Cultivating effective technical collaboration HYDROCARBON PROCESSING

Transcription

1 Culivaing effecive echnical collaboraion A GRANULAR LOOK AT HOW SOME HIGHLY SUCCESSFUL REFINERY PROJECTS ACHIEVED THEIR OBJECTIVES Shell Global Soluions

2 INTRODUCTION THE ARCHITECTURE I was ineresed o read a recen chief execuive officer sudy 1 by IBM ha found ha he world s op companies IBM called hem ouperformers are hose ha innovae wih exernal parners. The repor says ha he ouperformers are embracing new models of working ha ap Süleyman Özmen ino he collecive inelligence of an organisaion and is neworks o devise new ideas and soluions for increased profiabiliy and growh. This resonaes wih me because, a Shell Global Soluions, we also believe ha you can never bring oo many differen perspecives o bear on a problem. When we engage wih refiners o, for insance, help hem o improve undersanding of heir opions for processing a wider range of crude oils, enhancing heir yield of high-value producs, improving heir environmenal fooprin, or achieving long-erm susained operaional excellence, we follow a highly diagnosic proocol. This involves working wih he refiner a various levels in order o benefi from he experise, experience and fresh hinking ha exis wihin heir organisaion. There is evidence ha his collaboraive approach can help cusomers o achieve heir business objecives more effecively. This is highly relevan because he increasingly volaile naure of he refining indusry is aking is oll. According o one repor, nearly 40% of he European Union s 104 refineries are in need of refurbishmen. Analyss have prediced refinery closures of some 4.3 MMbbl/d of global disillaion capaciy beween 2009 and 2014, which is abou 5% of global capaciy. The mohballing of Peroplus s refining asses earlier his year ook he closures o abou 2.7%, according o Deusche Bank. Tough imes appear o loom for oher refiners. Moreover, wherever you are in he world, you are likely o find he financing marke excepionally difficul, wih capial shor and expensive. So, when you do commi o a projec, i is vial ha he iniiaive is he righ one, ha i delivers wha i was designed o and, equally, ha he objecives are robus o fuure marke changes. There is no room for error. OF VALUE CREATION A Shell Global Soluions, our approach o echnical collaboraion is designed o miigae hese risks. We call i co-creaion, bu here is a myh in he indusry ha co-creaion simply enails seeking he inpu of anoher pary. Co-creaion is much, much more han ha. I is a sophisicaed approach o echnical collaboraion ha brings ogeher disincive viewpoins o provide a fundamenally beer undersanding of a cusomer s problem and o evaluae he wides possible echnical soluion se. As shown in Figure 1, co-creaion is buil ino an archiecure ha hard-wires boh he owner s business objecives and he curren (or impending) exernal pressures affecing he owner ino he decision-making process. Anoher key feaure of his framework is ha i is designed o faciliae high-qualiy ineracion beween he owner and our consulans and echnologiss. This helps o merge he cusomer s highly valuable sie-specific knowledge and markeplace perspecives wih our global operaional and echnical experise. The exernal pressures and cusomer objecives feed ino four key decision-making areas. The firs is idenifying he projecs ha could achieve he refiner s objecives. Improvemens can be made righ across he value chain, from crude evaluaion and procuremen hrough operaions and processing o invesmen planning, and experience can help o idenify he wides possible echnical soluion se. We call his area value creaion. The second key area is opionaliy. There are usually many ways in which a soluion can be delivered. For insance, if he objecive is o enhance he yield of high-value producs, he refiner should evaluae he benefis of a uni revamp, he possibiliy of upgrading exising hardware and he feasibiliy of building a new uni. This is also he fundamenal poin of he Shell Global Soluions Muliplaform Penagon Model, which you may have heard me discussing in recen years. The hird area reflecs oday s rapidly changing world where he decisions we make oday will be measured in erms of how hey play ou omorrow. We someimes find ha execuives base heir decisions on a narrow premise ha reflecs he curren siuaion bu ha may be a odds wih eiher heir SHELL GLOBAL SOLUTIONS VALUE