Hydrocarbon Engineering February 2022

Page 1

February 2022

SK-501 Flex™ high temperature shift catalyst

Innovative new catalyst for hydrogen and ammonia plants The innovative SK-501 Flex™ HTS catalyst gives operators an opportunity to choose a more sustainable catalyst and at the same time improve the operation of the plant. A totally new catalyst formulation – the first in more than a century - this is the future choice for your HTS reactor.

SK-501 Flex™: • •

100% chromium free - no risk of Cr6 exposure during loading and handling Unmatched operational flexibility (operation at much lower steam to carbon ratios) - energy efficiency and/or production rate increase Exceptional activity – increased conversion rates

Scan the code or go to topsoe.com/products/catalysts/sk-501-flextm


CONTENTS February 2022 Volume 27 Number 02 ISSN 1468-9340

03 Comment

39 Small pipes, large benefits: gas measurement in tight spaces

05 World news 08 Australasia: oil market pre-existing conditions Dr. Nancy Yamaguchi, Contributing Editor, discusses how pre-existing conditions in Australasia, coupled with the effects of the COVID-19 pandemic, have left the region’s petroleum market facing a number of challenges.

17 In a world made of sulfur A.K. Tyagi, Nuberg EPC, India, explores the crucial role of sulfuric acid within the oil and gas industry, and places a spotlight on two projects that the company was involved in, in Saudi Arabia and Egypt.

21 The sulfur level challenge Richard G. Stambaugh, Merichem Company, USA, discusses the available technologies for lowering refineries’ sulfur emissions in order to meet tightening regulations.

25 Finding the right fit Peter Foith, CS Combustion Solutions, Austria, outlines the challenges faced when retrofitting an existing thermal stage of a sulfur recovery unit.

29 Maintaining sample heat integrity Rod Merz, AMETEK Process Instruments, Canada, explores the most common analytical heat integrity failure points seen with analysers in sulfur recovery units.

35 Delivering high-quality PTA Rhys Jenkins, Servomex, UK, considers the role of gas analysis in purified terephthalic acid (PTA) production.




Hydrocarbon Engineering

join Hydrocarbon Engineering

Tyler Schertz, Mettler Toledo, Switzerland, details the process of measuring small pipelines with tunable diode laser (TDL) using a multi-reflection folded optical path.

43 Gas recovery Marco Puglisi, AEREON Europe, Italy, examines applications, equipment and technical solutions for gas recovery packages in the oil and gas sector.

48 Sparking a safety culture Alec Cusick, Owens Corning, USA, explores the various considerations when designing a passive fire protection system.

54 The global hydrogen transition Rathishkumar Sukumar and Raghavendra Mahalingam, Baker Hughes, discuss the role that specialty valves will play in delivering affordable hydrogen energy in the quest for net zero emissions.

59 Calculating the cost of ownership Alejandro Plazas, ValvTechnologies, Latin America, explains why considering the total cost of ownership of valves will ultimately pay big dividends.

63 The challenges and opportunities of virtual engineering Carina Wegener, REMBE GmbH, Germany, discusses how to achieve the right installation torque with a virtual calculation.

65 Context-based predictive diagnosis A. Brighenti, C. Brighenti, M. Ricatto, and D. Quintabà, S.A.T.E. Systems and Advanced Technologies Engineering S.r.l., Italy, outline how to monitor and diagnose multiple plant units while considering operational contexts.

Safety is the top priority in any system design. In Owens Corning’s article (p. 48), discover considerations and approaches when selecting insulation and designing passive fire protection systems. Learn two common test standards, how to evaluate insulation properties, and why material matters.

2022 Member of ABC Audit Bureau of Circulations

Copyright© Palladian Publications Ltd 2022. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the copyright owner. All views expressed in this journal are those of the respective contributors and are not necessarily the opinions of the publisher, neither do the publishers endorse any of the claims made in the articles or the advertisements. Printed in the UK. CBP006075

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



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

callum.oreilly@palladianpublications.com EDITORIAL ASSISTANT Bella Weetch

bella.weetch@palladianpublications.com SALES DIRECTOR Rod Hardy

rod.hardy@palladianpublications.com SALES MANAGER Chris Atkin

chris.atkin@palladianpublications.com SALES EXECUTIVE Sophie Barrett

sophie.barrett@palladianpublications.com PRODUCTION Kyla Waller

kyla.waller@palladianpublications.com DIGITAL EVENTS COORDINATOR Louise Cameron

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

molly.bryant@palladianpublications.com DIGITAL ADMINISTRATOR Lauren Fox

lauren.fox@palladianpublications.com ADMIN MANAGER Laura White

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

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

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

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



or several months, ‘Partygate’ has been dominating the headlines here in the UK. Hardly a day goes by without fresh accusations of illegal gatherings within 10 Downing Street – the home and office of Prime Minister Boris Johnson – while COVID-19 restrictions were in place. From Christmas parties and leaving drinks, to birthday celebrations and ‘cheese and wine’ events, it seems that those in charge of making social distancing laws during the pandemic may have been wilfully breaking them at the same time. An internal inquiry is ongoing, and the Metropolitan police has launched a criminal investigation into the matter. Needless to say, the allegations have caused quite a storm, while also creating a worrying distraction from a number of major issues that are developing both domestically and internationally. At home, there is a cost of living crisis as the rate of inflation hits a 10-year high, driven by the rising cost of fuel and energy. And abroad, rising tension between Ukraine and Russia is causing anxiety across the world. In a recent report, Wood Mackenzie suggested that two key determinants of how gas prices will trend in 2022 are the timing of the start-up of Nord Stream 2 and winter weather dynamics.1 The company’s analysis suggests that at current levels of Russian exports and considering normal weather conditions, European storage inventories will fall to a record low of below 15 billion m3 by the end of March. While prices will eventually come down as the winter ends, requirements to refill storage facilities are expected to be 20 – 25 billion m3 more than last year. Wood Mackenzie suggests that the commissioning of Nord Stream 2 could be the only option to refill storage. However, the company’s Vice President, Massimo Di Odoardo, warns that commissioning of Nord Stream 2 could be stopped altogether if tensions between Russia and Ukraine escalate. He adds: “A cold winter in Europe and Asia, alongside continued uncertainty about commissioning of Nord Stream 2, could see prices increase further throughout 2022.” In its report, Wood Mackenzie notes five other trends to watch out for this year in the global gas and LNG industry. Firstly, the level of oil indexation in long-term LNG contracts is expected to rise, potentially reaching 12% on a weighted average basis. There is also increased momentum behind LNG projects, with 79 million tpy of additional LNG expected to take final investment decision (FID) over the next two years. Thirdly, the LNG sector is expected to focus on CO2 reduction across the value chain going forward, rather than carbon-offsetting. However, the report notes that more capital-intensive projects, including the use of low-carbon power and/or carbon capture and storage (CCS), remain at an evaluation stage. Another trend is likely to see global gas demand remaining resilient in the short-term, although the role of gas in the energy transition will come under pressure as prices remain high. Finally, Wood Mackenzie believes that the use of unabated natural gas in the EU is set to decline, even if the EU classifies investments in gas-fired plants as transitional investments. It’s clear that the road ahead for gas and LNG remains rocky, and one to keep a close eye on over the coming months. 1.

‘Global gas and LNG – 6 things to watch for in 2022’, Wood Mackenzie, (January 2022).


For the most absorbent polymers, the path to purity is crystal clear Fluid engineering makes the difference. How do you combine the highest purity with the highest yield? For manufacturers of superabsorbent polymers, the answer is falling film melt crystallization. First pioneered in 1989, this unique crystallization solution now produces 1.8 million tons of glacial acrylic acid every year, or 80% of total world production. That’s billions of diapers, personal care products and more, all produced with 99.9% purity to provide the high-performance absorption we rely on every day. Learn how smarter crystallization can make a difference for your manufacturing operations today at sulzer.com.

Learn why acrylic acid manufacturers worldwide rely on Sulzer’s innovative solvent-free melt crystallization


fuel production set to rise


fter declining in 2020, the combined production of US fossil fuels increased by 2% in 2021 to 77.14 quadrillion Btu. Based on forecasts in the US Energy Information Administration’s (EIA) latest ‘Short-Term Energy Outlook’ (STEO), US fossil fuel production is expected to continue rising in both 2022 and 2023, surpassing production in 2019, to reach a new record in 2023. Of the total US fossil fuel production in 2021, dry natural gas

accounted for 46%, crude oil accounted for 30%, coal 15%, and natural gas plant liquids (NGPLs) 9%. The EIA expects those shares to remain similar through 2023. US dry natural gas production increased by 2% in 2021, based on monthly data through October and estimates for November and December, while US crude oil production dropped slightly, by an estimated 1%. However, the EIA expects crude oil production to increase by 6% in 2022 and 5% in 2023.

Poland | SABIC

signs MoU with Saudi Aramco and PKN Orlen


ABIC has signed a Memorandum of Understanding (MoU) with Saudi Aramco and PKN Orlen for the exploration of potential petrochemical growth or expansion projects in Poland and Central and Eastern Europe. The agreement will enable several potential collaboration and investment opportunities to be studied, including a new chemical production facility in Poland, the expansion of several existing assets and the development of a new world-scale cracker.

Abdulrahman Al-Fageeh, SABIC’s Executive Vice President for Petrochemicals, said: “By bringing together the scale, expertise and technologies of three world-leading companies, this MoU enables us to identify and assess opportunities for ambitious and sustainable growth.” Upon completion of the exploration phase, and if the parties agree to pursue any potential petrochemical project, they will enter into a separate project joint development agreement (JDA).

USA | Shell

Oil Co. sells interest in Deer Park refinery to Pemex hell Oil Co. has completed the SRefining sale of its interest in Deer Park Ltd Partnership, a 50-50 joint

venture (JV) between Shell Oil Co. and P.M.I. Norteamerica S.A. De C.V. (a subsidiary of Petroleos Mexicanos, or Pemex), for US$596 million. The agreement covers the sale of Shell’s 50.005% interest in the partnership, and therefore transfers full ownership of the refinery to Pemex. Shell Chemical L.P. will continue to operate its 100%-owned Deer Park Chemicals facility, located adjacent to the site. “The completion of this sale marks the start of a new chapter of our history in Deer Park as we have worked closely with Pemex over the past few months to ensure a safe and responsible handover of operations for the refinery”, said Huibert Vigeveno, Shell’s Downstream Director. As part of its ‘Powering Progress’ strategy, Shell plans to consolidate its refinery footprint to five core energy and chemicals parks. These locations will maximise the integration benefits of conventional fuels and chemicals, and offer future potential hubs for sequestration.

UAE | TotalEnergies,

Masdar and Siemens Energy team up to produce SAF from hydrogen


asdar, Siemens Energy and TotalEnergies signed an agreement on the sidelines of Abu Dhabi Sustainability Week (ADSW) 2022 to act as co-developers for a demonstrator plant project, which will be established at Masdar City, Abu Dhabi. Having joined the initiative, the aim is for TotalEnergies to offer its

expertise in sustainable aviation fuel (SAF) production, offtake and supply to partner airlines. Since January 2021, the partners in the initiative have completed a range of evaluations on technology suppliers, feasibility studies and conceptual designs, while working with regulators on compliance issues.

The aim is to proceed to the FEED stage later in 2022. By leveraging their areas of expertise and their local and global market reach, the co-developers believe that the demonstrator project can pave the way to commercial production of SAF, helping to reduce production costs and making it commercially-viable. HYDROCARBON



February 2022

WORLD NEWS DIARY DATES 06 - 10 March 2022 AMPP Annual Conference + Expo San Antonio, Texas, USA ace.ampp.org

13 - 15 March 2022 AFPM Annual Meeting New Orleans, Louisiana, USA www.afpm.org/events

13 - 15 April 2022 24th Annual International Aboveground Storage Tank Conference & Trade Show Orlando, Florida, USA www.nistm.org

09 - 11 May 2022 Sulphur World Symposium Tampa, Florida, USA www.sulphurinstitute.org/symposium-2022

09 - 13 May 2022 RefComm Galveston, Texas, USA www.events.crugroup.com/refcomm

23 - 25 May 2022 StocExpo Rotterdam, the Netherlands www.stocexpo.com

23 - 27 May 2022 World Gas Conference Daegu, South Korea www.wgc2022.org

24 - 26 May 2022 Asia Turbomachinery & Pump Symposium Kuala Lumpur, Malaysia atps.tamu.edu

07 - 09 June 2022 Global Energy Show Calgary, Alberta, Canada www.globalenergyshow.com

08 - 09 June 2022 Downstream USA 2022 Houston, Texas, USA www.reutersevents.com/events/downstream

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

February 2022




Europe | BASF

to expand European production of specific petrochemicals


ASF has decided to build a new hexamethylene diamine (HMD) plant in Chalampé, France. The plant is set to increase the comany’s HMD production capacity to 260 000 tpy. Production is expected to start in 2024. Furthermore, the company will expand its polyamide 6.6 production in Freiburg, Germany, starting in 2022. The planned investments will further expand the polyamide 6.6

business that BASF acquired from Solvay in 2020. “With this new HMD plant in Chalampé and the expansion of the polymerisation in Freiburg, BASF ensures that customers can be reliably supplied with HMD and PA6.6, while also addressing increasing demand in the market,” said Dr Ramkumar Dhruva, President of BASF’s monomers division.

USA | Texas

LNG and Enbridge execute pipeline agreement


exas LNG Brownsville LLC, a controlled subsidiary of Glenfarne Group LLC, and Enbridge Inc. have executed a pipeline transportation precedent agreement for the expansion of the Valley Crossing Pipeline (VCP) to deliver approximately 720 million ft3/d of natural gas to Texas LNG’s export facility for a term of at least 20 years. Texas LNG is developing a 4 million tpy LNG export terminal in the Port of Brownsville, South Texas.

VCP consists of a 160-mile 42 and 48 in. dia. pipeline originating at Agua Dulce and extending to the Port of Brownsville. A 10-mile lateral will be built to extend the pipeline to Texas LNG’s facility, along with the addition of compression facilities on the existing pipeline. VCP’s pipeline header interconnects with 10 major gas pipeline systems, providing access to competitively-priced gas from the Permian and other major gas basins.

Worldwide | ExxonMobil

announces ambition for net zero emissions


xxonMobil has announced its ambition to achieve net zero greenhouse gas (GHG) emissions for operated assets by 2050, backed by a comprehensive approach to develop detailed emission-reduction roadmaps for major facilities and assets. The net zero ambition is contained in the company’s Advancing Climate Solutions - 2022 Progress Report. The net zero aspiration applies to Scope 1 and Scope 2 GHG emissions and builds on

ExxonMobil’s 2030 emission reduction plans, which include net zero emissions for Permian Basin operations and ongoing investments in lower-emission solutions in which it has extensive experience, including carbon capture and storage (CCS), hydrogen and biofuels. To help reach this goal, ExxonMobil has identified more than 150 potential steps and modifications that can be applied to assets in its upstream, downstream and chemical operations.

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

stable by nature

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



he COVID-19 pandemic is having monumental impacts on the global petroleum business. Momentarily in 2021, it seemed that the worst was over, but these hopes have been dashed as the year 2022 begins. Various levels of response throughout the world are cutting into economic activity, including in China, where lockdowns can cause serious knock-on effects across

February 2022




global supply chains. Cases are spiking around the world, led by the Omicron variant. In the Americas, rapid increases are being seen in the US, Mexico, Brazil, Colombia and Argentina. In Europe, cases are essentially rising in all countries, with surges seen in the UK, France, Italy, Spain, Greece, Germany and the Netherlands. In Asia, cases are rising sharply in Turkey, India, Vietnam and the Philippines. In Australasia, Australia’s

Nancy Yamaguchi, Contributing Editor, discusses how pre-existing conditions in Australasia, coupled with the effects of the COVID-19 pandemic, have left the region’s petroleum market facing a number of challenges.

situation is particularly heartbreaking because it had been so successful in its earlier efforts to deal with the virus. On 9 January 2022, confirmed cases topped 1 million. As recently as September 2021, cases had been below 100 000, and the nation appeared to have achieved a strong level of control over the virus. Infections in neighbouring countries also are rising, including the key Oceania countries of French Polynesia,

Fiji, New Caledonia, New Zealand and Papua New Guinea. These are major increases, but they pale in comparison to the situation in Australia. Oceania is a vast area, encompassing a variety of far-flung island nations dominated by Australia, New Zealand and Papua New Guinea. This has been a mixed blessing with regards to COVID-19, since many governments have found it simpler to protect their people HYDROCARBON



February 2022

Figure 1. Australian petroleum product sales, monthly bpd,

January 2019 – September 2021 (source: Australian Department of Industry, Science, Energy and Resources).

when quarantining smaller populations in an island setting. On the other hand, the region depends heavily on air transportation, and many of the countries suffer disproportionately when flights are cancelled and tourism slumps. Many types of businesses and industries face expansion-contraction cycles. These cycles have often been dramatic in the petroleum industry. In Australasia, however, there are underlying, long-term trends. In parlance that reflects current world events, these can be named as the following ‘pre-existing conditions’: nn Declining output from conventional oilfields. nn Weak demand. nn A shrinking refinery presence. nn Increasing regulatory costs. The COVID-19 pandemic did not cause these conditions, but it is accentuating them. This article examines the current situation in Australasia’s oil sector and highlights the impacts of the COVID-19 pandemic on oil refining and product trade in the region.

Pre-existing condition: weak demand

Figure 2. New Zealand oil product demand, recent quarters, ‘000 bpd (source: New Zealand Ministry of Business, Innovation and Employment).

Figure 3. Australia and New Zealand international aviation fuel sales, million l/month (sources: Australian Department of Industry, Science, Energy and Resources, and New Zealand Ministry of Business, Innovation and Employment).


