YoungPetro Magazine 22nd Issue - Spring 2019 by Young Petro

SPRING /â&#x20AC;&#x2030;2019

Editor-in-Chief Anna Orchel aorchel.youngpetro@gmail.com Deputy Editor-in-Chief Karolina Potasiak karolinka.potasiak@gmail.com Editors Patryk Bijak Filip Czerniawski Paweł Bielka Milan Zięba Piotr Froń Anna Orchel Filip Mazurkiewicz Monika Walczuk Graphic designer Agnieszka Zagraba Social media Ivan Gnilichenko

Marketing Anna Orchel Karolina Potasiak Ambassadors Chuah Kai Jie - Malaysia Mohd Zaidi Bin Jaafar - Malaysia Yohanes Nuwara - Malaysia Michael Aria Santoso - Malaysia Tita Oxa Anggrea - Indonesia Titania Nur Bethiana - Indonesia Totok R. Biyanto - Indonesia Publisher Fundacja Wiertnictwo - Nafta - Gaz, Nauka i Tradycje Al. Adama Mickiewicza 30/A4 30 - 059 Kraków, Poland www.nafta.agh.edu.pl

A disaster of Deep Water Horizons 6 Anna Orchel

Reducing pollutant emissions by using natural gas as fuel in car transport 9 Paweł Bielka

2018 Oil Price Overview 11 Filip Mazurkiewicz

The race for Arctic oil and gas reserves 13 Milan Zięba

Will gas hydrates cause ecological disaster? 15 Piotr Froń

Negotiation Game for Young Professionals – Model OPEC 22 Monika Walczuk

Optimization of Energy Consumption in Distillation Columns 24 Tita Oxa Anggrea, Titania Nur Bethiana, Totok R. Biyanto

Seismic Attribute Analysis Aa A Tool For Structural Interpretation 35 Yohanes Nuwara, Michael Aria Santoso issn

2300-1259

Published by

Foam Stability Performance Enhanced with Rice Husk Ash Nanoparticles 50 Chuah Kai Jie, Mohd Zaidi Bin Jaafar

¡¡

A disaster of Deep Water Horizons

Anna Orchel

A disaster of Deep Water Horizons Anna Orchel

9 years will soon pass since the most dangerous environmental disaster in US history. The explosion of the Deepwater Horizon oil rig took place on April 20, 2010 in the Gulf of Mexico. Deepwater Horizon was a dynamic, semi-submersible drilling platform, built in 2001 in South Korea, registered in Majuro in the Marshall Islands. It belonged to the Transocean Company that conducted the Mississippi Canyon 252 borehole ordered by the BP group (according to the agreement signed on August 2013). The platform was adjusted to drilling down to about 2,450 m (8,000 ft), and the maximum depth of the borehole was 30,000 feet (about 9,150 m). It also set the world record in terms of the real vertical depth (so-called TVD – " True vertical depth ") of the borehole (10,683 m), with a hole length of 10,685 m. Deepwater Horizon platform, cooperating with British oil company BP, was located about 80 kilometers from the cost of the state of Louisiana. On 20 April 2010 at 22.00 local time, there was an explosion on the drilling deck, in the Gulf of Mexico. In less than two days the drilling platform sank, triggering one of the biggest environmental disasters. The course of events ◀◀ April 20, 2010 – a rapid fire appeared on the Deepwater Horizon oil rig with 126 people on board. As a result of the explosion of methane and fire of crude oil,11 employees of the facility were killed (missing), 17 hurt. The rest of them were evacuated ◀◀ April 22, 2010 – sinking of the platform in the Gulf of Mexico, about 80 km from the Louisiana coast (USA);

Fig.2 The fire of the platform

◀◀

Fig.1 Gulf of Mexico

◀◀

◀◀ April 24, 2010 – detecting an underwater oil spot from drill pipes estimated at 150 m3/d at that time ◀◀ April 27, 2010 – spreading of the oil spill to the Louisiana coast. Several thousand volunteers together with coastguard built temporary dams in order to stop the further oil spillage; ◀◀ April 28, 2010 – US Navy sent special planes adjusted to spray chemical dispersing oil in water; ◀◀ April 30, 2010 – reaching the oil spill to the coast of Louisiana (USA), Causing the ecological damages to birds, fish and mammals not only in this particular region; ◀◀ May 2, 2010 – an unsuccessful attempt to stop the leakage by closing the blowout of pre-treatment head, and starting an additional drill that could decrease the tank pressure ◀◀ 7-8 May 2010 – an unsuccessful attempt to stop the leakage by placing a steel dome, allowing the collection and discharge of oil, which will soon clog and raise up by the decomposing hydrate of methane, to the tanker on the surface of the sea, ◀◀ May 16, 2010 – drilling a special rescue shaft near the whole with a constant leakage; ◀◀ 16-17 May 2010 – installing a crude oil pipe suc-

◀◀

king the oil from a damaged 1.6 km shaft transporting oil and gas to Discoverer Enterprise; May 19, 2010 – delivering a statement that the sucking installation sends about 5000 bbl * (795 m3 / d) per day to the surface May 21, 2010 – announcement delivered by Barack Obama – the President of the United States concerning starting an investigative commission to determine the causes of the disaster; May 26, 2010 – an attempt to reduce the leakage by injecting a thick scrubbing fluid into the well in order to cement it. The attempt failed due to too high pressure of the leaking oil; May 27, 2010 – an attempt to reduce the leakage by drilling of fragments of cut rubber and plastic in order to plug the borehole. The attempt also failed. June 1, 2010 – making the first sum up of the costs of the rescue operation, which showed that the expenditures already exceeded USD 1 billion; June 3, 2010 – installing the crude oil sucking dome at the base of the well with the system of pumping it into the tankers; June 4, 2010 oil – contaminating the fragments of the coast of two more American states, Mississippi and Alabama; 17 June 2010 – installing a crude oil extraction system from an eruption bore, enabling its outpouring to sea water by 12,720 m3/d; June 21, 2010 – estimating the cost of the previous rescue mission for 2 billion dollars;

◀◀ June 29, 2010 – another phase of the rescue mission, included finalizing the drilling of two side rescue shafts, reducing the reservoir pressure in the eruption borehole to a level that would allow the well to be clogged with self-drilling mud. It was completed on July 15; ◀◀ 5 July 2010 – engaging the rescue team of the ship A Whale , supertanker adjusted to collect oil from the surface of the water; ◀◀ July 12, 2010 – dropping a dome weighing 75 tons at the bottom of the Gulf of Mexico, the installation was supposed to protect the eruption opening after the leakage was stopped on July 15; ◀◀ On July 19, a new oil spillage was detected from the near the damaged well. The spillage started in September 2010, but it was not necessarily related to the above-mentioned disaster, because such bottom crude oil leaks are quite frequent, and they can occur in a natural way. ◀◀ July 27, 2010 – BP group informs that in October 2010 Tony Hayward will resign from the position of a general manager of the company. He was extensively criticized for the wrong decisions taken right after the disaster; ◀◀ September 2010 – the end of the process of stopping the leakage and continuation of cleaning of the sea surface and coasts from petroleum derivatives.

* 1 barrel of crude oil (abbreviation: 1 bbl) = 42 American gallons = 158.987 (~ 159 l), this is a standard unit of volume in the oil industry.

SPRING / 2019

8 Causes of the disaster The immediate cause of the disaster was the eruption of deposit products – crude oil and gas, their explosion on the oil platform, and later the fire which was impossible to put out. Indirect causes were, especially, the lack of an additional blow-out head, error of the apparatus concerning the pressure of deposit products, as well as the type of cement that does not provide resistance to the actual pressure of the oil and the use of six so-called well centralizers instead of the 21 recommended . Mechanical faults, wrong decisions made by employees, engineering errors or the way of introducing decisions also contributed to the accident. The effects During the Deepwater Horizon oil rig explosion on April 20, 2010, 11 people were killed, 17 injured, and the oil platform sank two days later. Almost 5 million barrels of oil,over 666,000 tons, poured into the ocean. 500 km of the US coastline were destroyed. The sea tourism in this part was also affected. Economic losses in tourism and fishing were caused by the oil spill, which, after emerging from the damaged borehole, covered a huge area of the sea. It resul-

A disaster of Deep Water Horizons

ted in ecological losses; it affected water birds and marine mammals, especially turtles and dolphins populations. It is estimated that a year after the environmental disaster life the Gulf of Mexico gradually normalized, and after five years tourist came back on the beaches of Alabama and the Louisiana. Mississippi Delta did not seem contaminated, and the local economy has already reached the level from before the catastrophe. While preparing the 2010 report, the PB Group stated that it has spent or is still obliged to pay a total of USD 40 billion, mainly as compensations. Among the mistakes made on the Deepwater Horizon platform, the investigators also pointed: bad risk assessment, ignoring the signals of approaching explosion and inappropriate security tests.

Fig.3 BP Logo

References: [1] https://en.wikipedia.org/wiki/Deepwater_Horizon_oil_spill [2] https://en.wikipedia.org/wiki/Deepwater_Horizon_explosion [3] https://publicintegrity.org/environment/eight-years-after-deepwater-horizon-is-another-disaster-waiting-to -happen/ [4] https://ocean.si.edu/conservation/pollution/gulf-oil-spill

Paweł Bielka

¡¡

Reducing pollutant emissions by using natural gas as fuel in car transport Paweł Bielka

Diesel oils, gasoline and LPG ( liquefied petroleum gas), are currently the most frequently used fuels. From one year to another we can observe an increase in pollutant emissions to the atmosphere from road transport. According to data from European agencies, about ¼ of carbon dioxide emissions in the countries of the European Union come from transport, that is why in the recent years the European Union has taken action to reduce emission of harmful substances from this sector. That is why in Europe natural gas vehicles (NGV’s) play a major role. A decisive reason for this is the dependence of most European countries from gas imports. In addition, there is a lack of infrastructure (e.g. fuelling stations). Contrary to Europe, in Latin American and Asian countries natural gas vehicles are widespread. Some countries foster natural gas vehicles because they have their own gas resources. Many countries must reduce the high air pollution in big cities. Environmental reasons are the main motive for using natural gas vehicles in Europe. In the last years, high oil Exhaust composition

prices stimulated the use of natural gas as fuel. However, the focus is set on hybrid technology and the electric car, which, unfortunately, still need further technical improvement. In contrast, the use of natural gas in conventional engines is technically mature. Additional gas imports can be avoided by further improvements of energy efficiency and the use of renewable energy. Natural gas as fuel The use of natural gas allows the construction of a vehicle with low emissions. This entails, however, a lot of complications that must be resolved. Natural gas has a much lower energy density, which means that it is in gaseous state while gasoline and diesel fuel are in liquid state. Lower Energy density requires customizing gaseous-powered vehicles which is connected with redesigning the drive unit. This implies that it is impossible to supply gasoline or diesel engines by natural gas without modification.

Value of exhaust emissions [g/kWh] engine powered by diesel

Nitric oxides

13,4

Carbon oxides

4,6

Particulate matter

0,3

Source: Monika Orzechowska, Dominik Kryzia, Analiza SWOT wykorzystania gazu ziemnego w transporcie drogowym w Polsce, Polityka Energetyczna, Tom 17, Zeszyt 3, 2014 r. Tab.1 Toxic compounds emitted from city bus

SPRING / 2019

Reducing pollutant emissions by using natural gas as fuel in car transport

Filip Mazurkiewicz

References: Diesel

Gasoline

Natural gas

Energy destiny

38 MJ/l

34 MJ/l

0,037 MJ/l

Air consumption rate

14,5 kg of air/ kg of fuel

15 kg of air/ kg of fuel

17,2 kg of air/ kg of fuel

Flash point

220 °C

650 °C

Flammability limit

0,6-6,5%

1,3-7%

5-15%

Density in gaseous

0,86kg/dm3

0,74kg/dm3

0,72kg/dm3

[1] Paweł Dorosz „Compressed and liquefied natural gas as an alternative for petroleum derived fuels used in transport”, Polityka Energetyczna, ISSN 1429-6675 [2] "WESTPORT HPDI 2.0 LNG engine". Retrieved 17 April 2015. [3] "Progress in Energy and Combustion Science”, Volume 64, January 2018, Pages 62-92 [4] Maciej Gniot "Analiza możliwości wykorzystania LPG i LNG w transporcie publicznym w województwie kujawskopomorskim” [5] https://en.wikipedia.org/wiki/Natural_gas_vehicle” [6] https://en.wikipedia.org/wiki/Compressed_natural_gas [7] https://en.wikipedia.org/wiki/Liquefied_natural_gas

state

Source: E. Król, M. Flekiewicz; Gaz ziemny jako paliwo do napędu pojazdów samochodowych – doświadczenia i perspektywy, Nafta-Gaz, Nr 7-8/1997 Tab.2 Comparison of the properties of the diesel, gasoline and natural gas However, natural gas under ambient conditions is c low energy density. That is the reason why it must be stored as gas pressurized over 200 bar (CNG – Compressed Natural Gas) or as the liquid gas (LNG – Liquefied Natural Gas). Differences between LNG and CNG fuels Though LNG and CNG are both considered natural gas vehicles, the technologies are vastly different. Refueling equipment, fuel cost, pumps, tanks, hazards, capital costs are all different. One thing they share is the fact that due to engines made for gasoline, both of them in order to control fuel mixtures require computer controlled valves which, often being proprietary and specific to the manufacturer. The on-engine technology for fuel metering is the same for LNG and CNG. LNG offers a unique advantage over CNG for more demanding high-horsepower applications by eliminating the need for a turbocharger. Because LNG boils at approximately −160 °C (−256 °F), by using a simple heat exchanger a small amount of LNG can be converted to its gaseous form at extremely high pressure with the use of little or no mechanical energy. A properly designed high-horsepower engine can leverage this extremely high pressure

energy dense gaseous fuel source to create a higher energy density air-fuel mixture than it could be efficiently created with a CNG powered engine. The end result when compared to CNG engines is more overall efficiency in high-horsepower engine applications when high-pressure direct injection technology is used. Conclusion -One of the alternative fuels that can be used for a more environmentally friendly vehicle supply can be natural gas in the form of LNG and CNG, due to the high availability and relatively low costs of expanding the existing gas stations for installations powered by natural gas. -Another aspect of using natural gas as fuel is the fact that natural gas reserves are much larger compared to crude oil reserves. With the depletion of crude oil reserves, the price of it will increase. As long as hydrogen technologies are not developed enough to be used on a massive scale and electric cars batteries will not be capacious enough, natural gas will be the only alternative to diesel oil and gasoline. However it is worth noting that the use of compressed and liquefied natural gas is causing some problems with exploitation.

¡¡

2018 Oil Price Overview Filip Mazurkiewicz

After the severe oil price crash in 2014 which ended up as a catastrophy for the Oil and Gas Industry all over the world. The price of one barel of oil plunged from over $100 to less than $40 within no more than an 18 month period. Since then the O&G Industry has slowly but steadily being recovering. Coming up with new, game changing technologies such as hydraulic fracturing to enhance global production and supply. Getting in to the full swing as the oil price exeeds the break-even point for fracking. All of a sudden, after almost 45 years the USA regained its place on top as the biggest oil producer in the world with 12.5 million barrels a day in its account. Thanks to the great minds of scientists and fortunate geology, the O&G Industry in the US is booming once again. The United States of America is facing unrepeatable chance to become an energetically independent country. But does it have what it takes? Let’s review some events from the recent past that began to show the impact on the oil price espe-

Fig.1 Crude oil brent cially in the fourth quarter of 2018. In May 2018 President Donald Trump announced that the U.S would withdraw from the Iran nuclear deal, which meant that he plans to impose sanctions on Iran for alleged development of nuclear weapon. Sanctions were about to come into effect in November 2018 obviously aiming at the most profitable industry in Iran – oil exports, with intention to force Iran to

SPRING / 2019

12 alter its policies. As a result the price of oil was rising higher, daily since September. When everyone interested: The O&G companies, professionals, speculators and traders were getting nervous about the missing Iranian barrels and supply shortage for the future. Oil prices hit a 4 year high at $87 per a barrel. Leveraging that situation, President Trump turned to Saudi Arabia to increase their output in order to make up for Iranian oil loss in the market. Meanwhile, Trumps administration is leading dangerous trade war with China which began in late January 2018 and only intensifies over time. Every month new tariffs are being imposed causing sufficient increases in prices of fundamantal materials such as steel or aluminium which are vital for the US industrial growth. It hurts the O&G sector, especially during ongoing shale boom. Therefore Trump, not leaving these matters unanswered, decided to impose tariffs on US oil. It forced China to cut out unprofitable US oil imports for now. As a result of this devastating economic war, the US lost exports of 0.5 million barrels of oil a day and created an excess of oil in American inventories which has proven multiple times to have a serious impact on oil prices. Thus, affected US oil producers badly.

2018 Oil Price Overview

Only a couple of days before sanctions were about to come into effect, President Trump granted waivers on imports of Iranian oil for its biggest customers, including China. This has left Saudi Arabia which has just increased production, stranded. It turned out that Iranian oil eventually found its market and won’t be that much of a loss in global oil supply. Oil prices bagan to drop immediately as there was no more threat of global supplies deficit. Furthermore, even a surplus of oil in the market has been created, being jointly a result of China trade war tariffs and granted waivers. These unpredictable Trump’s plays undoubtedly resulted in Iranian sanctions weakening, another oil price drop, lost of Saudis trust and what’s more displayed the US and Trump’s inconsistency. As a side effect it also made the US oil producers struggle financially which probably wasn’t the original plan of the president. Are these Trumps actions fulfilment of his presidential campaign promises and the American dream of cheap oil? Will we see a barrel of oil high above $80 again, soon? What other consequences will the US decision bring? Certainly the upcoming 2019 will blow some of these uncertainities away and provide answers.