CREATION ARCHITECTURE hard-wires boh he owner s business objecives and he curren (or impending) exernal pressures affecing he owner ino he decision-making process. longer-erm objecives or poenial fuure marke condiions. Reliabiliy consideraions also come ino play here, as hey help o deermine he plan s abiliy o generae reurns. These consideraions are designed o help preven regre decisions and are crucial in oday s fas-changing world. We call his area of he archiecure susainabiliy. The fourh area is execuion: ensuring ha he plan sars up on ime and on budge, and works in concer wih all he exising pieces of hardware and he inerfaces ha make he curren refinery work so well. Underlying all of hese facors are healh, safey and environmen consideraions. These issues are of grea imporance o Shell Global Soluions and are herefore cenral o our hinking. Ouperformers, IBM says, do no work in isolaion. They undersand ha good echnical ideas ofen bear frui hrough echnical alliances wih organisaions whose srenghs differ from heir own. As you will see on he pages ha follow, his is a view shared by Grupa LOTOS, Shell Scoford and Ukranafa and Shell Global Soluions value creaion archiecure helped hem o foser a rich exchange of knowledge, ideas and perspecives. Süleyman Özmen Vice Presiden, Refining and Chemical Licensing Shell Global Soluions Inernaional BV Reference 1 Leading Through Connecions: Insighs from he Global Chief Execuive Officer Sudy, IBM, 2012 EXTERNAL PRESSURES OWNER S OBJECTIVES Abou us LICENSOR S PERSPECTIVE VALUE CREATION SUSTAINABILITY OPTIONALITY EXECUTION OWNER S PERSPECTIVE Figure 1: Shell Global Soluions value creaion archiecure. n Shell runs more han 30 refineries and chemical plans, produces 3.2 MMbbl of gas and oil every day, and is acive in over 130 counries and erriories worldwide. n Shell Global Soluions is a sraegic licensor providing echnical suppor o over 50 sies worldwide and is par of Shell Projecs & Technology. Many of our professionals have operaional and echnical experience across a broad range of perochemical processing and producion indusries. Their experise and perspecives help o creae a deep level of undersanding of our cliens business challenges and enable us o provide effecive soluions. n Crierion Caalyss & Technologies (Crierion) has a rack record of over 50 years and is wholly owned by Shell. Is caalyss can be found in abou 60 of he world s hydrocracking unis and approximaely 850 hydroreaing unis. Crierion is also he marke leader in caalyss for diesel dewaxing. A large proporion of he ebullaed-bed residue upgrading unis around he world conain is caalyss, as do he majoriy of ail-gas reaing unis. n Zeolys Inernaional (Zeolys) is a global leader in commercial and specialiy zeolies. Is advanced hydrocracking caalyss are currenly insalled in more han 70 of he world s hydrocracking unis. CO-CREATED SOLUTION

3 HOW CO-CREATION CAN IMPROVE THE VALUE DELIVERED BY CAPITAL PROJECTS Only he bes planned and execued projecs successfully achieve all heir objecives. Someimes, even hose ha use sophisicaed risk analysis and modelling echniques only achieve parial success, and some become regre invesmens. Do you ever find ha iniiaives such as a revamp or he insallaion of a new process uni fall shor of delivering he promised arges? For insance, do hey fail o deliver he full yield or margin uplif ha was indicaed in he projec valuaion and feasibiliy sudies? Combining our cusomers srenghs in erms of heir undersanding of heir markeplace, business objecives, echnology and processes wih our echnical skills and global operaional experience can creae a coheren and compelling exchange of value. This is an area in which Shell Global Soluions leadership eam is acuely ineresed. For some ime now, we have been scruinising capial projecs around he globe and have idenified a series of key risks ha he op performers successfully miigae on heir way o achieving projec excellence. We describe hese laer in his aricle; see secion Why projecs ofen fail: Four key risks. Miigaing he risks is anoher maer, as, in our view, i requires he refiner o work wih is sraegic suppliers in a fundamenally differen way wih new levels of openness. A new paradigm is emerging. The convenional paradigm saw refiners issuing inviaions o bid o prospecive suppliers afer defining he business case and locking in he echnical