so the sample passes through it without damaging delicate instrumentation. It has full certification for ATEX Category 2 or 3 (Zone 1 or 2) hazardous areas regardless of the sample type, so the certification is not affected by the flammable gases used in PTA manufacture. It operates over a wide temperature range, and has a rugged construction. The SpectraExact 2500 can be used to measure water in liquid acetic acid as a quality control measurement, to assist with controlling the recovery and purification of the acetic acid solvent before recycling.

Other measurements While the reactor and crystalliser stages are the key measurement points in the PTA process for control, efficiency, and safety, there are other points throughout the process where gas analysis is highly advantageous. Ancillary plant operations, such as auxiliary boilers, may require measurements of O2 and combustibles in the flue gas, typically provided by a specialist combustion analyser, such as Servomex’s SERVOTOUGH FluegasExact 2700. Emissions monitoring may also be needed to demonstrate that the plant complies with environmental regulations. A variety of sensing technologies exist to support this application. Servomex’s SERVOPRO 4900 multigas provides continuous multi-component measurements of criterion pollutants, greenhouse gases, and reference O2, using configurable digital sensors to meet plant requirements. In addition to providing analytical instrumentation, Servomex Systems can deliver an entire, comprehensive gas analysis solution for PTA plants, with project management from design to installation and commissioning.

Solutions for PTA

Conclusion

Servomex has provided gas analysis solutions for PTA production since the infancy of the process in 1987. Dozens of PTA plants around the world use the company’s technology for process control, safety, efficiency, and emissions monitoring. For O2 measurements, Servomex supplies the SERVOTOUGH OxyExact 2200, a high-performance paramagnetic analyser with an optional solvent-resistant cell, delivering an accurate, reliable measurement that is unaffected by flammable and corrosive solvent traces.

The volatile nature of key reactions in the PTA process emphasises the need for a ‘safety-first’ mindset in the design and operation of the gas analysis system. By carefully understanding the risks in the process, and selecting both the appropriate measurements and best-suited sensor technology to apply to each point of analysis, a well-integrated gas analysis system will not only protect process and plant, but yield measurable efficiency and emissions monitoring benefits.

October 2021 40 HYDROCARBON ENGINEERING

Brought to you by

Co-Hosted by

129(0%(5 Hotel Melia Castilla, Madrid, Spain

(XURSH¬V /DUJHVW 0HHWLQJ 3ODFH IRU 'RZQVWUHDP /HDGHUV 5HWXUQV LQ 0DGULG

What Can You Expect: Expert Panels and Q&A Hear from over 100 leading industry experts in our high-level plenary, technical presentations and LQQRYDWLYH GLVFXVVLRQ IRUPDWV LQFOXGLQJ RXU «7DNLQJ WKH 3XOVH¬ VHVVLRQ *HW LQYROYHG LQ WKH GLVFXVVLRQV DQG ² QG RXW ZKDW \RXU SHHUV WKLQN ZLWK OLYH 4 $V DQG SROOLQJ

Speakers include: Josu Jon Imaz CEO Repsol

Antonio Joyanes 6HQLRU 9LFH 3UHVLGHQW 5H² QLQJ CEPSA

Specialised Technical Streams Identify cutting edge solutions to downstream’s most pertinent challenges including achieving FDUERQ QHXWUDOLW\ HQHUJ\ HI² FLHQF\ FOHDQ +\GURJHQ production and petrochemical integration

Philippe Boisseau

In-Focus Workshop Day

Berta Cabello

2XU H[SHUW OHG ZRUNVKRSV SURYLGH DQ LPPHUVLYH DQG intimate learning experience where you will focus on UHDO EXVLQHVV FKDOOHQJHV DQG EXLOG \RXU UH² QHU\ blueprint.

CEO CEPSA

+HDG RI 5H² QHU\ Transformation Repsol

Tomás Malango

Topics including:

Hydrogen Director Repsol

•

Delving into implications from the European Commission’s upcoming energy packages

Sølvi Storsæter Bjørgum

•

7DFNOLQJ WRGD\¬V (XURSHDQ UH² QLQJ LVVXHV DV ZH HPHUJH IURP WKH SDQGHPLF LQFOXGLQJ HQHUJ\ HI² FLHQF\ DQG RQVLWH RSWLPLVDWLRQ

(TXLQRU 5H² QLQJ 'HQPDUN

•

+RZ DUH \RX SUHSDULQJ \RXU UH² QHU\ IRU "

7R UHJLVWHU RU ² QG RXW PRUH YLVLW ZRUOGUH² QLQJDVVRFLDWLRQ FRP HUWF If you want to discuss how you can get involved in ERTC 2021, please contact:

Sandil Sanmugam, Portfolio Director, Europe, :RUOG 5H² QLQJ $VVRFLDWLRQ T: +44 207 384 7744 E: sandil.sanmugam@wraconferences.com

October 2021 42 HYDROCARBON ENGINEERING

Sami Tabaza, Atlas Copco Gas and Process, USA, explains why integrally geared centrifugal compression technology is increasingly being used in NGL plants alongside oil-flooded gas screw compressors.

as processing plants need to cope with varying and often uncertain conditions. Operators typically face challenges with the mode of operation (ethane rejection or ethane recovery) and feed gas uncertainty (rich gas or lean gas). In terms of equipment reliability, flexibility and efficiency, these circumstances also place significant requirements on the compressor technology used. The case study within this article will discuss the deployment of a mechanical refrigeration cycle using commercial-grade propane (95 – 98.5% propane, with the rest being heavy hydrocarbons, or HD5, or higher-grade propane). It will also examine the performance characteristics of compressor technology applied in gas processing plants, including oil-flooded screw, and integrally geared centrifugal.

Flexibility and performance Standardised gas processing plants up to 200 million standard ft3/d (~223 000 Nm3/hr) are typically built with a two-stage refrigeration cycle, using a flashing inter-stage economiser. Vaporising the commercial-grade propane refrigerant used in these plants in a two-stage process helps to improve

efficiency. The process is driven either by two compressors or, ideally, by a single compressor using two sections (i.e. two process stages). While the oil-flooded screw compressors are a widely used option in the gas processing industry, the oil-free integrally geared compressor (IGC) has been growing in popularity in recent years. Because of their efficiency, availability, reliability and a comparatively low CAPEX, integrally geared centrifugal compressors are increasingly being used in upstream, midstream and downstream markets. The technology’s compact size and ability to accommodate side streams also helps to enable flexible plant operations. Natural gas processing plants need the flexibility to handle both rich and lean feed gas without reducing performance. Plants must also be agile enough to operate in ethane recovery or rejection mode, and are often required to operate in extreme ambient (summer and winter) temperature conditions. Furthermore, on plants processing shale gas, such as several plants operating in North Dakota, US, with integrally geared technology, the compressor design must account for the variability of feed gas conditions. In other words, flexibility