The pandemic sharply reduced oil demand, and this has caused the industry a great deal of hardship. However, the pandemic was not the only cause of the decline in Australasia. Oil demand had already been stagnating, and most forecasts expected little or even negative growth. Gasoline demand was expected to continue on a downward trend. According to the Australian Department of Industry, Science, Energy and Resources, gasoline demand in the country fell from approximately 326 200 bpd in financial year (FY) 2010 – 2011 to 302 800 bpd in FY 2018 – 2019.1 The pandemic then slashed demand to 275 800 bpd in 2020 – 2021. Figure 1 presents the trend in Australia’s monthly oil product sales from January 2019 through to September 2021. Until the pandemic, sales had been roughly flat, at approximately 5000 million l/month. COVID-19 lockdowns caused a sharp drop in sales. Between March and April 2020, Australian gasoline sales dropped by 565.3 million l, kero/jet sales fell by 408.5 million l, and diesel sales fell by 484.4 million l. Sales crept back up, hitting a peak in May 2021 and tapering down again since then. The pandemic had the mildest impact on diesel sales, which bounced

Table 1. Author’s assessment of Australasian refinery capacity (‘000 bpd) Country






Cat ref

Alkyl/ poly










































Vitol Viva4











Marsden Point










Port Moresby



















New Zealand



Papua New Guinea Puma Energy6 Total

Notes: 1. Converted to product terminal in 2021; 2. Closed for extended maintenance during the COVID-19 pandemic, received government subsidy; 3. Closure scheduled in 2022; 4. Reduced throughput, received government subsidy; 5. Refinery will close in April 2022, Refining NZ will become Channel Infrastructure NZ; 6. Formerly InterOil.

back quickly, but sales of gasoline and aviation fuels have not fully recovered. Figure 2 shows a similar pattern in New Zealand’s fuel demand. Prior to the pandemic, quarterly sales were stable, at roughly 135 000 bpd. Demand fell to approximately 95 000 bpd in 2Q20. By 4Q20, demand appeared to be in recovery, but flattened during 1Q21 and 2Q21 and then declined to approximately 116 000 bpd in 3Q21. New Zealand’s COVID-19 response is internationally recognised as one of the most successful in the world, but by late 2021, cases were rising even there. The prospect of another COVID-19 surge in 2022 is one of the most significant threats to demand recovery. In late 2021, the Christmas holiday season brought with it a steep increase in infections, led by the new Omicron variant. Many people believed that the pandemic was either over, or over-rated, and were determined to travel. However, the spike in infections prompted a number of flight cancellations, among other responses. Figure 3 shows how the first wave of lockdowns in early 2020 caused a crash in aviation fuel sales in Australia and New Zealand. In March 2020, both countries closed their borders to non-resident travellers. In April 2020, sales of international aviation fuel in both countries had collapsed to approximately 23% of their levels in February 2020. As the chart illustrates, aviation fuel sales have climbed back marginally in Australia, and not at all in New Zealand. The official data currently extends to September 2021, and it is possible that 2022 will bring additional cuts in travel that could erode demand even further.

Pre-existing condition: declining crude production Another pre-existing condition is the decline in conventional oil production, which has forced local refineries to increase reliance on imported crude oil. According to the Australian Department of Industry, Science, Energy and Resources, during the decade from FY 2010 – 2011 to FY 2020 – 2021, Australian crude production was cut by more than half, falling from 297 000 bpd in FY 2010 – 2011 to 126 000 bpd in FY 2020 – 2021.1 This equated to a rate of decline of 8.2%/yr. Increases in condensate production helped February 2022 12 HYDROCARBON ENGINEERING

replace some of the lost crude volume, but the condensates were often remote and were not as well-suited as refinery feedstock for transport fuel production. In FY 2020 – 2021, indigenous hydrocarbons accounted for just 29.1% of Australia’s refinery inputs. The New Zealand Ministry of Business, Innovation and Employment reported that New Zealand crude and condensate production peaked at 61 100 bpd in 2008, then fell to 26 800 bpd in 2020.2 This was a decline of 7.4%/yr on average. Imports of crude and refinery feedstocks in 2020 amounted to 83 100 bpd – over three times as much as domestic production. In Oceania, only Papua New Guinea is producing enough crude and condensate to feed its small refinery (approximately 38 000 bpd of production vs a refinery size of 32 500 bpd). In general, when the cost of foreign refinery feedstocks rises more quickly than the value of refined products, refinery profits are squeezed. This situation had been developing in Australasia for years, and had already been forcing refining companies to rethink their operations. Some refineries changed hands, some closed, others waited and watched the market. The COVID-19 pandemic did not cause this situation, but it caused market upheaval, thereby accelerating decision making.

Pre-existing condition: Australasian refining in retreat The pre-existing conditions of weak demand and declining availability of inexpensive refinery feedstocks contributed to a lengthy retreat from refining in Australasia. As before, COVID-19 did not cause the shrinkage of the refining industry, but hastened it. Table 1 presents the author’s assessment of Australasian refinery capacity in 2021. There have been notable changes in recent years, with more on the horizon. In Australia, Caltex closed its Kurnell refinery at the end of 2014; BP closed its Bulwer Island refinery in 2015; Shell announced that it would convert its Clyde refinery to a product terminal in 2013. Additionally, Shell sold its Geelong refinery to the oil trading company Vitol in 2014; BP’s Kwinana refinery was converted to a product terminal in 2021; ExxonMobil announced that it would close its

2020 – 2021. In FY 2010 – 2011, refineries produced 716 000 bpd of finished products. This fell to 490 000 bpd in 2015 – 2016 following the closure of two refineries, then fell again to 376 000 bpd in 2020 – 2021 in response to the COVID-19 pandemic. Australian refinery output declined at a rate averaging 6.2%/yr over the decade. In 2018 – 2019, Australian refinery output was 3.3% LPG, 38.3% gasoline, 13.4% kero/jet, 31% diesel and 14% fuel oil and other products. In 2020 – 2021, refineries shifted output to 2.7% LPG, 39.8% gasoline, 3.9% kero/jet, 37.4% diesel and 16.2% fuel oil and other products. Kero/jet output of 68 000 bpd in 2018 – 2019 was cut to just 14 000 bpd in 2020 – 2021. Because New Zealand only has one refinery, refinery output had been relatively stable until the COVID-19 pandemic forced companies to reduce utilisation. According to the New Zealand Ministry of Business, Innovation and Employment, fuel production of 101 000 bpd in 2000 rose gently to 113 000 bpd in 2019 before the COVID-19 pandemic hit.2 In 2020, New Zealand’s refinery production fell to 81 000 bpd. Based on data for the first three quarters of 2021, New Zealand’s refinery output continued to decline, averaging 79 000 bpd. As noted, NZ Refining plans to close the refinery in April, 2022. If the refinery operates during 1Q22 at a level consistent with historic patterns, New Zealand’s refinery output could average in the range of 20 000 – 28 000 bpd for the year 2022, and it will fall to zero thereafter. In 2019, New Zealand’s refinery yield pattern was 29% gasoline, 35% diesel, 25% kero/jet and 11% fuel oil and other products. In 2020, the refinery shifted its output Figure 4. Declining refinery capacity in Australasia, 1990 – 2023 pattern to 33% gasoline, 43% diesel, 17% (estimated), ‘000 bpd (source: author’s own). kero/jet and 7% fuel oil and other products. Kero/jet production was cut nearly in half from 29 000 bpd in 2019 to 14 000 bpd in 2020. Altona refinery in 2022. Refining NZ will also close its refinery in 2022, leaving only two refineries in Australia, as well as the hydroskimming refinery in Papua New Guinea. Figure 4 displays the trend in Australasian crude refining capacity. Capacity peaked at 933 000 bpd in 2002. PNG built its refinery in 2004, partly offsetting capacity loss in Australia. Australia’s closures caused regional capacity to fall to approximately 611 000 bpd in 2015. Based on the planned dates of the current wave of closures, the author estimates 2021 capacity at 503 000 bpd, falling to 293 000 bpd in 2022 and 260 000 bpd in 2023. Australasian crude refining capacity in the year 2023 will be roughly one-quarter of what it was in 2002. Figure 5 shows the impact of Australia’s refinery capacity closures on refined product output between 2010 – 2011 and

Pre-existing condition: rising fuel import dependence

Figure 5. Australian refinery output, ‘000 bpd (source: Australian Department of Industry, Science, Energy and Resources).


When Australian refineries began to reduce capacity, the country started to increase its reliance on imported fuels. Figure 6 shows the growth in product imports. In 2010 – 2011, product imports totalled 300 000 bpd. Imports more than doubled by 2017 – 2018, reaching 626 000 bpd that year. However, imports then flattened, mainly because demand fell in response to the pandemic. Imports averaged 623 000 bpd in 2020 – 2021, and have grown at an average rate of 7.6%/yr over the decade. The fastest rates of growth were for diesel (9.9%/yr) and gasoline (11.4%/yr).

OGT | ProTreat® brings out













Kero/jet fuel imports had also been rising, but were sharply cut after the pandemic hit. Jet fuel imports of 106 000 bpd in 2017 – 2018 fell to 46 000 bpd in 2020 – 2021. Imports of fuel oil and LPG have declined. Australia is especially dependent on diesel imports, and the pandemic did not change this. In 2020 – 2021, diesel imports rose to 392 000 bpd, up from 152 000 bpd in 2010 – 2011. Australian refineries produced 140 600 bpd of diesel in 2020 – 2021, whereas demand was 520 200 bpd, rendering Australia dependent on imports for approximately 75% of its domestic demand. The COVID-19 pandemic has had a mixed impact on trade. Import dependence would have risen, had demand not fallen. It is logical to expect imports to rise in 2022 as refinery production falls and demand recovers. Figure 7 displays the trend in Australian refined product exports. Exports declined modestly from 2010 to 2015, averaging 85 100 bpd in 2015 – 2016. Exports then rose to 148 800 bpd in 2019 – 2020, before retreating to 127 200 bpd in 2020 – 2021. Most of Australia’s product exports are LPG. As with Australia, New Zealand has gradually become more dependent on imported products. Figure 8 presents the trend in imports by product type from 2000 to 2020, with the year 2021 estimated from the first three quarters. Imports trended up from 26 100 bpd in 2000 to 62 800 bpd in 2018. The pandemic then caused demand to fall, and imports declined to 57 700 bpd in 2020 and 64 100 bpd during the first three quarters of 2021. Diesel was the key import, accounting for 45.9% of imports in 2020 and 44.9% of imports during the first three quarters of 2021. Gasoline accounted for 38.7% of imports in 2020 and 40.7% of imports during this same period. New Zealand exports minor amounts of refined product. Exports of 3000 bpd in 2000 dwindled over the decade. With the sole refinery scheduled to close in April 2022, exports will vanish and imports will be equal to demand, in net terms, though it is always possible that some in-and-out trades will occur.




Conclusion This article has discussed several key trends in Australasia’s oil market, named as ‘pre-existing conditions’. The COVID-19 pandemic did not cause these conditions, but it has accentuated them. First, conventional crude production was waning. This was reducing the supply of local refinery feedstock, raising crude import dependence and squeezing refining margins. Second, oil product demand was fairly flat and gasoline demand was forecast to fall. The pandemic may have caused structural and behavioural change that will keep demand tamped down. Third, the refining sector was already in retreat, but the pandemic eroded profitability even further, hastening the decisions to close two additional refineries in Australia and one in New Zealand. Australia’s two remaining refineries have received government subsidies in order to remain in operation. The government decided

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

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

Figure 6. Australia’s refined product imports, ‘000 bpd (source:

Australian Department of Industry, Science, Energy and Resources).

Figure 7. Australia’s refined product exports, ‘000 bpd (source:

Australian Department of Industry, Science, Energy and Resources).

that the country had need for a ‘sovereign fuel supply’. Refineries will need to make additional investments to produce ultra-low-sulfur fuels. Fourth, Australasia already had been growing increasingly reliant on refined product imports, and imports should be expected to rise as capacity closures take effect. Product exports will be reduced, and in some cases they will vanish. Fifth, refining (and heavy industry in general) must comply with stringent environmental regulations. Australia lagged in producing ultra-low-sulfur gasoline, citing economic difficulties. When the government stepped in to support the refining industry, it also committed to helping Australia’s last two refineries make the investments needed to produce ultra-low-sulfur gasoline with a maximum of 10 ppm sulfur by 2024. Australia is also struggling to reduce carbon emissions. According to BP, Australia’s carbon dioxide (CO2) emissions totaled 370.3 million t in 2020, amounting to 1.2% of the global total. Yet Australia accounts for only 0.33% of the global population. The COVID-19 pandemic has rocked the global oil market. Recovery seemed underway in 2021, but the Omicron variant has eluded the defenses of key Australasian countries. At the outset of the pandemic, Australia and New Zealand were credited with having some of the best and most effective responses, keeping infections and fatalities under control. Now, cases are soaring in Australia and rising in other countries in the region. Most island countries rely heavily on imported fuels and upon trade and tourism. As Australasia sets off into the new year, there are many questions. Have COVID-19 infections peaked? Will renewed lockdowns be needed? Will structural change create a new normal where fossil energy use never returns to pre-pandemic levels? Because Australasia’s oil industry has already endured several stages of trimming and rationalisation, it can only hope that its actions were enough, and that 2022 will bring health and good fortune.


1. The Australian Department of Industry, Science, Energy and Resources, https://www.industry. gov.au/ 2. New Zealand Ministry of Business, Innovation & Employment, https://www.mbie.govt.nz/

Figure 8. New Zealand’s refined product imports, ‘000 bpd (source: Ministry of Business, Innovation & Employment).


Bibliography •

‘Coronavirus Pandemic’, Our World in Data, https://ourworldindata.org/coronavirus

A.K. Tyagi, Nuberg EPC, India, explores the crucial role of sulfuric acid within the oil and gas industry, and places a spotlight on two projects that the company was involved in, in Saudi Arabia and Egypt.


ife on Earth exists partly thanks to sulfur. It is one of the most prevalent chemical elements in nature. It is a pale yellow, tasteless and odourless brittle solid that is common in volcanic areas and hot springs. Its main source is as byproduct elemental sulfur recovered from natural gas and petroleum. Sulfuric acid (H2SO4), one of the most essential components that is utilised as an industrial raw material, is the most important derivative of sulfur. Dilute sulfuric acid, on the other hand, is formed naturally in the atmosphere when sulfur dioxide (SO2) is oxidised in the presence of atmospheric moisture. Sulfuric acid is in high demand as a result of its wide range of applications. Consequently, it is commercially produced through the contact process, which involves the reaction of sulfur trioxide (SO3) with water: SO3 + H2O ⟶



Global production and the growing role of sulfuric acid Sulfuric acid demand has fallen only once, during the global economic crisis in 2009. Between 2010 and 2012, however,

there was a positive trend. Global sulfuric acid production currently exceeds 270 million t, with this figure expected to rise. China tops the list of sulfuric acid producers, with over 74 million t of production output. The US comes in second with over 37 million t, followed by India (16 million t), Russia (14 million t) and Morocco (7 million t). In total, these five countries account for approximately 61.5% of all sulfuric acid produced worldwide. The global sulfuric acid market was worth US$10 billion in 2016. This figure is expected to increase to US$15 billion by 2025. Similarly, global demand is expected to rise exponentially, necessitating an increase in production to keep up.

Production of elemental sulfur India is one of the world’s leading sulfuric acid manufacturers. There is a positive relationship between a country’s sulfuric acid usage and its per capita income. The acidulation of rock phosphates and the generation of ammonium sulfate are both part of the process. Petroleum refining, steelmaking and other inorganic compounds are among its numerous industrial applications. In India, there are currently more than 65 sulfuric acid facilities.



February 2022

Sulfuric acid is made from a variety of basic materials, including elemental sulfur, hydrogen sulfide (H2S), pyrites, etc. Nearly all Indian manufacturers rely substantially on elemental sulfur as a major raw material source. On the other hand, the oil industry extracts sulfur from crude oil through the refining process. This is generally accomplished by hydrotreating, which produces H2S that is converted into sulfur in the sulfur recovery unit (SRU), thus limiting the SOX and NOX content as per regulations (typically a maximum of 8 mg/m3 of H2S). Refineries’ design capabilities meet both low-sulfur and high-sulfur crude oils and, with an SRU, they can produce elemental sulfur (99.9% recovery) and meet emission norms. The unit consists mainly of the separation section, reaction section, tail gas treating unit (TGTU) and incineration section. H2S + 1/2 O2 ⟶ 1/2 S2 + H2O


In the separation section, the feed gas – which consists mainly of sour and acid gases (sometimes 80% or more H2S) – passes through drums with the assistance of a mist eliminator, to avoid liquid carryover. The heart of the SRU is the reaction section and various options are available, sometimes with a high temperature reaction furnace to destroy ammonia, which generally creeps into the sour gases. The Claus reaction occurs in the reactor furnace at high temperatures (1300 – 1400°C). The sulfur produced is condensed and sent to storage. The exit gas from the condenser is reheated for the catalytic reaction stage, and the sulfur produced is again collected by condensing the gas. One can have multiple stages as well as a separate TGTU to accomplish product recovery. The treated gas from the TGTU is burned in the incinerator. Heat integration of the SRU is carried out by passing the appropriate stream through heat exchangers and using a waste heat recovery system to produce steam. With robust process design and established technology, an estimated 99.9% recovery of sulfur in the product can be achieved. Process optimisation and heat recovery, for the lowest CAPEX and OPEX, is accomplished through available licensed software tools. Utilities typically include refinery fuel gases and boiler grade feed water (high-pressure steam production).

The future of the sulfur market As per a report by Mordor Intelligence1, the sulfur market is predicted to grow at a CAGR of more than 5% from 2021 to 2026, with a market size of 61.88 million t in 2020. COVID-19 has caused a disruption in sulfur supply and a drop in demand from a variety of end user sectors, including metal manufacturing and chemical processing. Furthermore, the pandemic has entirely interrupted manufacturers and supply networks, posing a short-term threat to the market. However, the situation is likely to improve, restoring the market’s development trajectory over the second half of the forecast year.