specificaion hemselves. Then hey would go ou o bidders and compare heir proposals on he basis of price. The problems are, however, ha refinery projecs are complex and valuable invesmens, and ha much of he overall value a projec can deliver is buil in a he sar. The assumpions behind he business case and he decisions made during fron-end developmen can make or break a projec s effeciveness. In conras, he emerging paradigm sees he refiner including is echnology suppliers as an inegral par of he projec eam. The refiner shares is business drivers and assumpions wih he suppliers so ha, ogeher, hey can develop a cos-opimised specificaion. They can joinly explore echnical opions and bring o bear experience and experise from all paries o arrive a robus echnical soluions. Shell Global Soluions value creaion archiecure (see page 2) provides he plaform for his way of working, which we call co-creaion. Experience shows ha i can creae a coheren and compelling exchange of value, and i works because i has wo inheren philosophies. Firs, he refiner recognises ha, even hough i has a clear idea of where i wans o ge o, i may achieve a beer oucome by bringing in echnical experience from ouside. Working in andem wih an organisaion ha has probably seen and deal wih similar echnical challenges and complex operaing siuaions a oher faciliies can bring a differen level of undersanding o he problem, and muliple differen or novel echnical soluions can ofen be devised. Second, he supplier appreciaes ha he refiner has a vially imporan perspecive of is markeplace, business objecives, echnology and processes. So, over ime, we a Shell Global Soluions have changed he way we work wih our cusomers. Alhough he business acumen and echnical experise of our consulans and echnical eams remain imporan, we recognise ha heir qualiy of ineracion when rying o undersand cusomers wide-ranging business challenges as hey deliver ha experise is equally imporan. This is how we make sure we consider our cusomers underlying business objecives when devising echnical soluions. This approach of echnical collaboraion works paricularly well for innovaive, enrepreneurial, high-performing enerprises: businesses ha like o challenge he convenional wisdom and seek o idenify and evaluae he wides possible se of opions. Because hey embrace openness and accep he marke s volailiy, hey leverage exising experience and knowledge, and bes pracices and echniques ha are successful in oher faciliies. Crucially, as his approach hard-wires he cusomer s business objecives ino he decision-making process, i can help o make sure a projec successfully achieves is objecives. WHY PROJECTS OFTEN FAIL: FOUR KEY RISKS Shell Global Soluions believes ha refiners can heighen heir chances of achieving all of heir projec objecives by addressing he following risks. Key risk No. 1: Will he high volailiy of he curren fas-moving markes preven my projec from delivering he promised value? A projec may no deliver he promised value if an iniiaive is defined on a narrow premise ha underesimaes he poenial marke changes. The risk can be avoided by embracing he volailiy. For example, some refiners responded o he downurn by reducing invesmen levels, cuing coss and waiing unil normaliy reurned. Ohers, however, recognised ha he long-erm oil and gas fundamenals remained posiive and sough ou opporuniies o exploi hem. For example, a refinery in Asia incorporaed a winer diesel capabiliy: by invesing a relaively small amoun of capial, he refinery was able o creae addiional margin, which helped o make i a more robus business

4 Progressive companies recognise ha where here are challenges here are ofen opporuniies. These, however, will be specific o individual businesses. Unearhing hem will involve a horough examinaion of he model on which he business is run, along wih a reassessmen of is emphasis and direcion. Key risk No. 2: Will delays and cos overruns derail my projec? Sums of up o $1 million a day in los revenue can occur when sar-up is delayed on large projecs, and he broader impacs of delays on a projec s ne presen value are rarely fully facored ino he financial risk analysis. A hree-monh delay in sar-up can urn a business-criical projec ino a regre invesmen, so i is imperaive ha projec eams ake seps o guard agains plan underperformance during commissioning and sar-up. In addiion, here