HYDROCARBON 43

ENGINEERING

October 2021

is an essential requirement for enabling a refrigeration compressor to meet these different modes of operation. The ultimate target of a natural gas liquids (NGL) train is its ability to extract 95 – 99% of the heavy hydrocarbon (C2/3+) components of the feed, known as the Y grade. The North Dakota plants highlighted in this case study were designed for a nominal capacity of 20 million standard ft3/d (~223 000 Nm3/hr). For one of the plants, the objective was that it would start operations at 50 million standard ft3/d (55 800 Nm3/hr) and then increase the flow rate to 100 million standard ft3/d (111 600 Nm3/hr) and later even to 200 million standard ft3/d (223 000 Nm3/ hr) over the plant’s first year of operation. This rapid ramp-up in plant capacity had to be considered in the compressor design, as did the variation in suction pressure due to different modes of operation. The ability of integrally geared technology to flexibly adapt to such process conditions in the optimal manner made it a viable compressor solution for these specific plants.

system is not performed, or if unexpected performance issues arise, preventing the contamination of process gas (in this case, commercial grade propane refrigerant) downstream from the compressor becomes a challenge. If oil in the compression chamber is carried over with the process gas and flows downstream, contacting the refrigerant fluid, it can cause failures in the brazed aluminium heat exchanger or condensers. Additionally, if there is a bypass for the condenser, the leaked oil will also affect the accumulator. Because of these factors, screw compressors require more frequent maintenance than IGCs, both in terms of regularly changing the coalescing filter and oil filters, and for verifying that oil has not been carried into the process.

Suction pressure and flow control methods for extended compressor turndown

One of the key factors in any refrigeration application is the handling of process changes, which affects refrigeration duty. Variable diffuser guide vanes (vDGVs) provide greater Deployment of oil-flooded screw operational flexibility and help maintain stable compressor compressors in NGL plants operation over a wide range of conditions and at a constant When deployed in NGL plants, oil-flooded screw compressors discharge pressure. feature a design that enables them to function in a variety of In cases where air-cooled condensers are utilised, when a conditions, as outlined above. For example, these machines rise in ambient temperature occurs, refrigerant condensing can divide the total refrigeration duty in two (50% duty), three temperature also increases due to increasing pressure in the (33% duty) or four (25% duty) compressors. Depending on a cycle. To retain a consistent level of cooling, the system plant’s operating requirements and approach, this allows for requires higher flow at higher pressures. IGCs may incorporate anywhere from 10 to 100% turndown. Because oil-flooded two control types: vDGVs and inlet guide vanes (IGVs). vDGVs screw compressors are based on the principle of positive extend compressor turndown by up to 50% in both single and displacement, they are not unduly affected by fluctuations in multi-stage refrigeration applications. The combination of mole weight, suction pressure or temperature. Accordingly, aerodynamically engineered components such as impellers, these machines operate in the gas plant with a traditional vDGVs and volutes serves to boost efficiency in centrifugal efficiency of 65 – 70%. compressors. These aerodynamic components can be While oil-flooded screw compressors are generally matched by selecting the right gearbox and incorporating the considered easy to operate, they are subject to oil proper pinion speeds in the compressor design process.1,2 This can then be complemented with stable rotor dynamics to contamination of the propane, which reduces the reliability of achieve maximum efficiency. this solution, as reported by the designers and operators of When a change in a machine’s operational mode to the sites in study. If proper maintenance of the oil removal rejection mode is made, it will normally be with the lower head and smaller flow (assuming lower duty). This can be achieved as the refrigeration machine will adapt to that change by turning down the compressor’s vDGVs and adjusting the machine to the required discharge pressure. vDGVs help to optimise IGCs by enabling the machines to handle high settle-out conditions at start up. Figure 1. Integration of the integrally geared refrigeration compressor inside the gas By comparison, processing plant. screw compressors

October 2021 44 HYDROCARBON ENGINEERING

Global Publication Stay informed

Keep up to date with us to hear the latest World Fertilizer news

SPONSORS:

37th

Oil | Gas | Fertilizers | Metallurgy | Industrial

Sulphur + Sulphuric Acid 2021 Virtual Conference & Exhibition

1-4 November 2021

Stay connected and informed - from your plant, ofﬁce or home CRU’s Sulphur + Sulphuric Acid will run as a fully virtual event again in 2021 – connecting you with an even wider audience of sulphur and acid industry professionals from around the globe. Drawing on the success of the 2020 virtual event you can expect expanded content streams, enhanced networking opportunities and much more – all from the convenience of your plant, ofﬁce or home.

2020 Virtual Event Hightlights

+ 600 industry

professionals

countries represented

40 ++

250 +

exhibitors

operators OFFICIAL PUBLICATIONS:

30of hours presentations and Q&A sessions - live and on-demand

For more information or to book your place please visit: www.sulphurconference.com

bearings, which have less robustness, reliability, and lifespan characteristics. This design is complemented by dynamic dry gas seals backed by floating carbon ring seals, which results in oil-free operation in all conditions, even during emergency stops or if seal failure occurs. Atlas Copco Gas and Process is the first manufacturer in the world to employ dynamic dry gas seals in IGCs and expanders.

Efficiency characteristics As has been shown in numerous hydrocarbon installations, integrally geared technology can generate savings due to its overall efficiency characteristics, though this always depends on the specific application and case at hand.2,3 For propane refrigeration applications, for example, IGCs can attain Figure 2. Typical refrigeration compressor package polytropic efficiency in the range of 80 – 84% through using integrally geared technology. aerodynamic enhancements and stage optimisations. Depending on which process design is used on the NGL plant, this can generate total energy savings of between 10% and 12% compared to oil-flooded screw compressors. In one of the North Dakota plants, where the assumed polytropic efficiency was 80%, a three-stage IGC operating at full availability for 17 million Btu/hr (4 982 kW) chiller duty realised power savings of approximately 10% compared to a base oil-flooded screw compressor design. Because IGCs typically offer higher maintainability than oil-flooded screw compressors, they also generate maintenance cost Figure 3. vDGV typical performance map showing range of operation. savings. manage suction flow with sliding vanes, thereby using a different method to handle the same situations. While both processes are effective, vDGVs offer greater precision in terms of response time and amount of flow that can be controlled.

Compressor reliability in NGL plants The design of the IGC ensures that no oil can enter the compression chamber. As a result, IGCs provide 100% oil-free operation, increasing compressor reliability while also requiring less maintenance. These compressors provide continuous service with greater than 98% availability. To help safeguard reliability, the IGC units deployed in North Dakota feature a combination of tapered land thrust bearings with tilted pad radial bearings. This enables the machine to offer stable operation even during high settle out conditions and also when the thrust load is much higher than it would be during normal operating conditions. The thrust tapered land with tilting pad radial bearings employed in the North Dakota units help maintain stable operation even during high settle-out conditions, enabling high availability. Generally, the hydrodynamic bearings used in the North Dakota compressors offer advantages in terms of reliability and longer life compared to competitive technologies.3 Many screw compressors utilise anti-friction October 2021 46 HYDROCARBON ENGINEERING

Conclusion Given its ability to operate reliably and efficiently, integrally geared centrifugal compression technology is increasingly being used in NGL plants alongside oil-flooded gas screw compressors. From a CAPEX perspective, this technology provides an economically attractive solution, in addition to generating extensive OPEX savings over the course of the plant’s lifecycle. Furthermore, the compressor units installed by Atlas Copco Gas and Process at plants in North Dakota demonstrate how this technology can match the need for flexibility required by gas plant operations. Existing installations such as this show that IGCs can provide a reliable, efficient compression solution that offers the agility and flexibility demanded by the modern gas processing industry globally.