Sulfur in the oil and gas industry A large amount of sulfur byproducts are produced as a result of oil and gas operations. For example, Qatar, which has the February 2022 18 HYDROCARBON ENGINEERING

world’s third largest confirmed natural gas reserves, produces a lot of sulfur as a byproduct of its natural gas processing facilities. The amount of sulfur produced exceeds the amount that can currently be used in the country by a large margin. The Claus method is used to convert hydrogen sulfide in natural gas collected from Qatar’s North Field, to elemental sulfur. Across many nations that have oil and gas facilities, managing byproduct sulfur from natural gas processing is a critical part of economic development and environmental protection. As a result, new markets for sulfur must be discovered in order to avert disposal crises. A huge volume of sulfur is recovered each day by oil and gas corporations. The total amount of elemental sulfur produced across the globe in 2000 was 57.4 million t. Sulfur was collected at a rate of 103 000 tpd in January 2019, with Asia recording the highest rate. As more oil and gas is recovered from the ground, these figures are rapidly increasing. The massive amounts of sulfur produced as a byproduct of the sector can be utilised in a variety of ways including fungicides, gunpowder, fertilizers, rubber vulcanisation, matches, pyrotechnics, fumigants, insecticides, the treatment of certain skin ailments, etc. Sulfur has an extensive range of applications, and it has become a basic element due to its abundance, ease of extraction, and its ability to be directly mixed with most known elements.

Environmental effects of sulfuric acid production Sulfur depositions in landfills and on the outskirts of cities are becoming a major health and environmental issue. However, with more research and development, sulfur’s abundance can be employed to solve a variety of problems, or as a novel additive that boosts efficiency while using less product. This should be both environmentally and health-friendly, as well as cost-effective. Until now, industrial processes have been the primary cause of sulfuric acid related pollution. These include the manufacturing process, as well as its applications in metal smelting, petroleum refining, lead-acid accumulators and other operations. Sulfur dioxide is released into the atmosphere as a result of these processes, which can have disastrous consequences, such as acid rain. This can lead to the pollution of plant and animal life in the short-term. In the case of human contamination, sulfuric acid can cause significant burns if inhaled or absorbed via the skin. In contrast, recovery of sulfur by the SRU in refineries renders the flue gases more compliable to emissions standards. According to a report released by Greenpeace on 19 August 20192, rising emissions have pushed India above China, which has surpassed Russia as the world’s second-largest emitter as a result of its progress in decreasing emissions. China lowered sulfur dioxide emissions by enforcing strict emission regulations and using technology such as flue gas desulfurisation (FGD).

Case studies Since 1996, Nuberg EPC has successfully commissioned multiple sulfuric acid plants around the world, including Agrochem in Egypt and ADDAR in Saudi Arabia.

Performance Under Pressure

Turn to Elliott Group for a partnership you can trust. For nearly a century, Elliott steam turbines have earned a reputation as the most rugged, reliable, and versatile drivers in the industry. Customers choose Elliott for exceptional value and unmatched reliability in mechanical drive, power generation, and power recovery applications. Who will you turn to? n

Learn more at www.elliott-turbo.com

The World Turns to Elliott. C O M P R E S S O R S | T U R B I N E S | C R Y O D Y N A M I C S® | G L O B A L S E R V I C E

These case studies detail the learnings and obstacles encountered throughout the process.

Alexandria, Egypt

Agrochem contacted Nuberg EPC to express its ambition to build a new sulfuric acid plant in Egypt, with a capacity of 300 tpd (see Figure 1). With shop drawings and both basic and comprehensive engineering investigations of the proposed location, Nuberg EPC moved quickly. Following this, the company sought vendors in order to obtain the best supplies for the project. Process, technology know-how, basic and detailed engineering, design, procurement, production and supply of plant equipment – including complete installation, building, project management, commissioning and start-up – were all part of the project’s scope. The factory was successfully commissioned in 2012 after several months of hard labour, and it now uses double contact double absorption (DCDA) technology. The sulfuric acid factory, which was completed in 2012, was Nuberg EPC’s first turnkey project in Egypt. The factory generates sulfuric acid (98%) using DCDA technology, which is mostly used in the fertilizer business in Egypt. Nuberg EPC’s plant engineering and construction services helped Agrochem to run a safe, reliable and environmentally-friendly facility. Sulfuric acid is one of the world’s most widely utilised industrial chemicals. Fertilizers containing phosphates, such as diammonium phosphate (DAP), monoammonium phosphate (MAP), triple superphosphate (TSP) and single superphosphate (SSP), have fuelled expansion in the sulfuric acid market in recent years.

Figure 1. Sulfuric acid plant in Egypt.

Jubail Industrial Area, Saudi Arabia

Nuberg EPC was hired by Addar Chemicals Co. to build a new sulfuric acid and sulfolane acid facility in Jubail (see Figure 2). The latter, which has a capacity of 6000 tpd, was Saudi Arabia’s first greenfield sulfolane production factory, with a capacity of 80 tpd. As with the Agrochem project, process, technology know-how, comprehensive engineering, design, procurement, production and delivery of plant equipment, as well as installation, building, project management, commissioning and start-up of the plant, were all included in the scope of the project. The company complied with local standards set forth by the Royal Commission for Jubail and Yanbu (RCJY) and the Saudi Standards, Metrology, and Quality Organization (SASO). The plant in Jubail was completed in November 2018 and commissioned by the Saudi Arabian firm. The facility generates sulfuric acid (98%), which is used to make ammonium sulfate and super phosphate of lime, among other things. The sulfuric acid business has risen in parallel with rising demand for organic compounds, with fertilizer production accounting for over 70% of total production today.

Conclusion Sulfur is crucial for the existence of life on Earth. It has different derivatives, including sulfuric acid, which has an extensive range of applications in the oil and gas industry and is considered to be one of the most important industrial chemicals. Recently, COVID-19 has disrupted sulfur supplies and reduced its demand in a variety of end user industries. However, the situation is likely to improve as markets begin to recover and customers are express their ambition to build sulfuric acid plants. Nuberg EPC has signed multiple contracts amidst the pandemic. The 500 tpd Sprea Misr sulfuric acid plant project in Ramadan, Egypt, is one of them. On the contrary, industrial processes have been the primary cause of sulfuric acid related pollution. This includes its applications in metal smelting, petroleum refining, lead acid batteries and other operations. However, with more research and development, abundant sulfur can be used to solve a number of problems, or as a novel additive that reduces product consumption and increases efficiency. Supported by an international quality R&D centre for specialty chemicals, located in Sweden, Nuberg EPC has been improving its sulfuric acid technology. Since 1996, the company has effectively commissioned multiple sulfuric acid plant projects across countries such as Turkey, Egypt, Bangladesh and Saudi Arabia, and has maintained its commitment to providing safe, reliable and efficient facilities worldwide.

References 1. 2.

Figure 2. Sulfuric acid plant in Saudi Arabia.


‘Sulfur market – growth, trends, COVID-19 impact, and forecasts (2021 – 2026)’, Mordor Intelligence, (2021), https://www. mordorintelligence.com/industry-reports/sulfur-market ‘Greenpeace analysis ranks global SO2 air pollution hotspots’, Greenpeace, (2019), https://www.greenpeace.org/international/ press-release/23819/global-so2-air-pollution-hotspots-ranked-bygreenpeace-analysis/

Richard G. Stambaugh, Merichem Company, USA, discusses the available technologies for lowering refineries’ sulfur emissions in order to meet tightening regulations.


mong the 1.4 billion vehicles that are currently on the roads worldwide, more than 23 million use LPG or propane autogas as a fuel source. It is the third most common engine fuel behind gasoline and diesel, and the most popular alternative fuel globally. An environmentally-friendly fuel that is prevalent among fleet and public transportation, LPG produces 99% fewer particulate emissions than gasoline and diesel. It also produces 10 – 15% less CO2 and releases 50 – 60% less NO2 and other hydrocarbons into the atmosphere. Because the fuel systems of LPG vehicles are tightly sealed, there are no evaporative emissions while running or parked. Additionally, they do not significantly contribute to acid rain because of low sulfur content. Chemically, LPG is a mixture of propane and butane hydrocarbons that change to a liquid state at the moderately high pressures found in an LPG vehicle’s fuel tank. It is formed naturally, interspersed with deposits of petroleum and natural gas. Natural gas contains LPG, water vapour and other impurities that must be removed before being transported in pipelines as a saleable product. Approximately 55% of the LPG processed in the US is from natural gas purification. The other 45% comes from crude oil refining. There have been recent changes in US fuel standards complying with US Environmental Protection Agency (EPA) Tier 3, which is part of a comprehensive approach to reducing the impacts of motor vehicles on air quality and public health. It has forced many refiners to address their compliance methodology for fuel sulfur levels. Merichem Company provides a new approach to lowering the total sulfur in the LPG product stream. Its THIOLEX® RSH® removal system achieves product sulfur specification of less than 3 ppmw.

Challenge A major refinery in Midwest US (Padd II) was producing a high volume of finished products, including gasoline, with a throughput capacity of approximately 170 000 barrels per stream day (bpsd). Refinery personnel were evaluating capital scope options and requirements for manufacturing gasoline with 10 ppm sulfur to meet the EPA Tier 3 regulations. They determined that the total sulfur content of their butane/butylene (BB) stream would need to be maintained at less than 10 ppm. To allow for inert sulfur compounds already present in the hydrocarbon feeds, the sum of unextracted acidic sulfur and back-extracted disulfide oil (DSO) would need to be even lower – 3 ppm or less. The untreated BB stream had a design throughput of 13 000 bpsd and total mercaptans (RSH) of 222 ppmw as sulfur. While it is relatively easy to reduce extractable RSH in the product to 1 – 2 ppm as sulfur, one of the greatest challenges in caustic treating is preventing the resulting DSO from returning to the hydrocarbon feed. The DSO is typically both entrained and dissolved in the oxidised caustic. The colloquial term for the DSO which returns to the treated hydrocarbon is ‘re-entry sulfur’ or ‘back extraction’.

The technical solution Caustic treatments remove acidic sulfur species from LPG. The alkaline pH of the caustic solution reacts with the sulfur species – most notably, RSH – to form water-soluble, ionic compounds that preferentially move into the caustic phase. Once separated from the hydrocarbon phase, the aqueous sodium mercaptides are sent to an oxidative regeneration unit where heat, oxygen and catalyst are introduced. The mercaptides react with the oxygen to form mostly insoluble HYDROCARBON 21


February 2022

DSOs which are then removed from the caustic. The regenerated – or lean – caustic is then returned to interact with the hydrocarbon once more, extracting more acidic sulfur species. To achieve the ultra-low-sulfur (ULS) requirements of the refinery, Merichem Company introduced its REGEN® ULS technology, which significantly reduces DSO back extraction to achieve total sulfur product specifications below 10 ppmw. This new technology combines two separate existing technologies in such a way as to force the DSO to separate more completely from the regenerated caustic compared to a traditional REGEN® system.

The company paired a REGEN ULS that was specially designed for the refinery, with its high-efficiency RSH-removal system, called THIOLEX. The hydrocarbon stream is treated in the THIOLEX system using lean caustic from the REGEN ULS. THIOLEX technology utilises caustic soda as the treating reagent to remove acid gas and RSH compounds from liquid hydrocarbon streams. The technology is used with Merichem Company’s FIBER FILM® Contactor, a retaining cylinder packed with very fine, proprietary metal fibres. The large interfacial surface area, microscopic diffusion distance, and continuous renewal of the aqueous phase are combined to yield superior mass transfer efficiencies. Because the aqueous phase adheres to the fibres in the FIBER FILM Contactor rather than being dispersed into the hydrocarbon phase, aqueous carryover is virtually eliminated. The mercaptide-rich caustic is directed to the REGEN ULS system, where it is oxidised on a fixed catalyst bed. The resulting DSO is then removed from the caustic using bulk phase separation and decanting; solvent washing; and adsorption on a fixed bed to yield the ULS lean caustic that is recirculated to the THIOLEX unit for continuous treating. The technology super-regenerates the mercaptide-rich caustic and returns a lean caustic that is almost free of DSO.

Programme results Figure 1. A typical single-stage FIBER FILM Contactor

– the vertical pipe mounted on top of the horizontal phase separator. Untreated hydrocarbon and fresh treating solution enter at the top. Treated hydrocarbon and spent treating solution exit from the top and bottom of the separator, respectively.

Determined commitment to teamwork by both parties resulted in the refinery meeting its requirements for manufacturing 10 ppm sulfur product to meet EPA Tier 3 regulations for gasoline. The THIOLEX/REGEN ULS technology significantly reduced the sum of RSH and DSO as sulfur to lower than 2 ppmw in the product stream.

Figure 2. Mercaptan removal for light hydrocarbons with caustic regneration for ULS product.


DSO back extraction to just 1 – 2 ppm as sulfur, which created a much lower total sulfur product specification.


Figure 3. A typical multi-stage FIBER FILM Contactor and separator system treating light crude oil in a stick built unit. The unit handled turndown scenarios well and handled up to 20% C5+ material in the feed. The only impact of this was increased caustic consumption. On a day-to-day basis, operating involvement was minimised to monitoring levels, pressures, flows and temperatures, with occasional caustic batching and catalyst additions. Utilising THIOLEX technology, the BB stream was treated with regenerated caustic to extract acidic sulfur components. The REGEN ULS regenerated the spent caustic from the THIOLEX unit while removing almost all DSO, which resulted in ULS levels obtained in the treated butane product. The combined THIOLEX/REGEN ULS technologies reduced the total

Part of the differentiated, proprietary design of Merichem Company’s THIOLEX technology is the patented FIBER-FILM Contactor that enables more surface area for the mass transfer of hydrocarbon impurities with caustic soda. Due to this efficient mass transfer rate and low energy mixing, residence time is significantly reduced, and emulsions and caustic soda carry-over are minimised. This custom-designed solution was implemented at a lower CAPEX and OPEX with no reduction in octane. At approximately one-quarter of the cost of hydrotreating, the combined THIOLEX/REGEN ULS solution is far less expensive than hydrotreating or once-through caustic treating with no regeneration. Compared to a typical REGEN, the REGEN ULS cost modestly more (10 – 15%) but quickly paid for itself by yielding a much lower total sulfur hydrocarbon product. Regenerating and recycling caustic soda not only lowers OPEX but also helps protect the environment. Without a regeneration unit, spent caustic must be processed internally or manifested and shipped as hazardous waste to a third-party waste handler. However, the volume shipped out is much lower than it would be if the unit was a single pass with no regeneration unit. Alternative technologies such as hydrotreating are also significantly more energy intensive. The first REGEN ULS unit was installed in Taiwan more than 10 years ago. It has been successfully operating and meeting product specifications since then.


LaserGas™ III SIL2 analyzer combines the highly selective and sensitive TDLAS measurement principle with explosion proof (Ex-d) and safety (SIL) designs.

Performance You Can Trust www.neomonitors.com

Peter Foith, CS Combustion Solutions, Austria, outlines the challenges faced when retrofitting an existing thermal stage of a sulfur recovery unit.


or the retrofit of a 700 tpd Claus plant with specific challenges, CS Combustion Solutions utilised its accumulated experience in designing and engineering combustion systems for sulfuric acid production, based on combustion of sulfur containing liquid and gaseous streams, spent acid regeneration units, and hazardous waste incineration, to supply new equipment. This included a low pressure drop swirl burner with staged combustion of Claus gas, a reactor utilising a Blasch VectorWallTM, and a new waste heat boiler with a focus on minimising the pressure drop to meet the clients’ requirements.

Initial situation Burner

The retrofit project started with the company analysing the existing equipment and engineering new equipment to overcome the various challenges. The original burner onsite was a multi-flame burner with low pressure drop on the Claus gas thermal stage, with undefined mixing of combustion air and Claus gas, which leads to low flame stability in low load cases as well as during start-up. The reason for the undefined mixing was found in the design of the original burner, as the HYDROCARBON 25


February 2022

underlying construction of the multi-flame burner, with 48 individual ceramic blocks for each of its 48 flames, is prone to poor Claus gas distribution in low load cases if the ceramic blocks are not centrically placed. In addition to this, corrosion was prevalent due to condensing of Claus gas in these situations.


The original reactor was equipped with a checkerwall in the first third of the chamber. However, even though a checkerwall generally helps to improve temperature distribution in the reactor, in this case the temperature distribution was not homogenous and hot spots would lead to refractory damage, as it would burst over time due to stress in the material. Another major reason for using a checkerwall is to facilitate the

conversion rate of hydrogen sulfide to sulfur, as a high conversion rate is a key factor for an efficient Claus plant. Unfortunately, the checkerwall did not lead to satisfying results, as the conversion rate was inefficient and below expectations. As there was the additional problem of the long burner flames coming into contact with the checkerwall, it was surmised that the position of the checkerwall was not ideally selected, which could have been the reason for the aforementioned problems in the reactor.

Waste heat boiler

The original waste heat boiler was a smoke tube boiler with an internal bypass, with the steam drum directly installed on the boiler. The total design steam capacity of the equipment was 75 tph, and the steam had a temperature of 250°C at a pressure of 36 bar(g). In comparison with the reactor, the waste heat boiler operated more effectively and with fewer issues. However, a recurring problem with welding cracking was prevalent with the boiler and arose due to thermal stress caused by an inhomogeneous temperature profile.


Figure 1. CS Combustion Solutions’ low pressure

swirl burner. S1= secondary air, S2= primary air, S5= Claus gas first stage, S6= Claus gas secondary stage.

Figure 2. VectorWall.