Fig.2

Milan Zięba

¡¡

The race for Arctic oil and gas reserves Milan Zięba

Arctic is a natural resource treasure-trove. There are seven countries: Canada, Greenland (Denmark), Iceland, Norway, Russia and the United States, who have a legitimate stake to it. Two-thirds of the Arctic area is the offshore continental shelf, whose depth usually does not exceed 500 meters. A certain part of the Arctic waters is currently covered with ice, but the surface of this ice cap has significantly decreased in recent years. It presented the possibility of exploration and exploitation of the resources behind the Arctic Circle and brought the challenges--- which we may soon have to face. Arctic resources From time to time, information about the need of drawing the borders on the Arctic Ocean appears in the world media. There is no doubt what the main reason for this state of affairs is. It is estimated that an area of about 6% of the Earth's surface can have more than 20% of the world's undiscovered but recoverable oil and gas resources, so we are talking about 412 billion barrels of total energy worth $ 10.5 trillion. Even the most pessimistic data indicate that under the Arctic ice 90 billion barrels of oil and over 1.669 trillion cubic feet of natural gas are buried. Such amount of oil resources is more than three times higher than current oil reserves in the United States or 13 times higher than known oil reserves in Norway. In the case of natural gas resources, these values are even higher and constitute the equivalent to 500% of the known natural gas reserves in the US and more than 2,300% of gas reserves in Norway.

Arctic brings challenges The most obvious difficulty in extracting hydrocarbons in this region is environmental conditions. Low temperatures in the winter, even down to -50 degrees Celsius can cause mechanical defects and freezing of shipping channels. Hence, difficulties to deliver goods and supplies can occur. Almost total winter darkness may have a significant impact on the work of the crew. Moreover, potential candidates wanting to start mining in the Arctics have to face billions of dollars for investment. There is no proven technology to work in such harsh conditions, and access to existing infrastructure is very limited. In addition, due onshore swamp formation in the summer, road transport and drilling can be almost impossible. Floating ice floes and icebergs can pose a permanent threat to ships. High costs and extremely long-lasting projects, such as the start of mining in the Arctic, carry an additional risk of economic downturns and exceeding the assumed costs. This means that obtaining financing for such projects can be difficult. Arctic influences politics From seven countries mentioned above, the US and Russia are most eager to start up drilling in the polar region. Undoubtedly, Russia has economic and military domination in this region of the world, because more than half of the Arctic coastline is located along--- its northern shores. On the other side, Donald Trump conducts aggressive politics and puts much ---effort to drill for oil in the Arctic. Trump is trying to make it easier for oil companies to develop in the U.S. Arctic and in the same time, he is placing new sanctions on Russia, to make it

SPRING / 2019

The race for Arctic oil and gas reserves

Piotr Froń

Fig.1 Arctic region

harder for them to develop in Russia. At the beginning of the 2018, ExxonMobil announced it was withdrawing a valuable venture with Rosneft and it is going to look for new opportunities in the United States. Despite this, Russia continues the aggressive conquest of the Arctic, which is important to it for economic and reputational reasons. Russia, with each subsequent year, launches oil fields that are located more and more to the north, at the same time building new polar military bases.

Economic and military power of China can be sufficient to start up in the competition for resources and influence in the polar region. China has already undertaken few arctic-aiming projects such as launching its first domestically built polar ice-breaker and investing in Russia’s Yamal liquefied natural gas project in the northern port of Sabetta. China’s interest in the Arctic extends beyond resources. China conducts multi-billion-dollar project to create new shipping routes and new ,,Polar Silk Road” can be a part of it.

In this race, which seems to have two favorites, the most confusion can be caused by a country, that does not even have territorial claim to the Arctic. References: [1] https://www.cnbc.com/2018/02/06/russia-and-china-battle-us-in-race-to-control-arctic.html [2] https://newrepublic.com/article/148095/trump-putin-race-arctic-oil. [3] https://www.businessinsider.com/how-gigantic-arctics-undiscovered-oil-reserves-might-be-2016-4?IR=2 [4] http://www.ourenergypolicy.org/wp-content/uploads/2013/09/Arctic_oil_and_gas.pdf [5] https://www.theguardian.com/environment/ng-interactive/2015/jun/16/drilling-oil-gas-arctic-alaska

¡¡

Will gas hydrates cause ecological disaster? Piotr Froń

„Burning ice” The main sources of energy used by the people are fossil fuels: hard coal, lignite, oil and natural gas. Anticipating further increase in energy needs and rising oil prices, alternative solutions are being sought. One of these sources may be methane hydrates. Methane hydrates (also known as clathrate hydrates) are crystalline water-based solids, in which small molecules (gases) or big molecules with large hydrophobic are trapped inside "cages" of frozen water molecules. Most low molecular weight gases, including O2, H2, N2, CO2, CH4, H2S, Ar, Kr, and Xe, as well as some higher hydrocarbons and freons, will form hydrates at suitable temperatures and pressures. Hydrates are created by the natural gas components like : methane, ethane, propane, isobutane, as well as by: nitrogen, carbon dioxide (CO2) and hydrogen sulphide (H2S). They are compounds

of the summation formula CH4 × 6H2O. Clathrate hydrates are not officially chemical compounds because related molecules are never associated with the network. Clathrate hydrates were first documented in 1810 by Sir Humphry Davy who found out that water was a primary component of what was earlier thought to be solidified chlorine, burning with an even red flame, leaving water. Hydrates can take the form of one of two structures crystal lattice [1]: ◀◀ type I – crystal lattice built of 46 water molecules. There are six large pores in the interior of the structure and two smaller pores ◀◀ type II – crystalline network consisting of 136 water molecules (16 small and 8 large cage) Pure methane hydrates present type I structure.

SPRING / 2019

Will gas hydrates cause ecological disaster?

Piotr Froń

Fig.1 Molecular structure of gas hydrates [2]

In nature there is also, although very rare, type III (H). In addition to methane, there are other hydrocarbons in it, such as n-pentane.

Fig.3 Three-phase (liquid water + hydrate + vapor) stability conditions: (a) in the permafrost and (b) in the ocean [4]

Stability of gas hydrates A binder that determines the formation of methane clathrate, are hydrogen bonds between water molecules and strength of van der Walls catenations – between water molecules and molecules of methane. This affects the stabilization of the crystal structure hydrate. A compulsory condition for this phenomenon to exist is the right temperature and pressure. At a temperature of a few degrees Celsius and under a pressure of about 50 bar, methane ,appearing in the ocean at a depth of about 500 meters, forms with water colorless, glassy solids, in appearance resembling ice. According to literature [3], a parameter contributing to the formation of hydrates methane, except pressure and temperature, is the so-called temperature subcooling. Its average value during the formation of methane hydrates is from 3°C to 6°C. There are four main factors determining the stability of hydrates: the right temperature, the right pressure, the right gas composition and the right water, in which the sediment, the hydrate could be created. Methane hydrates are stable at temperatures up to + 18 ° C at high pressure, prevailing at a depth of 1500 meters. At a temperature of + 2 ° C, a much lower pressure, such as at a depth of 300 meters, is sufficient to maintain stability.

Fig.2 Typical phase diagram for simple gas hydates (in this case methane clathrates)

1. Starting from the ocean floor into the depths of the earth, the temperature rises (geothermal gradient) until it reaches the level at which methane can no longer exist in the form of hydrates (although it can exist in gas form). Low temperatures in polar regions ensure the possibility of hydrate deposits even at small depths. The upper limit of the hydrate stability zone occurs at a depth of approximately 200 meters. The maximum depth of this zone is limited by the geothermal gradient.

within the upper 2000 m of the Earth's surface, and the wide geographical distribution of gas hydrates. Natural gas is widely expected to be the fastest growing primary energy source in the world over the next 20 years. It is estimated that worldwide gas consumption will increase to almost double to 162 trillion cubic feet in 2020 from 84 trillion cubic feet (standard conditions) in 1999. Given the attractive features of gas hydrates, and the growing demand for natural gas, it seems reasonable to conclude that gas hydrates could serve as a future energy resource

[2] Moreover, from one cubic meter of methane hydrates we can get even 160 cubic meters of natural gas. How much is this on a global scale? The range of estimates is huge: from 3,000 trillion cubic metres up to over 140,000 trillion cubic metres. But the most common is the size of 20 000 trillion cubic metres – that is what US Department of Energy says. For comparison, the available resources of conventional natural gas and shale gas are in total 640 trillion cubic meters. And in 2011, the global gas consumption amounted to approximately 3.4

2. The upper limit in oceanic sediments depends on temperature, best is near 0 °C and at a water depth of 1200m. The lower limit of the hydrate stability zone is bounded by the geothermal gradient. The appropriate depth of the water body also allows a proportional pressure increase that will allow us to form a hydrate. Gas hydrates are attractive as potential energy recourse because of two factors: the huge volumes of methane that is apparently trapped as clathrate

Fig.5 Map of gas hydrate locations

SPRING / 2019

18 trillion cubic meters. Methane clathrates can be naturally found on Earth on the seafloor, in ocean sediments, in deep lake sediments as well as in the permafrost regions. The search for hydrate deposits is facilitated by the fact that sound waves diverge in them twice as fast as in ordinary bottom sediments. The largest ones so far are found in the depths of Blake Ridge off the coast of North Carolina, rich deposits are also found in the Gulf of Mexico and the Nankai Trough off the coast of Japan. Japanese success You can believe it or not, but so far nobody has benefited from the resources stored in the methane hydrate deposits on a mass scale. However, the Japanese achievement can potentially change this state of affairs. Hydrates are now extracted primarily by injecting hot water into the deposits to melt the ice, and drilling wells to lower the pressure. However, this method is inefficient and chaotic. It is more efficient to use a small burner in the well when controlling the size of the flame by limiting the oxygen supply. Methane is burned then in 10%, but the rest is usable. The use of microwaves of a certain frequency to heat hydrates is the most effective method because it allows warming hydrates, not surrounding rocks. Japan is currently the largest single importer of liquefied natural gas. Developing your own technology that will allow you to use the rich resources around the Japanese islands can be a solution to their energy problems. It is very likely that gas extracted from methane hydrates will cause more shuffle on the gas market than shale gas. It all depends on how profitable it is to exploit this type of deposits. According to the state-owned company JOGMEC, only Nankai Trough deposit, about 50 km from the coast of Honshu (1.1 trillion cubic metres trapped in gas hydrates) can satisfy gas consumption in Japan for 11 years, while all the estimated resources of methane hydrates in the waters surrounding the archipelago can satisfy the demand of the Land of the Rising Sun even for 100 years. In the work on the operation of this energy source, Tokyo has spent an average of USD 100 million a year on implementing a program for

Will gas hydrates cause ecological disaster?

the use of methane hydrates as a source of energy. Japanese got them for the first time in 2013. Drills were carried out near the south-eastern coast of the largest Japanese island of Honshu. They were stopped suddenly after six days due to clogging of the pipes and blocking the drill through the bottom sand. Over the next four years, Japanese scientists and engineers tested further, improved versions of the drill, until finally at the beginning of 2017 year, it was decided that the big research vessel "Chikyū", equipped with a drill that could drill holes 7 km in the seabed, would flow again near Atsumi peninsula and once again will try to get to the hydrates. The depth of the sea is about 1000 m here and the deposit is another 300 m below the bottom. At the beginning of June 2017 year, "Chikyū" drilled two holes in the bottom, and in each tested a different raw material extraction technology. In mid-June, the Japanese Agency for Natural Resources and Energy, which finances the project, apparently satisfied with the fact that both methods work and methane flows from the bottom, communicated that their goal is to launch commercial hydrocarbon extraction by private companies in the middle of the next decade. [5] Potential ecological hazards ◀◀ Accelerated temperature increase In connection with global warming, potential threats posed by clathrates are considered. Methane is a greenhouse gas whose heat-retaining capacity (greenhouse potential) is twenty times higher than in the case of carbon dioxide.. Increase of global temperatures could destabilize natural gas hydrates causing the release of trapped gases from the oceans and permafrost into the atmosphere and then increased greenhouse gases could --- cause further increase of global temperatures causing ongoing destabilization of gas hydrates on earth. It is suspected that the increase in methane concentration caused a rapid increase in temperature by 7 ° C in the late Paleocene 55 million years ago, which led to the extinction of many species of marine organisms. Paleobiologists from the Institute of Paleobiology PAS put forward the hypothesis that methane clathrates are responsible for the majority of rapid climate changes in the history of Earth.

Piotr Froń

◀◀ Poisoning with hydrogen sulfide Methane hydrate deposits are destabilizing with temperature growth. Initially, methane is released and goes to the ocean. The released methane starts to collect oxygen from water, according to the reaction: CH4 + 2CO2 → CO2 + 2H2. Bacteria are used in the oxidation of methane, using methane as a source of energy. They absorb 90% methane, which is released from hydrate deposits. Intensification methane emissions from the ocean floor means deoxidation oceanic depths, which in turn will be conducive to development sulfur bacteria and lead to the reproduction of conditions characteristic of the above issues. Consequently, this can lead to large amounts of hydrogen sulphide – which will consequently cause the local extinction of most of the marine fauna. ◀◀ Tsunamis Another potential danger is the violation of the stability of continental slopes. Geologists emphasize that many hydrate deposits are located in such a way that if they were removed, the undersea slopes would immediately lose their balance and go down. Such subsea landslides could be really huge and "run" gigantic tsunamis. As a result of an increase in temperature at the ocean floor, for example caused by volcanic eruptions, methane hydrate deposits are starting to heat up at a fairly rapid rate. They accelerate the penetration of temperature into the deposit, also facilitating the release of methane. This causes the formation of cracks and crevasses, and consequently can lead to the displacement of rock masses of the Earth's crust, resulting in landslides of many land areas.

As a result of a huge amount of sedimentary rocks breaking with the hydrates from the seabed, tsunamis can be generated, like in the case of a catastrophe, around 6100 BC. The disintegration of the clathrate deposits led to a mass transfer of rocks from the continental slope to the Norwegian Sea, with a volume estimated at 5,300 km³ by 800 km, which caused a massive wave. Its effects are still noticeable in the north of England . Among other things, the Bahamas are at risk because their east coast falls 5,000 m into the depths of the ocean, while the clathrats are the binder that keeps them. ◀◀ Explosions of gas aggregates However, it must be admitted that even tsunamis are not as frightening as other events accompanying the destabilization of the deposit. Methane is lighter than air, but mixed with drops of water has a higher density and accumulates near the surface. After mixing with air, at a concentration of 5-15%, methane forms an explosive mixture (similar to an air-fuel bomb). Ignition of the released methane would release energy equal to the explosion of 100 gigatonnes, which is 1000 times larger than the largest hydrogen bombs (100 megaton), but without radioactive fallout. Even a smaller deposit would be enough to cause problems on a continental scale. Despite these potential threats, as well as the very high costs of exploratory research, many countries – including Japan, the USA, China and India – are trying to get to methane hydrates. The hunger for energy encourages risky steps. Let’s hope that this hunger will not lead to a cataclysm on the world scale. Clatrates and the Bermuda Triangle

Fig.7 Potential scenario whereby dissociation of gas hydrates may give rise to subsea slope failure and massive methane gas release

There are many proposals for explaining the secrets of the Bermuda Triangle: from atmospheric anomalies, through trumpets and large tidal waves, black holes, and finally--- "arrivals from another planet". An attempt to explain this phenomenon may be presence of methane hydrates that form the bottom of the oceanic depths near Bermuda. There is a hypothesis that in the area of Bermuda Triangle, methane bubbles change density of water, which makes ships sink. Ship flowing over the break of the

SPRING / 2019

20 ocean floor from which methane would come out, would be serious danger. A huge wave would hit the bottom of the ship compressed gas. The density of water would drastically decrease and it would not be able to displace a huge mass ship. The ship would start sinking very quickly, thus falling to the bottom of the ocean or the sea. After a few minutes the sea would calm down and on its surface there would not be the slightest sign of the ship. The released methane can also be an explanation for disappearing aircraft. As a result of constant withdrawal methane to the atmosphere, after some time the gas would be found much higher than flying planes. Then the engines of any plane that would fly into a clean cloud methane, would immediately stop working – due to lack of oxygen. In turn, fires of aircraft in the air ---may be the result of the plane's contact with the cloud flammable methane. An explosion could cause the smallest spark, and thus the hot exhaust of aircraft engines. A powerful explosion would rip every plane to shreds. The remaining of the plane and its crew would immediately disappear under water without leaving--- the surface traces on the basis of which it is possible to reconstruct the cause of the disaster. The legend of the Bermu-

Will gas hydrates cause ecological disaster?