References 1. 2. 3. 4.

RINALDO, J., ‘Gearbox Specs – Getting Them Right’, Turbomachinery International Handbook, (2017). BEATY, P., EISELE, K., and MACEYKA, T. et al., ‘Integrally-geared API 617 process gas compressors,’ 29th Turbomachinery Symposium, Houston, Texas, US, (2000). PATEL, T. and STRUCK, H., ‘Ensuring integrally-geared compressor reliability with API 617’, 10° Fórum de Turbomáquinas, Rio de Janeiro, Brazil, (28 - 30 November 2017). PATEL, T. and RENKEN, H., ‘Achieving greater agility in LPG recovery with integral-Gear Compression Technology for Propane Refrigeration’, Abstract prepared for the World Gas Conference 2018, (2017).

The future of the energy industry depends on great ideas. At Atlas Copco Gas and Process, we transform ideas into solutions for tomorrow. We are committed to helping our customers achieve a decarbonized future by providing centrifugal compressors, expanders and pumps that can immediately assist in lowering your greenhouse gas emissions today, and we have the ingenuity and engineering prowess to design hydrocarbon-free machinery that will be needed in the future.

Learn more at www.atlascopco-gap.com

In the second of two installments, Ralph H. Weiland and Nathan A. Hatcher, Optimized Gas Treating Inc., explore case studies to show how simulation is used in troubleshooting.

HYDROC HYD ROCARB ROC A ON AR ARB 48 HYDROCARBON Octo Oc October tobe to berr 20 be 22021 021 48 ENGINEERING EN NGI G NE NEER E IN ER ING G

imulation tools are used extensively in a wide range of activities from plant design, unit revamps, and troubleshooting, to plant optimisation and unit monitoring. Confidence and trust in a simulator are usually gained by running simulations against measured plant-performance data. But what happens when measurement and simulation do not align? Is it possible to learn anything from such experiences? This is the second of a two-part series of case studies to illustrate examples of how simulation is used in troubleshooting (part one of this article was published in the May 2021 issue of Hydrocarbon Engineering).1 The two further cases discussed here were each initiated by disparity between simulation and plant data too large to be attributed to just modelling and measurement errors. The examples are cases attributed to poor temperature control and to a problem with the model itself. Each case compares simulation results with measured performance metrics, and each one uses a combination of data and simulation to logically deduce a diagnosis and resolve the issue.

Case study 1 A Middle Eastern ammonia plant was experiencing very erratic operation of its CO2 removal system with CO2 concentrations in the treated gas ranging from 10 to 10 000 ppmv in a seemingly random fashion. The unit was using piperazine-promoted MDEA solvent with a split flow design. Figure 1 shows a process flow diagram (PFD) of the bulk removal and polishing absorbers which were packed with RVT Hiflow® Rings #50 – 5 and #25 – 5, respectively. The unpredictable nature of the product gas CO2 content threw the rest of the process into chaos and demanded a rapid resolution.

Speculation about the cause of the instability included the possibility of foaming, tower flooding, and hysteresis in control elements such as flow control valves. However, pressure drop measurements gave no indication of any hydraulic anomalies, and solvent flowrates showed completely steady behaviour at least to within the measuring ability of instrumentation. Ultimately, recourse was had to simulation. The absorption system, sectioned off from the rest of the CO2 removal process, was set up in the ProTreat simulator using the actual tower diameters, bed depths and type, and size of packing contained in each of the two beds in each column. Temperatures, pressures and flowrates of all inlet streams were set at the nominal values at which the plant was known to be operating. It had been observed that the problem was less severe during winter months and on colder nights during the summer, and it was known that the fin-fan coolers on the semi-lean amine stream were operating at their limit for current production rates. In fact, the licensors of the process technology had originally recommended that the semi-lean amine temperature be kept below 70°C, whereas under hot ambient conditions the temperature of this stream was sometimes as high as 75 or even 80°C. It was implausible that temperature variations of a few degrees could cause CO2 in the treated gas to escalate to 10 000 ppmv; nevertheless, a simulation study of sensitivity to operating parameters was done. Figure 2 summarises the results. A temperature of 74°C produces synthesis gas below 1000 ppmv CO2 and one quarter of a degree cooler drops the CO2 level by a factor of 10. Semi-lean temperature only 2°C higher elevates the treated gas to almost 10 000 ppm. The simulations indicate that 74°C is a critical temperature whose HYD HYDROCARBON YD DROCAR RBON 49 RBO

EN ENGINEERING NGI G NEER RIN ING

Octo Oc Octo October tobe berr 20 22021 21

Figure 1. PFD of the ammonia plant’s CO2 absorption system.

Figure 2. Treating sensitivity to semi-lean temperature.

Figure 3. Response of polishing column to semi-lean temperature

October 2021 50 HYDROCARBON ENGINEERING

vicinity shows extreme sensitivity. The problem reveals itself in the polishing column but its cause is in the bulk column. The maximum possible net loading of solvent in the bulk column depends on temperature, and this goes down as temperature goes up. Any CO2 not removed in the bulk column spills over into the polisher and must be removed there; however, the polisher also has CO2 capacity limits. At 73°C, the approach to equilibrium loading at the rich end of the polisher is 80% and the lean solvent loading controls treating. But at 74˚C the rich solvent loading is at 98% of equilibrium, so now it is the rich solvent loading that controls treating. The bulk removal column is so close to capacity that only a 1°C change in semi-lean temperature throws the polisher from lean-end to rich-end control. Figure 3 illustrates the extreme sensitivity of the polishing column to any increase in demand on the bulk column’s performance. It should be noted that this is an optimally designed set of columns and each one in the pair is balanced against the other. Therefore, near the critical temperature of 74°C, even a slight increase in semi-lean temperature translates into a decrease in the maximum possible net loading. The bulk column simply does not have the capacity to function properly with a lower net loading, so the excess CO2 necessarily has to spill over into the polisher. However, it is also at its limit, meaning that there is no choice but for the excess to pass through and leave the polisher as an off-specification product gas. For completeness, the plant’s sensitivity to fully-lean and semi-lean solvent flowrates and CO2 loadings were also assessed, and in the vicinity of 74°C, small fluctuations in these variables also cause marked changes in treated gas CO2 content. Around 74°C, both columns are near their capacity for holding CO2 and the whole system is in a delicate balance. This is one of the risks of operating a process plant near its stability limits.