Together with specified expectations from the client, the initial situation highlighted certain issues, resulting in challenges that led to various complexities. These were met with the following necessary engineering efforts: nn As the focus of the retrofit was only on the replacement of the existing thermal stage, it was necessary to ensure that all process parameters remained unaffected, so that the equipment upstream and downstream of the thermal stage continued to work as intended. nn The existing multi-flame burner was required to have a low pressure drop on the Claus gas side, as the available supply pressure was very low. The available combustion air pressure was also low. This meant that the new burner required a low pressure drop for Claus gas and combustion air, while ensuring a high enough turbulence for proper mixing of both in order to achieve a stable flame for low and high load cases. nn A further requirement was to ensure a high hydrogen sulfide to sulfur conversion rate to meet the client’s specifications and expectations. nn The available space for the new thermal stage equipment remained the same, as the upstream and downstream equipment was unaffected by the retrofit. This meant that the overall size and dimensions of the burner, reactor and waste heat boiler also remained the same. Addressing these challenges meant the target for the maximum pressure drop through the complete thermal stage unit (burner, reactor and waste heat boiler) was 100 mbar(g). The aim for the burner was a low pressure drop design on the Claus gas and combustion air side to meet the issue of low pressure availability. The burner design also needed to ensure flame stability by generating a sufficiently high turbulence and mixture of combustion air and Claus gas across all load cases. Additionally, the aim was to not only get as close as possible to the theoretical maximum achievable hydrogen sulfide to sulfur conversion rate of 70% at these process parameters of 1338 K in order to meet the guaranteed



MOVE TOWARDS MODULARIZED PACKAGING In response to this new shift to modularization as it relates to a compressor package and its surrounding process and auxiliary equipment in the downstream oil and gas sector, MHI developed Mitsubishi Compressor Smart Packaging, or MPAC. The key parameters that MPAC improves for the end user are total installation cost, lead time, ease of maintenance, logistical feasibility, construction risk and site restrictions. MHI Compressor’s modularized MPAC includes the entire centrifugal compressor train along with the lubrication and dry gas seal systems on a single skid with all of the associated interconnecting piping and wiring. It is compact enough to be suitable for ground transportation. Visit us at mcoi.mhi.com/insights to learn more today, or contact sales@mhicompressor.com for inquiries.

MCO-I Pearland Works 14888 Kirby Dr. Houston, Texas 77047 Tel: 1-832-710-4700 mcoi.mhi.com

conversion rate of 65%, but also to increase the maximum sulfur capacity to 120%.

Meeting the challenges and requirements Low pressure swirl burner

Based on existing and proven CS Combustion Solutions burners, a new low-pressure, double-staged swirl burner (see Figure 1) was designed to meet the requirements of low pressure drop on the Claus gas and combustion air side, while ensuring proper mixing.

New reactor

The existing reactor made use of a checkerwall, however due to the requirements for this project, the company decided to utilise a VectorWall (see Figure 2) for the new reactor, which has the same dimensions. This VectorWall was specifically designed with a focus on low pressure drop, and to promote the swirl

and turbulence inside the reactor. The vector tiles (see Figure 3) incorporated into the VectorWall are another advantage of this technology, as they generate separate mixing zones for an even distribution across the entire cross section. Multiple computational fluid dynamics (CFD) analyses were carried out in order to verify that the burner and reactor design, as well as the positioning of the VectorWall, led to homogenous temperature distribution (see Figure 4), to ensure that the guaranteed hydrogen sulfide to sulfur conversion rate (see Figure 5) of 65%, as well as the increase of the sulfur capacity to 120%, was met.

New waste heat boiler

As with the existing waste heat boiler, the new waste heat boiler was a smoke tube type boiler with the steam drum installed on top, and was supplied by a partner. The engineering focus for the boiler, as well as the design of the heat exchangers, bypass and internal piping, was to keep the pressure drop as low as possible whilst ensuring that the necessary steam generation capacity was met.


Figure 3. Vector tiles for separate mixing zones.

Figure 4. Temperature distribution.

Figure 5. Hydrogen sulfide to sulfur conversion.


Each section of the original equipment – the burner, the reactor and the waste heat boiler – had issues. Due to construction issues, the burner had undefined mixing of oxidation and Claus gas; the flame stability was poor during start-up and lower load cases; and there was corrosion due to Claus gas condensation. Even though the reactor was equipped with a checkerwall, it had inhomogeneous temperature distribution, hot spots damaging the refractory, and problems with long burner flames, as the checkerwall was positioned near the burner. The biggest issue with the waste heat boiler was welding cracking, which arose due to thermal stress caused by an inhomogeneous temperature profile. To meet the client’s requirements – no impairment of process parameters, low-supply pressure on the Claus gas and combustion air side, high hydrogen sulfide to sulfur conversion rate – CS Combustion Solutions took the following measures: nn The original burner was replaced with a low-pressure double staged swirl burner to minimise the pressure drop over the burner and ensure proper mixture of Claus gas with oxidator. nn A new reactor with the same dimensions as the original one but equipped with a VectorWall with low pressure drop instead of a checkerwall was used. Swirl and turbulence was increased, as the VectorWall generates separate mixing zones for an even distribution across the entire wall. nn A new waste heat boiler was specially designed by a partner to keep the pressure drop as low as possible. nn Multiple CFD analysis instances were carried out to verify the design and to ensure that the guaranteed hydrogen sulfide to sulfur conversion rate of 65% or above, as well as an increase in the maximum sulfur capacity to 120%, was met. The result of the retrofit was a completely new thermal process stage consisting of burner, reactor and waste heat boiler for a Claus plant with a sulfur output of 700 tpd.

Rod Merz, AMETEK Process Instruments, Canada, explores the most common analytical heat integrity failure points seen with analysers in sulfur recovery units.


he number one analytical concern with analysers used in a sulfur recovery unit (SRU) is sample heat integrity. Heat integrity is the most common non-electronic related cause of analyser problems in an SRU. There are several analytical points of interest in a typical SRU and amine-based tail gas treating unit (TGTU). Each analytical point has a unique purpose and unique measurement challenges. The analytical measurement points, as shown in Figure 1, include: the feed forward analyser (AT1), the sulfur pit analyser (AT2), the air demand or tail gas analyser (AT3), the possible gas measurement points within the TGTU (AT4/AT5/AT6), and the thermal oxidiser’s continuous emissions monitoring system (CEMS) analyser (AT7).

Feed forward/acid gas analyser The feed forward analyser (AT1) is found in a refinery and gas plant SRU where the acid gas concentration may vary, or the hydrocarbon concentration variations from upstream process upsets can cause difficulty in controlling air demand. The acid gas sample stream at this point in the process consists of the main constituents of hydrogen sulfide (H2S), carbon dioxide (CO2), water (H2O) and, to varying degrees and depending on the process, ammonia (NH3), amine, and carried over hydrocarbons. While the acid gas stream is lower in temperature than the rest of the SRU sample stream analytical points, it is always considered ‘saturated’ with H2O. This means that, from the sample take off point to the sample return point, the sample temperature must be kept above the sample’s actual dew point, or the sample will condense. This condensation can cause fouling of optical windows, plugged or percolating sample lines, fouled filters, corrosion, as well as other analytical maintenance issues. An analytical best practice is to operate the sample system and analyser 10 – 20°C or higher than the process HYDROCARBON 29


February 2022

Figure 1. SRU and TGTU analytical points. sample temperature. As an example, if the acid gas stream is 30°C, the sample probe and sample line, as well as all wetted analyser components, should be at a minimum of no less than 40 – 50°C. In addition, for sour water stripper gas (SWAG), the sample must be maintained 10 – 20°C above the ammonia salt formation point (typically heated to 80 or 90°C). This fine detail is often overlooked by an inexperienced sample system designer who only looks at the sample dew point on the data sheet. While most installations are designed with good intentions, there are a few common issues that are found in the field. In colder climates, the sample nozzle should be insulated, and if the nozzle is longer than 6 in. (150 mm) off the process pipe, the nozzle should be heated – either jacketed or heat traced. The sample probe should also be insulated and have its own heater built in to maintain it above the sample dew point. Connections between sample system components need to be looked at closely for possible condensation points, including unions between the probe and sample line, transitions into analyser enclosures, and internal analyser connections. All of these are locations where sample condensation can wreak havoc with an analytical measurement.

Air demand/tail gas analysers The air demand analyser (ADA), also known as the tail gas analyser, and the pit gas analyser can be treated as identical systems since the sample composition and dew points are similar enough. In this analytical application, the most significant problem observed is sulfur coming out of the vapour phase and plugging off the sample handling and analytical components. From a heat integrity point of view, the ADA is similar to the CEMS unit in that it has the greatest plant impact due to heat integrity issues. With the sample stream temperature from the SRU’s final condenser optimally being only a few degrees above the freezing point of sulfur, any cold spots with the ADA and its sample handling components will lead to the occurrence of solid sulfur deposition, which will inevitably cause the analyser system to plug. The following proactive measures will keep the analytical system in optimal operating condition: February 2022 30 HYDROCARBON ENGINEERING

Proper insulation of the sample nozzle

The sample nozzle itself is a significant point of risk and a common trouble spot. It must be properly insulated regardless of length. If the sample nozzle is longer than 6 in. (150 mm), it must also be heated (see Figure 2). It is not adequate to try and wrap the nozzle in stainless tubing using steam as the heating medium, with insulation over the top. The actual surface contact point of tubing to nozzle surface is very small, so no useful conduction of heat occurs. Any heat transfer is mostly by radiation and it does not have the efficiency to transfer the heat properly. Using a steam or glycol jacketing system such as ControTraceTM (Figure 3) or ControHeatTM is one solution. If low or medium pressure steam is the heating medium, it needs to be at the appropriate temperature/pressure and adequately trapped and dry. Wet steam is not going to meet the manufacturer’s temperature design criteria. This tends to be one of the more common problems that service technicians see. A means to quickly determine whether there are nozzle temperature issues is to take a piece of sulfur and scrape it on the surface of the nozzle. It should melt.

Sample line transitions

As mentioned previously regarding the feed forward/acid gas analyser, attention should also be paid to any sample line transitions into enclosures that could be a cold spot to ensure that they are not a point where sulfur deposition could occur. In addition to cold spots, excess heat can also be problematic for analyser sample lines. ADAs that have high temperature sample lines can have issues if the sample inlet and sample return lines are not properly spaced during installation. In this case, the issue is excess heat buildup where the sample lines are too close and can cause premature failure of the sample line’s heater circuits.

Protection from external elements

The analyser also needs to be protected from prevailing wind, rain and snow. Rain or snow falling on a hot nozzle can quickly

WHEN YOU SEE ARIEL, you know your operation is built for unequaled reliability and environmental sustainability. Our compressors go the distance, just like our people. You can’t spell reliable without Ariel. Visit www.arielcorp.com/pledge

drop the surface temperature and result in sulfur deposition. A simple three-sided shelter can prevent weather-related problems in temperate climates, but a proper environmentally controlled shelter will be needed in both cold and hot climates. Excess heat is a significant factor in shortening the lifespan of analyser electronics. It is important to ensure that the analyser is protected from direct sun and is in an environment that meets the manufacturer’s specifications for ambient temperature operation.

Amine-based TGTU analysers

Figure 2. Poor attempt to use steam heated tubing; insulation removed.

Most amine-based TGTUs will only have one of the analyser tags shown in Figure 1 and will be located on either the quench outlet (AT5) or the absorber outlet (AT6). Occasionally, there will be an analyser at the hydrogenation reactor outlet (AT4), but it is typically considered a problematic or compromised sample point in comparison to the other two locations. In general, everything that was discussed with the feed forward analyser or acid gas analyser also applies to this analytical application – the sample gas will be saturated and therefore will require all analyser sample components to be heated to 10 – 20°C above the sample dew point. The only special consideration that is uniquely applicable to the TGTU is the potential to form sulfur in the TGTU analyser probe/filter and plug it off if sulfur dioxide (SO2) has broken through. The residual tail gas H2S can react with the SO2 breakthrough and form sulfur in the analytical system. If sulfur formation occurs in here, more significant problems will occur with the acidification of the quench water system, plugging of the quench water circuit, and the formation of heat stable salts in the amine absorber.

CEMS analysers

Figure 3. ControTrace prior to installation.


The CEMS analyser (AT7 on Figure 1) is a governmental compliance requirement for SRU SO2 emissions in most jurisdictions. In the majority of countries, there are strict penalties for not meeting emission rules and regulations, so the CEMS analyser should be the most operationally significant analyser and operational downtime must be minimised. From a heat integrity point of view, the components that need to be considered include: the sample probe, heated sample lines, and the sample conditioning system if the CEMS is considered for cold/dry extractive systems or dilution-based analytical systems. The most common problem is a result of sulfur trioxide (SO3) concentrations present during SRU upset conditions and the resulting elevation of the acid dew point. SO3 is typically present in concentrations of between 2 – 10% of the SO2 concentrations in SRU incinerator emissions. The SO3 reacts with water molecules and forms sulfuric acid (H2SO4) – a strong acid. When SO3 concentrations are high enough and the acid dew point is reached, the acid forms on surfaces within the sample probe and sample lines. The formation of ‘green slime’ can be physically seen on metal surfaces as this occurs. The green colouration comes from the leaching of chromium and other elements from the stainless steel components into the deposited acid film. The ensuing problems that occur can range from sample system plugging, to a noticeable chromatographic effect where sample and calibration gases diffuse in and out of the slime film in the sample lines, increasing sample response time. Corrosion issues and damage to the analytical sample system are also possibilities.

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

© 2021 by AMETEK Inc. All rights reserved.

The CEMS sample probe is where the issues are often centred, and sample lines follow close behind. If the probe requires unusually frequent maintenance, analysis should be carried out on the sample’s acid dew point and the operating temperature of the probe. As indicated above, sample nozzles are notorious for being bad actors, so a probe tube heater may be required to help keep the probe tube’s temperature above the acid dew point. Acid knockout pots and sacrificial filters built into the probe can also be used to mitigate the acid formation in other sample system components such as the sample lines. The CEMS sample lines can also be a source of temperature issues. Excessive sample line length is often the precursor to heat integrity problems. Depending on both the power supply and sample line manufacturer design, there is a maximum heated sample line length for a single circuit. Where sample line runs exceed a single circuit length, segments of sample line need to be coupled together. This introduces increased risk of losing heat integrity at unions, not to mention the addition of multiple circuit elements for potential failure. For cold/dry extractive CEMS, the points mentioned above are important and include the sample conditioning system (SCS) located in front of the analyser(s). Heat integrity concerns are usually related to the sample line entry into the SCS and all connections and elements in front of the moisture removal device – usually a Peltier or vortex driven sample cooler. There are otherwise good designs where the heated sample line terminates to an unheated solenoid or valve, which is an excellent point for the sample gas to condense.

Dilution-based CEMS problems typically relate to probe temperature concerns where the orifice may be below the acid dew point, resulting in the formation of acid slime on it. Another concern is during upset conditions where, due to the elevated SO3 concentrations, the acid dew point is still reached even though the system design is supposed to dilute the sample enough to reduce the acid dew point below the system components’ operating temperature. Hot/wet extractive systems have the benefit of providing an unaltered sample to the analyser but, in this case, everything including the analyser must operate above the acid dew point. All of the identified potential condensation concerns listed above must be considered, and, in situations where SO2 excursions go beyond the normal operating range and are expected, the use of acid knock-out systems at the probe can really help.

Conclusion This article has focused on the most common analytical heat integrity failure points that are observed in the field. While not discussed, electrical heater failure is also a potential concern; however, typically this is not considered an analytical design failure. The points that have been outlined are not just applicable to SRU and TGTU applications. The principles discussed can be applied to any analyser system on any process. In summary, when looking at the reliability of an SRU analytical system, it is important to consider the following question: where am I at risk of losing analytical heat integrity?

THE FUTURE OF CORROSION PREVENTION Introducing AMPP, the Association for Materials Protection and Performance. After more than 70 years, NACE International and SSPC have united as the global authority for corrosion prevention knowledge. We are creating the future of materials protection and performance and leading the way for the industries we represent. AMPP members protect people and places around the world from corrosion, and together we are building a safer, protected, and preserved world.

Learn more about AMPP.



Rhys Jenkins, Servomex, UK, considers the role of gas analysis in purified terephthalic acid (PTA) production.

urified terephthalic acid (PTA, sometimes referred to as polymer-grade terephthalic acid) is an important chemical component in the plastics industry. Together with ethylene, it is used to produce the high-demand plastic polyethylene terephthalate (PET), widely used as a polyester fibre and for recyclable food and drink containers, such as plastic bottles. There is a continuing and growing demand for PTA throughout the world, particularly in fast-expanding economies

such as those found in Asia. Approximately 80 – 100 million tpy of PTA is produced, the vast majority coming from Asian plants, while China alone makes up 60% of the global demand. The production of PTA requires expert gas analysis for process control, efficiency and safety, as well as quality monitoring and environmental compliance. In order to deliver this wide range of measurements, a variety of sensing technologies are required. Not only do these sensors have to be accurate and react quickly to changing gas



February 2022

Figure 1. A PTA production plant. concentrations, but they also have to be able to operate in the challenging – and often corrosive – conditions of the manufacturing process.

The PTA process and key gas analysis points PTA is manufactured from the aromatic hydrocarbon p-xylene, which undergoes air oxidation in a reactor at high pressure and high temperature. Liquid acetic acid is used as the solvent for this reaction. The crystalline PTA product is separated off in crystallizer vessels, recovered and purified. The oxidation reactors and the crystallizers are key points for gas analysis in the PTA process. At the beginning of the PTA manufacturing process, air is passed into the oxidation reactors. This oxidises the p-xylene to form terephthalic acid while also generating carbon dioxide (CO2) and carbon monoxide (CO). Some oxygen (O2) is likely to remain unreacted – it is critical to monitor this residual O2 level in the off-gas to ensure that it stays between 4 – 5%. If the level gets too high, sudden runaway oxidation of all the flammable materials could occur, leading to an explosion. However, if the level drops too low, insufficient oxidation occurs, reducing efficiency and lowering product yield. Monitoring of the air inlet to the oxidiser, and the level of unreacted O2, can be reliably handled by paramagnetic O2 analysers. Speed of response and reliable accuracy are critical at this point. As such, to ensure reactor safety, multiple analysers are typically used, installed in a voting system. A voting system uses multiple analysers and depends upon the measurement of the majority. In a three-analyser system, for example, if one analyser detects a significant change, it is outvoted by the other two and no action is taken. However, if two (or all) of the analysers detect a change, this reading is seen as correct, and action may be taken, such as informing the operator or automatically halting the process. This approach provides an extra layer of reliability for safety measurements, and also helps detect analyser problems at an early stage – if one device is continually giving a different result, it can be investigated and corrected. Measuring the off-gas CO2 level offers more information about the progress of the oxidation. This can be accomplished using an infrared (IR) analyser configured for CO2. February 2022 36 HYDROCARBON ENGINEERING

Next comes the crystallizer stage, where acetic acid is driven off as the PTA product crystallises out of the solvent liquor. This vapour is extremely flammable, so a measurement of residual O2 – again made possible by paramagnetic sensing – is essential to warn of any explosion risk. Measurement of CO2 in the vapour also provides an indication of any post-oxidation reaction. IR gas analysis can be used for this application, and also to measure water in liquid acetic acid, helping to control the recovery and purification of the solvent before it is recycled back into the process. The nitrogen inert blanket in the PTA product driers and storage bins should also be monitored for very low O2 levels, using a certified paramagnetic O2 analyser. In addition, PTA plants may use auxiliary boilers for heating parts of the process – the combustion reaction within these boilers can be monitored for reaction efficiency, control and safety by a combustion analyser measuring both O2 and combustibles.