Piotr Froń

da Triangle enjoys a special interest among many people; starting with scientists and ending with ordinary people. A lot of books and press articles were created and radio and television programs also appear from time to time. All reports agree that something extraordinary is happening in this area. The future Now we are hoping that in five years time we will start acquiring gas from "burning ice" on an industrial scale. Such a scenario would mean a complete revolution: Japan from a country totally dependent on gas and oil imports would become self-sufficient in terms of energy. Other countries also count on similar benefits of "burning ice". But the road to the hydrometan revolution is full of obstacles. The two most important are ecological threats and – for now – no technologies that make profitable operation possible. Hopefully, all future activities will be planned and well thought out and most importantly – they will not cause any negative ecological effects and will allow us to live in peace in a new, better world.

References: [1] Molenda J.: Gaz ziemny. WNT, Warszawa 1996 [2] https://www.pet.hw.ac.uk in applying the enhanced coalbed methane recovery process; conference paper, 2005. [3] Lubaś J.: Doświadczalno-teoretyczne studium zjawiska powstawania i dysocjacji hydratów gazu ziemnego. Prace Instytutu Górnictwa Naftowego i Gazownictwa, nr 117; 2002 [4] E. Dendy Sloan ,Carolyn A. Koh and Amadeu K. Sum Gas Hydrate Stability and Sampling: The Future as Related to the Phase Diagram Center for Hydrate Research, Chemical Engineering Department, Colorado School of Mines, 1500 Illinois Street, Golden, CO 80401, USA; 2010 [5] https://wysokienapiecie.pl [6] Henriet, J.-P. and Mienert, J. (eds.), Gas Hydrates: Relevance to World Margin Stability and Climatic Change, Geological Society of London Special Publication 137; 1998 . [7] Max, D.: Natural Gas Hydrate in Oceanic and Permafrost Environments, Kluwer Academic Publishers, Dordrecht, Netherlands; 2000. [8] Sloan, E.D.: Clathrate Hydrates of Natural Gases, Marcel Dekker Inc., New York; 1998 [9] Kusche L.D.: Trójkąt Bermudzki. Zagadka rozwiązana., Warszawa 1993 [10] Waite, W.F.; Santamarina, J.C.; Koh, C.A.; Sloan, E.D., Jr.; Grozic, J.L.H.; Hester, K.C.; Howard, J.; Mahajan, D.; Priest, J.; Rydzy, M.; Seol, Y.; Winters, W.J.; Yun, T.S.; Inter-laboratory comparison of wave velocity measurements. In Proceedings of the Seventh International Conference on Gas Hydrates, Edinburgh, UK, July 2011. [11] Anna Rabajczyk Zagrożenia dla środowiska wynmikające z eksploatacji klatratów metanu – studium oceny oddziaływania na środowisko Polska Akademia Nauk – Oddział w Krakowie; 2009 from http://journals. bg.agh.edu.pl [12] Luff R., Wallmann K.: Fluid flow, methane fluxes, carbonate precipitation and biogeochemical turnover in gas hydratebearing sediments at Hydrate Ridge, Cascadia Margin: numerical modeling and mass balances. Geochim. Cosmochim. Acta 67(18), 2003 13. Beerling D.J., Lomas M.R., Gröcke D.R.: On the nature of methane gas-hydrate dissociation during the Toarcian and Aptian oceanic anoxic events. American Journal of Science; 2002

SPRING / 2019

¡¡

Negotiation Game for Young Professionals – Model OPEC

Monika Walczuk

Negotiation Game for Young Professionals – Model OPEC Monika Walczuk

For an ambitious student who is more interested in commercial and business than the technical side of oil and gas industry, there is no better way to try out knowledge and negotiation skills than participating in Model OPEC. This is an unique opportunity to become for a while a representative of an oil-exporting country in negotiations based on real scenarios and face the problems of global decision makers. Last year the 1st Polish simulation of OPEC Meeting was organized by WPC Young Professionals Poland (part of WPC Polish National Committee). Event was dedicated to Young Professionals. OPEC – Organization of Petroleum Exporting Countries, founded in 1960, is an international cartel that controls global supply of crude oil. It was responsible for i.a. oil shock in 1973 that resulted in economic crisis. Nowadays there are 14 member countries (among others, Saudi Arabia, Iran, Iraq, Venezuela, Kuwait). Organization’s main goal is to maintain high level of oil prices through setting production quotas. The decisions are made during meetings of OPEC members’ representatives. Model OPEC 2018 was a simulation of such meeting. The event was conducted in the form of 12-round game – simulation of the oil market in years 2016-2076 (12x5 years). Participants, grouped in 2-person teams, played the role of representatives of member countries of OPEC. In each round teams negotiated with each other on quotas of crude oil production and settled amount of production, oil stocks and some economic variables. The results of their actions influenced market and country situation in the next round. The game mechanism has been designed to reflect a real petroleum market, market power of OPEC and specific situation of each country. Hence, the participants had to take into consideration not only market factors but also an economic situation of their country. Moreover,

OPEC’s market power is not visible if there is no cooperation between the teams. However, there was also an element of competition between the teams – each one had to retain as much profits from oil production as it was possible. This factor was used to choose a winner. Model OPEC 2018 was held during the Session of Youth and Innovation as part of the 12th Polish Congress of Oil and Gas Industry Professionals (https://www.sitpnig.pl/12-pknig-en) – one of the largest oil conferences in Poland. The participants of the simulation also had the opportunity to participate in paper sessions and workshops during the Congress on May 17-18th in Cracow, Poland. There were 11 participating teams – 10 from Poland and 1 from Iran, representing student scientific associations focused on energy. They came from not only technical universities (AGH University of Science and Technology Krakow) but also financial schools (Warsaw School of Economics) and legal faculties ( Jagiellonian University). Participants were selected on the basis of their answers to questions about future of oil industry that were given to contestants 2 months in advance of the event. During the simulation they had to recognize results for the oil market of such events as: “OPEC+” countries changeable policy, overestimates of lithium reserves in Bolivia, pirates attacks in Middle East, depletion of Norwegian Oil Shelf reserves, and many others. Event turned out to be a huge success, a lot of fun and a chance to gain knowledge. That is the reason why organizers have decided to arrange a second edition of Model OPEC during 10th East meets West International Student Petroleum Congress & Career Expo in Cracow. Do not hesitate to join, register as a participant to check your market intelligence and negotiation skills in April 2019!

SPRING / 2019 LOTO-046_Rafineria_176x250 EN v01.indd 2

20.03.2018 15:20

¡¡

Optimization of Energy Consumption in Distillation Columns

Optimization of Energy Consumption in Distillation Columns Tita Oxa Anggrea, Titania Nur Bethiana, Totok R. Biyanto

** Institut Teknologi Sepuluh Nopember ÞÞ Indonesia

Distillation is the most process study in industrial application and academic. It is due to the huge energy consumption in this operation unit at evaporation and condensation parts. Nowadays, there are around 40 000 distillation columns to serve the separation and refining process. It expense about 8 million USD in capital expenditure. Distillation column consume amount 3% of the total national energy consumption in US, therefore, most of the research on distillation columns aims to reduce the energy consumption. Generally, more than half of the heat in the columns is used for the heating process in distillation column. The addition of heat energy at the columns re-boiler will be used to evaporate the liquid and push it to the top of the columns. Meanwhile, the vapor will be condensed again at condenser. It’s necessary to reduce energy consumption such as using forward-backward heat integration configurations and Fully Thermally Couple (FTC) or Petlyuk distillation column.FTC could save the energy consumption significantly, around 10-40 % compare to the conventional columns configuration. FTC was introduced by Wright in 1947. Further developments in theoretical equations have been performed by Petlyuk in the 1960s. According to the petlyuk’s theory, the inefficiency can be mitigated by combining the re-boiler and condenser at pre-fractionators with the main column. FTC column use only one condenser and one re-boiler to separate

more than two component. Therefore, it could save the energy consumption and the capital cost due to reducing the number of heat exchangers. Halvorsen & Skogestad proved that FTC could reduce energy consumption in multi component separation process. In design stage of petlyuk column, the approach that can be used namely as sloppy sequence distillation column, The results of the design stage will be used to determine the parameter of petlyuk configuration i.e. the minimum number of trays by utilize the VLE method to determine the minimum energy consumption. In this paper, two methods of Petlyuk configuration design were compared i.e., Vapor-Liquid Equilibrium (VLE) approach and the proposed method that incorporating Duelist Algorithm Optimization (DAO) method and VLE approach. Optimization was performed by considering techno economic point of view. The objective is minimizing the total annual cost for operating period over 10 years. Optimization results show that the maximum saving about 34,38% was obtained the actual number of trays at pre-fractionators and at main columns are 3 and 109, respectively. 1. Introduction Distillation process is the most important process in industries. Distillation process is a technique conducted in order to separate mixture consists of two or more components. Separation process in distillation columns can be occurred due to heat added into re-boiler causes vapor flow produced in re-boiler goes through column and contacted with feed and condenser that flow down to the column. Distillation columns consume large ener-

Tita Oxa Anggrea, Titania Nur Bethiana, Totok R. Biyanto

gy, about 50% of industrial energy needs used as cooling and heating process in distillation columns. Distillation columns consume large energy in order to convert fluid mixture from liquid phase to vapor or gas phase and re-convert the vapor or gas phase to the liquid phase in condenser. In general, more than 50% heat distributed in a plant used to supply needs from re-boiler for heating process or evaporation stage in distillation columns (Kunesh J, 1995). Humphrey (Humphrey, 1995)estimated that there are about 40,000 distillation columns in USA and 90% used in separation and refinement process with cost around USD 8 billion. Based on data from Mix et al (Mix T, 1978), Soave & Feliu (Soave G, 2002), it is obtained that calculation of distillation columns system consume about 3% from total energy consumption in USA, in which this value is equal with 2.87x1018 J (2.87 million TJ) per year, or it is equal with power consumption of 91 GW, as well as 54 million tons of crude oil. Thus, most of studies conducted to the distillation columns intended to reduce energy consumption. Study to the distillation columns design or operation is always developed due to it has great effect if it is seen from economic perspective. Study about short cut design method can be used to determine minimum energy needs in separation process of distillation columns. This method aims to analyze stream application point with tray in distillation columns and determine minimum energy needed from tray analysis.(Avami, Marquardt, Saboohi, & Kraemer, 2012). However, the configuration design of distillation columns still in conventional design. Then, in the next study is conducted by(Uwitonze, Han, Kim, & Hwang, 2014), distillation columns design using thermally coupled distillation column proven to be able in saving energy and capital cost compared to traditional or conventional distillation columns. This design has large degrees of freedom. Number of tray in each section column can determine basic operation and product specification. Procedure in this method can be separated from the role of interlinking tray simulated in rigorous. The equation used in this method is Fenske-Underwood-Gilliland equation to result actual reflux ratio, number of tray, and placement for every feed tray. This research completed in which the distillation

columns designed with fully thermally (Uwitonze, Hwang, & Lee, 2015). Coupled distillation column should be considered the benefit in energy and capital cost aspect as well as compared to conventional distillation columns. Theory used in this method mainly based on flow rate (liquid and vapor) in each section column and component equilibrium constants (K-value) uses FUG equation. Study in designing distillation columns keeps developed to the multicomponents separation process in which it needs an approach to determine geometry of separation process. Geometry of separation process determined by number of tray or stage used in feed tray location, tray for distillate product output, tray for side product output and tray for bottom product output (Adiche & Aissa, 2016). The newest study that been conducted is about implementation of shortcut method and rigorous simulation in non -ideal condition that can be applied in industries. This method enable to determine mass balance and thermodynamic from separation process and from minimum reflux value. Based on minimum reflux value, preliminary design can be determined and applied to be initiated by using a rigorous simulation. From the result of rigorous simulation, it is showed that this method gives better approach than FUG shortcut method. This method gives specification to build production segment for each output in order to be suitable with the desired composition as well as it is considered about its energy cost and efficiency (Worms, Meyer, Rouzineau, & Brehelin, 2017). The most important thing in designing stage is minimum number of tray. Petlyuk optimized with Duelist Algorithm (DA) has objective function to minimize Total Annual Cost (TAC). It’s about 34.38% maximum saving produced by actual number of tray in prefractionator and in main columns of 3 and 109.(Biyanto, et al., 2016). Based on many studies above, configuration design of distillation columns mostly studied is Thermally Coupled Distillation (TCD) that been developed since 20 years ago. It is due to the configuration design has been proven to save energy significantly, which is about 10-40% from conventional column design (Biyanto, et al., 2016). Thermally Coupled Distillation (TCD), first, found by Wright in 1947 (Wright, 1947)and proven theoretically by Petlyuk

SPRING / 2019

Optimization of Energy Consumption in Distillation Columns

in 1965 (Petlyuk FB, 1965). Based on theories formulated by Petlyuk, the inefficiency can be solved by removing heat exchanger and create thermal coupling between columns. The use of non-conventional distillation columns, such suggested by Petlyuk, will be able to save significant energy consumption and, therefore, significantly reduce purchasing cost, installation cost, operating cost and reduce energy consumption up to 30% than conventional distillation columns (Salinas, 2014). The study conducted by Halvorsen and Skogestad has proven that minimum energy consumption in a mixture with ideal multi-components is always obtained by using Fully Thermally Coupled (FTC) configuration (Halvorsen IJ, 2003). Thus, in order to answer question about energy needs and high cost in distillation process, then it needs Petlyuk distillation column design and optimization by using appropriate method. Study that would be conducted in this paper was distillation column design optimization by using Petlyuk configuration includes pre-design consists of configuration fragmented into few sequential sets of distillation columns. The result obtained in pre-design phase would be used to determine initial design parameter of Petlyuk configuration. Four design methods used in this study are: first, by using Vapor-Liquid Equilibrium (VLE) method; and second, by using Vapor-Liquid Equilibrium (VLE) method optimized with Duelist Algorithm. 2. Theory 2.1 Distillation columns Distillation process is a separation process which often used and substantial in large or small scale industries. Distillation column has a function to separate component from a mixture. This separation occurs based on volatility value of the component on the mixture. Component that has greater relative volatility will be easier to separated. The fluid to be processed is usually called feed and fed into the tray, which then called feed tray. Feed tray divides the column into two parts, they are rectifying section and stripping section. On each tray, vapor and liquid contact based on mass and energy balance.

Mass balance F = D+B Component mass balance F.Xf = D.XD + B.XB Energy balance F.hf – D.hD – B.hB + Qr – Qc = 0

(1) (2) (3)

Where:

F : Feed rate (kg/s) D : Distillate rate (kg/s) B : Bottom rate (kg/s)

Fig.1 Schematic diagram of distillation column There are several trays inside of distillation column, each tray have two lines on each side which are called down comers. Liquid fluid falls through down comers from one tray to another tray. Trays in the distillation columns have some functioning hole that works for steam flow rate. Steam flows lead to the top column and forced to pass the liquid through the aperture on each tray. When the hot vapor passed through the liquid from one tray to another tray, the vapor provides heat transfer to liquid so that some of the vapor is condensed and add fluid to the tray. The rest of the vapor than comes out through the top column and cooled by condenser. Some of the liquid produced re-enter to the top column and it is called reflux, while the others were expelled from system and it is called distillate or top product. Heat energy is supplied to the re-boiler for produces vapor. Vapor re-inserted into the system through the bottom column. Liquid output from re-boiler is called bottom product.

Xd : Composition in distillate (Mole) Xf: Composition in feed (Mole)

2.3 Design of Petlyuk Distillation Column In the distillation column design, there are two important things that required to be considered i.e. the number of minimal plate on columns for the component separation process on total reflux condition, and the number of minimum reflux that needed for the separation process to produce the desired product. An empirical method that normally used to calculate those two things is Fenske-Underwood-Gilliland (FUG) Method. The distillation columns design using this methods consist of the structure making of equivalent column distillation with a Petlyuk columns.

XB : Composition in bottom (Mole) hf : Liquid enthalpy on feed ( Joule/kg) hd : Liquid enthalpy on distillate ( Joule/kg) hb : Liquid enthalpy on bottom ( Joule/kg) Qr : Heat given by re-boiler ( Joule/kg) Qc: Heat given by condenser ( Joule/kg)

However, further study of control system on this configuration that has a lot of degree of freedom is necessary. The main reason of this petlyuk configuration making is to avoid thermodynamic losses on flow mixing process with different feed tray. The equation of mass and energy balance on each tray using the petlyuk configuration can be written as follow: (4)

Fig.3 The arrangement of columns which equivalent with Petlyuk configuration

For each column designed with short cut method based on fenske-underwood-gilliland (FUG) theory, through equation as follow:

Mathematic model of distillation column generally written on mass, component mass, and energy balance equation. Mathematical model was derived from physical model system. Mathematic model of distillation column at steady state condition can be written as follows: 2.2 Petlyuk Distillation Column The first petlyuk column configuration was created in 1965. Petlyuk configurations consist of a pre-fractionators and main columns. Based on the theory, this configuration design could reduce the energy usage for one re-boiler and one condenser.