Case study 2 The final case relates to the simulation of packed absorbers in the CO2 removal units of two LNG plants. In one plant, the absorber contained a 250-size structured packing and in the other it contained 50 mm random packing. Both were performing more or less according to design with treated gas at just below 1 ppmv CO2 and a temperature bulge of just under 85°C in one case and at 8 ppmv CO2 and 75°C in the other. The purpose of the ProTreat® simulations was to validate the simulator against known performance data. The operators of both plants use thermal imaging and scanning extensively in troubleshooting and monitoring of the units. Trayed columns are naturally discretised by the actual trays they contain. However, because packed columns are continuous contacting devices, they must be discretised somewhat artificially for simulation because simulators are digital, not analogue, and must deal with continuous variables by discretising them. Figure 4 shows the effect of discretisation on the simulated temperature profiles for the two columns. Comparing with the measured data, discretisation of Columns A and B into 37 and 20 segments, respectively, gives a very satisfactory comparison between model and data. It is important to note that the data are column skin temperatures. Depending on ambient temperature and wind speed, these

may be several degrees cooler than might be measured inside the towers. These degrees of segmentation appear to represent the behaviour of these absorbers quite well. Segmentation (or the height of segments) is related quantitatively to: The total depth of packing in the column. The packing size. The ProTreat simulator automatically defaults to the correct Figure 4. Simulated vs measured temperatures in two LNG absorbers segment size based on heuristics (Column A – left, and Column B – right). developed from comparisons against a wide range of unit performance data from numerous columns containing a variety of both random mixing of vapour with vapour and liquid with liquid. Without and structured packings. But what does segmentation really axial mixing, the two phases would retain all of their heat, represent? simply carrying it up and down the column. As a result, the full A column described by a single segment is essentially heat of reaction would heat both phases to the maximum equivalent to a well-mixed vessel. One with a very large extent possible. number of segments has both phases in plug flow. Real packed A simulator that produces temperature profiles like the columns operate somewhere between these extremes. As blue lines in Figure 4 is using incorrect parameter values, which Figure 4 shows, ProTreat simulation with 200 segments invalidates the results because it disregards the fact that there produces a very broad flat bulge of high temperature which is significant axial mixing in real packed columns. Based on returns to known end conditions only very close to the top thinking along lines parallel to numerical integration for the and bottom of the columns. The heat being carried by the gas area under a curve, at one time OGT advised using fine and liquid in plug flow would have little or no opportunity to enough segmentation to obtain results that gradually became disperse across the column like it would with longitudinal stable against ever finer segmentation. This was bad advice

COMPLETE PROCESS LINES FOR THE SULPHUR INDUSTRY SBS STEEL BELT SYSTEMS is your reliable partner for continuous process plants and VROLGLÀ FDWLRQ KDQGOLQJ OLQHV HTXLSSHG ZLWK À UVW TXDOLW\ VWDLQOHVV VWHHO EHOWV SBS STEEL BELT SYSTEMS S.r.l T. +39 0331 864841 3LRQHHUV LQ WKH GHYHORSPHQW RI SDVWLOODWLQJ XQLWV info@steelbeltsystems.it ZLWK RXU 5ROOGURS DQG $FFXGURS SDVWLOODWRUV www.steelbeltsystems.com 6XSSOLHUV RI FRPSOHWH WXUQNH\ SODQWV IRU WKH 2LO *DV DQG 3HWURFKHPLFDO ,QGXVWULHV

SBS will be a bronze sponsor of the next Sulphur + Sulphuric Acid 2021 Virtual Conference and Exhibition ZZZ VXOSKXUFRQIHUHQFH FRP

based on faulty thinking. The development team went through a thorough review of all the elements in the rate model including detailed assessment of vessel heat-loss expectations, packed-bed heat transfer and mass-transfer correlations for packing, reaction kinetics, and vapour-liquid equilibrium. The analysis was conducted based on feedback from several customers, and it led to the conclusion that ever finer segmentation corresponded with perfect plug flow of both phases; whereas real packed columns exhibit significant axial dispersion that is masked and eventually negated by ever-finer segmentation. An interesting conclusion is that without axial mixing, LNG absorbers would have to be operated quite differently – it is mixing that keeps temperatures within acceptable limits. Axial mixing or dispersion is critical to the success of gas treating with large heat effects. Without axial mixing it would be very difficult to keep temperatures low enough to avoid serious solvent degradation and to allow the desired treating to be achieved with reasonable solvent flow rates. On the other side of the axial mixing coin, maldistribution of either phase will lead to greater axial mixing, poorer performance and cooler temperature curves. Measured temperatures that fall too much below simulation are indicative of maldistribution or packing that has been poorly installed or that has lost some of its integrity, perhaps leaving large voids in the bed. Axial dispersion can also take place in trayed columns. OGT has recently extended axial dispersion considerations to the separation performance of trays by

allowing for specifiable levels of weeping and entrainment from each tray in a column.

Summary Without simulation, a defect can hide and go uncorrected for a long time. What is worse, using a less-than-rigorous simulator with buried tuning parameters can lead to some fruitless and costly adjustments to operating conditions, hardware replacements, misdiagnosis of the true problem, or the resigned acceptance of reduced processing rates and lost revenue. Troubleshooting is not always as straightforward as in some of the examples here, but it can be made a lot easier when a reliable simulator is used to assess the validity of measured data. It is almost always important in assessing lean loadings to account for the HSS concentrations, simply because HSS levels can account for the lean loading value itself. In treating to low residual acid gas levels, HSSs can determine the treat actually achieved. If the simulation tool is reliable, it is almost always possible to pinpoint the cause(s) of poor performance because a reliable simulator tells what the plant as-specified should be doing. It is up to the engineer to use the simulator thoughtfully to understand why expectations are not being met.

Reference 1.

WEILAND, R.H. and HATCHER, N.A., ‘Simulation in a world of trouble: part one’, Hydrocarbon Engineering, (May 2021), pp. 44 - 46.

T +49 2961 7405-0 T +1 704 716 7022

Your Specialist for

PRESSURE RELIEF SOLUTIONS Consulting. Engineering. Products. Service.

Gallbergweg 21 | 59929 Brilon, Germany F +49 2961 50714 | hello@rembe.de | www.rembe.de

9567 Yarborough Road | Fort Mill, SC 29707, USA F +1 704 716 7025 | info@rembe.us | www.rembe.us

Bill Johnson and Chuck Baukal, John Zink Hamworthy Combustion, a Koch Engineered Solutions Company, USA, evaluate the impact of tramp air leakage on process heater operation, as well as corrective and preventative actions.

rocess heaters are typically designed to operate with a negative pressure (draft). Most process heaters are natural draft, which means there are no fans or blowers because the air for combustion is pulled in by the negative pressure inside the heater. The draft in a heater is directly related to the difference in pressure between the inside and outside of the heater. Under normal operating conditions, the pressure inside the heater is significantly lower than outside the heater due to the hot combustion products which are very low density and buoyant. As they rise in the firebox, they create a negative pressure or draft. The draft in a heater is a function of the firebox temperature, the height of the stack, and the location in the firebox. Since the firebox is less than atmospheric pressure, the combustion air is pulled into the heater through the burners. The draft is the highest (most negative) at the floor, which is typically where the burners are located. This maximises the amount of air that can be pulled into natural draft burners. Not only will air be pulled through open burner air registers, but it will also enter the firebox at any location that is not adequately sealed. This includes, for example, open air registers on burners that are out-of-service, tube penetrations, open sight ports, leaky explosion doors, and cracks in the heater shell. Most heaters have been in service for many years, which means

that it can often be very challenging to seal up unwanted air leaks. It is not unusual to find that levels of excess O2 anywhere from 1 to 3% or higher enter the heater through leaks in the convection section. Air that enters the heater at locations other than the operating burners is often called ‘tramp’ air. Too much tramp air leakage can be very detrimental to process heater operation and potentially dangerous.