Key gas analysis sensing technologies Paramagnetic

Ideal for reliable, accurate measurements in flammable or corrosive gas mixtures, paramagnetic technology provides fast, accurate and sensitive measurements of percentage levels of O2. The paramagnetic sensor cell consists of two nitrogen-filled glass spheres, mounted within a magnetic field, on a rotating suspension, with a centrally-placed mirror. Light shines on the mirror and is reflected onto a pair of photocells. As O2 is naturally paramagnetic, it is attracted to the magnetic field, displacing the glass spheres and causing suspension rotation, which is detected by the photocells. A current is applied through a feedback coil present in the magnetic field to provide sufficient torque to return the suspension to its original position. The magnitude of this current is directly proportional to the O2 present in the sample gas mixture. Unlike electrochemical sensing technologies, a paramagnetic cell never requires changing and its performance does not deteriorate over time, reducing ongoing maintenance requirements and delivering a long operational life.


IR sensing is a flexible, widely-used measurement technology based on the unique light-absorbing properties of some gases. It delivers non-contact, real-time detection of the selected gas’ concentration in a mixture. IR sensors focus an IR light source through a sample cell holding a continuously flowing sample of the gas mixture, and onto a detector after wavelength selection. The property of some gases to absorb unique light wavelengths can be used to detect the concentration of a selected gas in a mixture. Depending on the intended application, this concept can be applied in two ways: single beam, single wavelength (SBSW) and single beam, dual wavelength (SBDW). SBSW delivers fast, stable, and accurate real-time measurements with low maintenance requirements, and is used where a small transducer is required. The IR light source is electronically modulated, removing the need for a motor and rotating filters.

The SBDW method uses a pair of optical filters mounted on a rotating disc, which pass through a beam of IR light alternately. The measure filter is chosen to pass light only at a wavelength that the gas to be measured absorbs, while the reference filter has a light passed through it at a wavelength unaffected by the gas to be measured. The difference in absorbance is measured by the detector and provides a direct output of the gas concentration.

Gas filter correlation

Gas filter correlation (Gfx) technology is an enhanced version of the photometric analysis used in IR technologies. It is effective for applications where extremely accurate, low-level measurements are needed, or where background gases may interfere with the measurement. Gfx sensing exploits the ability of gases to absorb unique light wavelengths in order to detect the concentration of a selected gas in a mixture. Two gas-filled cuvettes are mounted on a rotating disk, each passing through a beam of light alternately. The measure cuvette is typically filled with nitrogen, while a second cuvette – the reference cuvette – is filled with a sample of the gas to be measured. Light is passed through the gas to be

Figure 2. The PTA process.





The new SERVOTOUGH SpectraExact 2500 upgrades the photometric gas analysis of its trusted predecessor, delivering an advanced solution to your process, in an easier-to-use package, leading the way for application solutions in a range of industries. • Ethylene production • Carbon capture and storage • Ethylene dichloride

• Direct reduction iron production

Get the latest launch news: servomex.com/signup

measured: the difference in absorbance is measured and provides a direct output of the gas concentration. Gfx measurements provide a real-time measurement response, unaffected by background gases, and the technique is virtually immune to obscuration of the optics. This prevents sensor drift, reducing calibration frequency.

Case study Servomex was selected by Jiaxing Petrochemical Co. Ltd to supply a complete analyser system for the second-phase development of its PTA production facility in Zhejiang province, China. The plant was supplied with three SERVOTOUGH Oxy 1900 paramagnetic O2 analysers and three SERVOTOUGH SpectraExact 2500 IR gas analysers, all integrated into a bespoke analyser house. The SERVOTOUGH Oxy 1900 uses paramagnetic O2 sensor technology to deliver accurate and stable measurements of O2. With a resilient enclosure for its transmitter unit, the analyser is optimised for hazardous applications including safety-critical oxidation, and meets Zone 1/Division 1 hazard ratings. A heated sample gas compartment improves measurement performance, while the non-depleting measurement technique means the analyser requires minimal calibration, ensuring a long lifetime of ownership and low maintenance costs. Combining gas filter correlation and IR sensing technologies, the SERVOTOUGH SpectraExact 2500 analyser is designed for a wide range of demanding process applications. Its robust sample cell is separated from the electronics, and is heated with solvent-resistant O-rings and calcium fluoride windows, making

it suitable for the high-temperature, acidic environments encountered in PTA production. For 2022, Servomex is releasing a new SpectraExact 2500, which will have upgraded electronics and a new user interface consistent with existing products in the company’s industrial process and emissions range. The all-digital design will have ATEX hazardous area certification (pending) and other global certifications. Jiaxing Petrochemical Co. Ltd selected these analysers because of their strong track records in delivering accurate, stable results in challenging environments, such as those found in PTA production. Servomex had also supplied a gas analysis system for the first phase of the Zhejiang project, and the reliability of this existing system – along with the service support offered in China – were major factors in the company being selected for the next stage of the development.

Conclusion PTA plants require a comprehensive gas analysis solution, with project management from design to installation and commissioning. To meet individual plant requirements for gas analysis, this may range from O2 sampling and analysis for quality and safety in the oxidation reactor, through to a complete solution for the plant that encompasses CO2 and CO process control analysis, combustion control and continuous emissions monitoring. In this way, manufacturers can ensure that a high-quality PTA product is delivered in an efficient and safe manner, while meeting all necessary environmental regulations.

Delayed Coking | Fluid Catalytic Cracking | Sulfur Production & Processing | Solvent Deasphalting ®

REFCOMM 2022 Training, Conference, & Exhibition May 9-13, 2022 • Galveston, Texas, USA

Returning to Galveston LIVE and in person I picked up a lot of good information. I would definitely encourage those that are interested to participate and not to miss this conference. Gregg Lorimor, Sr. Engineering Specialist, HollyFrontier Tulsa

Reasons to attend: Learn from 50+ technical presentation in the multi-track agenda Agenda covers coking, cat cracking, sulfur and SDA Network with a large audience of refiners and technical experts Develop practical solutions for optimising your coking unit with our fundamentals and advanced training courses

500 180



*Stats from Galveston 2019

Discover more at: www.refcomm.com




Tyler Schertz, Mettler Toledo, Switzerland, details the process of measuring small pipelines with tunable diode laser (TDL) using a multi-reflection folded optical path.

he chemical and petrochemical industries require real-time information on the dynamics of the production of an end product, in order to control the process and improve both safety and efficiency. To conform to these aspects, significant layers of automation are implemented to report the dynamic readings in the process, by means of various devices that detail the process variables such as pressure, temperature and concentrations.

Measurement techniques Concentration readings are provided by an analyser or chemical sensor. One notable example of this automated measurement is the assurance that oxygen does not ingress into a flare pipeline

between the knock-out vessel and the liquid seal to prevent explosive mixtures from forming. To obtain these measurements, a large number of classical analytical methods have been adapted and packaged so that the measurement of a gas such as oxygen can be taken extractively or in situ in a process stream without the need for laboratory analysis by taking an aliquot. Analytical technologies such as paramagnetic (oxygen measurement), nondispersive infrared sensor (NDIR), ultraviolet-visible (UV-vis) and electrochemical are used extensively in the aforementioned industries. Of these analyses, many relate to classical absorption spectroscopy and, more recently, tunable diode laser absorption spectroscopy (TDLAS or TDL).



February 2022

TDL is a mature analytical method and although it was primarily an extractive methodology and continues to be used as such, in recent years, multiple instrumentation vendors have established in situ measurements as cross-stack and probe designs. Mettler Toledo has also introduced a wafer process adaption for measurement in extremely small pipeline diameters of 5 – 10 cm (2 – 4 in).

The Beer-Lambert law Absorption spectroscopies in the domain of process industries include UV-vis, infrared (IR), NDIR, TDL and Raman, and are a consequence of the fact that molecules absorb light in a specific fashion. For example, light of specific frequencies is absorbed by a specific molecule in an exact and repeatable manner. The implication here is that a particular molecule at a specific concentration will absorb light of the same wavelengths and the resulting absorption of light will be proportional to the concentration. The physical law that is at the heart of this concept is the Beer-Lambert law that follows as equation 1: I = I0 e-acL

Figure 1. Spectroscopic differences between UV-vis absorption (top) and TDL absorption (bottom). Note the contrast in wavelength specificity.

Figure 2. The conceptual approach to path length multiplication by the White cell.



The absorption of light by molecules can be widely different. UV-vis absorptions are quite broad and use light in the UV and visible wavelengths. In contrast, near-infrared (NIR) spectroscopy, where TDL resides, has extremely narrow wavelengths and this is a major benefit to selectively choosing an absorption line that is unique to one analyte in a multicomponent process stream. Equation 1 relates the absorption (I) by the analyte to the total light emitted (I0). This absorption is proportional to the absorption coefficient (a) multiplied by the path length and the concentration. The absorption coefficient itself is dependent upon the pressure and temperature of the gas. Fundamentally, the Beer-Lambert law is summarised as: the absorbance of light energy is proportional to the path length of light passing through the sample, multiplied by the concentration of the species. It is easy to observe from equation 1 that low concentrations will present a very small absorption of light and that to increase the absorption of light, one must increase the optical path length (OPL) presented for the laser beam to transit or increase the concentration of the absorbing species. This article will further discuss the means by which a measurement for very low concentrations of analyte with TDL can be improved by increasing the OPL through the use of unique optical component arrangements, and the pros and cons of each.

Absorption spectroscopy Returning to the presentation of absorption spectroscopy and the major difference between TDL and others such as UV-vis, note the spectral wavelengths presented for a typical absorption in the graphs in Figure 1. Clearly, the advantage with TDL is the narrow absorption line(s) which is often unique for a particular analyte of interest and fits perfectly with the narrow spectral range presented by laser light. Other technologies such as paramagnetic analysers are susceptible to interferences, measure indirectly and cannot measure in the process itself. The Beer-Lambert law clearly implies that one of the major limitations to obtaining low level (ppm and ppb) measurements by TDL is potential limitations to the overall OPL. A general constraint of obtaining a TDL measurement is related to the commercial availability of lasers of the required wavelengths. Oxygen is a notoriously weak absorbing species with respect to the NIR range lasers on the market today. This critical gas measurement relies greatly on the ability to increase the OPL.

The process industries clearly demonstrate that the limitation here is most commonly the physical dimensions of the process piping in which the gas sample is contained for in situ measurements. For flue and other ducts of 1 – 3 m dimensions, path length does not become a major obstacle to the measurement in situ, even for weakly-absorbing species such as oxygen. However, when the diameter of the pipe is less than 38 cm (15 in.), the direct OPL for an in situ measurement by cross-pipe or probe design becomes significantly more difficult when measuring for oxygen below 5000 ppm (0.5%). TDL measurements in the latter example present only a few notable designs on the market for increasing the OPL of extractive measurements, and for in situ TDL there is only one prominent and unique process adaption by which to measure in small pipe diameters for oxygen. Additionally, when the cross stack is mounted longitudinally, the purge gas protecting the optics from contamination leads to the dilution of the sample stream. This is a significant consideration for the accurate measurement of the lower explosive limit (LEL) of oxygen in a process stream. Common methods to increase the OPL typically fall under the following three categories: nn White cell. nn Herriott cell. nn Optical prism arrangement (multi-reflection device).

White cell

The White cell (Figure 2) was first detailed by John White in 1942 in the Journal of the American Optical Society as a

means to increase the OPL using a compact arrangement of three spherical mirrors with the same radius of curvature.1 In this arrangement, the primary mirror is diametrically opposed to the secondary mirrors to provide an increase in path length that is always four-fold the distance between the primary and secondary mirrors. Although this accomplishes the goal of increasing the path length to achieve the desired detection sensitivity, it suffers from placing the optics in contact with the process gases, which can be caustic or contain contaminants such as dust and condensable materials. The other detriment to using this arrangement is that aligning three mirrors and spacing them accordingly can be rather difficult, and the precision wave front (smoothness) of the mirrors is an added expense and is easily destroyed by contamination from dust and condensates.

Herriott cell

The Herriott cell is another popular option with some TDL manufacturers to increase the OPL.2 This method relies on two spherical, highly-reflective and polished mirrors to propagate an incident laser beam between two spherical mirrors, as shown in Figure 3. Depending on the angle of the incident beam through a small hole on the primary mirror, path lengths of many metres are possible, providing very low detection limits. The Herriott cell has an advantage over the White cell in that the mirror arrangement is typically more stable and easier to align, with the primary factor being that there is one

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

© REMBE® | All rights reserved

Your Specialist for

PRESSURE RELIEF SOLUTIONS Consulting. Engineering. Products. Service.

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

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

Figure 3. A general perspective of the Herriott cell and the propagation of the beam line.

Figure 4. The optical arrangement for the multi-reflection prism arrangement.

Another significant form factor in the product portfolio is the unique process adaption aptly named the wafer, which functions as a spool piece within the pipe without obstructing the process flow. This can be fitted to pipe diameters as small as 10 cm (2 in. OPL). To address the severely shortened OPLs in these measurements, a new optical means was developed to increase the OPL in situ and fit with the form and function of the product line – the multi-reflection device (MRx). This new, commercially-available means to increase OPL is unique compared with the Herriott and White cell arrangements in that it does not rely on mirrors in a properly conditioned sample stream. The laser beam path in this optical arrangement is simply doubled or tripled by utilising optical prisms, as shown in Figure 4. The secondary optic (prism) in this in situ measurement is not in contact with the process gas, which is a significant benefit for use in harsh processes that could introduce optical fouling. It is also a significantly more compact technique than the other methods to increase the OPL, and this can benefit installations where the external physical dimensions are limited. The same secondary optical prism is used with both the two-fold (MR2) and three-fold (MR3) OPL adaptions. To achieve the three-fold reflection, an uncoated sapphire ‘cat eye’ lens is used, which is not only extremely durable and chemically-compatible, but also has the distinct advantage of eliminating flat reflective surfaces that could induce etalon effects and therefore improve the baseline noise that is inherent to direct absorption spectroscopy.

Conclusion less mirror in the arrangement to carefully position. Here again, however, there are a few disadvantages in that the optics are in contact with the process gases and the size and materials of construction generally relegate this to an extractive measurement. The disadvantages of both methods are significant considerations when the majority of industries utilising TDL consider in situ measurements as a significant factor for form and function of an analyser.

Optical prism arrangement

In the last two years, a new process adaption has been introduced into the TDL marketplace. The in situ self-aligning probe design offered by Mettler Toledo addresses customary pain points. Namely, it addresses the need to make a measurement within the process at critical measurement points for process safety and control, and provides a modular, universal design that is easy to maintain. One of the limitations to enclosing the optical components in a rigid body as designed by the company for in situ measurements is that often the pipeline diameter for the in situ measurement of oxygen, such as a flare header pipe, is limited to very small diameters. This often limits the insertion depth of the probe to 20 cm and thus restricts the lower detection limit of oxygen to thousands of ppm. February 2022 42 HYDROCARBON ENGINEERING

TDL spectroscopy is often considered by the process industries based on its long life, analyte specificity and capability to make fast in situ measurements for reasons of process safety and control with oxygen as the primary species of measurement. Oxygen unfortunately suffers from being a weakly-absorbing species in the perspective of the choice of adequate laser diodes. As such, means to increase the OPL are critical in TDL measurements of oxygen. A number of methods with regards to increasing the OPL have been introduced in this article. Two of the aforementioned methods suffer from notable drawbacks for use in harsh industrial processes. However, the MRx from Mettler Toledo helps to meet the needs and long operational life of TDL in chemical, petrochemical and other industries, while addressing the critical need to fit within compact pipeline diameters to measure low concentrations of oxygen for safety and process control. In the near future, it is anticipated that other path length increasing optical methods that correspond with the form and function of today’s needs and desires within the scope of TDL measurements will become available.

References 1. 2.

WHITE, J, U., ‘Long Optical Paths of Large Aperture,’ Journal of the Optical Society of America, Vol. 32, No. 5 (1942), pp. 285 – 288. HERRIOTT, D., and SCHULTE, H., ‘Folded Optical Delay Lines’, Applied Optics, Vol. 4, No. 8 (1965), p. 883.

Marco Puglisi, AEREON Europe, Italy, examines applications, equipment and technical solutions for gas recovery packages in the oil and gas sector.


n recent years, the oil and gas industry has been under a lot of pressure from environmental regulatory agencies worldwide to sensibly reduce the emissions of pollutants into the atmosphere. Regulatory attention has been focused not only on the reduction of the emissions of toxic and hazardous substances, but also on the reduction of greenhouse gas emissions – such as methane and carbon dioxide (CO2) – that in the long-term have been proven to have a harmful effect on the environment. The development and installation of gas recovery packages in various upstream, midstream and downstream applications of the oil and gas industry is an initiative in the aforementioned direction. Gas recovery packages are process units aimed at recovering and re-using the gas that would otherwise be either emitted to the atmosphere or thermally destroyed in a flare or thermal oxidiser.

Applications In the upstream sector, gas recovery units find application in well-head compression facilities. They allow for the continuous and stable production of compressed gas delivered to the sales pipeline, and enhance the life cycle of the well. The benefits of the installation of a gas recovery unit in a well or group of interconnected wells are evident, as it

provides a positive return on investment (ROI) thanks to the economic value of the recovered gas. Moreover, its presence allows for the operation of sites where flaring is restricted or not permitted. In the midstream sector, gas recovery units are utilised to recover the flash gas that originates from heaters-treaters, phase separators or during crude oil stabilisation process/storage. Typical examples are the dehydration process of well gas with ethylene glycol, and the light ends (C1, C2) removal from crude oil during the stabilisation process. In the downstream sector, gas recovery units are typically utilised in refineries or petrochemical plants to recover gas that otherwise would be flared; for this reason, they are called ‘flare gas recovery units’. A flare is a piece of equipment that is considered as a ‘safety device’ in refineries, petrochemical plants and oil and gas production/transformation sites. It provides safe and effective disposal of toxic and harmful gas during plant emergencies or upset conditions, such as fires, blocked outlets, control valves failures, cooling water failures, electric power failures, instrument air failures or runaway chemical reactions. A flare produces continuous flaring by collecting and disposing of process gas vented during start-up and shutdown, process transitions, and pressure safety valve (PSV) vents.