Tita Oxa Anggrea, Titania Nur Bethiana, Totok R. Biyanto

(5) Where:

If we have found the specification value of the fraction of light key (rLK) and heavy key (rHK), it will be able to find the minimum number of stage (Sm). Fig.2 Petlyuk Column Configuration

(6)

SPRING / 2019

Optimization of Energy Consumption in Distillation Columns

The calculation of the minimum reflux ratio is sought by using the Underwood equation. (7)

The Kirk bride equation is based on comparison of the number of stages on the rectifying (Noverfeed) section with the stripping section (Nunderfeed). (12)

Where q is the value of the quality of the vapor fraction, q will be 0 if the mixture of the components of vapor phase is perfect and is worth 1 if the phase is perfectly liquid, and φ is the roots of the Underwood equation to be searched and put in equation (8) to find the minimum value of the vapor flow (Vmin) on the top product. (8) After obtaining the value of the minimum stage number (Sm) and the value of the minimum reflux rate (Rmin), the next is to find the value of the actual total number (S) and the actual value of reflux (R) by using the formula developed by Eduljee based on the correlation Gilliland, as in the following 9 equation. (9)

In 1958, a Winn equation developed from the Fenske equation was to find the minimum number of stages. When the value of relative volatility varies considerably in a column with a cascade configuration and consists of several stages, the Fenske equation tends to be less accurate in guessing the number of stages that will result in too many stages. Under varying volatility conditions, the Winn equation will yield more accurate results.

d. The calculation of the cost of the design of the distillation column consists of two things that need to be taken into account the cost of purchase and installation costs.

Therefore, the formula Kirk bride for feed tray position (NF) is:

NF = 1 + Noverfeed + Nunderfeed

(13)

The design of a distillation column can be done by decreasing the Vapor-Liquid Equilibrium (VLE) equation. In this method will be calculated bubble point and dew point using equation of Antoine temperature and equation of UNIQUAC (Universal Quasi Chemical). The method of bubble point and dew point calculation is used to determine product quality in both distillate and bottom flow. Furthermore, calculations of the amount of reflux flow and the calorific value of the condenser and re-boiler are required. 2.4 Total Annual Cost (TAC) 2.4.1 Capital Cost The design costing steps in the distillation column according to Douglas as follow (Douglas, 1980): a. It is assumed that the column diameter is 1.5 meters or 5 ft. b. Tray used is plate type with material made of stainless-steel. c. The M&S index is obtained through the following figure:

(10)

Purchased Cost,

) (14)

While the formula of purchase cost calculation is as follows: Installed Cost,

(16) Where, Qcond is the amount of heat needed by the condenser and Qreb is the amount of heat required by the re-boiler. 3. Methods

Where: D = Diameter (ft) H = Height (ft) Fc = Fm.Fp

H Fc)

(15)

Where : D = Diameter (ft) H = Height (ft) Fc= Fs + Ft + Fm 2.4.2 Operational Cost Operational cost calculation is obtained from the calculation of the amount of condenser energy and the re-boiler needed to produce a product that is suitable for the purpose to be achieved. Here is an equation for calculating operational costs according to Jimenez (Gutierrez, 2010):

3.1 Design of Distillation Column Petlyuk The collection of data from several distillation columns is done by taking data from the Process Flow Diagram (PFD) of the distillation column used as the object in this final project, i.e. de-ethanizer column and de-propanizer column having conventional configuration. These data are used as reference in designing distillation columns that have Petlyuk configuration. The data of fluid properties and the feed compositions are tabulated in appendix. The short cut method is conducted by designing the Petlyuk distillation column using the HYSYS V8.8 ASPEN software. The simulation process applied by short cut column on HYSYS is based on mathematical calculation of mass and energy balance to determine the composition of interconnection stream and Fenske-Underwood-Gilliland equation to determine minimum tray, minimum reflux, and actual tray number. The arrangement of Petlyuk distillation column configuration using short cut column can be shown in Figure 5 below.

Fig.5 Design of Petlyuk distillation column by using short cut column

(11) Where: x = The mole fraction of the component is in liquid form. θLK=Relative volatilities of light key component (β and θ constantare an empirical constant that has been determined by the range of pressure and temperature in the column).

Tita Oxa Anggrea, Titania Nur Bethiana, Totok R. Biyanto

Fig.4 The M & S factor on the cost of designing the distillation columns (Douglas, 1980)

SPRING / 2019

Optimization of Energy Consumption in Distillation Columns

Composition of components A, B, and C (ethane, propane, and n-butane) targeted at distillate product, side product and bottom product are adjusted to PFD, ethane = 0.05775 (5.775%), propane = 0.94925 (94.925%), and n-butane = 0.38469 (38.469%). While the composition of components A, B, and C in interconnection stream 1 (Distillate 3) and interconnection stream 2 (Bottoms 3) is obtained through mass and energy balance equation as shown in Table 1. Component

Interconnection

Stream 1

Stream 2

Ethane

0.0561

0.0001

Propane

0.0298

0.3315

n-Butane

0.0001

0.2544

Tab.1 Composition of interconnection stream 3.2 Optimization of Petlyuk Distillation Column Design Using Duelist Algorithm The stochastic algorithm optimization is applied to the VLE method to optimize the design parameters of the distillation column design by the VLE method. The optimization technique used is Duelist Algorithm (DA). The objective function in this optimization is to minimize total annual cost (capital and operational cost) of the distillation column design:

J min = Total Annual Cost

(17)

3.2.1 Duelist Algorithm DA is a new optimization algorithm inspired by human fighting and learning capabilities. states that the individual in DA are known as duelist. All those duelists would fight one by one to determine the duelist’s classification either as champions, winners or losers. There is a probability that the weak one would be lucky enough to win. Therefore, there are two different treatment implemented to the duelist based on their classification to improve each duelist which involved innovation and learning. (Biyanto 2016).

4. Results and Discussions 4.1 Design Stage. Petlyuk distillation column is designed using 2 methods, namely VLE and VLE optimization using Duelist Algorithm. The design of the distillation column in the VLE method is based on the equilibrium of steam and liquid. The first thing determined on the VLE method is the saturated pressure (Psat) and the saturated temperature (Tsat). The Psat and Tsat values of each component are influenced by the Antoine constants (A, B, and C) on each component. The value of Psat and Tsat is used to determine bubble point and dew point. Bubble point is the temperature at which the liquid begins to form a steam wave according to the applied pressure, while the dew point is the temperature at which the vapor/gas begins to condense according to the applied pressure. The calculation of the enthalpy value is also performed to determine the enthalpy of the vapor and liquid. The UNIQUAC equation is used in this method to determine the composition of each stream. The binary mixtures constant value of the three main components (ethane, propane, and n-butane) shows the composite interactions with each other. The stochastic algorithm optimization is applied to the VLE method to optimize the design parameters of the distillation column design by the VLE method. The objective function used in this optimization is to minimize the total annual cost (sum of capital cost and operational cost) from the design of the distillation column design. The objective function value of DA during iteration is shown in Figure 6 4.2 Simulation of Operational Stage Column design parameters such as minimum number of trays, actual number of trays, and optimal feed stage from design result of Petlyuk distillation column from short cut method applied to validation of Petlyuk distillation column design using rigorous method under HYSYS V8.8 software. Validation results show the value of Q condenser and Q re-boiler required on the design of Petlyuk distillation columns. 4.3 Capital Cost Capital cost can be obtained through equations 14

Tita Oxa Anggrea, Titania Nur Bethiana, Totok R. Biyanto

Methods

Parameter

Value

VLE

Number of tray pre-fractionator

Number of tray main column

Optimal feed location of pre-fractionator

Optimal feed location of main column

9 and 35

VLE-DA

Number of tray pre-fractionator

Number of tray main column

Optimal feed location of pre-fractionator

Optimal feed location of main column Optimal side stream location

and 15 which results can be shown as in Table 5 below. Table 4.Capital cost of each method The decrease of capital cost value in Table 5 can be clearly seen in the bar chart shown in Figure 7 below. Methods

Purchased Cost

Installed Cost

Capital Cost

(million $)

Conventional

2,47

0,94

3,41

15 and 44

VLE

0,96

0,285

1,245

VLE Optimization

0,891

0,264

1,155

using DA

Tab.2 Result from designing stage Tab.4 Capital cost of each method

4 3.5

Fig.6 Results of VLE optimization using Duelist Algorithm (DA)

`~éáí~ä=`çëí(million $/year)

Conventional

3 2.5 2 1.5 1 0.5 0

Methods

Heat flow at

Heat flow

condenser

at re-boiler

(kW)

VLE

31220

1204

VLE + DA

30640

1144

Tab.3 The result of operational stage

Optimization DA

VLE

Methods

Fig.7 Capital cost

4.4 Operational Cost The operational cost in the design of the distillation column can be determined through equation 16, from the equation the parameters affecting the operational costs of the distillation columns are Q condenser and Q re-boiler. The operational costs of designing the Petlyuk distillation columns of each method are as follows.

SPRING / 2019

Optimization of Energy Consumption in Distillation Columns

Re-boiler

Condenser

Operational

Cost (million

Cost(million

$/year)

Conventional

166,99

13,01

180

VLE

37,75

63,89

101,64

VLE Optimization using DA

35,87

62,7

98,57

Tab.3 The result of operational stage The decrease of operational cost value in Table 6 can be clearly seen in the bar chart shown in Figure 8 below. 200

million$

150 DA

VLE

100

Methods

Fig.8 Operational cost 4.5 Total Annual Cost The calculation of Total Annual Cost (TAC) of each method used in the design designs of the Petlyuk distillation column can be shown in Table 6 below.

◀◀ Optimization of the VLE method using duelist algorithm can optimize the number of tray in pre-fractionators and main columns, hence minimizing the energy required of re-boiler and condenser, capital cost, operational cost and improve product quality were obtained. ◀◀ The design of the Petlyuk distillation columns using the optimized VLE using the DA provides the Total Annual Cost value of 45.76% per year compare with the conventional columns.

22.6

Heat Capacity

kJ/kg 0C

2.51

Heat Flow

-178424

Mass Density

kg/m3

56.59

H2S

0.00

Vapour Fraction Composition

CO2

2.74

Nitrogen

1.80

Methane

83.02

Ethane

5.55

Propane

4.04

0.21

102,885

n-hexane

0.12

99,725

n-heptane

0.03

n-octane

0.01

n-nonane

0.00

n-decane

0.00

n-C11

0.00

n-C12

0.00

n-C13

0.00

200

Fig.9 Total Annual cost

42.2

Barg

0.34

of each method

Methods

Pressure

n-pentane

VLE Optimization using DA

Temperature

1.10

VLE

150667

i-pentane

183,41

VLE

Feed

kg/hr

1.03

Conventional

100

Unit

Flow rate

i-butane

Total Annual Cost(million$)

150

Specification

n-butane

Metode

Tab.6 Total Annual Cost

million$

Design optimization of Petlyuk distillation columns using duelist algorithm have been performed, and the conclusions as follow:

Appendix

i DA

References:

5. Conclusions Methods

Tita Oxa Anggrea, Titania Nur Bethiana, Totok R. Biyanto

n-C14

0.00

n-C15

0.00

n-C16

0.00

n-C17

0.00

n-C18

0.00

n-C19

0.00

n-C11 +*

0.00

PVT-2 C20*

0.00

PVT-4 C20*

0.00

BHS-1*

0.00

H2O

0.00

Total

100.00

[1] Adiche, C., & Aissa, B. A. (2016). A Generalized Approach for The Conceptual Design of Distillation Columns with Complex Conffiguration. Chemical Engineering Research and Design, 150-170. [2] Avami, A., Marquardt, W., Saboohi, Y., & Kraemer, K. (2012). Shorcut Design of Reactive Distillation Collumns. Chemical Engineering Science, 166-177. [3] Biyanto T R, Fibrianto H Y, Nugroho G, Listijorini E, Budiati T and Huda H 2016 Duelist algorithm: an algorithm inspired by how duelist improve their capabilities in a duel, Int. Conf. in Swarm Intelligence: Advances in Swarm Intelligence 2 pp 39-47 Biyanto, T. R., Rahman, J. A., Sarwono, Roekmono, H, N. L., Abdurrakhman, A., et al. (2016). Techno Economic Optimization of Petlyuk Distillation Column Design Using Duelist Algorithm. Engineering Physics Department. Faculty of Industrial Technology. Institut Teknologi Sepuluh Nopember, 520-527. [4] Douglas, K. (1980). Summary of Cost Correlation. New York: McGraw-Hill. [5] Gutierrez, A. J. (2010). Optimum Design of Petlyuk and Divided-Wall Distillation Systems Using A Shortcut Model. Instituto Tecnologico de Celaya, Departamento de Ingeniería Quimica, Celaya, Gto. 38010, Mexico. [6] Halvorsen IJ, S. S. (2003). Minimum Energy Consumption in Multicomponent Distillation. 1. V-min Diagram for a Two-Product Column. Industrial & Engineering Chemistry Research. Industrial & Engineering Chemistry Research, 42(3): 596-604. [7] Humphrey, J. (1995). Separation processes: playing a critical role. Chemical Engineering. Chemical Engineering Progress, 91(10):43-54. [8] Kunesh J, K. H. (1995). Distillation: Still Towering Over Other Options. Chemical Engineering Process, 91(10). [9] Mix T, D. J. (1978). Energy Conservation in Distillation. Chemical Engineering Process, 74(4). [10] Petlyuk FB, P. V. (1965). Thermodynamically Optimal Method for Separating Multicomponent Mixtures. International Chemical Engineering, 5(3): 555-561. [11] Salinas, G. Z. (2014). Modified Method to Improve The Design of Petlyuk Distillation Columns. Chemistry Central Journal, 8: 41. Soave G, F. J. (2002). Saving Energy in Distillation Towers by Feed Splitting. Applied Thermal Engineering, 28(80: 889. [12] Uwitonze, H., Han, S., Kim, S., & Hwang, K. S. (2014). Structural Design of Fully Thermally Coupled Distillation Column Using Approximate Group Methods. Chemical Engineering and Processing, 155-167. [13] Uwitonze, H., Hwang, K. S., & Lee, I. (2015). A New Design Method and Operation of Fully Thermally Coupled Distillation Column. Chemical Engineering and Processing Process Intensification, 47-58. [14] Worms, G., Meyer, M., Rouzineau, D., & Brehelin, M. (2017). The Production Zone Method: A NonIdeal Shortcut Method for The Design of Distillation Columns. Separation and Purification Technology, 404-423. [15] Wright, R. (1947). Fractination Apparatus. US Patent.

(TAC)

SPRING / 2019

Optimization of Energy Consumption in Distillation Columns

Yohanes Nuwara, Michael Aria Santoso

¡¡

Seismic Attribute Analysis Aa A Tool For Structural Interpretation

And Mapping Possibility Of Hydrocarbon Accumulation In Unconventional Oil Reservoir Yohanes Nuwara, Michael Aria Santoso

** Bandung Institute of Technology ÞÞ Indonesia In recent days, the accelerating quest for oil for a nearly primary source for energy requires innovation to maintain the sustainability of oil and gas production worldwide. In many cases, the production is announced to an end when the reservoir is declared not economical due to decreasing recoverable oil. Enhanced Oil Recovery is a strategy to increase the oil recovery factor so that the exploration and production carries on by planning another new potential unconventional reservoir and stimulating the well using many options, such as water and CO2 flooding. As a target of EOR, any play in the field can be the reservoir for oil stimulation. However, the challenge is to comprehend the structure of our target that is previously known little. Granite wash is a potential reservoir target for EOR in our research in South Sumatera basin, Indonesia. This granite wash potentially contain hydrocarbon which has not been recoverable in unknown structure. Therefore, structural interpretation is required to characterize the unknown structure of granite wash. In this research, we interpret the faults and horizon from post-processed 3D seismic data volume in the field through sets of processes including well-toseismic tie. From this

process, we run seismic attributes that consist of structural dip and azimuth, and structural curvature to verify the structural interpretation. In the end, we extract amplitude attributes on our granite wash horizon. As a result, seismic attributes provides more meaningful insights that go beyond looking through the unknown prospect. Consistent dip and curvature attribute provides insight about the fault orientations that match with regional geology data, and amplitude attributes provides the ability to map potential accumulation of hydrocarbon in the granite wash structure. In conclusion, seismic attribute analysis gives an ability to unlock the structure and possible hydrocarbon accumulation in any unconventional reservoir. Keywords: granite wash, enhanced oil recovery, reservoir characterization, seismic attribute, structural interpretation Introduction The global oil discovery began to decline after the year of 2012. One of the reasons are that explorers are finding less oil resources per field. More problems arrive with the demand of global markets for oil consumption which are increasing and expected to accelerate to 101.3 millions barrels per day (BPD) at the end of 2018. While the world production or supply for oil is not capable to ba-