are many examples of projecs whose coss have spiralled ou of conrol owing o exernal variables such as he price of seel, labour or financing. However, experience shows ha he effecs of exernal evens can someimes be miigaed wihou affecing a projec s overall value. For example, he global economic crisis challenged many refiners plans for capial projecs bu hese companies were able o proceed by reacing quickly o he changing markeplace. Reaining a high level of flexibiliy afer he final invesmen decision is a key aribue. For example, when he projec economics of a Chinese refiner s in-developmen hydrocracker began o look less favourable, managemen was quick o ake acion o bring he coss down. The refiner brough ogeher expers from is business wih represenaives of Shell Global Soluions o discuss how he projec could be configured differenly. I was hrough his echnical collaboraion ha he refiner idenified ha i was possible o achieve he same capaciy from wo rains insead of hree wihou affecing he uni s oupu. Key risk No. 3: Will he high capial cos make my projec unviable? In oday s financial markes, here is always he risk ha you will be unable o secure invesmen for your business-criical projec, a leas on favourable erms. However, here are usually many alernaives ha could mee he needs of he marke. When a business seeks he perspecive of an experienced consulan o help i challenge convenional wisdom, working ogeher in his way ofen co-creaes innovaive, highly effecive echnical soluions ha neiher would have idenified on heir own providing here is high-qualiy ineracion. Take, for insance, a European cusomer ha waned o build a mild hydrocracker complex o process a mix of medium and heavy vacuum gas oil (VGO), and also o hydrorea he gas oil from a crude disiller. Many refineries achieve his hrough wo separae sandard designs. However, because capial was a major issue for his refiner, i was keen o invesigae novel soluions. As a resul, a combined eam of he clien s senior personnel and saff from Shell Global Soluions devised an innovaive soluion: an inegraed reacor secion ha would squeeze down he capial cos. The hydrodesulphurisaion (HDS) reacor and he hydrocracking reacor were combined wih a common separaion, compression and fracionaion secion. Because he cusomer only needed one uni insead of wo, he capial cos was significanly lower. Key risk No. 4: Will my new asses operae reliably enough o generae sufficien cash o cover he financing coss? Even wih exensive process modelling, you canno be compleely sure ha your new echnology will deliver wha i was designed o do. Will i hi is arges in erms of cycle lengh and yield profile? Wha impac will i have on your environmenal performance? Are you confiden ha i will no rigger any unanicipaed effecs in he unis up- or downsream? Tha was he issue encounered by a US refinery. I had a long rack record of processing difficul, discouned crude oils o reurn an impressive produc slae of clean fuels. However, a move o impor more opporuniy crudes severely curailed he cycle lengh in a kerosene hydroreaer. The refiner had o increase he operaing emperaure in a kerosene hydroreaer by 8 11 C o mee he ulra-low-sulphur-diesel (ULSD) specificaion, and he caalys deacivaion rae ripled. A eam of hydroprocessing specialiss from he refinery and Crierion examined he composiion and properies of he feed and formulaed he bes operaing sraegy for he hydroreaing uni and he upsream crude disillaion uni. As a resul of he changes ha hey made, which included insalling a new caalys as well as process changes, he cycle lengh was doubled. Graham Wood Soluions Developmen & Markeing Manager Shell Global Soluions Inernaional BV STRAIGHT TALK ON NEXT-GENERATION HYDROCRACKING TECHNOLOGY Shell s new deasphaled oil (DAO) hydrocracker a Grupa LOTOS SA s Gdañsk refinery in Poland is processing abou 50% DAO and 50% VGO a up o 85% conversion direcly ino je fuel and Euro 5 diesel. In an inerview, wo senior managers a he refinery reveal o Hydrocarbon Processing how he uni has performed over is firs year of operaion and he effec ha i is having on he faciliy s economics. AT A GLANCE The Grupa LOTOS 10+ projec The Gdañsk refinery is highly complex and he second larges refinery in Poland. When increasingly sringen produc specificaions hreaened is compeiiveness, i ook decisive acion by launching he 10+ projec, a massive $2 billion upgrade of he refinery. The iniiaive, which was compleed in 2011, has helped Grupa LOTOS o: n raise refining capaciy by 75%; n focus producion on higher-margin diesel fuels; n increase marke share; and n enhance margins by $5 per barrel. Shell Global Soluions DAO hydrocracker was key o he scheme. The refiner also insalled an inegraed crude disillaion vacuum disillaion uni, an HDS uni, a solven deasphaler (SDA), an amine sulphur recovery uni, a hydrogen manufacuring uni, a refinery harbour produc pipeline, produc anks and pumping saions, and revamped is hydrogen recovery uni. Couresy of Grupa Loos SA