February 2022

How it works – compression technology The principle is similar with all gas recovery units. The flare gas is withdrawn upstream of the liquid seal, compressed through a positive displacement compressor, discharged into the separator, and then routed back to the plant. The liquid seal maintains a positive pressure on the flare header to prevent air ingress. If the flare gas stream exceeds flare gas recovery system (FGRS) capacity, excess gas will flow to the flare. If the flare gas stream is less, then the FGRS will turndown by staging compressors and/or recycle.

The heart of the unit is the compressor Figure 1. AEREON gas recovery unit in the upstream sector.

The compressor is a rotating machine, usually driven by electric power. There are two main families of compressors: positive displacement machines and dynamic machines. In positive displacement equipment, the gas is compressed at the pressure that is found in the downstream environment, whether it be a pipeline network or a vessel. In this family of machines, the most common technologies applied in a flare gas recovery (FGR) application are outlined below:

Liquid ring

The majority of FGRS installed worldwide are equipped with liquid ring compressors, and it has therefore became a sort of standard for FGR. They can guarantee discharge pressure up to 10 – 12 barg, but their efficiency in the compression process is low (25%). Sealing of the cylindrical impeller towards the casing (oval) is done with a ring of liquid (seal fluid), usually water, and therefore this machine requires a separator downstream. The eccentricity between the moving impeller and casing provides the compression of the gas, ensuring a balance of stress and forces. These compressors are flexible with changes in gas composition, can tolerate presence of particulate, and are reliable and robust if built following the API 681 standard.

Figure 2. AEREON gas recovery unit in the midstream


Continuous flaring not only has the negative effect of increased emission of pollutants, but it also has other disadvantages, such as noise, smell, and visible, smoky flames that can disturb surrounding neighbourhoods. It also increases steam consumption (for steam assisted flares) and electric power consumption (for air assisted flares). As such, a unit that reduces gas flaring and enhances the lifetime of the flare tip is preferable. Moreover, the recovered flare gas often has a considerable calorific value and can therefore be reused as fuel in the furnaces of the plant or process feedstock, reducing the usage of natural gas or LPG. Refinery flare gas often contains a sensible concentration of hydrogen sulfide (H2S). Before being reused in the fuel gas network in the furnaces, it must undergo a process of H2S removal through amine absorption. This purification process sets constraints on the minimum pressure at which the recovered gas needs to be available, in the range of 7 – 10 barg. February 2022 44 HYDROCARBON ENGINEERING

Sliding vane

Sliding vanes are mainly used in simple gas recovery packages for upstream and midstream applications. The maximum discharge pressure is approximately 10 barg and they are built with limited turndown. The compression process in these machines is very efficient (approximately 70%), resulting in lower power consumption. Disadvantages lie in the limited range of material of construction and their inability to follow API manufacturing standard.

Oil flooded screw

These compressors can reach discharge pressures of up to 20 barg; their compression efficiency is higher than that of liquid ring compressors (60%). The main disadvantage is that lubrication oil can come in contact with process gas and may become contaminated if not kept at a high temperature. They are typically found in well-head compression facilities.

Dry screw

These compressors can reach discharge pressures of 10 – 14 barg. The heat of compression remains in the process

Delivering sulphur solutions for a more sustainable world

Decrease OPEX

Reduce emissions

Our large range of technology solutions ensures our customers keep their costs low, reduce their carbon footprint and meet or exceed their sulphur recovery targets.

Optimize capacity

Learn more with one of our experts email: comprimo@worley.com Improve uptime


They are very efficient and reliable, with a narrow range gas composition (sensitive to surging), and are therefore not suitable for FGR applications. They are found in refineries’ fuel gas and/or hydrogen networks, where gas composition is well defined.

FGR design

Figure 3. AEREON FGR unit with liquid ring compressor.

The first step in designing a useful and efficient FGR system that is tailored to a customer is to carry out a FEED study. This should evaluate the following points as a minimum: nn The existing flare system (liquid seal depth). nn Flow metering of the flare gas for an extended period (one month) in two different periods of the year. nn Alternative uses for the recovered flare gas. nn Flare gas availability (cost). nn Flare gas composition range. nn Select compressor technology. nn System cost/benefit. nn Payback time and ROI of the FGR system as a function of the selected capacity vs availability of flare gas determined at flow metering (bullet point 2). Moreover, the selection of the compressor technology should take into account the following factors: nn Process inlet and outlet pressures and temperatures. nn Maintenance requirements. nn Plot plan availability. nn Utilities availability (power, water, steam). nn Payback time and ROI as a function of the chosen technology.

Figure 4. AEREON FGR unit with liquid ring compressor. gas and its temperature increases significantly; to keep the process safe and effective, multiple compression stages with intermediate intercooling are required. Variable speed drives (VSD) are used to control the flow and suction pressure; turndown is very high, at 10:1. These machines have high CAPEX.

Reciprocating compressors

Reciprocating compressors are typically not preferred in FGR applications, mainly due to their high maintenance requirements and complexity of construction and assembly. However, their advantage is that they can reach very high discharge pressures of more than 50 barg. They are typically found in the hydrogen network of refineries, where hydrogen needs to be compressed at high pressure in hydrotreating units to remove sulfur. In dynamic machines, the higher-pressure gas is produced by the dynamic action of the machine itself.

Centrifugal compressors

Centrifugal compressors are dynamic machines. Higher-pressure gas is delivered through a high revolution speed impeller with blades. They are run by VSD, which increases the turndown and optimises power consumption. February 2022 46 HYDROCARBON ENGINEERING

Auxiliary equipment to operate the gas recovery unit depends on the selected compressor technology. For any compressor where service liquid (water or oil) is mixed with the process gas, a separator vessel is required in the outlet line of the compressor. This provides liquid/gas separation and recycle of the gas to keep the suction pressure stable (turndown). The service liquid takes the heat of compression that needs to be removed. This can be done with an air cooler (axial fans on finned tubes) or a shell and tube heat exchanger, using cold water. In dry screw or reciprocating compressors, the heat of compression is taken by the gas, and interstage cooling is required. Depending on the compression technology, the gas inlet requirements may vary. For instance, gas must be clean and dry for screw and sliding vane compressors, but this is not critical for liquid ring compressors. In some cases, knock-out drums or scrubbers may be necessary at suction and discharge of the compressor.

Conclusion Gas recovery units provide the benefit of conserving resources and reducing emissions by recovering the process vent gas, which often has a considerable heating value. It is a process unit that controls pollution with excellent payback. However, an optimised gas recovery unit design and the selection of the right compression technology for a given application is not a straightforward and easy task; it requires a FEED study that takes into account several intercorrelating factors and should be carried out by qualified and experienced vendors.

Electric actuators for all types of industrial valves Reliable and long-term service. AUMA offers a comprehensive portfolio. Q

Customised solutions thanks to the modular scheme




Corrosion protection with offshore certification


Temperatures up to +75 °C, down to –60 °C


Integration into all conventional distributed control systems


Worldwide certifications and vendor approvals


Service worldwide

Discover our solutions for the oil and gas industry



Alec Cusick, Owens Corning, USA, explores the various considerations when designing a passive fire protection system.


herever flammable materials are present, so is the risk for a fire. Facilities that work with flammable petrochemicals or hydrocarbons must look beyond detection systems such as alarms, and even active systems such as extinguishing technologies, to consider the role of passive fire protection. In industrial environments that use, process or store flammable materials, an ignition source or spark may be all that is needed to trigger a fire. Ignition sources are abundant in industrial facilities. Welding, maintenance, electronic malfunctions, static electricity, lightning or autoignition can all trigger a fire. Passive fire protection systems support both detection

and active systems and bring an additional safety element to these sites when applied to critical assets such as piping, equipment, structural elements, tanks and storage vessels. While the word ‘passive’ typically does not imply ‘urgent’, passive systems are a non-negotiable factor for supporting safety in any environment where flammable chemicals are present. Considering that structural steel can lose up to 50% of its load bearing ability when heated to 538°C (1000°F), the importance of protecting the structures supporting volatile processes is evident. A loss in load bearing capacity could cause storage containers and equipment to buckle or collapse, potentially adding additional fuel to a fire.



February 2022

What components make up a passive fire protection system? Many consist of single or multi-layer jacketed insulation systems to slow the spread of heat into the material that is being protected (i.e. pipes/vessels/ structural beams). This strategy may allow additional time for occupants to evacuate the site or for employees to safely shut down equipment. Providing additional time may also support emergency responders’ efforts to fight the fire.

Figure 1. The furnace-based UL 1709 test establishes how long an insulating system is effective for, in terms of protecting the underlying structure and delaying heat penetration.

Figure 2. Jet fire testing determines a material’s resistance to high velocity flame and a rapid change in temperature.

When selecting insulation and designing passive fire protection systems, several elements should be considered. These factors include the performance of a material using standards-based testing methods, including but not limited to combustibility; smoke production; and flame spread characteristics.

Testing standards Two commonly referenced standards for passive fire protection systems are UL 1709, the standard for rapid rise fire tests of protection materials for structural steel, and ISO 22899, which determines the resistance of passive fire protection materials and systems to jet fires.1,2 UL 1709 measures how long a given system can offer protection to underlying equipment during a fire. Initially used with structural steel, this test has been expanded to address process piping, equipment and storage tanks or vessels. During the test, thermocouples are attached to a steel substrate and the passive fire protection system that is being tested is installed over the thermocouples and steel. The structure is placed in a furnace and heated to 1093°C (2000°F) over 5 minutes. The test stops when either a group of the protected thermocouples reaches 538°C (1000°F) or a single thermocouple reaches 649°C (1200°F). ISO 22899, sometimes called the ‘jet fire test’, examines the ability of a system to offer protection to underlying equipment during high velocity jet fire scenarios. During testing, thermocouples are again attached to a steel substrate under the passive fire protection system that is being evaluated. A directional, high intensity flame is then applied to the test specimen and time is recorded. The test stops when a single thermocouple reaches 400°C (752°F) above the initial ambient temperature at the beginning of the test. The presence of passive fire protection systems may also impact sizing requirements for relief valves used within a facility. Because keeping hydrocarbons in a safe, liquid form means that they are at extremely low temperatures, in many instances depressurising systems need to be incorporated into the designs of tanks and vessels. This allows for the safe release of pressure that could build up from the vaporisation that occurs during a fire. The amount of protection provided by passive fire protection may be considered when deciding how large a relief valve is needed for the depressurisation system. Relevant standards exist to aid the design of these relief valves, including API 2000 – venting atmospheric and low-pressure storage tanks – and NFPA 30 – the flammable and combustible liquids code.3,4 These standards provide equations to help establish the thermal credit provided by insulation, should a fire occur.

Evaluating insulation properties

Figure 3. Mineral wool insulation has been evaluated for use in passive fire protection systems because it is non-combustible and has a high service temperature.


When evaluating insulation for use in passive fire protection systems, it is important to consider several qualities. Typically, a material intended for use in a passive fire protection system should be examined for combustibility. It should also be checked for flame spread and smoke development properties in the event of a fire. The intention is to find materials that will not burn,

HELP PROTECT YOUR TANK BASE SYSTEM WITH FOAMGLAS® CELLULAR GLASS INSULATION Selecting the proper insulation for your tank base is critical to staying operational, safe, and high-performing. FOAMGLAS® High-Load-Bearing (HLB) Insulation provides constant thermal efficiency, boasts superior compressive strength, and is moisture impermeable and noncombustible. From cryogenic to hot tank base applications, FOAMGLAS® Cellular Glass Insulation has been trusted globally for decades.

Now Offering XL Block Sizes

Designed for ease of installation and more efficient applications on tank bases, FOAMGLAS® HLB Insulation is now available in larger sizes. FOAMGLAS® HLB Insulation is offered in six standard grades to meet the loading requirements for various tank base system designs. XL formats are available in FOAMGLAS® HLB 800, 1000, and 1200.

X-LARGE FORMAT 24 x 36 in 600 x 900 mm



18 x 24 in 450 x 600 mm Standard sizes still available.

www.foamglas.com 1-800-327-6126 © 2021 Owens Corning. All Rights Reserved. © 2021 Pittsburgh Corning, LLC. All Rights Reserved.

will not allow flames to move across a surface, and will not generate dense smoke, which could obscure escape paths and increase risk of smoke inhalation. Other elements to evaluate include the thermal conductivity of a material – how efficiently heat is able to pass through it. The specific heat of an insulation is also important to know, as it depicts how much heat must be applied to raise one unit of the material by one unit of

temperature. For example, the specific heat of water is 1.001 Btu/lb·°F (4.187 J/g·K) at 60°F (15°C). The maximum service temperature of the insulation material is also an important factor for design. However, the service temperature range on product data sheets often reflects the fitness for use of insulation for process control, condensation control or personnel protection, and not for performance of the material for passive fire protection. While a high service temperature could be beneficial for passive fire protection, it is not uncommon for this published service temperature to be exceeded in the event of a fire. In this case, the material may still provide passive fire protection as a sacrificial material. When evaluating a material for use for passive fire protection, one may prioritise finding insulations that have low thermal conductivity, high specific heat and fitness for use as passive fire protection.

Material matters when insulating passive systems Figure 4. Insulated systems can provide piping and

structural elements protection against both rapid rise and jet fire type situations.

Figure 5. Combining insulation types helps create passive fire protection systems that can address more than one facility challenge.

Insulating materials deliver different performance properties that affect their ability to function in different applications, including passive fire protection. Mineral wool has long been renowned for its fire resistant properties. The composition of the material helps explain its performance in passive fire protection systems. For example, Owens Corning’s mineral wool insulation is an inorganic fibre made from rock and recycled slag. This composition contributes to an insulation that is non-combustible, with a maximum service temperature of up to 649°C (1200°F) and flame resistance of 1093+°C (2000+°F). The material also has low flame spread, is water resistant, provides acoustic dampening, and is available in a range of shapes, sizes and flexibilities to promote easy application. Another non-combustible material is cellular glass insulation. In addition to its non-combustible properties, cellular glass insulation is also trusted in industrial and mission-critical applications, as it is impervious to moisture. Again, material composition contributes to performance. For example, FOAMGLAS® cellular glass insulation is a lightweight insulation comprised of completely sealed glass cells. The material is inorganic, non-permeable and has a wide service temperature range from -268°C to 482°C (-450°F – 900°F). It is dimensionally stable, non-combustible, will not give off smoke and has a high compressive strength. Additionally, it is easy to cut and handle on the jobsite and can be prefabricated into multiple shapes for easy application. The insulation(s) selected are typically paired with additional materials, such as jacketing. In some hybrid systems, the insulation is combined with an intumescent coating to provide extra protection. When exposed to high levels of heat, intumescent coatings char and expand to further slow heat reaching the underlying structure.

Tested systems Figure 6. Hybrid systems can also be used to provide passive fire protection at facilities with fire concerns.


Studies designed to evaluate the material performance validate use for passive fire protection. A series of insulation systems incorporating cellular glass and/or

mineral wool insulation along with jacketing and other accessories was tested using either ISO 22899 or UL 1709. Findings offer insight into the types of time ratings achieved by different passive system designs. For example, insulation intended for use on ambient or below ambient equipment, including piping, tanks, or structural steel, could consist of one layer of FOAMGLAS OneTM cellular glass insulation with PITTSEAL® sealant, PITTWRAP® B100 jacketing and 0.016 in. stainless steel cladding. When tested using UL 1709, this set-up provided up to 30 minutes of protection. The same insulation system with 0.027 in. stainless steel cladding also achieved 30 minutes of protection when tested according to the ISO ‘jet fire test’. Beyond providing passive fire safety, insulating systems can also address other issues at a facility, such as noise reduction. A different type of passive fire protection design combined cellular glass insulation with mineral wool pipe insulation, mass loaded vinyl, and stainless steel cladding. In addition to providing 120 minutes of protection in a jet fire situation and 180 minutes in a rapid rise fire, the insulating system has been tested to ISO 15665 for sound isolation performance.5 Additionally, passive fire protection systems can combine different materials to optimise performance. Owens Corning’s hybrid design combines cellular glass insulation, PC® 88/62 adhesive, PITTWRAP B100 jacketing, PPG Pitt-Char® NX and carbon mesh. This hybrid system can be adapted for both below and above ambient conditions, and provides up to 240 minutes of protection in a jet fire situation.

Vacuum Systems Process-integrated solutions for Ejector Vacuum Systems

Conclusion Safety will always be a top priority in any system design. The considerations discussed in this article can spark approaches to support life safety via passive fire protection systems and potentially deliver ancillary benefits such as noise reduction. It is important to seek technical expertise on insulation systems when designing a passive fire protection system, as results will vary with each unique project.

References 1.