SPRING / 2019

Seismic Attribute Analysis Aa A Tool For Structural Interpretation

lance the number, this could lead to a tightening in supplies. While there have been some notable successes throughout the year for oil production, the circumstances that the low discovered volumes on global level represent a serious threat to the supply levels for the next years down the road. Therefore, Enhance Oil Recovery could be a solution. Enhanced Oil Recovery (EOR) is one of the method to recover all the crude oil possible from an oil reservoir. There are a number of techniques currently used for Enhanced Oil Recovery, each of which has varying implications on cost, efficiency, and safety. Oil production can be broken down into three phases: Primary, Secondary and Tertiary. Each phase requires different amount of technology to be used to recover the oil. This, in turn, increases the cost of well production. With increasing global demand and oil prices the costs associated with more expensive techniques are fast becoming less of an issue. Primary recovery relies on naturally occurring pressure within the oil reservoir to drive oil to the surface by extracting a small amount of oil compared to the maximum potential of reservoir. When primary recovery cannot be done anymore because the natural pressure within the reservoir is not sufficient to push the oil to emerge from the depth, then secondary oil recovery or EOR is needed. The technique works by injecting water (waterflooding) or by pumping compressed gasses into the reservoir. After the secondary oil recovery is not working anymore either. Then the tertiary oil recovery can be applied. It works by chemical injection into the reservoir. In this paper we present our analysis about how seismic attributes give us insight about how we locate the target of reservoir for EOR development in a depleted oil field in one of the sedimentary basins in Indonesia located in South Sumatera. It is known that the recovery factor of this sedimentary basin is 17.5%, meaning that 82.5% of total oil produced still remain inside. Additionally, the field has already produced oil for decades and its rate of production is beginning to fall, hence the need of EOR is inevitable. Our target is a layer of Granite Wash rock that potentially contain hydrocarbon because of the cracks which give this rock a perme-

able characteristic. Granite wash is relatively new in prospects since not so many study have discussed about the prospectivity. It has been found that there is some fault structure seen in the layer. We present a structural interpretation of the horizon and faults of the layer from post-processed 3D seismic data volume in the field through sets of processes. Including well-to-seismic tie to know where the Granite Wash horizon is. There are 5 seismic attributes we used to identify the structure and the hydrocarbon potential, each of them will give different kind of results. This results will help us to fully understand the character of our target of reservoir. Regional Geology South Sumatera basin has five sub-basins, namely Jambi, North Palembang, Central Palembang, South Palembang, South Palembang, and Bandar Jaya. Some parts of the structures of South Sumatera Basin are located on a basement high named Pendopo High. According to Pulonggono (1992), the tectonic process occuring in South Sumatera basin consists of three major phases, namely the first compressional, tensional, and second compressional phase. Occured during early Jurassic to Cretaceous, the first compressional phase created dextral stikeslip Lematang, Kepayang, Saka, and Pantai Selatan fault oriented WNW – ESE and granite intrusion. This phase also created Musi lineament and same lineaments with N–S orientation. The tensional phase happened from late Cretaceous to early Tertiary which produced predominantly normal faults oriented N – S and WNW – SE. In this phase, the sediment filled in the basin and overlay the basement in the basin. This sediment made the proportion of Lahat Formation (LAF). The third phase is second compressional which happened from Miocene to recent (Plio– pleistocene). During this times, the basin was uplifted and followed by deposition of clastics which created Talang Akar (TAF), Baturaja (BRF), Gumai (GUF), Air Benakat (ABF), and Muara Enim Formation. ABF and Muara Enim deposits were then eroded and overlied by Kasai Formation. Oriented at NW – SE direction, the compression folded the basin, which is called basin inversion tectonics. Also during this phase, faults were reactivated. Figure 1 clearly shows the whole tectonic process started

Yohanes Nuwara, Michael Aria Santoso

from Baturaja deposition in early Miocene and then inverted after late Miocene. According to Bishop (2001) and Margaretha et al (2006), the structural geology of South Sumatera basin is generalized into three common structures, related to the whole tectonic processes. The area is dominated by NW – SE oriented thrust faults and anticlines caused by compressional subduction directed perpendicular to the faults in Miocene to Pliocene, NW – SE oriented normal faults reactivated by Setiti–Tembesi dextral strike-slip fault in the same era, and older NNE – SSW oriented horst and graben caused by tensional phase during Eocene to Oligocene. There are two major strike-slip faults, namely WNW – ESE dextral Lematang Fault and SW – NE sinistral Setiti–Tembesi Fault. The stratigraphy of South Sumatera Basin is divided into three categories of deposits. The first or bottom deposit is categorized as syn-rift. Syn-rift deposit (Lemat Formation) was created during Eocene to Oligocene and composed of alluvial, fluvial, lacustrine, and deltaic sediments in Lematang trough

Fig.1 Tectonic process of South Sumatera Basin from early Miocene to present day

Fig.2 Location and structure map of South Sumatera Basin

SPRING / 2019

Seismic Attribute Analysis Aa A Tool For Structural Interpretation

and Limau graben. Second deposit is post-rift. The post-rift sediment (Lahat and Talang Akar) was formed during Oligocene to early Miocene and composed of gradually differentiated fluvio-deltaic sediments in braided system (Lower TAF member) to shallow marine (Upper TAF member). The uppermost deposit is syn-inversion, which was created during middle Miocene to present day as the result of late basin inversion. The deposits are predominantly tuff volcanic sediments of Kasai Formation. Figure 3 clearly represents the stratigraphy of South Sumatera basin. The granite wash deposit is found overlying the pre -Tertiary metamorphic and granite basement. Granite wash is composed of reworked conglomerate basalt and overlain by sandstone, amber fragments, conglomerate, and volcanic rock fragments from Lahat or Lemat Formation. This deposit has been renounced to contain certain composition of oil in Lemat or Lahat Formation. Methodology The work is divided into 6 processes; well-to-seismic tie, fault picking, horizon picking, seismic attribute analysis, time structure map, and interpretation. Uncorrelated data between seismic and well log data must be correlated since the rock properties information obtained from resistivity, gamma ray, neutron, and sonic P and S wave log must represent the change of amplitude responses in seismic data. Therefore, in order to obtain correct interpretation of seismic data guided by

Yohanes Nuwara, Michael Aria Santoso

well logs data, the seismic data in depth domain is correlated to well log data in depth domain using well-to-seismic tie process. We use Hampson-Russell software to perform this process. The post-processed seismic 3D cube or volume has seismic inlines ranging from 10003 to 10338 and seismic crosslines ranging from 2087 to 2807. There are 2 wells surveyed in the field, each well penetrates at different depth. Well 1 penetrates to 802.5 meter and Well 2 penetrates to 822 meter. The lithological markers which distinguish one different lithology from another are provided within the wells. Some of the formations are encountered in the wells, including the granite wash which is found at the bottom depth. Granite wash is encountered at depth 673.6 meter in Well 1 and 714 meter in Well 2. Well-to-Seismic Tie The well logs data contain RHOB density, NPHI neutron, resistivity, gamma ray, P and S sonic wave log, and vertical seismic profiling (VSP) data. The VSP data is first applied to the logs as checkshot data to achieve correction of depth and time among sonic log and density log. Next, since we will have to create synthetic seismogram, wavelet is created to be then convolved with the reflection coefficient (RC). The RC has been already resulted from impedance contrast calculated from sonic (VP) and density log ( ). We try varied wavelets including Ricker, bandpass, and statistical wavelet, each with different parameters. These wavelets are correlated

Fig.4 Workflow of seismic interpretation

with the seismic data. The best wavelet is chosen if high correlation between the wavelet and the synthetic seismogram is achieved (calculated by the software). Table 1 below represents the best wavelet we choose. The final process of well-to-seismic tie is stretching and squeezing of the wavelet to make the wavelet similar to the synthetic seismogram by certain degree of similarity, which is called correlation coefficient. Correlation coefficient of 0.548 and time shift 0 ms is achieved. The result of well-to-seismic tie is represented in Figure 6. After the well-to-seismic Parameter

Value

Wavelength

100 ms

Frequency

40 Hz

Sample rate

2 ms

tie, the well logs and seismic are now correlated, including the depth time of P-wave is now updated. Fault and Horizon Picking After both seismic and well log data is correlated, fault picking and horizon picking is ready to be applied. We use Petrel software to do fault and horizon picking. First, the threedimensional seismic volume and well logs data are inputted into Petrel. Since both data is still uncorrelated, we input the P-wave depth-time result of previous well-to-seismic tie in HampsonRussell as checkshot data to correlate both data.

Ricker Wavelet

Fig.3 Stratigraphic column of South Sumatera Basin (de Coster, 1974) y

Tab.1 Ricker wavelet for well-to-seismic tie

Fig.5a One of the wells penetrates on the structure of anticline as seen in seismic data

SPRING /â&#x20AC;&#x2030;2019

Seismic Attribute Analysis Aa A Tool For Structural Interpretation

Fig.5b The result of well-to-seismic tie. Blue, red, and black traces in the right side is wavelet, synthetic seismogram, and seismic data respectively.

The fault picking process is done per 10 slices in both inlines and crosslines. The picking method is manual picking. Fault in a seismic section is identified as a disrupted or discontinued feature, and always appears as coherent seismic trace. Other indicators that show fault on the section are amplitude anomalies, poorly-organized apparent reflectors, the distributed reduction in amplitude on stratal reflectors and reflector planes, and pseudo-continuity on stratal reflector which often comes as apparently continuing but in fact reduced amplitude. The succeeded faults are followed by horizon picking in gra-

Fig.6 The generated seismic volume nite wash horizon. Based on the depth information of Top of Granite Wash from two wells, e.g. Well 2, which is 714 m below the surface (Figure 5a), and matched with the position of Well 2 both in inline and crossline position in seismic data, the picking must match. We are aware to every process that will make our later interpretation accurate and precise.

Recalling the workflow of seismic interpretation as pictured in Figure 4, the picking process is guided with seismic attribute. According to Barnes (1999) and Brown (2000), seismic attribute is a qualitative and quantitative information derived from seismic data. This information gives clearer insight of anomalies in our seismic data which are not clearly visible in the conventional seismic. The attribute is widely used for structural interpretation such as discontinuity features (faults, folds, collapse, or dome in diapir cases) and facies interpretation which aids stratigraphy analysis.. Throughout this paper, we focus on structural interpretation and reservoir properties inference. We use some attributes such as … Using Petrel, we generate variance, curvature, and consistent dip seismic volume from conventional seismic for structural interpretation in sections and timeslices. The timeslice interpretation helps structural analysis laterally which shows the orientation of faults. After all faults and horizons guided by seismic attributes are completed, we generate the time structure or surface map of granite wash horizon. Time structure map shows two-way travel seismic time which then infer the topography of horizon. Seismic attributes, namely RMS amplitude, envelope, RMS amplitude, and relative acoustic impedance to infer our reservoir properties. Seismic Attribute Analysis Throughout this paper, there are two aims of using

Yohanes Nuwara, Michael Aria Santoso

seismic attribute analysis for interpretation of our seismic data. First, we would like to characterize the structures (including faults, collapses, and their orientation as well) in the seismic data. Second, we are interested to infer the reservoir properties, including the possibility of hydrocarbon accumulation. Five attributes are used, namely consistent dip, curvature, variance, root mean square amplitude, and relative acoustic impedance. The first three attributes are aimed for structural interpretation, and the latter two are for reservoir properties inference. We use three steps in seismic attribute analysis. At first, we use structural seismic attributes to guide us in picking faults and granite wash horizon in our seismic data. This could be validating the results. Next, we generate the seismic timeslice at depth time nearly corresponding to the depth of granite wash (650 ms) and apply structural attributes to the timeslice for interpretation of structures orientation. Finally, we generate the two-way time surface map of granite wash horizon that we pick previously and extract. Figure 7 gives brief representation of our seismic attribute analysis methodology. ◀◀ RMS amplitude reveals bright spots and amplitude anomalies in seismic data, equivalent to reflection strength (Koson et al., 2014). The computational method is done as presented in Equation 1. In contrast with reflection strength, the window length can be altered to set the resolution we need. Longer windows produce a smoother amplitude estimation, which at certain times useful. This attribute is useful to highlight coarser-grained facies, compaction related effects, for instance in marl and limestone, and also unconformities.

Eq.1 RMS Amplitude (Alifudin et al., 2016) Where N = number of data and a = seismic amplitude. ◀◀ Relative acoustic impedance indicates impedance changes in a relative way by calculating the running sum of the trace to which a low cut filter is applied (Subrahmanyam, 2008). In typical impedance data the low cut filter is applied to remove the DC shift. But if the value of the low cut filter is zero, then one cannot applied it. The calculated trace is gain by integrating the seismic trace. This attribute will give a better view of the acoustic contrast, sequence boundary, unconformity stratigraphic, and discontinuity. It can also indicate fluid content and porosity of the rocks. ◀◀ Variance attribute estimates the similarity of adjacent waveforms or traces given over lateral and/or vertical windows (Koson et al., 2014). Hence, it can be used to image discontinuity or incoherency of seismic data related to faulting or stratigraphy. Variance attribute is a very powerful tool for edge detection and delineation of a certain structure like fault or stratigraphic feature such as channel. It gives the best view for both horizontal slices and vertical seismic profiling. Because it can be used on wide area this attribute can be used to display major fault zones, fractures, unconformities and even the major sequence boundaries (Pigott et al., 2013).

Eq.2 Variance Where S = Variance, n = number of data, x = variable value, and k = window length.

Fig.7 Three steps of seismic attribute analysis and the attributes used in each step

SPRING / 2019

Seismic Attribute Analysis Aa A Tool For Structural Interpretation

◀◀ Curvature of a surface is a measure of curvedness (reciprocal of radius, see Equation 3) analyzed on marked reflector in case of seismic. Curvature estimation is done after analyzing the reflectors dips and azimuths in seismic cube. Although curvature attributes can be measured in different directions, the two method of curvature; most positive and most negative, are found to be useful in imaging structural features.

Eq.3 Curvedness (Asif et al., 2015) The method of most positive and most negative curvature, both of them can produce a polygonal like appearance of faults and lineaments. Concept of curvature can be explained in Figure 8. Anticline and syncline will give a positive curvature and negative curvature respectively. Flat terrain such as dipping plane or flat ground will give a zero curvature. In Cartesian coordinates X-Y, curvature (k) can simply be defined as a second derivative in equation 4,

Eq.4 Curvature (Asif et al., 2015)

constraints. Conventionally, local dip estimations is calculated through cross-correlation or gradient based method (Iske and Randen, 2005). In order to make these local estimates look consistent and spatially continuous, sometimes the local estimates are then smoothed using a mean filter, for example. Consistent dip works by doing estimation of the dip volume iteratively, hence it is called Volumetric Dip estimator in the industry, of a seismic volume to improve what other conventional dip estimation methods are lacking. For each iteration the consistent dip method will check for reciprocity, causality, consistency, and vertical and lateral continuity for local dip estimation which is specified by the user constraints Time-Structure Map Time structure map can be created and obtain after performing fault and horizon picking to estimate the contour of the elevation based on time from seismic data. The idea for making a time structure map is to display the elevation of a particular rock layer beneath the surface, which proved to be useful at helping interpretation. This contour of elevation depict geological formation or structure such as faults, folds, and other type of geological structures and can gives a clear display of it. The horizon are picked with time constraint based on 3D post- or 3D prestack time migration from the seismic volumes. Results and Discussion Structural Interpretation in Inline and Crossline Seismic Section The fault picking process results number of faults growing on the seismic section. We discover 46 faults in crossline seismic section and 7 faults in inline seismic section, consisting of major (big) faults and minor faults.

Fig.8 Curvature in two dimensions Consistent Dip provides an estimate of local dip in seismic section. Consistent dip is a relatively new attribute that solves basic problem that local (instantaneous) estimates of dip can only be calculated without taking into account global consistency

Figure 9 above is the result of fault picking in inline seismic section directed SW - NE. The major feature appeared in the seismic section is folding, thrust faults, and normal faults. From Figure 9a, we find out normal fault feature characterized by the appearance of major synthetic fault (cyan colored) and its antithetic fault (light-green). Thrust fault

Yohanes Nuwara, Michael Aria Santoso a)

Fig.9 Faults picked on Inline seismic section, each has different interpretation corresponding to (a) and (b), and (c) is the NW – SE and N – S oriented faults visible on Inline on timeslice (z=350 ms)

(dark-green) is visible in front of fold feature in both Figure 9a and 9b. The fold characteristic is faultpropagation anticline fold. In Figure 9b, the light-yellow fault is characteristic of roll-over antiform (product of normal fault). The top of granite wash horizon picked is appeared higher in NE and sloping steeply to lower part in SW. Figure 9c is the timeslice section which shows faults oriented SW-NE and N-S using variance attribute. Each fault with different colors in vertical seismic section will be shown on timeslice with same colors. From these results, we find out that SW-NE oriented faults in timeslice is shown as thrust faults, whereas the N-S faults is shown as normal faults. This clearly shows two different regimes; compressional and extensional regime. The SW-NE faults are product of compressional regime since this regime is always associated with the existence of thrust faults, similar to N-S faults as product of extensional regime. Variance attribute on timeslice shows very clear and distinct fault features, and also tells the orientation, which could not be shown in conventional seismic on timeslice section. We will cover this discussion later on. a)

This result matches with the geology of South Sumatra Basin presented previously that thrust faults and anticlines from Miocene to Pliocene age are oriented NW-SE whereas normal faults from Eocene to Oligocene are oriented N-S. Therefore, the N-S faults are relatively older than NW-SE faults. The Figure 10 above is corresponding geological interpretation of the Inline seismic section; SW-NE fault-propagation anticline fold (Figure 10a) and N-S roll-over antiform (Figure 10b). In the crossline seismic sections, we find out that most of the minor faults are located in the granite wash horizon. This can be interpreted as intensely fractured granite wash due to intense geological processes. We only show three of the overall crossline seismic slices with very distinct features. In Figure 11a above we find out 4 minor faults crossing the granite wash horizon and 6 major faults in the upper part. We can see that the faults are typical of half-graben faults. In different Crossline slice in Figure 11b, there are 3 minor faults in granite wash horizon b)

Fig.10 The corresponding geological interpretation of the seismic section on Inline

SPRING / 2019

Seismic Attribute Analysis Aa A Tool For Structural Interpretation

Yohanes Nuwara, Michael Aria Santoso

Fig.13 (a) Original post-stack seismic section, (b) RMS amplitude attribute, and (c) relative acoustic impedance attribute c)

Fig.11 Faults picked on Crossline seismic section, each has different interpretation corresponding to (a), (b), and (c). Figure (d) is the NW â&#x20AC;&#x201C; SE oriented faults visible on Crossline

and 3 major faults in the upper horizons. Similar to normal fault case in the previous Inline slice (Figure 9a), the light-green colored fault can be identified as synthetic and the light-blue colored fault is the antithetic. Additionally, another Crossline seismic section clearly shows horst and graben blocks (Figure 11c). These faults are characteristic of extensional faults. As it is seen on the timeslice variance section (Figure 11d), the faults picked in the seismic section shows a uniform orientation in NE-SW. This surely matches the result of regional geology of South Sumatera Basin which discusses that structures oriented at NE-SW are associated with horst and graben as product of extensional phase in Eocene to Oligocene.