Effects on operations Tramp air can impact the heater operation in many ways, each of which is briefly considered below, including how they impact the operation of the heater.

Reduced heater efficiency Air that leaks into a heater is cooler than the flue gas. Any air above that needed for combustion is absorbing energy, some of which is in the flue gas leaving the radiant firebox. The higher the flue gas temperature, the higher the sensible energy loss due to the tramp air leakage. This means a portion of the fired duty is being used to heat this tramp air instead of heating the process. This reduces the energy efficiency of the process. It also indirectly increases pollution emissions because the heater consumes more energy than needed when the thermal efficiency is reduced due to tramp air. HYDROCARBON 53

ENGINEERING

October 2021

Increased NOX levels

Tramp air leakage may increase NOX levels. Since air is cooler and heavier than the flue gas, some of the tramp air will actually fall to the floor of the heater. This can increase the local concentration of oxygen near the burner gas tips. This can result in higher amounts of NOX formation because NOX increases with excess O2 during typical operating conditions. If a very large amount of excess air is used, NOX levels on a concentration basis actually decline because of dilution and lower flame temperatures. However, the thermal efficiency also declines with higher levels of excess air, which again indirectly increases NOX as more energy is needed per unit of production.

High carbon monoxide and combustibles Figure 1. Air leaking around a process tube and header box.

Tramp air may produce high levels of carbon monoxide (CO) and possibly even combustibles. Depending on where the tramp air is coming into the heater, the measured excess O2 levels, which should be measured at the top of the radiant section (also known as the arch), may be within target levels and yet the burner flames may be starved for air. Another possibility is that tramp air infiltration near the excess O2 sampling location can produce a false high O2 reading. This means there is actually less excess O2 in the heater than is measured. Either of these conditions can result in the burners being operated with a reduced amount of air, which can lead to increased levels of CO. If the flames are very starved for air, combustibles may also be high, which should not happen under proper operating conditions. While some of the CO and combustibles may be consumed as tramp air leaks into a heater at higher elevations, they are not nearly as efficiently consumed as in the flame due to much lower temperatures in the firebox compared to the flames.

Afterburning in the convection section Possible afterburning in the convection section is related to high levels of CO and combustibles. Because tramp air can result from the burners being starved for air, the flue gas entering the convection section may contain high amounts of CO and combustibles. The convection section is normally where much of the tramp air enters a heater. This is due to the many tube presentations and the presence of header boxes. The oxygen in the tramp air results in oxidation of the CO and combustibles, which increases the temperature in this part of the heater. Combustion in the convection section can damage the finned tubes, tube supports, and tube sheets. It will also likely reduce the thermal efficiency as there may not be enough residence time in the convection section to transfer as much of the heat before the combustion products exit the stack. This can result in high flue gas temperatures, which is an indicator of reduced thermal efficiency.

Flame impingement on the process tubes

Figure 2. Sealed sight port design.

October 2021 54 HYDROCARBON ENGINEERING

If the burners are short of air, then the flames tend to get soft and lazy. Flames that have lost some of their mixing energy are influenced more by the flue gas currents in the furnace. This condition can lead to the flames impinging on the process tubes. Additionally, flames that are starved for air tend to get longer due to the lack of air needed to complete combustion. Those longer flames may impinge on the convection section tubes. Prolonged flame impingement can lead to process tube

damage ranging from coking inside the tubes, to tube leaks, and to the worst-case condition of a tube rupture.

Unstable flames A final detrimental effect of significant tramp air leakage is also potentially the most dangerous, which is unstable flames. If the process burner flames are lacking enough combustion air, the

flames could be operating near the upper flammability or fuel rich limit. Operation near that flammability limit is dangerous because if the flames get close enough or go above that limit, they could become unstable and even flame out if there is not enough oxygen for combustion. Unstable flames may bounce or pulse (‘huff’). If a flame goes out in a heater that is above the ignition temperature of the fuel, the flame could be reignited when sufficient air is found such as by tramp air leakage. This re-ignition could lead to equipment damage and personnel injuries.

Corrective/preventive actions

Figure 3. Engineered tube seals (image courtesy of Thorpe Corp., Houston, Texas, US).

Figure 4. Thermal image of a partially open explosion

door.

Figure 5. Burner out-of-service with an open air

damper.

October 2021 56 HYDROCARBON ENGINEERING

The first step in reducing the amount of tramp air is to locate areas where air is undesirably entering the heater. The most common places for air to leak into a heater are: the convection section; header boxes; process tube penetrations (Figure 1); partially or fully open sight, access, or explosion doors; and burners out-of-service with the air registers open. Performing a visual inspection around the heater will normally show many areas that need attention. Some of the common materials used to seal around tubes and sight doors are ceramic blanket, engineered tube seals, braided rope gasket, and high temperature (500°F) silicone sealant. If the heater casing has developed cracks, these should be welded closed to prevent air infiltration. Sight doors and inspection ports should be closed when they are not being used. The heater should never be operated with the sight doors open to get enough air for combustion. There are some sight port designs (Figure 2) that are sealed with a shutter that opens inside the heater which dramatically reduce tramp air leakage. When a heater is down for maintenance, smoke testing is a common way to determine where air is leaking into a heater. This can be done with smoke bombs placed inside the heater or with smoke cartridges placed in a blower inlet. The stack damper should be closed so that a slight positive pressure is created inside the firebox to force smoke out of the heater at the leak locations. Tube penetrations can be a large source of tramp air. Engineered tube seals (see Figure 3) eliminate tramp air from getting into the heater around the tubes while also allowing the tubes to grow as the temperature increases. Explosion doors should be sealed to eliminate tramp air. Figure 4 shows a thermal image of an explosion door that was not well sealed. This can allow a significant amount of tramp air leakage. It is not uncommon to find burner air registers open on burners that are not in operation (see Figure 5). This may be an oversight or for convenience when the burner is brought back into service. However, open air registers on burners that are out-of-service can be a very large source of tramp air leakage because the burners are often located where the draft is the highest. Note that there will still be some air leakage through closed air registers as the seal is not designed to be air tight. The effort and labour put into eliminating tramp air from the heater can result in better efficiency, higher yields, improved flame quality, lower NOX and CO levels, lower tube temperatures, and improved safety. As far as possible, tramp air leaks should be minimised or eliminated. The reduction in fuel costs alone will normally far offset the cost of the inspection and repairs.