‘Standard for rapid rise fire tests of protection materials for structural steel (UL 1709)’, Underwriter Laboratories, (2017), https://standardscatalog.ul.com/ProductDetail. aspx?productId=UL1709 ‘Determination of the resistance to jet fires of passive fire protection materials – Part 1: General requirements (ISO 22899-1:2007)’, International Standards Organization, (2007), https://www.iso.org/standard/39750.html ‘Venting atmospheric and low-pressure storage tanks (API 2000- 7ed)’, American Petroleum Institute, (2014), https://www.apiwebstore.org/publications/item. cgi?78c81e7e-5e5d-4d09-a5bf-c0d3a65b868e ‘Flammable and Combustible Liquids Code (NFPA 30-2021)’, National Fire Protection Association, (2021), https://www. nfpa.org/codes-and-standards/all-codes-and-standards/listof-codes-and-standards/detail?code=30 ‘Acoustic insulation for pipes, valves and flanges (ISO 15665:2003)’, International Standards Organization, (2003), https://www.iso.org/standard/28629.html

GEA supplies steam jet vacuum systems and hybrid vacuum pumps, optimizing production processes to reduce both costs and carbon emissions. With over 90 years of experience and thousands of references from satisfied customers from diverse industrial sectors all over the world, GEA is the partner you are looking for. Get in touch: gea.com/contact


ydrogen is one of the most talked about – and fastest growing – clean sources of fuel, and it is leading the way in the energy transition. However, while hydrogen is growing at an incredible pace as a fuel, it is not a new resource. Hydrogen production originally started more than a century ago, and has been supported for decades by Baker Hughes’ valves in the refining industry, primarily as a reactant feed to treat unrefined oil and gas products. Many of today’s wells are pulling heavy crude oil, which contains a high percentage of sulfur. However, the end customer markets are demanding improved diesel fuel with lower sulfur content. By virtue of this conflict of supply and demand, new refinery improvements have soared over recent decades, with expansions and new greenfield projects adding hydrotreating and hydrocracking units that inject hydrogen into the process to support this low sulfur content conversion. In addition, many refineries now include catalytic reforming – a chemical process used to create high octane products that generate hydrogen as a byproduct. As global demand for hydrogen increases, these proven and cost-effective methods of hydrogen production should remain a constant for many years to come.

Decades of process improvement The oldest, but still most common hydrogen production method, is steam reforming of natural gas. This moderate pressure production technology has been around for generations and has led to many developments and improvements in hydrogen processing, including enhanced specifications such as the National Association of Corrosion Engineers (NACE), to address hydrogen embrittlement. As hydrogen has an incredibly low molecular weight, its tiny


molecules can severely attack materials by easily penetrating voids, impacting castings, polymer diaphragms and other porous material surfaces if the materials are not properly specified. Further, as temperature increases, these molecules will diffuse into the steel at an even faster rate, combining with the carbon within the steel to form methane, leading to accelerated wear from corrosion. NACE hardness and radiographic quality specifications have emerged over the years to address embrittlement and corrosion from hydrogen. Within the steam methane reformer (SMR) there are several harsh applications that require severe service valve solutions. One example is when the process condensate lines with carbon dioxide (CO2), which creates a highly-corrosive carbonic acid requiring exotic trim materials depending on the level of concentration. Other applications such as the feed gas compressor anti-surge and recycle valves, and the carbon monoxide (CO) shift converter start-up vent valves, are examples of high-pressure reduction applications where rapid gas expansion will lead to high velocity and vibration-induced damage if not properly designed with multi-stage low noise trim. The pressure swing adsorption (PSA) plants add enhanced hydrogen purification, bringing the level to 99.99% purity – ideal for transportation and other energy uses. The greatest challenge within the PSA units is reliability due to continuous cycling, where valves are expected to exceed more than 100 000 cycles. Thorough laboratory validation is essential to ensure that the product is fit for the life cycle within these units.

Rathishkumar Sukumar and Raghavendra Mahalingam, Baker Hughes, discuss the role that specialty valves will play in delivering affordable hydrogen energy in the quest for net zero emissions.



February 2022

Coal gasification, which is mostly found in countries with scarce supply of natural oil and gas resources, is another source of hydrogen production, and has several unique valve application challenges that are critical to keeping the plant running. The gasifier units produce a synthesis gas, which is processed through a series of scrubbers to remove ash or slag from the coal. This black water fluid includes highly-erosive entrained solids, which require sweep angle letdown valves coupled with tungsten carbide trim to avoid rapid failure and unit shutdown. Baker Hughes has developed proprietary, dual grade materials using additive manufacturing to blend ductile properties of stainless steel with the erosion resistance of a hardened tungsten carbide to create a custom material for longer lasting service. Other challenging applications are the air separation units (ASU) that operate under cryogenic temperatures to separate oxygen from air. As oxygen is highly combustible, care must be taken to select materials, as having low spark tendencies, such as Alloy 625 or Monel 400/K500, followed by the removal of foreign particles, could lead to ignition. UV cleaning in a certified clean room is absolutely required to ensure safety of the plant and operators.

The transition to blue hydrogen The adoption and transition of hydrogen as a primary fuel source is largely limited to the economic challenges when competing against gasoline. The traditional hydrocarbon-based processes remain the most efficient and cost-effective methods of hydrogen production, however their production generates CO2. As such, they are not considered a truly clean fuel source. The use of carbon capture technologies followed by geosequestration is required to transition these processes to a carbon-free ecosystem. The capture of CO2 emissions is of critical strategic importance for sustainable large-scale production of hydrogen from natural gas or coal, without increasing greenhouse gas emissions. Present day analysis shows that producing blue hydrogen with carbon capture and sequestration can be an economical option to match the price range of gasoline or diesel fuels. However, carbon credits from government policies are still essential to offset this balance. The advancement of carbon capture technology in both greenfield construction and the retrofitting of existing assets is critical to delivering CO2 emissions reductions needed to meet global 2050 net zero emissions targets.

Figure 1. The hydrogen liquefaction and transportation process.


Present day carbon capture technologies are either amine-based or novel salt-based absorptions that rely on commodity chemicals. Baker Hughes currently has a licensing agreement in place that leverages SRI International’s Mixed-Salt Process (MSP) for CO2 capture with benefits of a low-carbon manufacturing footprint, reduced energy consumption and greater efficiency. The technology also differentiates itself from other amine-based carbon capture technologies by negligible solvent-degradation, reduced steam and water use, and low reboiler duty for solvent regeneration, and seems more promising for the future than conventional amine-based processes. The captured CO2 is compressed (>130 bar) before delivery to storage sites, such as depleted oil reservoirs, or may be used for an enhanced oil recovery process. Injection pressures may exceed what is noted above, due to the fact that voids in oilwells gradually fill as time progresses. Compressor anti-surge recycle valves for such high-pressure drops expose the potential for CO2 icing, due to the Joule-Thomson effect, inside the active trim pressure drop stages. Multi-stage or multi-path trim designs may be needed to handle these issues in order to avoid large single step drops that can fall below the critical temperature limits.

Liquefaction and transportation Traditional refinery production of hydrogen has often led to onsite consumption to satisfy local needs. The increased demand for this clean fuel is driving a greater need for the transportation of gas through pipelines or liquefaction of the gas followed by condensed product transportation using portable vessels such as tankers. Pipeline transportation remains attractive over long distances within a continent, and more than 4000 km of hydrogen pipeline exists worldwide today (2000+ km in the US alone). Yet even with this benefit of existing infrastructure, there remains an abundance of challenges in transporting this highly-volatile fluid. First, the explosive nature of hydrogen must always be considered. Safety measures must be taken, and even more so when the fluid is compressed or heated, as this may lead to violent combustion or explosion. After proper safety measures are in place, additional factors must be considered, such as pipeline material selection. Hydrogen will degrade mechanical properties of most metals and leaks three times faster than natural gas. Hydrogen blending with natural gas is an alternative method to change the chemistry for transportation and reduce degradation rates. For these pipeline services, rotary valves are commonly used for isolation and control. When the hydrogen percentage of the fluid is high, the standard valve seals made from materials such as viton, nitrile rubber (NBR) or neoprene rubber will have high permeation and may later be at risk of explosive decompression (ED). Under pressure, the hydrogen atom can permeate through elastomer materials, and upon release of the pressure it can rapidly escape and cause ED. For these applications, ED-resistant materials are required to prevent a catastrophic failure. Hydrogen liquefaction is one of the most significant and challenging processes in the entire transportation system, but it does enable more range for cross-ocean transportation by tanker. Storing hydrogen as a liquid for compact

Global presence, local support Valve failure can be catastrophic for people and processes. However, routine valve maintenance can be costly — especially if you don’t know what parts or repairs you’ll need.

Optimize Service With Baker Hughes

Our valve lifecycle management tools will diagnose problems before they arise and identify performance issues ahead of milestone maintenance events, improving your planning and inventory capabilities. For unparalleled OEM support within hours, take advantage of our extensive field service, inventory and product networks, as well as our 80 field support and maintenance facilities across 30 countries. You’ll also enjoy factory-certified commissioning, inspection, repairs and training, quick access to parts and ValvFAST 24-hour delivery.

To learn more, visit valves.bakerhughes.com.

bakerhughes.com © 2021 Baker Hughes Company. All rights reserved.

transportation requires cryogenic temperature reduction to below -253°C (-423°F). The use of liquid hydrogen originally started in the space industries, and it has been used as a fuel for several decades. For over 40 years, Baker Hughes has supplied cryogenic control valves for liquid hydrogen service, especially for the bench testing of cryogenic engines. The company’s long-standing relationship with the cryogenic industry is based on the use of a single-seated valve made from specially selected materials, with the packing separated from bonnet by means of an extension. To simplify the problems of control and installations in liquid hydrogen application, valves are specially made and must meet the following industrial requirements: nn Minimum material (to reduce inflow of heat by conduction) in the cold zone. nn Quick and easy access to the seat area, body and plug, without having to remove the insulation located inside the cold box.

nn Live-load packing located remotely from the cold zone with fugitive emission standards. nn Bonnet (and gasket seal) away from the cryogen cold zone to prevent any leakage into the insulated zone. nn Simple assembly and ease of maintenance.

The future of green hydrogen Today’s ongoing race is to establish a truly clean and cost-effective process to produce hydrogen through electrolysis of water. The benefit of this process is that it only produces hydrogen and oxygen, without any CO2 or methane byproducts. However, generating large volumes of hydrogen through electrolysis requires large amounts of power, which currently increases the cost of the production dramatically compared to traditional hydrocarbon-based methods. Hydrogen production powered by renewable energy sources, such as wind or solar energy, would truly be the cleanest option available without any trace back to carbon-based output, but this currently comes at a high cost, requiring effective government subsidisation. Present prominent electrolysis technologies use either liquid alkaline electrolyte bases or membrane technologies such as a proton exchange membrane (PEM) or solid oxide electrolysis (SOE). Despite their simplicity, these electrolysis technologies consume a large amount of power and remain very costly and uncompetitive compared to alternative methods. Typically, within moderate sized electrolysis units (PEM cells), there are approximately 100 control valves on various systems, including process water, cooling water, hydrogen purification, oxygen purification, nitrogen purging, and many others. Alternatively, there is the high temperature electrolysis process, where heat from industrial processes, such as in a nuclear reactor, can be used to improve the efficiency of electrolysis to produce hydrogen. By increasing the temperature of the water using super-heated steam, the decomposition potential (voltage) of water is decreased. As such, less electricity is required to split it into hydrogen and oxygen, reducing the total energy required. The current trade off for all the cases is the balance between cost and carbon-based output.


Figure 2. The Masoneilan 41005 Series is ideal for high pressure reduction, large temperature variations and cryogenic service, and is available up to 30 in. in size.


Though the processes for production and use of hydrogen are not something new, the global need for clean energy is growing at an unprecedented rate. The energy transition is driving the need for use of hydrogen as a cleaner fuel for vehicles and power generation, in turn driving an increased demand for hydrogen specialty valves and application-based knowledge. In petroleum refining applications, NACE standards address material selections, but new hydrogen designs within the industry still have no such standard for material selection. A considerable amount of experience is needed in this space to specify valve solutions to materials. For several decades, Baker Hughes has supplied valves for standard hydrogen applications in process industries and to the extreme cryogenic application in the aerospace industry. Today, the company continues to charter for innovation to bring energy forward, making it safer, cleaner and more efficient for people and the planet.

Alejandro Plazas, ValvTechnologies, Latin America, explains why considering the total cost of ownership of valves will ultimately pay big dividends.


ll industries calculate the total cost of ownership, particularly when they plan new projects or evaluate unit maintenance costs. Unit by unit, they evaluate equipment, seeking the lowest lifetime cost. The

challenge is to realistically and reliably apply the concept to different processes, industries and equipment. Thinking of a valve as an investment rather than a cost changes perspectives. This article discusses how to calculate the



February 2022

total cost of valve ownership, specifically when it comes to isolation valves. The process entails considering the cost of equipment acquisition, maintenance, repair and replacement, all during a specified time.

nn Valve locking or stroke issues, mainly on emergency shutdown valves (ESDVs) and blowdown valves (BDVs). If a valve that isolates a process does not move, the consequences can be catastrophic.

Steps to take

Next, it is important to evaluate the total cost of ownership through the following five steps:

The first step is to define isolation valve failures and consider the following risks: nn Leaks through the stem create fugitive environmental emissions. nn Internal leaks through the seats of a closed valve. These leaks degrade process safety, availability and efficiency.

Define the time range

To fully analyse the economics of a valve purchase, the use of two operating windows or scheduled plant shutdowns is recommended.

Identify the valves to analyse

For simplicity, no more than four valves are recommended at one time. Valves that have some relationship to each other should be chosen, enabling you to fully understand their operational effect on the plant.

Calculate total maintenance costs

Figure 1. Comparison between the valves’ initial purchase,

maintenance and replacement costs during the specified time range.

Include the valves’ initial purchase price, installation cost, preventive maintenance costs, repair costs (multiplied by the number of repairs expected during the valves’ operational life) and replacement costs (multiplied by the number of replacements expected during the identified period). This is shown in Figure 1. Installation costs are sometimes negligible and sometimes quite high. In offshore applications such as FPSOs and platforms, installation and transportation costs can approach the purchase price of a new valve. To evaluate a valve purchase, it is important to consider not only the aforementioned maintenance costs, but also the costs of valve failures, including the following:

Unscheduled shutdown costs

Valve failure can cause an unscheduled shutdown if it creates a safety risk to people or the environment, or damages equipment. First, it must be determined whether the process needs to be stopped completely to replace the valve, thereby reducing production, or if it is a batch process that requires an extended cycle time due to the intervention. Second, costs needed to be quantified. Start with the per-day cost of a unit shutdown; divide it into an hourly cost; then multiply by the number of hours needed to repair or replace the valve (see Figure 2).

Efficiency-related costs

This analysis is one of the most complex because it requires a high level of process knowledge. Some specific examples include:

Figure 2. Comparison between initial purchase, maintenance and replacement costs, plus the costs of unscheduled shutdowns to change valves.


Increased fuel consumption If drain and vent valves in boilers, generation and cogeneration plants leak, the energy required to



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

ABC. See it. Believe it. Trust it.


heat the water is lost, since the steam never reaches the turbine. This reduces the plant’s thermal efficiency and increases its fuel costs. These losses and increases can be calculated using the heat rate and the cost of fuel.

Many other costs can be associated with isolation valves, however the determination of the aforementioned costs permits a useful analysis.

Lengthened operating cycles for batch applications

This article will now evaluate two possible solutions for calculating the total cost of ownership of a catalyst withdrawal application in an FCC unit.

If an isolation valve of a multitrain decoking system leaks, the pressure available for the cutting water system declines, increasing the time required to decoke the reactors. This longer coking/decoking cycle shrinks overall production. To calculate this, multiply the additional time by the hourly cost calculated for unscheduled shutdown costs.

Increased downtime due to leaking valves In refinery fractionator-tower bottoms systems with two pumps, pumps must be periodically shut down for maintenance. In this case, the failure of an isolation valve will considerably escalate downtime.

Table 1. The cost of ownership

Calculating the total cost of ownership

Solution 1

nn Two gate valves. nn Valve 1 must be replaced annually, and valve 2 every two to three years. nn Operational window: five years. nn Shutdown cost: US$250 000/d. nn Time required to change valve 1: 1 hour. nn Time required to change valve 2: 36 hours. nn Valves 1 and 2 begin to leak after six months of operation. n Catalyst loss per day: 1 kg/d average during six months.

Solution 1

Solution 2


Valve purchase price

US$20 000

US$210 000 -US$190 000

Total purchase, maintenance and replacement costs*

US$317 398

US$699 574 -US$382 176

Total purchase, maintenance and replacement costs, plus cost of unscheduled shutdowns*

US$3 502 719

US$699 574 -US$2 803 145

Total purchase, maintenance, replacement and unscheduled shutdown costs, plus cost of lost catalyst*

US$3 717 197

US$699 574 -US$3 017 623

Note: *Total costs calculated for 10 years ** Negative value means solution 2 is more expensive than solution 1 when only considering purchase, maintenance and replacement costs

Solution 2

n Two ValvTechnologies zero-leakage ball valves for isolation, and a gate valve for throttling. n Operational window: five years. n Shutdown cost: US$250 000/d. n Replace gate valve every two years. n Replace valve 1 every five years. n Replace valve 2 every 10 years. n Time required to change gate valve: 1 hour. n Time required to change valves 1 and 2: none, because it is performed at the scheduled shutdowns. n No catalyst loss. n Analysis time: 10 years – equivalent to two operational windows.

Conclusion For solution 2, the decade-long total cost of ownership is just 25% of solution 1 (see Figure 3), despite its higher initial investment. The difference lies in eliminating costs for unscheduled shutdowns and catalyst losses. Similar analysis has shown comparable results in a variety of processes and industries, including: n PIG launchers and receivers. n Delayed coking units. n Dehydration with molecular sieve. n Instrumented safety systems. n Fractionation-tower bottoms. n Catalytic cracking units. n ESDVs and BDVs. n Gas injection (API 10 000).

Figure 3. Comparison between initial purchase, maintenance, replacement and unscheduled shutdown costs, plus the cost of catalyst lost from valve leaks.


Using these tools to calculate the total cost of ownership before making your next valve purchase will likely pay big dividends.


Carina Wegener, REMBE GmbH, Germany, discusses how to achieve the right installation torque with a virtual calculation.

etermining the required installation torque for a leak-tight flange system is a daily challenge for industrial valve manufacturers and plant operators alike. All of the components in the flange system have individual installation requirements, and environmental considerations add further difficulty to the task. The collaboration between the individual component manufacturers and the plant operator is crucial to everything working in harmony. This article will describe the challenge in greater detail, and outline the possible solutions that are available.