5 Dariusz Kruk Grzegorz Orzeszko Dariusz Kruk, Manager, Hydrocracking Operaions, Grupa LOTOS SA, has 14 years of experience in hydrocracking and has overall responsibiliy for he uni s operaions. Grzegorz Orzeszko, Depuy Manager, Hydrocracker Operaions, Grupa LOTOS SA, played a key role hroughou all he projec phases, from he iniial basis of design hrough he invesmen process o sar-up. Clearly, he caalyss are doing heir job very well. Bu i is also reassuring ha Crierion s echnologiss are monioring he uni and advising on he poenial of newly developed caalyss ha could help o opimise he uni even furher. Q: The global refining indusry is exremely ineresed in your hydrocracker because DAO has radiionally been used as a fluidised caalyic cracker feed. How has i performed over is firs year of operaion? Orzeszko: I am pleased o repor ha i is operaing exremely well: here have been no process upses and, since sar-up, all he performance guaranees have been me. There have been no produc qualiy issues, and i is generaing an exremely high yield of premium producs. Q: A hydrocracker can provide a refinery wih much-needed flexibiliy in erms of feed and produc slae. Have you aken advanage of his? Kruk: Absoluely. The hydrocracker s flexibiliy has been of special significance. We have been able o increase he conversion level from he planned 60% o 80 85%. We also rouinely fine-une he DAO/VGO balance beween 40 and 60% in response o processing objecives. The DAO we currenly process is of a beer qualiy han he much heavier design feed, which was 100% DAO derived from 45% amospheric residue and 55% vacuum residue. A he momen, we have no need o process he design feed because he refinery has been reconfigured. Since sar-up, our focus has been on opimising he profiabiliy of our overall invesmen, so all our effors have been on increasing conversion and processing addiional DAO. Q: Wih he markes for fuel oil diminishing, refiners residue sraegies can have a subsanial boom-line effec. Wha do you do wih he unconvered residue from he hydrocracker? Orzeszko: We reprocess i as much as possible. Some sreams are roued back o he hydrocrackers, which increases overall conversion, bu mos go o our lubrican oil block, some via bach vacuum disillaion, o produce high-qualiy base oils. These are usually Group II, bu we expec Group III oils o be possible in he fuure. Q: And wha abou oher refinery residue sreams? Kruk: Currenly, we sell he residue as fuel oil in winer, and, in summer, we use i o produce asphal, which is currenly in high demand in Poland. Bu we recognise ha we need o reduce he amoun of residue we produce. We had originally planned o insall a gasifier, bu he projec coss became asronomical owing o he rising coss of seel, engineering and so on. Insead, we worked wih Shell Global Soluions o phase he invesmen. So, firs we buil he hydrocracker ogeher wih an SDA and a hydrogen plan o supply he hydrocracker. Now, we are considering insalling a coker plan. This will mach he overall configuraion of he refinery, as i will enable he hydrocracker o be run consanly a higher conversion and a high hroughpu. This will deliver a beer feed han we are geing from he SDA only. Moreover, wih boh a coker and an SDA plan, we will be able o swing conversion beween hose wo unis and srike a balance beween minimum coke yield and opimal hydrocracker feed qualiy. Q: Since i came on sream, have you aken any seps o opimise he performance and operaion of he hydrocracker? Kruk: Yes indeed. There are small weaks ha you can make o improve efficiency, especially when he feed qualiy and he severiy of he process have changed. Many seps have been aken. For insance, because he hroughpu was lower, here has been exra room for opimisaion, especially