Fig.12 The corresponding geological interpretation of the seismic section on Crossline

Figure 12 above is the full scale representation of an ideal extensional basin. Three parts in the extensional basin which infer the seismic section we discussed previously are highlighted here. Up to this point, we find out that variance timeslice validates the result of our fault picking and matches with the regional geology. Generating seismic attribute volume to analyze the structures is a powerful tool. Seismic Attribute for Validating Fault and Horizon Picks Result Each seismic attribute has its own advantage to be used for helping interpret seismic data by enhancing certain parameters. We use RMS amplitude to determine fault like structure and stratigraphic sequence boundary when normal seismic cross section are lack in resolution (Figure 13a). To pick the most accurate time for determining layer boundary, we use the relative acoustic impedance (Figure 13c). Seismic attributes are useful while

picking faults and horizon in our seismic section. In seismic data, faults often come as pseudo-continuity of stratal reflector. Pseudo-continuity makes an appearance of continuing amplitude, whereas in fact it is a fault zone. This pseudo-continuity can lead to ambiguity of fault picks and later, wrong interpretation. As in Figure 13a, several pseudo-continuities can be found in the seismic section. After applying RMS amplitude (Figure 13b), the pseudo-continuities become separated. Therefore, RMS amplitude can eliminate the pseudo-continuity in conventional data and distinguish faults. Result is also fairly good using acoustic impedance (Figure 13c). Additionally, the strong reflectors can be distinguished by high amplitude (yellow) using RMS amplitude, but poor in continuity of amplitudes. With acoustic impedance, the continuity of stratal reflectors is perfect, but it has too much multiple reflections which become individual stratal reflectors. This could be useful in horizon picking, although each attribute has its advantages and limitations. Comparison of Different Seismic Attributes for Structural Interpretation As discussed previously, interpretation on timeslice section from our seismic volume using variance attribute gives excellent result on orientation of faults in the seismic data that perfectly matches with regional geology of South Sumatera basin. Structural seismic attributes highlight structural features such as faults and collapses which we could

Fig.14 Original post-stack seismic data at time slice z = 650 ms (green arrow points northward)

not spot easily in conventional seismic. Comparing the conventional timeslice seismic section from our seismic volume data (Figure 14 above) and the seismic attributes applied to timeslice sections (Figure 14 to 15 to 17), we find out distinct features of both coherent and incoherent features, including faults. Generally similar to variance attribute, we can easily find three orientations of faults (SWNE, SE-NW, and N-S). We compare several other structural attributes as well. Variance attribute (Figure 15) gives the best result for structural features in the timeslice so that we can estimate the different orientations of faults. Red co-

SPRING /â&#x20AC;&#x2030;2019

Seismic Attribute Analysis Aa A Tool For Structural Interpretation

Fig.15 Variance attribute at timeslice z = 650

Figure 16 is the result of curvature attribute applied to timeslice. Brighter color indicates positive curvature, whereas darker indicates negative curvature. The faults in different orientations are visible using curvature, which comes in dark color because the existence of faults in seismic data is identified as negative curvature. We could see that this attribute also provides sense of “roughness” which makes the structural features more real. The dark chaotic feature adjacent to the great SE-NW fault is interpreted as damaged zone. Vertical radius parameter of curvature also has influence on smoothing the rough features. Lower vertical radius (Figure 16a) results a more rough feature than higher vertical radius (Figure 16b).

Yohanes Nuwara, Michael Aria Santoso

Seismic Attribute for Mapping Hydrocarbon Accumulation in Granite Wash

The final result of structural characterization of our top of granite wash horizon is done by extracting structural seismic attributes from the seismic data to the horizon. The above variance and curvature horizon gives an excellent and very realistic image of a horizon. Variance gives the best image since the faults and topographic features are very clear, including the great SE-NW thrust fault in red color and collapse feature that coincides with lowest depression from the previous time surface time. Figure 19b depicts the curvature extracted on the horizon. Curvature gives a fairly good image of the structures on the horizon.

ms (green arrow points northward)

lor indicates the highest variance, darker indicates high variance, and brighter color indicates small value of variance. The great fault oriented SE-NW and N-S appeared dark with chaotic red patterns nearby, clearly show as faults and damaged zone in front of the fault. There are smoother dark lineaments oriented perpendicularly, indicates SW-NE faults. a)

Fig.16 3D Curvature (Most Negative Method) attribute at time slice z = 650 ms (green arrow points northward) (a) With vertical radius = 5, (b) With vertical radius = 20

Fig.18 Time surface map of top of granite wash

Fig.17 Consistent dip attribute at timeslice z = 650 ms (left: consistent dip inline, right: consistent dip crossline)

As seen in Figure 17 above, consistent dip enhances different perspective of faults relative to trace movement in inline or crossline. Consistent dip is a modern attribute that is regarded as dip attribute. As dip attributes do, consistent dip give an estimation of local dip. We know that the distinction between dark and bright color is positive and negative value of dip, respectively. In consistent dip inline, we see that the great SW-NE fault is bright or “turned on” whereas SENW faults are dark or “turned off ”. In contrast, we see that in consistent dip crossline the great SW-NE fault becomes dark and inversely, the SE-NW becomes bright. Therefore, each of these two is inverse to the other one. Consistent dip gives the interpreter a preference to view the structures from different perspectives, either from inline or crossline sight.

The result of time surface map can be seen in Figure 18 above. It shows a clear depiction of topographic feature of the granite wash horizon top. There is a distinction between higher part in the northeast and lower part in the southwest. The distinction is caused by different two-way travel time. Travel time will be shorter to reach the higher part than the lower part. This feature coincides with the geology of the basin, where the higher part is associated with Pendopo High. The lowest depression (purple-colored) is associated with the existence of N-S oriented half graben. a)

Fig.20 Consistent dip inline and consistent dip crossline three-dimensional granite wash horizon

Another structural attribute that we use, consistent dip, is extracted on the granite wash horizon with distinct image between consistent dip crossline and consistent dip inline technique that we discussed earlier. The distinction between these two is the imaging of one technique of some uniformly oriented faults appearing as bright lineament which appears dark in another technique. After processing the granite wash horizon with extracted structural attribute, the reservoir propera)

Fig.19 a) Variance, and b) curvature three-dimensional granite wash horizon

Fig.21 a) Relative acoustic impedance surface map, and b) RMS amplitude surface map

SPRING / 2019

Seismic Attribute Analysis Aa A Tool For Structural Interpretation

ties of top of granite wash is characterized by extracting root mean square (RMS) amplitude and relative acoustic impedance (AI). The result of both attribute is presented in Figure 21. In relative AI surface map, we can distinguish two zones with different value of AI; high AI (red) in higher topography in the northeast and relatively low AI (blue) in the lower part of the granite wash horizon top. We have already known that AI corresponds to degree of consolidation of rocks in the horizon, the rock types, and fluid content. Therefore, the two different zones can be said having different properties. Additionally, we know that higher AI often indicates to hydrocarbon content. Hence, the possibility of hydrocarbon can be indicated in the higher zone, which previously coincides with Pendopo High, which is the basement of our reservoir. Sedimentation can also be inferred from this result. From the RMS amplitude surface map presented in Figure 21b, there are varied values of RMS amplitude distributed on the Top of Granite Wash. Using our previous knowledge that RMS amplitude corresponds to reflection strength, we can understand that red-colored zone has high reflection strength, and blue-colored is low reflection strength. Reflection strength indicates the contrast of rock types. Therefore, there are two different rock types in the reservoir. This high RMS amplitude may be caused by several possible causes, such as high hydrocarbon saturation, high porosity, or low density. Useful Tool for Prospect Evaluation and Planning in Enhanced Oil Recovery We find one interesting feature in the granite wash horizon. Well 2 in the field penetrates the “red” zone

or high acoustic impedance in the relative acoustic impedance surface map of granite wash, as it is presented in Figure 19. In the previous discussion, the high AI zone probably corresponds to hydrocarbon content. Therefore, we can say that there is prospectivity in Well 2, or in other words Well 2 have already encountered a possible prospect. Surely, the possibility must the be validated with other data. Additionally, there have not been so many wells penetrating in the other red zones nearby Well 2. Hence, a recommendation could be made to consider the prospectivity in this zone for future target. Up to this point, we can say that the result helps in prospect evaluation that most oil industry do in development. It also helps decisionmaking in planning the target for Enhanced Oil Recovery. After the reservoir been characterized, a static model can be built afterward. This static model provides information about both structural features, such as faults, and suggests the possibility of hydrocarbon accumulation. The model can then be used for petroleum engineers to set up a simulation of fluid in a dynamic model of reservoir. In waterflooding for instance, the structures must be known so well that sealing and leaking is an important key to success waterflooding. Using seismic attributes for dip estimation are beneficial for mapping faults structure. Sometimes it is a must to know what kind of structure we are dealing with before drilling pipes for EOR or injecting some substance to deep reservoir. Because, as we know, some structure like juxtapose faults can make a good caprock sealing for hydrocarbons since the structure can support to be a fine seal. But only after we identify the quality of the seal first (Bai et al., 2017) and for that seismic attributes would be a very helpful tools to identify it. One of the reason, and the practical, why one should investigate it first because we should know whether the faults would become a sealing one or a leaking one. All in all, leaking faults must be prevented for success waterflooding. Conclusion

Fig.19 One of the wells penetrates the “bright spot” feature on relative AI surface map of granite wash horizon

In this research, we apply seismic attribute analysis for structural interpretation and mapping

Yohanes Nuwara, Michael Aria Santoso

possibility of hydrocarbon accumulation in a granite wash unconventional reservoir targeted for Enhanced Oil Recovery process based on 3D seismic volume data of an unconventional reservoir in South Sumatera. From the manual picking of faults and horizon using the guidance of seismic attribute on seismic section namely root mean square amplitude, variance, and relative acoustic impedance, there are 53 growing faults which consists of major and minor thrust and normal faults. The granite wash is appeared to be intensely fractured since we find out numbers of minor faults in the horizon. Structural seismic attributes such as variance, consistent dip, and curvature are added to timeslice interpretation show clearly different orientations of faults

49 which is not visible in our conventional seismic data. The orientation matches with geology; N-S and SE-NW normal faults, and SW-NE thrust faults. The three structural attributes is compared to observe the sensitivity to map the features. Variance attributes comes as the best. Finally, the time structure map clearly shows the topography of the horizon which comes as higher part in the northeast coinciding with Pendopo High and lower part in the southwest as the product of normal faulting. Time structure map is the most realistic depiction of our reservoir, where we extract seismic attribute to analyze the structures and some reservoir properties directly on the granite wash. Seismic attribute analysis is a powerful tool to evaluate the prospect in our unconventional reservoir.

References: [1] A. Asif A., Rehman E.A., Shoaib K., Anjum A.G. Mustafa R.K. 2015. Discontinuity Attributes, their Visualization and Seismic Interpretation: Case Studies from Indus Basin, Pakistan. Oil and Gas Research. Volume 2, Issue 1, 1000107 [2] Alifudin, Ridho F., Lestari, Wien, Syaifuddin, Firman, dan Haidar, M. Wahdanadi. 2016. Karakterisasi reservoir karbonat dengan aplikasi seismik atribut dan inversi seismik impedansi akustik. Jurnal Geosaintek. 2 (2), 107-112 [3]Bai, Bing, Qifang Hu, Zhipeng Li, Guangzhong Lü, and Xiaochun Li. 2017. Evaluating the Sealing Effectiveness of a Caprock-Fault System for CO2-EOR Storage: A Case Study of the Shengli Oilfield. Geofluids, Volume 2017, Article ID 8536724, 17 pages [4] Bishop, M. G. 2001. South Sumatra Basin Province, Indonesia: The Lahat/Talang Akar-Cenozoic Total Petroleum System. Open-File Report USGS 99-50-S [5] Chopra, S., K. F. Marfurt. 2015. Seismic Attribute for Fault/Fracture Characterization. Search and Discovery AAPG Article #41539 [6] Bushara, M. N., R. F. Cox. 2007. Guided Facies Modeling using 3D Seismic and Well Data: Soft-Conditioning to Geomorphologic Objects. International Journal of Petroleum Science and Technology 1 (1), 51–60. http://www.ripublication.com/ijpst.htm [7] Iske A., T. Randen. 2005. Mathematical Methods and Modelling in Hydrocarbon Exploration and Production. Berlin : Springer [8] Koson, S., P. Chenrai, M. Choowong. 2014. Seismic Attributes and Their Applications in Seismic Geomorphology. Bulletin of Earth Sciences in Thailand. 6 (1), 1-9 [9]Odoh, B. I., J. N. Ilechukwu, N. I. Okoli. 2014. The Use of Seismic Attributes to Enhance Fault Interpretation of OT Field, Niger Delta. International Journal of Geosciences 5, 826-834. [10] Pigott, John D., Moo-Hee K., Hyun-Chul H. 2013. First order seismic attributes for clastic seismic facies interpretation: Examples from the East China Sea. Journal of Asian Earth Sciences. 66 (2013) 34–54 [11] Sukmono, S. 2006. Integrating Seismic Attributes for Reservoir Characterization in Melandong Field, Indonesia. The Leading Edge 25 (5). 532-538 [12] Roberts, Andy. Curvature Attributes and their Application to 3D Interpreted Horizons [13] Subrahmanyam, D., P.H.Rao. 2008. Seismic Attributes – A Review. 7th International Conference & Exposition on Petroleum Geophysics [14] Widodo, R. 2012. Integrating Wells and 3D Seismic Data to Delineate the Sandstone Reservoir Distribution of the Talang Akar Formation, South Sumatra Basin, Indonesia. Search and Discovery Article AAPG #50748 [15] Zhenga, Zhihong, Tan, Jieqing, Liuc, Kang. 2014. Most Extreme Curvature and Its Application to Seismic Structural Interpretation

SPRING / 2019

¡¡

Foam Stability Performance Enhanced with Rice Husk Ash Nanoparticles

Foam Stability Performance Enhanced with Rice Husk Ash Nanoparticles Chuah Kai Jie, Mohd Zaidi Bin Jaafar

** Faculty of Engineering Universiti Teknologi ÞÞMalaysia

Rice husk ash (RHA) has been recently used as a source of silica (SiO2) production due to its high silica content. Besides, high purity silica nano-powder has been successfully synthesised from RHA and employed in various industries including electronic component manufacturing and fillers in polymers. Meanwhile, silica nanoparticles has been widely used in the application of Enhanced Oil Recovery (EOR). This is due to its ability in enhancing the foam stability besides modifying the wettability of the rocks in the formation. However, the synthesis of silica nanoparticles from RHA for the application in big scale operation such as EOR using conventional method is energy and time consuming. Therefore, the objective of this work is to study the effectiveness of using nano-sized rice husk ash (nanoRHA) as an additive to stabilise normal gas generated surfactant foam used in EOR. In order to decrease the size of the RHA into nano range, planetary ball mill was used in both dry grinding and wet grinding. Different surfactants including anionic and non-ionic were then used to study the polydispersity index of the dispersion and the hydrodynamic diameter using dynamic light scattering in dilute suspension. Besides, the nanoRHA was also characterized using FESEM, EDX, XRD and the change in specific area after grinding process was studied using BET. The foamability of different surfactants were then studied using minor concentration of nanoRHA. Next, the concentration

of the nano-RHA was varied from 0.1wt% to 0.9wt% in normal gas bulk foam stability test using the suitable surfactant, the texture of foam was observed as well. Apart from that, the effect of oil on bulk foam was also studied. Finally, the result was compared using pure silica nanoparticles as the foam addictive at the same variation of concentrations. Dispersion stability tests showed that both anionic and non-ionic surfactants can be used to disperse nano-RHA in water. Moreover, in the presence of 0.9wt% of nano-RHA concentration, the bulk foam stability test results revealed that the sodium dodecyl sulfate (SDS) bulk foam half-life increased by 6.21% without the presence of oil, and gave an increment of 23.5% half-life in the presence of oil. Therefore, the study showed that there is a potential of utilizing nanoRHA in stabilising bulk foam. Keywords: Foam stability, nano-rice husk ash, ball-milling

Introduction The total volume of conventional oil discovered in year 2016 plunged to the lowest annual yield in the six decades, for a merely 9 bnBOE. Besides, it also marked the sixth consecutive year of decline (Flowers, 2017). Despite the recent global crude oil price crisis started in year 2015, there were only six times since year 2000 in which the industry has delivered more than 30 bnBOE in a single year. The recent oil discovery performance becomes the concern and the industry is conceding the necessity to address the declining