Selecting the installation torque Generally, there are different approaches when it comes to selecting the required installation torque. Alongside manufacturer information for individual components, there are

calculation standards such as DIN EN 1591-1 or the AD 2000 set of rules, with which conventional flange systems can be calculated analytically. According to these standards, a conventional flange system (see Figure 1, left) is defined as a flange inlet, flange outlet, gasket and connecting elements (e.g. bolts). In practice, a flange system is often expanded by additional components (see Figure 1, right). This can be a pressure protection mechanism, such as a rupture disc. A rupture disc is usually installed with a mounting unit, consisting of a holder inlet and holder outlet. A second gasket is also required. The conventional calculation standards become invalid for these expanded flange systems, as the equations underlying these standards do not take any additional components into account. Metallic sealing surface contact has also been excluded up until now, as have systems whose stiffness HYDROCARBON 63


February 2022

Figure 1. Comparison of a conventional flange system (left) and a flange system with rupture disc and holder (right). fluctuates greatly over the width of the gasket. Notably, metallic sealing surface contact will be integrated into future requirements due to the amendment of the Technical Instructions on Air Quality Control (TA Luft). From the perspective of the rupture disc manufacturer, the requirements for the design of the flange connection are primarily determined by the functionality of the rupture disc. With regards to the connection between the holder and rupture disc (see Figure 1, right, connection disc – holder), the contact pressure must be sufficient in order to hold the rupture disc and prevent it from being pulled out during the process. The system must also be tight so that the medium does not leak out. That being said, the contact pressure cannot be so high that it leads to the destruction of the rupture disc material. As a result of these factors, there is a range of permitted surface pressures in this connection, which must be set via the installation torque. A gasket manufacturer focuses on the sealing point between the flange and gasket (see Figure 1, connection flange – gasket). Every gasket manufacturer specifies a minimum and maximum sealing surface pressure, creating a range which must also be set via the installation torque. Two different requirements for the necessary installation torque are produced: in most applications, rupture discs are manufactured from stainless steel, whereas gaskets are often made from non-metallic materials. Different materials with different properties each have a specific required surface pressure. To adjust this surface pressure, however, only one shared installation torque is to be set for both cases. This creates an optimisation problem. The solution lies in the geometric alteration of the contact surfaces between the rupture disc and holder, and is therefore down to the rupture disc manufacturer. Due to the metallic sealing in this connection (see Figure 1, connection disc – holder), the required surface pressures in this range are usually higher than the permitted gasket compression. With targeted design measures, excessive tension is created in the sealing surface between the rupture disc and holder. This allows the optimum surface pressure to be set in the sealing surface without the surface pressure in the gaskets increasing to an inadmissible extent. February 2022 64 HYDROCARBON ENGINEERING

In addition, when calculating the installation torque, it is important to investigate whether the components of the flange system are overloaded in the respective constellation. Installing a rupture disc and a holder in a flange system causes the mechanical properties of the overall system to change. This results in greater flange blad tilt and tension in the flange. If the rupture disc is triggered by an overloading of the process, it must then be replaced along with the gaskets. However, the flange and holder are often used multiple times and must retain their function-securing geometry after repeated dismantling and installation. Permanent plastic deformations in the loading conditions installation, testing and operation are therefore not permitted. To ensure this, a strength test must be carried out for the configuration, which ensures a sufficient distance from the material’s elasticity limit for all loading conditions.

Conclusion All of the required calculations for configuring the installation torque cannot currently be performed with sufficient precision, using analytical calculation principles. Having taken into account the complexity and the ever-greater challenges of the configuration of the installation torque, REMBE uses virtual engineering with the finite elements method. This makes it possible to depict the flange system – including all additional components – as a digital twin, and thus simulate its mechanical behaviour before manufacture and commissioning. With this digital twin, the mutual influences of all of the components are taken into account in advance, before being brought together in reality. The simulation results and empirical knowledge are then used to optimise the individual components virtually until all of the components are in harmony with their specific installation requirements. Only after this are the components manufactured or ordered, and then installed. REMBE offers such calculation services. For plant operators, there is no longer the need to involve an engineering firm or similar service providers. The customer receives a complete solution with real added value, including reliable simulation results calculated by specialists, which then flow into the customer-specific manufacture of the rupture discs.

A. Brighenti, C. Brighenti, M. Ricatto, and D. Quintabà, S.A.T.E. Systems and Advanced Technologies Engineering S.r.l., Italy, outline how to monitor and diagnose multiple plant units while considering operational contexts.


ny kind of complex system presents the personnel responsible for checking performance with demanding tasks. Asset performance management (APM) software is often used to help with the supervision of operations within plants that reportedly also use methods based on artificial intelligence (AI) algorithms. However, while these systems are able to detect the present status of a plant, most of them are still lacking reliable failure prediction capability or the remaining useful life (RUL) at subsystem or component level. Different operational structures may determine different normal behaviour patterns that must be distinguished in order to reliably identify anomalies and predict failures. This article describes the approach and the typical results obtainable by a set of algorithms resulting from two decades of

R&D and application experience by SATE1 - 6, implemented as a suite of software components and referred to as Diagnostic Kernel Modules (DKM). They are now the core of a number of diagnostic applications being proven in the operational environment of space satellites constellations (CASTeC4, 6), industrial vehicle fleets2,3, and hydrocarbon production facilities, the latter of which is under an ongoing technology transfer and demonstration project with the cooperation of a leading international oil company. Despite the very different operational environments, system characteristics and dynamics, the approach towards detecting novel or anomalous systems behaviour – which eventually become failures – to form a ranking of likelihood of possible failure modes and a prediction of unacceptable functionality (i.e. failure), relies on common data-based and/or HYDROCARBON 65


February 2022

In this scenario, the typical methodologies to investigate the presence of anomalies in operational telemetry data are based on fixed thresholds.7 - 9 These techniques are able to assess if the system remains within the allowed boundaries, but do not detect trends of incipient anomalies, which may result in The common problem unexpected system behaviour during operations. During operations, Satellite Flight Control Engineers (FCEs) and Another limitation of the traditional checking approach is industrial Plants Operation Supervisors (POS) are in charge of that these techniques do not allow for the detection of checking the behaviour of their assets – satellites of a contextual anomalies, i.e. telemetries behaviours that are constellation and equipment and machinery, respectively – anomalous under certain operational contexts. The aiming to detect anomalous behaviour early enough to context-based approach that is part of the DKM suite provides implement the right counteractions for mitigating the related significant improvement in data-driven anomaly detection. risks and costs. Contexts are defined based on ambient or operative conditions that occur several times during a system’s lifetime. For instance, context can be related to positional information, e.g. the satellite eclipse condition, or to a functional condition, e.g. specific configuration imposed by the operators. In hydrocarbon processing plants, contexts could be specific operational phases or season dependent conditions, process control settings, or normal start-up or shutdown. An example is provided in Figure 1, which shows a parameter time series with two different nominal ranges depending on the status of subsystem A (on or off). When subsystem A is off, the parameter typically ranges within the Figure 1. Parameter time series with two different nominal ranges depending blue bounds (bottom right plot), and on the status of subsystem A. These two different configurations identify two when subsystem A is on, the parameter different contexts. typically ranges within the green bounds model-based methods. The DKM suite is an example of the successful transfer of technology from ground industrial systems (fixed and mobile) to space systems, and vice versa.

Figure 2. Time series of the evolution of the DKM health index (top) as a result of parameter behaviour (bottom), with its traditional nominal bounds (dashed red lines). The vertical red bar shows the time at which traditional anomaly detection methods raise an alarm.



LNG application field

Gas pipeline feeding. Gas power generation stations. Gas peak shaving plants and LNG transfer in small to medium size shore terminals. The experience acquired through the complete pumping systems supply launched Vanzetti Engineering in the development of systems and components dedicated to any type of LNG and LBG applications.

VANZETTI ENGINEERING. Widening the horizons of LNG sustainability. www.vanzettiengineering.com

(bottom left plot). In this case, the use of traditional fixed thresholds would not detect an anomalous evolution of the signal when a specific status of subsystem A is active. The influence of context could impact a large set of telemetries. In a hydrocarbon processing plant, for instance, the pressure control setting in a reactor unit fed by a gas compressor influences the operating point position on its operating envelope (i.e. its characteristics map). It is well known that when this is close to the surge limit line, flow measurement and antisurge valve position oscillations are possible and normal, within certain limits. On the contrary, they are less affected by oscillations when the operating point is further to the right of the surge limit line. Therefore, for sensors or compressor diagnostics using flow meter, pressure, temperature and valve position as input, and that use symptomatic features, the oscillating patterns or value thresholds of these signals should take into account the

influence of the downstream unit setting (i.e. the compressor operating context). This makes a context-based approach even more powerful. Another important aspect to highlight is that the large amount of data that is generated throughout the system design, testing and operation phases can be exploited to extract knowledge that can be used to improve the latter phases (OPS). In the space sector, in particular, increasing attention has been paid to learning the behaviour of satellites from operations data, and improving monitoring and diagnostics as a result.1, 10, 11 From the operational data, new contexts may be identified, leading to the definition of more detailed contextualised operative nominal ranges. This same approach may well be applied to other types of system, such as automotive power train systems and hydrocarbon processing plants. New strategies can be envisaged to improve anomaly detection and investigation during systems operations, characterised by: nn Context-based reasoning. nn Operational data exploitation. nn Predictive capabilities to detect trends of degradation and anomalies well before these evolve into more severe events. In the following section, the first and the last points will be briefly discussed.

Context-based approach

Figure 3. Fault Isolation Module (FIM) result – example of power train diagnostics.

Figure 4. Example of four correlated parameters.


The context-based approach is part of the DKM software tool set that can provide: nn Early identification of anomalies in the behaviour of the system relative to a contextualised standard, characterised by specific operations conditions, e.g. configuration, manoeuvres, process setting, season or specific known fault modes.




TRI-BLOCK Double Block and Bleed Valve

nn Identification of root causes or correlated events, enabling fault isolation and mitigation strategies definition. nn Identification of critical operative conditions that may affect a system’s service performance. The DKM diagnostic approach is based on the eventually combined analysis of telemetries and on a set of features extracted from the raw data, which can be simple statistical quantities such as average, maximum, minimum of the parameters in certain time windows, or more innovative or complex features. From the analysis of these features, DKM first characterises the expected nominal behaviour and then uses this characterisation to compute an index that measures the degree

of anomaly at parameter level, subsystem level and unit level (e.g. a satellite of a constellation or a plant unit). Figure 2 shows telemetry raw data (blue in the bottom plot) of a power train emission control system, with its traditional nominal bounds (dashed red lines). This signal shows a normally oscillating behaviour associated with the engine power (left part of the plot, until Aug 27). Then, the range of the signal oscillation changes, but in small increments and within its nominal upper bound. The DKM provides a system status index, called health index (HI), where 0 = faulty and 100 = healthy, based on one or several features calculated from the raw data. In this case, the features calculated by DKM and used to evaluate the HI determine its gradual decrease (upper plot in Figure 2) and, based on defined crossing thresholds of the HI, an early alert (Aug 31) that largely anticipates the fault detection alarm raised by the traditional onboard diagnostics (Sep 18). The DKM provides a sorted list of relevant events that may need further investigation by in-field or specialist engineers. This list contains all of the analysed parameters, sorted by HI in increasing order, so that the most critical telemetries are at the top of the list. This list allows engineers to execute further investigation, as they can filter the parameters, access details of the symptoms, and investigate the reasons for the detected anomalies. The HI is computed not only at parameter level but also at subsystem and unit level, and produces both detailed and global indications of the health status of the subsystems and system (e.g. all compressors, pumps or filters of a plant). HI values of each analysed parameter can be visualised through a heat map panel that shows all telemetries grouped by subsystems in a topological representation, such as the one used in the software CASTeC, applied to satellite constellations.4 - 6

Anomalies correlation and fault isolation Often, the effects of an anomalous event can be observed in multiple parameters, and can belong to different subsystems. For this reason, in addition to the identification of anomalies through the analysis of the parameters, the DKM suite includes tools for the fault isolation, i.e. the identification of the most likely root causes (if known a priori, as shown in Figure 3) or the detection of correlated anomaly events or telemetries, which are highlighted to the engineer. The engineer can then explore raw data and telecommand information in dedicated panels. The advantage of an approach that combines early anomaly detection and correlated events detection is that it also allows for the extraction of new knowledge. In Figure 4, a set of parameters from a real satellite mission are shown. These parameters present some anomalies in the periods highlighted in red and yellow points in the plot. These anomalies were identified by DKM as correlated, which was of unexpected relevance and interest to the engineers to whom these results were reported, as no correlation was expected among those parameters.

RUL estimate Figures 5. These four graphs depict DKM failures prediction based on the HI trend.


Anomaly detection and fault isolation are important capabilities of diagnostic tools to be integrated in APM systems, as they allow for the reduction or elimination of useless maintenance and replacement actions and subsequent costs.


International Aboveground Storage Tank Conference & Trade Show April 13-15, 2022 Rosen Shingle Creek Hotel | Orlando, Florida

The conference focus is field-erected bulk petroleum storage tank systems, but it includes field-erected and shop-fabricated aboveground tanks and piping containing petroleum, hazardous substances, and food and vegetable oils. NISTM Conferences also specialize on the environmental issues associated with ASTs, while covering the important operational, management, and corrosion prevention aspects of terminal operations. This conference is for anyone working with AST standards and systems, and it provides an excellent opportunity to network with individuals involved with operations, construction, compliance, regulation, management, and spill prevention and response.




www.NISTM.org | 800.827.3515

International 011.813.851.1700

However, the next and most demanded feature of diagnostic systems is the ability to predict when a failure is going to occur, i.e. the RUL of the component, subsystem or system for which the anomaly has been detected (prognostics). The four graphs in Figure 5 show a real situation whereby DKM was applied to vehicle power trains, with no loss of generality. The top plot shows the HI time history with its reduction and trespassing of the alert threshold, down to reaching a critical condition of the subsystem observed (in this case, the pollutants absorption system of an internal combustion engine). The middle plots show the predicted time and confidence interval of the critical event, at two subsequent times and HI thresholds passing, when DKM is used for real time predictive diagnosis. The bottom plot shows the last predicted and the actual failure times (in green and red, respectively). It is clear that DKM detected the anomaly in due time, and anticipated the actual event. In this case, the failure did not compromise the use of the vehicle, however the operation of the emission reduction system was unacceptable. The vehicle operated for several hours before the stopping at a workshop, after which it recovered the nominal condition (HI ≈ 100%).

Conclusion The context-based telemetry data checking approach is based on DKM – a tool that provides predictive alerts on the status of a plant, vehicle or spacecraft system or its subsystems, and produces a priority list of anomalies, possible root causes and RUL estimate. The advantage of this tool is that it implements a fully context-sensitive, interpretable, data-driven approach which is beneficial for understanding the reasons behind a detected

Page Number | Advertiser 61 | ABC 33 | AMETEK Process Instruments 34 | AMPP 31 | Ariel Corp. 47 | AUMA 57 | Baker Hughes 45 | Comprimo 19 | Elliott Group 02 | Eurotecnica 53 | GEA IBC | Global Hydrogen Review IFC | Haldor Topsoe 13 | Merichem Company 27

Mitsubishi Heavy Industries Compressor International


anomaly. This interpretability is a feature that is hardly covered by state-of-the-art deep learning or, more generally, by AI techniques typically exploited in this field. Finally, it does not require experts’ knowledge to be configured, but simply the knowledge of the relevant operational contexts, i.e. specific system operational conditions.

References 1.

BIANCAT, J., BRIGHENTI, C., BRIGHENTI, A., MARTÍNEZ-HERAS, J. A., DONATI, A., and EVANS, D., ‘Priority scores based on novelty detection to improve the efficiency of ground-operations’, BiDS 2016 Conference, (March 2016). 2. BRIGHENTI, A., ‘Model-Based Diagnostics - A real opportunity for efficient vehicles management’, SAE On-Board Diagnostics Symposium, (September 2009). 3. BRIGHENTI, A., et al, ‘Exploiting existing information through a vehicle network to identify faults & improve maintenance’, Mobility, No. 15, (2009). 4. BRIGHENTI, C., AMORUSO, L., EVANS, D., BRIGHENTI, A., MORETTO, D., RICATTO, M., FERRARI, F., and CARBONE, M., ‘Advances in context aware spacecraft telemetry checking’, IAC 2018, (October 2018). 5. BRIGHENTI, C., RICATTO, M., ZORZI, A., BRIGHENTI, F., BARISON, M., AMORUSO, L., and CARBONE, M., ‘Constellation Monitoring With CASTeC’, International Astronautical Congress – IAC 2021, (October 2021). 6. ‘CASTeC™ – from space systems to production plants’, (October 2021), https://www.hydrocarbonengineering.com/special-reports/05102021/ castec-from-space-systems-to-production-plants/ 7. MARTINEZ-HERAS, J. A., et al, ‘New Telemetry Monitoring Paradigm with Novelty Detection’, SpaceOps 2012 Conference, (June 2012). 8. HUNDMAN, K., CONSTANTINOU, V., LAPORTE, C., COLWELL, I., and SODERSTROM, T., ‘Detecting Spacecraft Anomalies Using LSTMs and Nonparametric Dynamic Thresholding’, KDD ’18: The 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, (August 2018). 9. VARUN, C., BANERJEE, A., and KUMAR, V., ‘Anomaly detection: A survey’, ACM computing surveys (CSUR), Vol. 41, No. 3, (2009), pp. 1 – 58. 10. MARTINEZ, J., LUCAS, L., and DONATI, A., ‘Dependency finder: surprising relationships in telemetry’, SpaceOps 2018 Conference, (2018). 11. MARTINEZ, J., and DONATI, A., ‘Novelty Detection with Deep Learning’, SpaceOps 2018 Conference, (2018).

AD INDEX 11 24 23 71 15 OFC & 51 07 38 41 37 04 61 67 OBC 69

| Mokveld Valves BV | NEO Monitors | NEUMAN & ESSER | NISTM | Optimized Gas Treating, Inc. | Owens Corning | Paqell | RefComm | REMBE® GmbH Safety+Control | Servomex | Sulzer Chemtech | ValvTechnologies | Vanzetti Engineering | Zeeco Inc. | Zwick Armaturen GmbH

Global Publication

A new magazine focused on the global hydrogen sector

Subscribe for free: www.globalhydrogenreview.com