Eric Pratchard, Zeeco, USA, evaluates the challenges to consider when transitioning to firing hydrogen.

ising fuel costs, new regulations requiring carbon footprint reductions, and global net zero carbon initiatives continue to pressure the refining and chemical industry markets. The two main methods to reduce CO and CO2 emissions are to either capture and sequester the carbon in the fuel gas or to remove carbon from the fuel before firing. Many organisations are considering refuelling existing fired equipment with sustainable, low carbon fuels. One such fuel is hydrogen (H2), which can be produced through renewable energy sources (green hydrogen) or by reforming natural gas (blue hydrogen). Removing carbon before combustion eliminates the need for costly equipment to capture and sequester the carbon during firing. Whether produced through green or blue sources, or recovered from existing plant processes, H2 can be injected into existing fuel gas networks to produce high hydrogen blends or used in pure form for fuelling fired heaters and process furnaces. As H2 displaces hydrocarbons in the fuel composition, the number of carbon atoms decreases. A fuel stream composed of 100% H2 cannot generate CO nor CO2 as a by-product of combustion due to the lack of carbon in the combustion reaction. Therefore, the higher the H2 content of a fuel, the lower the overall CO and CO2 emissions.

Due to the vastly different combustion characteristics of hydrogen compared with conventional hydrocarbon fuel gases, there are several challenges that need to be considered by plant owners and operators when transitioning to firing hydrogen.

Design challenges Most fired heaters and process furnaces that exist today were designed for firing natural gas or refinery fuel gases that contain a high proportion of saturated hydrocarbons with a make-up of hydrogen, inert gases, and traces of other compounds. Typical hydrogen content for refinery fuel gas may vary between 20 and 40% by volume. For hydrogen refuelling, H2 concentrations of 90 to 100% are probable. This quantity of hydrogen changes the operating parameters of the burner, so special attention must be paid to the burner design to ensure heater operation is not detrimentally affected. Firstly, the laminar flame speed of hydrogen is significantly higher than hydrocarbon fuels, promoting a more rapid combustion process and increased heat release per unit volume. Combining that fact with higher adiabatic peak flame temperatures leads to elevated temperatures local to the flame, which directly increases NOX emission rates by up to a factor of three. Employing ultra-low-NOX burner technologies can help HYDROCARBON 57

ENGINEERING

October 2021

Retrofit burners for high H2 firing To utilise high H2 fuel and remain within legislative NOX emission requirements, it is necessary to utilise ultra-low-NOX burner technologies. Diffusion type burners present different challenges than pre-mix burners when firing high hydrogen fuel; each type of burner will be discussed separately.

Ultra-low-NOX diffusion burners

The ZEECO ultra-low-NOX FREE JET burner utilises internal flue gas recirculation for reconditioning of the fuel gas with inert products of combustion prior to mixing with combustion air. The reconditioned fuel mixture prolongs the combustion reaction, thus reducing peak flame temperature and thermal NOX production. This burner can achieve NOX emissions of less than 50 mg/Nm3 on 90% hydrogen fuel without steam injection or post-combustion emissions Figure 1. Isometric and cutaway views of a ZEECO FREE JET burner control. The principle relies on the conversion of showing staged fuel risers to create the necessary turbulence and momentum of fuel gas ejected from the tips to mixing zone for counteracting the high speed of hydrogen flames. entrain flue gas. To achieve this, discrete high-velocity jets of gas are injected through a ring to remain within legislative NOX emission requirements when of gas tips on the outside of the tile. The gas jets help to firing high H2 fuel. counteract the high flame speed of hydrogen flames, ensuring a For example, the flame speed in H2 combustion is stable and robust flame over a wide operating range. approximately 5.7 ft./sec., while the flame speed of natural gas Another consideration when firing high hydrogen fuel gas is is significantly slower at only 1.3 ft/sec. H2 firing has a higher using a burner with low mass gas tips. In the FREE JET burner stoichiometric adiabatic flame temperature of 3960°F, while example, the gas tips protrude through the furnace floor by natural gas has an adiabatic flame temperature of 3518°F.1 These approximately 25 mm, so the thermal intensity over the gas tip significant differences in combustion characteristics require profile is significantly reduced. This means the gas tips are engineers to evaluate the materials used in burner construction suitably designed to withstand the elevated temperatures and the type of burner used. characteristic of hydrogen firing with an extended operating life. Typical burner construction includes various metal Ultra-low-NOX pre-mixed radiant wall components and a refractory throat or tile. The increased flame burners temperature of H2 requires upgrading the steel used for nozzle Pre-mixed radiant wall burners, as commonly used in ethylene construction, throat construction, and flame stabilisers to a cracking furnaces, present an altogether different challenge higher grade stainless or alloy. Refractory used within the burner when firing high hydrogen fuels due to the propensity for should be carefully evaluated and its composition modified to flashback. Because of this, burner designers must consider the withstand the elevated temperatures characteristic of H2 firing. The steel used in burners firing H2 should not be susceptible flammability window and flame speed of each specific fuel to hydrogen embrittlement and high-temperature hydrogen composition. The flammability window of pure methane is attack. Both phenomena can prematurely degrade an between 5% and 17%, with a flame speed of 1.3 ft/sec. This improperly chosen steel, leading to early failure of the burner means that when mixed with air at a concentration between 5% parts. and 17%, methane will support combustion at that speed. The Increasing hydrogen content in the fuel gas lowers the flammability window for hydrogen is between 4% and 74%, with specific gravity of the fuel, causing the fuel gas mass flow rate to a flame speed of 5.7 ft/sec. decrease. Consequently, it is often necessary to increase the fuel As the industry pushes hydrogen concentration higher, it gas pressure to achieve the same burner heat release. Therefore, becomes increasingly difficult to design burners with exit burner gas tip design and fuel gas piping hydraulics should be velocities that can overcome this increase in flame speed, assessed and re-sized as necessary. especially considering a common requirement for the burner to Additionally, existing heater safety interlocks and trip operate on both a natural gas and high hydrogen blended fuel. If settings need to be reviewed and amended as appropriate for the designer gets this balance wrong, the flame will propagate high hydrogen fuel. For example, furnace TDL systems and flame back inside the burner. Known as flashback, this can be scanners with UV/IR detection may no longer be suitable and detrimental to the mechanical integrity of the burner alternative technologies capable of detecting high hydrogen components and thermal NOX emissions levels. To overcome this technical challenge, Zeeco has developed flames may need to be considered. These aspects are covered a radiant wall burner design capable of firing fuel gas through heater impact studies. October 2021 58 HYDROCARBON ENGINEERING

F I R IN G H Y D R O G E N ? N O P R O B L E M.

END TO END PROVEN HYDROGEN FIRING SOLUTIONS. Zeeco is the world leader in innovative hydrogen firing firring onn, solutions. From burner retrofits to new construction, trust the name known worldwide for dependable technology even under the toughest circumstances. ess. Firing hydrogen safely and efficiently requires specific eccific changes to pilots, flame scanners, and control systems in addition to burners. Zeeco can providee a complete solution backed up by our worldwide experience and regional expertise. Redefine your fuel capabilities and transition onn to hydrogen with confidence. Zeeco - Redefining Combustion.

Learn more at zeeco.com BURNERS

FLARES

ZEECO INC.