for hea inegraion. We have changed he operaing parameers of he hea exchangers in boh he reacor and he fracionaion secions. As resul, we have cu he plan s energy consumpion by some 25%. Q: Have you encounered any issues caused by increasing he conversion beyond he design level? Orzeszko: Yes, here were some, bu we had anicipaed hose anyway. We work closely wih our counerpars a Shell Global Soluions and Crierion; hey help us hroughou he run cycle. They are fully aware of our operaing sraegy wih he uni and wha we are rying o achieve. Before changing any process parameers, we always consul hem and ha approach is paying dividends. Combining heir process and caalys experise wih our sie-specific knowledge is an exremely powerful way of meeing our operaional objecives. Kruk: One of he issues ha occurred was an increase in polynuclear aromaics (PNA), which can lead o hea exchanger fouling and caalys deacivaion. Nohing is free, and his is he price of higher conversion and hydrowax recycling. For sable hydrocracking operaions, he PNA level mus be kep wihin a cerain window. So, wih Shell Global Soluions, we analyse feed samples and creae a PNA balance for he plan s feed sysem. This has helped us o idenify he mos severe PNA sreams, and now we preferenially roue hose away from he hydrocracker feed. Consequenly, we have sabilised he PNA levels. Q: Has he difference in operaing duy affeced he caalys s performance? Kruk: Even hough he caalys was seleced on he basis of a differen feedsock and for 60% conversion raher han 80 85% conversion, i is operaing exremely well. The seleciviy for middle disillaes is very good. Of course, we consanly monior he uni o make sure ha increasing he conversion does no cause complicaions wih he caalys sysem or affec he caalys s performance. Convenional hydrocracking parameers such as nirogen remain imporan and he large quaniies of DAO in he feed mean ha he meals conen and he Conradson carbon level mus be carefully moniored. Orzeszko: Hydrogen consumpion is also very imporan, as we are hydrogen consrained here a Gdañsk like mos oher European refineries. We monior hydrogen consumpion and, so far, he levels have been wihin he performance guaranees. Clearly, he caalyss are doing heir job very well. Bu i is also reassuring ha Crierion s echnologiss are monioring he uni and advising on he poenial of newly developed caalyss ha could help o opimise he uni even furher. Q: This is an exremely large and complex hydrocracker, and you were performing he fron-end developmen and implemenaion phases during he global economy s crisis of Did you see ha as a risk? Orzeszko: The global economic crisis inroduced a layer of uncerainy bu we had o remain focused on he long-erm picure. We always believed ha he indusry fundamenals remained unchanged and he invesmen is now proving robus. I is rue ha elsewhere in he indusry projecs were being pu on hold or cancelled. Bu we were deermined o work wih Shell Global Soluions o adjus he scope of our projec o process a differen feed han he one ha we had originally designed i for. Tha has allowed us o phase he projec, and we are ready for heavier feedsock in fuure. Those changes, which included opimising he conversion level and phasing he projec, enhanced he projec s viabiliy and enabled i o go ahead. And now we are seeing he benefis of ha. I have no doub ha we would be far less compeiive if we had no invesed in his projec. I has been a remendous success

6 MANAGING COMPLEXITY, OPTIMISING INTEGRATION AND REASSURING INVESTORS The value of employing an EXPERIENCED sraegic licensor As a sraegic licensor ha has operaional experience, Shell Global Soluions srives o ake a holisic approach during projecs because i sees opporuniies o be he archiec of value creaion across he business, no jus wihin an individual projec or aciviy. Our value creaion archiecure (see page 2) provides he plaform for his. A wide-ranging porfolio of refinery processes is also a prerequisie for his approach, which is why, a Shell Global Soluions, we acively seek sraegic echnical alliances wih specialis companies ha offer leading-edge echnologies in areas where we do no have a proprieary echnology. A good example of his is he Shell Sulphur Technology Plaform hrough which we license a suie of echnologies ha can handle sulphur in any form: in crude, producs and emissions, and