Chuah Kai Jie, Mohd Zaidi Bin Jaafar

exploration in technical and economic performances. Hence, it is crucial for the industry to recover considerably more oil from the current producing assets when the challenge in making consequential new discoveries continues to rise. It is estimated that even a 5% increase in the average global recovery is comparable to the new reserves of all the future exploration activities (Florida, 2013). Meanwhile, enhanced oil recovery (EOR) has been identified as the key effort to further improve oil recovery from the remaining 60% or more of the oil after the secondary recovery process. Currently, EOR contributes about 3% of the worldwide production (Abu El Ela, 2008; Taber et al., 1997). Miscible gas flooding increases from 118 projects in 2004 to 174 projects in 2014, becoming the most widely implemented EOR technique in worldwide (Abu El Ela et al., 2014). Gas flooding provides a higher microscopic sweep efficiency compared to water-flooding. However, the efficiency is affected by viscous instability, reservoir heterogeneity and gravity segregation (Lake, 1989). A need for mobility control in gas flooding has led to the use of foam for sweep improvement and profile modification. Foam is a two-phase fluid system in which the gas phase is discontinuous, separated by thin liquid film known as lamellas (Hirasaki, 1989). Unlike gas flooding which is influenced by viscous fingering and gravity override, foam has great specific gravity-independent properties. With the controlled mobility using foam, the gas can be channelled from high-permeability areas to the lowpermeability areas. Thus, enabling the injected gas to reach the poorly swept sections of the reservoir. Also, foams performs better in controlling the fingering problem as compared to WAG (Zhu et al., 1998). The key parameter to define a good quality of foam is the foam stability (Tan, 2017). In other words, foam needs to possess a better lifetime depending on its purpose of applications, usually it is desired to be stable over the distance of propagation through the well. In foam flooding, surfactants are generally used to generate and stabilize the foam lamellas by adsorbing themselves to the fluid interface. Foamability and the foam stability are enhanced with incre-

51 asing surfactant concentration (Belhaji, 2015). To produce the stable foams, the concentration used should be at least maintained at the surfactant critical micelle concentration (CMC). Surfactant foam, however, is unstable when in contact with oil (Farzanehand and Sohrabi, 2013). Moreover, surfactant-stabilized foam is also thermodynamically inferior (Kaptay and Babcsán, 2012) and prone to destabilization in harsh conditions such as high salinity (Borchardt et al., 1988) and high reservoir temperature. Consequently, greater amount of surfactants as compared to laboratory scale is required to produce surfactant-stabilized foam for mobility control (Yin et al., 2009). Meanwhile, the use of solid particles in nano-size provides solution to the problems discussed above. Khajehpour et al. (2016) reported nanoparticlesstabilised-foam is more stable than surfactant -stabilisedfoam. Compared to surfactant molecules, due to the higher adsorption energy of the nanoparticles, they are able to adsorb irreversibly at the fluid interface (Binks and Horozov, 2005). Besides, they are able to stabilize foam even under extreme temperature and high salinity conditions (Zhang et al., 2011). Hydrophobicity is known to be the key property of nanoparticles because it has significant effects on their reaction behaviours (Handy et al., 2008). Generally, nanoparticles surface is coated to alternate the surface chemistry and it will influence the effects of the nanoparticle core. Polyethylene glycol (PEG)-coated hydrophilic silica nanoparticles generated stable supercritical CO2 water foam in glass beads (Espinosa et al., 2010). On the other hand, Worthen et al. (2013) reported partially hydrophobic nanoparticles performed more effectively than the PEG-coated ones as foam stabilizers. It is commonly surface-active or modified nanoparticles were better foam stabilizers than surfactants. However, nanoparticles available at commercial are usually non-surface active with extreme hydrophobicity (Cui et al., 2011). Surface modification of nanoparticles using surfactants has shown potential method to stabilize foams. The synergy of between nanoparticles and surfactants has been studied over years. Both Zhang et al. (2008) and Cui et al. (2010) reported the stabilization of aqueous foams using a mixture of non-surface ac-

SPRING / 2019

Foam Stability Performance Enhanced with Rice Husk Ash Nanoparticles

tive nanoparticles and surfactants. Furthermore, Binks et al. (2015) even generated ultrastable foam with more than one-year time foam half-life using CaCO3 nanoparticles and sodium stearoyl lactylate (SSL) at high concentration. Using fly ash nanoparticles, Singh et al. (2015) reported the synergy with anionic surfactant which enhanced foam stability in porous media. Besides, J. Wong (2017) achieved remarkable stable sodium dodecyl sulfate (SDS) bulk foams with the addition of fly ash nanoparticles. Rice husk (RH), an agricultural waste is vastly available in rice producing countries such as China, India and Indonesia. Incineration of RH from the RH biomass power plant produces rice husk ash (RHA) as waste product. According to World Rice Statistics (2014), the total RHA world production increased from 28 Mt in 2010 to 29.6 Mt in 2014. India alone accounted for 20.6% (6.1 Mt) of RHA in 2014, second highest after China 27.9% (8.3 Mt). Inappropriate way of disposal will undoubtedly cause long term environmental issues due to the low bulk density of RHA (Pode, 2016). Positively, RHA is used in coatings, cement industry, insulator and electronics (Soltani et al., 2015). The laboratory synthesis of SiO2 nanoparticles (SiO2-NP) from RHA was also studied by Thuadaji and Nuntiya (2008). However, strong acid leaching pre-treatment of RH, is significantly hazardous to environment and human life, and results in an increase of the SiO2-NP process cost (Soltani et al., 2015). The feasibility of using RHA-nanoparticles (RHANP) in EOR application, precisely in stabilizing surfactant foams should be studied. Therefore, this paper establishes a novel method to use RHA-NP as an additive to stabilize surfactant foams. Without using any chemical pre-treatment, the size range of RHA was decreased by planetary ball mill grinding. The importance of this study will contribute another effort of utilizing the sustainable use of waste materials, besides providing an idea of the possibility to substitute hydrophilic SiO2-NP (without surface modification) with RHA-NP since SiO2 nanoparticles are commonly used in EOR application and they are considered one of the best nanoparticles (with dichlorodimethyl silane surface coating) to stabilize foam (Ogolo et al., 2012).

Chuah Kai Jie, Mohd Zaidi Bin Jaafar

Objective The main objectives of this research are as followed: 1. To establish a novel study of the stability performance of rice husk ash nanoparticles (RHA-NP) in surfactant air foam. 2. To compare the performances of surfactant air foam stabilized by RHA-NP and hydrophilic SiO2 nanoparticles (SiO2-NP). 3. To study the stability performance of RHA -NP surfactant air foam in the presence of oil, and without oil.

Fig.1 Left: before thermal treatment; Right: after thermal treatment (lower carbon content, higher silica content).

Fig.4 EDX analysis result of RHA-NP showing

Methodology Material Rice Husk Ash (RHA RHA was obtained from a rice mill. The as-received RHA (AR-RHA) was physically black in colour with some grey particles (Fig. 1, left). It was due to incomplete carbon combustion during the incineration of RH (Della et al., 2002). The AR-RHA was sieved (AS200; Retsch, Germany) and the average size obtained was at a range between 88µm to 105µm. It was then treated under 700°C for 4 hours using an electric furnace (Thermolyne 30400 Laboratory Furnace; Barnstead, United States). This step was to reduce the carbonaceous materials present in the samples, thus increasing the relative amount of silicon dioxide content (Della et al., 2002). The thermal-treated RHA (TT-RHA) was observed to be grey in colour with abundant white particles (Fig. 1, right). Next, TT-RHA was then milled in order to reduce its particle size (the grinding process will be further discussed in the following section). Upon getting the desired size range, it was then characterized by morphology study using FESEM (Fig. 2&3) along with its elemental composition analysis by EDX (Fig. 4) (SU8000; Hitachi; Japan). The phase analysis was studied using XRD (Fig. 5 & Table 1) (SmartLab; Rigaku, Japan). Preparation of Rice Husk Ash Nanoparticle (RHANP) The size of the foam additives has to be small enough to avoid them from plugging the pores of the rocks in the formation, in which typically around

the ash comprised mainly of silicon and oxygen 8.0e+003

Intensity (counts)

Meas. data:Rice Husk Ash-101-2017Nov19 _1

6.0e+003

4.0e+003

2.0e+003 0.0e+000

Fig.2 Morphology of a bulky ball-milled RHA studied by FESEM (in 3µm).

100

2-theta (deg)

Fig.5 XRD anlaysis of RHA-NP. The curve shows two peaks, which represent silica and zeolite. Heating under 700°C produced amorphous SiO2 as shown by the curve.

Fig.3 FESEM image of RHA-NP (in 1µm). The

Phase name

Content (%)

Silicon dioxide (SiO2)

99.82

Zeolite (Na2Al2Si3O102·H2O)

0.17

Others

0.01

shapes of the nanoash particles are irregular and they tend to agglomerate.

Tab.1 XRD result showing the phases detected in RHA-NP

SPRING / 2019

Foam Stability Performance Enhanced with Rice Husk Ash Nanoparticles

few microns. TT-RHA was too large to be feasibly used as foam additive. Therefore, it was submitted to ball milling process in order to produce the highest degree of fineness. It was performed by a two-stage grinding process using a planetary ball mill machine (PM100; Retsch, Germany). The method of ball milling was adopted from the guide by Retsch Germany. The grinding process involved in both dry and wet medium. First, dry grinding was performed to get a homogenous size of TT-RHA. The 250 mL tungsten carbide (WC) grinding jar was loaded with 66.7% by volume of WC mill balls (approx. 150 mL, measured using a measuring cylinder), of which 3 mm in diameter; and 33.3% by volume of TT-RHA (approx. 83 mL). It was grinded for one hour at a speed of 450 rpm. Dry grinding was intended to achieve a more homogenous particle size and to reduce the size above the average distribution. The second stage grinding was performed in wet medium to achieve nano range. To prevent the agglomeration of the particles, ethanol was added as a dispersing agent. Ethanol was added until it covered the ash and the balls mixture. The ash and the mill ball volume ratio was kept constant. Wet grinding was run for four hours at a speed of 450 rpm. The samples were then collected and dried in an oven under 90°C for at least 15 hours. Eventually, the size of the TT-RHA after wet grinding was determined by hydrodynamic light scattering using particle size analyser (Zetasizer Nano ZS; Malvern Instrument, United Kingdom). The result gave dn50 reading of 140 nm. BET surface area analysis (3Flex; Micromeritics, United States) gave readings of 12.4506 m2/g before grinding and 23.7660 m2/g after grinding. Hydrophobicity Test In order to study the characteristic of the hydrophobicity of RHA-NP, the contact angle measurement method was conducted. RHA-NP was dispersed in deionized water (DI water). Few drops of the concentrated RHA-NP dispersion was allowed to dry under room temperature on a sterile glass slide. Then, a drop of DI water was placed on the surface of the dried RHA-NP dispersion. The shape of the water droplet was captured and ImageJ software was used to analyse the contact angle.

Dispersion Stability Test The test was conducted to identify suitable dispersant for RHA-NP. Several anionic surfactants and a non-ionic surfactant were tested. Table 2 below shows the types of the surfactants used: Type Anionic

Name (Label) Sodium Dodecyl Sulfate (SDS) Sodium Dodecylbenzenesulfonate (SDBS) Alpha Olefin Sulfonate (AOS)

Non-ionic

Triton X-100

Tab.2 Types of surfactants used in dispersion stability test.

0.5 wt% of RHA-NP was diluted in 0.2 wt% of surfactant solution. Mixture was stirred by a magnetic stirrer for an hour, followed by an hour of ultrasonication to promote dispersion stability. Samples were taken from the middle of the vials after letting three hours of settling time. Next, the samples were characterized by dynamic light scattering using particle size analyser, Zetasizer Nano SP. Poly-dispersity index was also taken to measure the degree of homogeneity in respective surfactants. Foamability Test Compatible surfactants (0.2 wt%) from the dispersion stability test were then mixed with 0.5 wt% of RHA-NP. The foamabiliy of these surfactants in the presence of RHA-NP were studied by measuring the height of the foam produced. 4 mL of the mixture was shaken vigorously 10 times in a centrifugal tube. The height of the foam for each mixture was taken at the initial time. Foam Stability Test The most suitable surfactant, sodium dodecyl sulfate (SDS) was chosen as the dispersing agent in this foam stability test based on the previous dispersion stability and foamability tests. Despite the better result from the previous tests, according to Esmaeilzadeh et al. (2014), SDS also works well with nanoparticles when injected as part of an EOR process. Foam stability test was conducted using dynamic foam analyser (Krüss DFA100; Krüss, Germany). The concentration of the brine solu-

Chuah Kai Jie, Mohd Zaidi Bin Jaafar

tion was kept constant at 2 wt% to simulate the formation water salinity ( J. Wong, 2017) whereas the concentration of the SDS surfactant was kept at its critical micelle concentration (CMC) value. The concentrations of RHANP were manipulated from 0.0 wt% (without RHA-NP) to 0.9 wt%. It was intended to determine the trend of the effect of different RHA-NP concentrations against the SDS foam stability performance. The mixture was prepared using DI water as the solvent. The mixture was stirred using magnetic stirrer for at least 16 hours, then ultrasonicated for two hours to get better dispersion stability. The mixture was then injected carefully into the glass column of the foam analyser using a 50 mL syringe without pre-wetting the glass above the liquid level. This is crucial as pre-wetting can affect the adhesion characteristics of the foam. The volume of each mixture injected was kept at 30 mL. Besides, the continuous phase used to generate the foam in this experiment was normal air. The normal air flow rate was set at 0.3 L/ min for 15 seconds. Thus, giving a foam quality of 71.4%. In fact, stable foam can be formed through quality of 50% – 85%, while foam formed with less than 50% quality is considered relatively unstable (Friedmann and Jensen, 1986). The half-life of the foam was used to define the foam stability, in which the foam height and the time taken for the foam to disintegrate to half of its initial height were recorded using the foam analyser software. Using the optimum concentration of RHA-NP, the experiment was repeated with hydrophilic SiO2-NP without surface modification (Product # 6808NM, 99.5% pure, 20 nm, non-porous,) obtained from SkySpring Nanomaterials, Inc. The SiO2-NP was varied at the same concentrations as the ones previously used in RHA-NP. This step was done to compare the performances of both RHA-NP and commercial SiO2-NP in stabilizing surfactant foam. The experiment was repeated by adding oil to study the foam generation in the presence of oil. Heavy oil, nhexadecane was used as a resemblance of crude oil. Moreover, the foam is also characterized by the foam textures, taken by the camera of the dynamic foam analyser, Krüss DFA100. The ma-

terials and formulations used were summarized as followed: Material

Function

Composition

Deionised water

As liquid

80 mL

(DI water)

solvent

Sodium chloride

As brine

2 wt%

As surfactant

0.23 wt% (CMC

(NaCl) Sodium dodecyl sulfate (SDS)

value)

Rice husk ash

As foam

0.1 wt%, 0.3 wt%,

nanoparticles

additives

0.5 wt%, 0.7 wt%

(RHA-NP) Normal air

and 0.9 wt% As foam

0.3 L/min, density

booster

of 1.293 kg/m3

Tab.3 Materials and formulations used in the foam stability test.

Results And Discussion Hydrophobicity Test The contact angle obtained from ImageJ software is 14.6°. This implies that the RHA-NP is hydrophilic in nature. Image below shows the screenshot from ImageJ software analysing the contact angle.

Fig.6 Screenshot of ImageJ software analyzing the contact angle of RHA-NP.

Dispersion Stability Test The results were tabulated in Table 4. SDS, SDBS and Triton X-100 have PDIavg of lower than 0.200, which implies a good dispersing stability among

SPRING / 2019

Foam Stability Performance Enhanced with Rice Husk Ash Nanoparticles

Type Anionic

Surfactant SDS

PDIavg 0.127

RHA-NP stabilized foam

Surfactant

Foam height reading (scale division of the centrifuge tube)

SDS

12.5

SDBS

6.5

AOS

2.5

Triton X-100

5.0

150 145 140 135 130 125 120 115 110

145 140

123

125

128

wt%

0.132

159.2

AOS

0.220

184.5

Triton

0.116

162.8

X-100

Tab.4 Results of dispersion stability test. Foamability Test This test was conducted to study the performance of RHA-NP on the foaminess of different surfactants (Fig. 7). The higher the foam height, the better is the foaminess. The height of each foam was tabulated (Table 5). Results show that SDS had the best foaminess with the highest foam produced whereas the foaminess of AOS was the lowest. Based on the observation, the bubbles foam generated in SDS with RHA-NP were denser and smaller in sizes. Thus, making SDS the best surfactant among the four.

RHA-NP.

Following the foamability test, SDS was chosen to proceed with the foam stability test. Foam Stability Test Effect of RHA-NP Concentration on the Foam Stability Performance The stability of SDS foam was first tested without the addition of RHA-NP. It acted as the control of the experiment. Then, RHA-NP was added and varied with 5 concentrations from 0.1 wt% to 0.9 wt%. RHA-NP were added into 80 mL of DI water accordingly to the weight concentration as shown in the Table 6 below:

Concentration (wt%)

Weight (g)

0.1

0.08

0.3

0.24

0.5

0.40

0.7

0.56

0.9

0.72

RHA-NP

SiO2-NP

Increment

(min)

percentage (%)

0.00

0.30

0.70

Nano-

0.1

125

128

2.40

0.3

128

130

1.56

0.5

132

134

1.52

0.7

140

142

1.43

0.9

145

154

6.21

ced by RHA-NP.