THERMAL OXIDIZERS

+1 (918) 258-8551 | sales@zeeco.com

VAPOR CONTROL

RENTALS

AFTERMARKET

prior to combustion. When using high hydrogen pilots, the chance of flashback inside the pilot tube increases. It is crucial to employ pilot designs capable of firing on pilot gas compositions of up to 90% hydrogen without flashback to avoid damage to the pilot itself and any internal components. For example, some pilot designs utilise an adjustable air door to manually control airflow and prevent flashback when firing high Figure 2. Examples of ZEECO pilots firing high hydrogen fuels. hydrogen fuels. As the hydrogen content increases, the air door combinations of more than 90% hydrogen whilst producing less must be closed further to maintain the jet speed and prevent than 100 mg/Nm3 NOX emissions levels. This burner design flashback. utilises a proprietary fuel staging design that produces two Pilot flame detection separate mixture zones exiting the tip. This combination allows Many burner pilots used on fired heaters are equipped with the burner to generate the exit velocity needed in the lean zone ionisation rods for pilot flame detection. Flame rods work to prevent flashback while delaying combustion in the rich zone through the ionisation/rectification process to complete an long enough to mix fuel with inert flue gas products. electrical circuit. When the flame rod is energised, the current This burner can be retrofitted in existing ethylene furnaces produces a positive charge that attracts negative ions in the for firing high H2 fuels even in challenging furnace applications with extremely tight burner-to-burner and burner-to-tube flame. The positive ions produced in the combustion process spacing, still meeting the <100 mg/Nm3 emissions are attracted to the grounding area of the pilot tip. By attracting requirement. more positive ions to the ground, the flow of electrons is rectified and flows in one direction. This produces a direct Flame scanners current signal that indicates the presence of flame. Conventional flame scanners are configured to detect For the ionisation system to function properly, a significant combustion radiation in the UV, visible, and IR spectrum. The number of ions must be present in the flame. Hydrogen flames exact spectral ranges are based on the wavelengths produce few ions compared with organic compounds, so a characteristic of the combustion of hydrocarbon fuels. The UV weak current is generated that cannot be detected by the and IR components vary depending on the specific type of fuel flame ionisation module. Therefore, flame rods are not a used; therefore, conventional scanners have wide spectral suitable method of pilot flame detection for high hydrogen ranges, allowing operation on various gaseous and liquid fuels. fuels, and alternative means of pilot flame detection must be When firing high hydrogen fuels, the absence of carbon considered. means that the spectral range of radiation produced by the One method of pilot flame detection is the use of a flame combustion process is significantly narrowed and shifted more scanner that is mounted to the rear end of the pilot. The flame towards the UV spectrum. Therefore, scanners that rely on UV scanner is sighted on the pilot tip for detection of the flame. and IR spectral ranges experience weakened peak flame signals, The pilot tip shield can be specially modified so the flame leading to nuisance trips. scanner does not detect a signal from the main burner flame. For scanners utilising the UV spectral range only for flame Instrumentation and controls detection (no IR component), peak response occurs at the considerations OH-radical remission wavelength. Therefore, the absence of When utilising H2 as a fuel source, the final topic to be carbon in the flame does not impede the scanner’s ability to considered is the controls and instrumentation required for safe detect radiation from the target burner. One example is Zeeco’s firing. Any burner designed to have a varying fuel composition ProFlame scanners, which can reliably detect 100% hydrogen spanning from natural gas to high H2 content should have a fully flames, a critical furnace safety interlock when firing high metered combustion control system coupled with a Wobbe hydrogen fuel. Index meter or specific gravity meter in some cases. The Wobbe The burner flame may become much shorter when firing Index meter monitors the varying fuel stream composition and high hydrogen fuels due to increased flame speed. Therefore, it provides the necessary input to the control system to properly is essential to sight the flame scanner properly to ensure it is adjust the fuel/air ratio control in the combustion control aligned with the target flame. This will reduce the chance of system. The inability to monitor the fuel stream composition unwanted background signals being detected by the flame and adjust the combustion control system to those changes can scanner. lead to a potentially unsafe, fuel-rich condition.

Pilots

Most pilots used on fired heaters are the self-aspirated type, meaning the air is naturally induced for mixing with fuel gas October 2021 60 HYDROCARBON ENGINEERING

Reference 1.

GLASSMAN, I., ‘Combustion’, Second Edition, (1987).

Stuart Morstead, Honeywell Connected Industrial, presents an overview of the main trends shaping the future of the industry, and outlines the need to connect processes, people and assets to meet new challenges.

2020 Honeywell survey of 3000+ managers in multiple industries found that a majority of respondents said they were “worried they may be too late with their digital transformation efforts and will fall behind their competitors.”1 New challenges, including macro volatility, continue to drive the need to transform, and nowhere more so than industrials that have been historically sheltered with fewer

transactions and lower user experience expectations than consumer oriented sectors. There are wide-ranging emerging industrial trends and diverse stakeholder needs that have to be met, including sustainability, product mix and financial performance. These trends and stakeholder needs represent a significant challenge to both conservative and aggressive operators alike.

HYDROCARBON 61

ENGINEERING

October 2021

The good news is that one digital lever in particular can play a role in meeting a broad set of transformation drivers: design, build, operate and optimise.

The big three dividers Increased volatility High volatility in supply and demand severely tests operators’ abilities to hit their volume and mix targets – all while maintaining profitability. Both sides of the supply/demand equation are seeing increased complexity, which can be an impediment to productivity gains. The global regulatory environment, the cyclicality of commodity prices, and the impact of generational events or black swans have all convened to increase volatility, especially in supply chains. While not necessarily fragile, supply chains in many industries are not as resilient as they need to be, and the interdependencies can result in major problems when a crisis hits. On the flip side of the supply chain, the demand patterns in end markets are rapidly creating new product mixes for operators with which to contend. Many industries are dealing with new demands for new types of products in their portfolio. For instance, automotive is facing a demand for lighter materials and electric vehicles (EVs), as well as autonomous vehicles. In construction, the demand for pre-fab and 3D printed modules has increased. Manufacturing and refining are facing their own new product trends and demands. As demand shifts in consumer end markets, operators are struggling with a flexible manufacturing response, and it is challenging to deliver a more diverse product set.

Sustainability as a core imperative The second urgent need relates to the global call for decarbonisation, methane reduction, and green/blue/grey hydrogen issues. In short, the need for real movement on operational sustainability. Many organisations are asking themselves how they can properly respond to these societal demands. How do we all balance profit and efficiency against being a responsible steward of the environment? And the challenges are different for each industrial factor. For instance: Refining: irrespective of recovery pattern, the large volume of emissions contributed by refined products must be offset through efficiency or capture. Chemicals: plastics recycling promises the creation of a circular economy, but investments, partnerships and infrastructure are all still required. Mining: CO2 and water challenges are this sector’s biggest environmental, social, and governance (ESG) issues to solve in order to earn ‘the right to play’. Regardless of the specific sector – improved monitoring of CO2 emissions, better management of energy usage, new implementations of carbon capture technology, and process information-sharing between partners – all of these challenges require digital solutions to properly execute. October 2021 62 HYDROCARBON ENGINEERING

The demographic cliff The third set of challenges involve the impending ‘demographic cliff.’ There are two factors at play in the creation of this ‘cliff.’ First is an ageing workforce which has started to retire – taking decades of expertise with them. Going through this ‘great crew change’ has impacted manufacturers who must increase their training programmes while enduring a decrease in productivity at the same time. The second factor is the ‘deskless employee.’ A lack of remote access to key data can greatly impair productivity across the entire organisation. This gap is especially evident in departments that operate across a far-flung plant – such as maintenance, where personnel may need to be in one location but not have access to accurate data about that equipment and processes. Solving these workforce challenges truly depends on having the right information and right expertise for frontline workers – all at the right time. A successful industrial digital transformation solves all these issues and brings with it the following benefits: Enhanced situational awareness. Minimised human exposure to safety risks. The ability to make better, faster decisions. Enhanced process safety and execution. Improved knowledge capture and management. Valuable, actionable insights and proactive guidance via the use of machine learning (ML)/artificial intelligence (AI). A paradigm shift in training and skills-upgrade, enabled by augmented reality (AR)/virtual reality (VR).