162.5

SDBS

The commercial SiO2-NP gives slightly better foam stability with the increment percentage as shown below:

132

Fig.8 Stability performance of SDS foam enhan-

davg (nm)

Tab.5 Height of the surfactant foams with

Non-ionic

Half-)life (minutes

the particles in the solution. Higher value of PDIavg indicates agglomeration. Therefore, that particular surfactant is no longer suitable to be used to disperse RHA-NP. Generally from the results, higher PDIavg also gives higher davg reading. As such can be seen in AOS, it gives the highest values of both PDIavg and davg, thus, it is considered ineffective in dispersing RHA-NP as compared to the other three surfactants.

Chuah Kai Jie, Mohd Zaidi Bin Jaafar

Figure 8 above shows the trend of the stability performance with the addition of RHA-NP. The increase in RHA-NP concentration leads to the increase in the foam half-life. The highest concentration used, i.e. 0.9 wt% successfully enhanced the half-life of the original SDS foam by 18%. It is believed that the hydrophilic nanoparticles increase the stability of the foam lamellas by adsorbing themselves on the liquid phase, thus reducing the drainage time which promotes longer halflife. Further studies on varying the concentrations of RHANP should be done in order to identify the trend beyond 0.9 wt% and to find the optimum concentration of RHANP in stabilizing SDS foam. Comparison between RHA-NP and SiO2-NP in Foam Stabilizing Performance The experiment was repeated using SiO2-NP (without surface modification) with the same concentration used in RHA-NP. The figure below shows the comparison of the foam stability performance between RHA-NP and SiO2NP.

Tab.7 Comparison of RHA-NP and SiO2-NP performances.

With the addition of 0.9 wt%, SiO2-NP only gives an increment in the half-life at around 6.21%. Below 0.9 wt%, the increment is less than 3% for all the concentrations tested. Therefore, it generally shows that RHA-NP is delivering remarkable foam stabilizing ability as compared to commercial SiO2-NP. Effect of RHA-NP on Foam Texture The foams structure and shape distribution were observed and compared for SDS generated foam, SDS with 0.9 wt% RHA-NP at 60s, 3600s, and 8700s. Figures below show the respective snapshots obtained from the foam analyser software.

Comparison between RHA-NP and SiO2-NP 160 140

Half-life (minutes)

125128

128130

132134

0.10

0.50

140142

154 145

0.9

120

Tab.6 The concentrations of RHA-NP used and their respective weight needed.

Fig.7 Observations of foaminess of surfactant foams with RHA-NP.

100 80 60 40 20 0

Nano-RHA

SiO2 NP

Fig.9 Comparison between RHA-NP and SiO2-NP on foam stability performance.

Fig.10 Bubbles texture and size distributions of SDS foam; at 60s.

SPRING /â&#x20AC;&#x2030;2019

Foam Stability Performance Enhanced with Rice Husk Ash Nanoparticles

Fig.11 Bubbles texture and size distributions of SDS and 0.9 wt% RHA-NP foam; at 60s, stable foam with small bubble sizes.

Fig.12 Bubbles texture and size distributions of SDS and 0.9 wt% RHA-NP foam; at 3600s, liquid drainage occurring, foam started to coalesce.

reas weak foam bubbles are polygonal in shape and appear thinner. Thin lamellas are thermodynamically unstable. The foam generated with SDS only without additives were comparatively bigger in size compared to that of RHA-NP added. As the foams disintegrated, showing the bubbles size distribution varies from 0-10 000 px. Over time, more liquid drainage will happen, causing the lamellas to become thinner and the foam become unstable which eventually collapse. The smaller the bubbles sizes, the more stable the foams are. In porous media, smaller bubbles will be able to pass through the pore without any division occurring whereas bigger bubbles will undergo bubble division. It may cause the lamellas to coalesce in the process due to liquid drainage (Ransohoff and Radke, 1988). Effect of Particle Sizes in Foam Stability Performance The foam stability was investigated using RHA with original size as to compare with RHA-NP at the same concentration, i.e. 0.9 wt%. Results show that using 0.9 wt% of original size RHA slightly decreased the half-life. On the other hand, the stability of the foam was higher with RHA-NP as an additives. The result was expected because according to Hunter et al. (2008), the particle size of the additive is one of the key parameters in stabilizing foam.

Chuah Kai Jie, Mohd Zaidi Bin Jaafar

the same procedure. 27 mL of this mixture was injected into the glass column, followed by 3 mL (10 v/v%) of n-hexadecane. The oil was added carefully onto the liquid surface without contacting the glass column surface above the mixture level. The experiment was repeated with same concentration of SiO2-NP. The results (Fig. 15) show that foams stabilized by RHANP and SiO2-NP. Both decreases in their half-life. Bulk foam stability is poor in the presence oil. Positively, the trend increases with the addition of RHA-NP and it managed to increase the half-life of the SDS foam by 20.7%. On the other hand, SiO2-NP performed better than RHA-NP by a merely 4.76%. The characteristics of the detrimental effect of oil on foam has been mentioned by Zanganeh et al. (2009) and it is very complicated. It is however, the effect of oil on foam generation is unknown (Farajzadeh et al., 2012). 120

105

100

Half-)life (minute

110

80 60 40 20 0

0wt% RHA-NP0 .9wt% RHA-NP 0.9wt% SiO2-NP

Half-life (minute)

-life 150 145 140 135 130 125 120 115 110 105

145

Conclusion 123

SDSR

121

HA+SDS

Fig.13 Bubbles texture and size distributions of SDS and 0.9 wt% RHA-NP foam; at 8700s, fragile foam with severe liquid drainage.

The bubble size distribution histogram was plotted beside the image captured. The bubble size of the foam plays an important role in defining a stable foam. Stable and strong foam bubbles are generally spherical in shapes and appear to be denser. Whe-

Fig.15 Effect of oil in bulk foam generation.

0.9wt% NanoRHA+SDS

Fig.14 The effect of additives size on the foam stability performance. .

This work presents the idea of utilizing RHA as an additive to enhance the foam stability used in EOR. Despite high purity SiO2-NP can be effectively derived from RH; using RHA without any chemical processes such as acid leaching can save the costs. Besides, using sustainable waste material also meant

to prevent pollution of the environment. Here are a few conclusions made based on the objectives of the research: ◀◀ RHA-NP behaves as hydrophilic particles in DI water. ◀◀ Dispersion stability test showed that SDS, SDBS and Triton X-100 are good dispersing agent for RHA-NP in DI water. ◀◀ SDS air foam was successfully enhanced by RHA -NP using minor concentration. ◀◀ The performance of the foam stability using RHANP is comparable to that of using commercial hydrophilic SiO2-NP without surface modification. ◀◀ RHA-NP stabilized foam possesses good foam texture qualities. The main highlight of this research is the comparison performance of RHA-NP and commercial hydrophilic SiO2-NP without surface modification. Both solutions were under same salinity conditions and other parameters such as foam quality and concentration. Thus, the result from the comparison shows possibility of using RHA-NP as an alternative source in replacing hydrophilic SiO2-NP. In conclusion, this work serves as a preliminary study on the feasibility of using RHA-NP as an additive to stabilize air foam. It is however, further studies have to be done in order to find a better dispersing agent for RHA-NP in high salinity condition, and also to identify the optimum concentration of RHA-NP in stabilizing surfactant foam. Acknowledgements The authors are grateful towards the financial support from the Malaysian Ministry of Higher Education, Universiti Teknologi Malaysia (grant references vote number: 4F954).

References: Effect of Oil on the Bulk Foam Generation The experiment was run with RHA-NP in the presence of 10.0 v/v% n-hexadecane, C-16 (Acros Organics, 99% pure). Mixture of 0.9 wt% RHA -NP and surfactant solution was prepared using

[1] Abu El Ela, Mahmoud et al. “Thermal heavy‐oil recovery projects succeeded in Egypt, Syria“ OGJ, Volume 106, Issue 48, December 22, 2008. [2] Abu El Ela, M., Sayyouh, H., and El Tayeb, E. S. (2014). An Integrated Approach for the Application of the Enhanced Oil Recovery Projects. Journal of Petroleum Science Research, 3(4), 176. doi.org/10.14355/ jpsr.2014.0304.03.

SPRING / 2019

Foam Stability Performance Enhanced with Rice Husk Ash Nanoparticles

[3] Belhaij A, Mahdy OA (2015) Foamability and Foam Stability of Several Surfactants Solutions: The Role of Screening and Flooding. J Pet Environ Biotechnol 6:227.doi:10.4172/2157-7463.1000227 [4] Binks, B. P., Campbell, S., Mashinchi, S., & Piatko, M. P. (2015). Dispersion Behaviour and Aqueous Foams in Mixtures of a Vesicle-Foaming Surfactant and Edible Nanoparticles. Langmuir, 31 (10), 2967-2978. [5] Borchardt, J. K., Bright, D. B., Dickson, M. K., and Wellington, S. L. (1988). Surfactants for Carbon Dioxide Foam Flooding. In Surfactant-Based Mobility Control (Vol. 373, pp. 163-180–8). American Chemical Society. doi.org/doi:10.1021/bk-1988-0373.ch008. [6] B. P. Binks, T. S. Horozov, Angew. Chem. 2005, 117, 3788 –3791. [7] Cui, Z. G., Cui, C. F., Zhu, Y., & Binks, B. P. (2011). Multiple phase inversion of emulsions stabilized by in situ surface activation of CaCO3 nanoparticles via adsorption of fatty acids. Langmuir, 28 (1), 314-320. [8] Cui, Z. G., Cui, Y. Z., Cui, C. F., Chen, Z., & Binks, B. P. (2010). Aqueous foams stabilized by in situ surface activation of CaCO3 nanoparticles via adsorption of anionic surfactant. Langmuir, 26 (15), 12567-12574. [9] Esmaeilzadeh P, Hosseinpour N, Bahramian A, Fakhroueian Z, Arya S. (2014). Effect of ZrO2 nanoparticles on the interfacial behavior of surfactant solutions at air–water and n-heptane–water interfaces. Fluid Phase Equilibria. 361(0):289-95. [10] Espinosa, D., Caldelas, F., Johnston, K., Bryant, S., & Huh, C. (2010, April). Nanoparticle-stabilized supercritical CO2 foams for potential mobility control applications. SPE 129925, Proceeding of the SPE Improved Oil Recovery Symposium, Tulsa. [11] Farajzadeh, R., Andrianov, a., Krastev, R., Hirasaki, G. J., and Rossen, W. R. (2012). Foam-oil interaction in porous media: Implications for foam assisted enhanced oil oil recovery. Advances in Colloid and Interface Science, 183–184, 1–13. doi.org/10.1016/j.cis.2012.07.002. [12] Farzaneh, S. A., and Sohrabi, M. (2013). A Review of the Status of Foam Applications in Enhanced Oil Recovery. In EAGE Annual Conference and Exhibition. [13] Floride, A. G., (2013, September). Oil and Gas Financial Journal. 8-11. [14] Flower, S., (2017, June 06). Conventional exploration – playing devil’s advocate. Retrieved from https://www. woodmac.com/news/the-edge/conventionalexploration-devils-advocate/ [15] Friedmann, F. and Jensen, J. A. (1986). Some Parameters Influencing the Formation and Propagation of Foam in Porous Media. Paper SPE 15087. [16] Hirasaki, G.J. (1989). Supplement to SPE 19505, The Steam-Foam Process—Review of Steam-Foam Process Mechanisms. Society of Petroleum Engineering. [17] Hunter, T. N., Pugh, R. J., Franks, G. V. et al. 2008. The role of particles in stabilising foams and emulsions. Advances in Colloid and Interface Science, 137(2): 57–81. doi.org/10.1016/j.cis.2007.07.007. [18] J. Wong, S. K., (2017). Fly Ash Nanoparticles Enhance Foam Stability Performance in SDS-Stabilized Foam. [19] Kaptay, G., and Babcsán, N. (2012). Particle Stabilized Foams. Foam Engineering: Fundamentals and Applications, 2000, 121–143. doi.org/10.1002/9781119954620.ch7. [20] Khajehpour, M., Etminan, S. R., Goldman, J., Wassmuth, F., and Technology, A. I. (2016). Nanoparticles as Foam Stabilizer for Steam-Foam Process. In SPE EOR Conference at Oil and Gas West Asia (p. 14). [21] Lake, L. W. (1989). Enhanced Oil Recovery. Prentice Hall. [22] Lee, D., Cho, H., Lee, J., Huh, C., & Mohanty, K. (2015). Fly ash nanoparticles as a CO2 foam stabilizer. Powder Technology, 283, 77-84. [23] Ogolo, N. A., Olafuyi, O. A., and Onyekonwu, M. O. (2012). Enhanced Oil Recovery Using Nanoparticles. In SPE Saudi Arabia Section Technical Symposium and Exhibition. [24] Pode, R. (2016). [25] Potential applications of rice husk ash waste from rice husk biomass power plant. Renewable and Sustainable Energy Reviews, 53, 14681485. doi:10.1016/j.rser.2015.09.051. [26] R.D. Handy, R. Owen, E. Valsami-Jones, The ecotoxicology of nanoparticles and nanomaterials: current status knowledge gaps, challenges, and future needs, Ecotoxicology 17 (2008) 315–325. [27] Ransohoff, T. ., and Radke, C. (1988). Laminar flow of a wetting liquid along the corners of a predominantly gas

Chuah Kai Jie, Mohd Zaidi Bin Jaafar

-occupied noncircular pore. Journal of Colloid and Interface Science, 121(2), 392–401. doi.org/10.1016/00219797(88)90442-0. [28] Singh, R., Gupta, A., Mohanty, K. K., Huh, C., Lee, D., & Cho, H. (2015, September 28). Fly Ash Nanoparticle-Stabilized CO2-in-Water Foams for Gas Mobility Control Applications. Society of Petroleum Engineers. doi:10.2118/175057-MS. [29] Soltani, N., Bahrami, A., Pech-Canul, M.I., & González, L.A. (2015). [30] Review on the physicochemical treatments of rice husk for production of advanced materials. Chemical Engineering Journal, 264, 899-935. [31] Tan, X. K., (2017). Nanosilica-Stabilized Supercritical Carbon Dioxide Foam for Enhanced Oil Recovery Application. [32] Thuadaji, N., & Nuntiya, A. (2008). Preparation of nanosilica powder from rice husk ash by precipitation method. Chiang Mai Journal of Science, 35 (1), 206-211. [33] World Rice Statistics, International Rice Research Institute(IRRI);2014 http://ricestat.irri.org:8080/ wrs2/entrypoint.htm. [34] Worthen, A. J., Bagaria, H. G., Chen, Y., Bryant, S. L., Huh, C., & Johnston, K. P. (2013). Nanoparticlestabilized carbon dioxide-in-water foams with fine texture. Journal of colloid and interface science, 391, 142-151. [35] Yin, G., Grigg, R. B., and Svec, Y. (2009). Oil Recovery and Surfactant Adsorption During CO2-Foam Flooding. In 2009 Offshore Technology Conference. doi.org/10.4043/19787-MS. [36] Zanganeh, M. N., Kam, S. I., LaForce, T. C., and Rossen, W. R. (2009). The Method of Characteristics Applied to Oil Displacement by Foam. In SPE EUROPEC/EAGE Annual Conference and Exhibition. [37] Zhang, S., Sun, D., Dong, X., Li, C., & Xu, J. (2008). Aqueous foams stabilized with particles and nonionic surfactants. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 324 (1), 1-8. [38] Zhang, T., Espinosa, D. A., Yoon, K. Y., Rahmani, A. R., and Yu, H. (2011). Engineered Nanoparticles as Harsh-Condition Emulsion and Foam Stabilizers and as Novel Sensors. In Offshore Technology Conference (pp. 1–15). [39] Zhu, T., Strycker, A., Raible, C. J., and Vineyard, K. (1998). Foams for Mobility Control and Improved Sweep Efficiency in Gas Flooding. Most, 277–286. doi.org/10.2118/39680-MS.

Authors Information and Contact Details Chuah Kai Jie is currently a final-year student pursuing Bachelor of Petroleum Engineering in the Universiti Teknologi Malaysia. His current interests include reservoir simulation, enhanced oil recovery, and applied nanotechnology in EOR. Department of Petroleum Engineering, School of Chemical and Energy Engineering, Faculty of Engineering, Universiti Teknologi Malaysia. Email : chuahkj15@spe.petroleum.utm.my Contact : +60125472552

Ir Dr Mohd Zaidi Bin Jaafar received his BSc. Petroleum Engineering in University Missouri Rolla, USA, MSc. and PhD in Petroleum Engineering in Imperial College London. He is currently a senior lecturer in Department of Petroleum Engineering, Faculty of Chemical and Energy Engineering, Universiti Teknologi Malaysia. His current interests are downhole monitoring, electrokinetics, smart well and EOR.

Department of Petroleum Engineering, School of Chemical and Energy Engineering, Faculty of Engineering, Universiti Teknologi Malaysia. Email : mzaidi@utm.my Contact : +60162642573 SPRING / 2019

Chuah Kai Jie, Mohd Zaidi Bin Jaafar62