A collective shift in operating context Collectively, the three main drivers create a dramatic shift in the operating context for industrials. Figure 1 demonstrates how the operating context has shifted in recent history, driving the need for digital transformation. The bottom row illustrates how, in the last few decades, industry has already undertaken several shifts driven by the operating context. This encompasses the internal environment of the physical plant and process, the external environment, and the responsibilities that lie with the social contract (worker safety to water/energy consumption, etc.). In the next row, it is shown how these challenges can be managed, with tools that can be easily implemented, such as: preventive maintenance/computerised maintenance management system (CMMS) protocols, automation systems, isolated data analysis, and digital point solutions. But comprehensively, these have not resulted in an integrated view of the industrial enterprise, and so the promise of the ‘smart plant’ or of ‘industry 4.0’ remains elusive. While the ultimate destination in the future may be autonomous, self-optimising plants, the journey to cover the ‘last mile’ for operating technology, especially in chemical processing plants, is a difficult one and still lacks critical connectivity. Connecting assets and connecting processes, highlighted in the third row of Figure 1, results in

connected plants and enterprises – in other words, plants that can be optimised across the production process using technologies such as process digital twins, asset performance management, and data orchestration. But there is a missing piece of this connectedness: ‘connected workers’, the people who ensure the work gets done the right way, safely and efficiently. By combining the top two rows, along with a robust, interoperable connected data solution, Enterprise Performance Management for Industrial can be created.

Figure 1. This table demonstrates how the operating context has shifted in recent history, driving the need for digital transformation.

The ‘last mile’ Enterprise Data Management for Industrial (EDM) is key to a safe, successful, efficient digital transformation. It means collecting and analysing disparate data from the plant floor, other systems, and orchestrating it across multiple applications in those connected processes, assets and worker suites. Ultimately, it provides visualisation and insights into enterprise performance. It is the foundation of the system and is not sufficiently available today for operators.

Honeywell asserts that, on a global scale, the lack of EDM today is holding operators back. There are various digital point solutions for asset monitoring, or isolated process management, or enabling knowledge workers. Data storage and historians are readily available. But to deliver digital transformation – to connect IT to OT – industrials must cross the last mile to that future with true EDM. EDM for manufacturing is like the enterprise resource planning (ERP) systems that are used by finance functions today, or the customer relationship management (CRM) systems that have revolutionised sales teams.

October 12-13, 2021 Moody Gardens Convention Center GALVESTON, TX Register now at:

gulfcoastconference.com

Questions? (281) 256-8807

True industrial EDM provide an integrated management of operations in real time with cybersecurity, data connectivity, orchestration, contextualisation and visualisation as the foundational elements. EDM can be thought of as a new System of Record for IT and OT – just as ERP is a system of record for finance and accounting. And like ERP, it has the potential to unlock enormous enterprise value. Over the last 30 – 40 years, the industry has invested in foundational technology, including automation to improve profitability and run safer, reliable operations. Today, as we move out of the COVID-19 pandemic, the industry faces many challenges and disruptions – from an ageing workforce, to sustainability/net zero/decarbonisation needs, to a volatile supply and demand. Process industries are in the early stages of digitising manufacturing operations, with increased focus on monetising data and the orchestration of information. Achieving both of these goals will maximise performance from assets, processes and people. After a history of point solutions and individual transformation initiatives, the time has come for the process industries to pivot to a new digital future. By adding a system of record (SOR), coupled with visualisation and advanced analytics in a complete EDM solution, Honeywell aims to drive ‘the last mile’ around connecting processes, people and assets. This will address new industry challenges and maximise profitability for customers in a safe, reliable and sustainable way.

The path to safe, autonomous, optimised plants is an SOR that covers the following. Connected assets: Maximise asset performance and optimise for reliability. Manage planned and unplanned maintenance activities. Plan, track, optimise energy and emissions. Connected processes: Optimise processes and production performance across enterprise with integrated planning. Dynamically adapt processes to changing plant conditions. Provide workforce competency and productivity improvements. Connected workers: Manage operational risks and ensure safety. Enable new ways of working with people away from work sites, regardless of which of these needs is driving digital transformation. The time to start transforming is now. Focusing on this last, crucial link is the only way to guarantee that a plant is fully optimised, truly efficient and completely connected.

Reference 1.

Multiple sources including Gartner articles 2020 - 2021, Honeywell analysis.

AD INDEX Page Number | Advertiser 21 15 OFC & 47 32 11 02 41 04 OBC 63 IFC 55 17

October 2021 64 HYDROCARBON ENGINEERING

31 | NAPCON 35 | NEO Monitors 27 | Optimized Gas Treating, Inc. IBC, 31, 39 | Palladian Publications & 45 23 52 21 51 24 07 45 13 59

The premier source of the Theinformation premier source latest for of the informationenergy for the renewable renewable industry, energy industry covering solar,covering wind, solar, wind, bioenergy, bioenergy and storage. storage and more Subscribe online at: www.energyglobal.com/magazine

OH YEAH, WE HAVE A FILTER FOR THAT.

ÇɁ ǁɁʍƹɽ ƃƹɁʍɽ Ȉɽӝ Śȃƃɽǉʤǉɨ ɥɨɁƺǉɰɰ ʰɁʍԇɨǉ ɨʍȶȶȈȶǼӗ ʥǉ ȃƃʤǉ ƃ ˎȢɽɨƃɽȈɁȶ ƃȶǁ ɰǉɥƃɨƃɽȈɁȶ ɰɁȢʍɽȈɁȶ ɽȃƃɽ ʥȈȢȢ ɰƃʤǉ ʰɁʍ ɽȈȴǉ ƃȶǁ ȴɁȶǉʰ Ȉȶ ɽȃǉ ȢɁȶǼ ɨʍȶӝ ȶǁ Ȉǹ ʰɁʍɨ ɥɨɁƺǉɰɰ ǁǉɰȈǼȶ ȃƃɰ you stretching out into new territory, don’t worry. New territory is kind of our thing.

FILTRATION CHALLENGES? RADICAL SOLUTIONS.

LIQUID / SOLIDS

GAS / LIQUID

GAS / SOLIDS

LIQUID / LIQUID

713.849.0849 | ftc-houston.com

Hydrocarbon Engineering October 2021

Articles inside

The last mile

H2 ready

The problem with tramp air

Contactless combustion analysis

Trading risk for reliability, fl exibility and effi ciency

The critical measurements

Simulation in a world of trouble: part two

A delicate return

Green SRU in biofuels

World news

Comment