also as a solid. Many of hese are proprieary echnologies, bu we also have agreemens wih, for example, Cansolv Technologies Inc. (now owned by Shell), Merichem, Praxair, Sandvik and Sulzer o complee our porfolio. Cusomers also receive consisen guaranees ha are aligned wih heir overall objecives, which can be exremely reassuring, especially o financiers. This can also minimise he poenial for liabiliy dispues. I heard of a refinery ha insalled a Claus uni ha was oo big: hey could no urn i down sufficienly for i o operae. I seems ha he licensor, which was no an operaor, designed in coningency capaciy in all he upsream unis ha resuled in a delivered uni so large ha here was no enough hydrogen sulphide in he refinery o fill i, even a maximum urndown. Inegraing new process unis is rarely sraighforward, bu experience can ofen prove invaluable. Employing a sraegic licensor ha has iself operaed large, complex indusrial faciliies migh also help you o address invesor concerns and, as a resul, secure preferable erms for finance and insurance. In financiers eyes, i helps o miigae he compleion risks associaed wih complexiy and inegraion; i is much less risky han having muliple parners. Refinery projecs can encouner problems if hey become unexpecedly complex or fail owing o insufficien inegraion beween he disillaion, conversion, reaing and emissions conrol processes. The impac on a business can be devasaing bu, by following a differen approach, such issues can usually be avoided. Asounding levels of complexiy can be encounered during he insallaion of new process unis owing o he overwhelming number of inerfaces wih oher pars of he plan. For insance, he Pearl GTL developmen in Qaar involved 800 pipelines crossing beween conracs, and here were over 2,600 cases where i was necessary o pass process daa among conracors. There were more han 3,600 consrucion inerface poins, including hose for he exchange of schedule informaion and overlapping responsibiliies for cross-conrac sysem hydroesing. This is clearly no an assignmen for he fain-heared. As a general rule of humb, every inerface increases projec complexiy by a square facor; i is no unusual for a projec s complexiy o mushroom. There are various forms of inerface, including echnical inerfaces ha occur beween srucures and process unis, and organisaional inerfaces. Invesigaion revealed failures in he laer caegory o be he cause of he Three Mile Island nuclear inciden in A Qaar, he projec schedule was mainained because he complexiy was miigaed wih ried-and-esed echniques. Seamless inegraion among conracors was achieved hrough sophisicaed work processes ha faciliaed communicaion and daa exchange, and by an inerface daabase ha consolidaed all he inerface poins and issued weekly saus repors. On highly complex capial projecs, he value of experience and bes pracices canno be oversaed. Decisions made during he fron-end developmen phase can have a major influence on complexiy. For insance, I have heard of refiners ha have licensed muliple echnologies from differen suppliers only o find ou someimes when i was oo lae ha hey do no mach a he inerfaces. Ofen, he cause is ha hey were designed o differen sandards or philosophies. Sourcing from muliple suppliers is he radiional approach, as i enables soluions o be cherry-picked and he various suppliers o be beaen down on price. Increasingly, however, he progressive refiners are realising he subsanial value ha is a risk wih his approach: delays, rework coss and plan underperformance. Sourcing from a limied number of licensors can help o avoid his. For insance, an owner can license a package of residue upgrading echnologies ha combines he high vacuum uni, he hydrocracker and he solven dewaxing uni from one supplier. Many cusomers find his o be a much beer way of execuing a projec because heir dealings are only wih one pary he sraegic licensor. I hen becomes he sraegic licensor s responsibiliy o ensure, for insance, ha he numerous inerfaces mach and ha he schedule is opimised. I also promoes safey. The Grupa LOTOS DAO hydrocracker demonsraes he cusomer value of such an alliance. As discussed in he aricle on page 7, he qualiy of he DAO feed from KBR s SDA uni o Shell Global Soluions DAO hydrocracker was pivoal. In heory, Grupa LOTOS could have licensed a sandalone SDA uni, bu ha opion would have missed he opporuniy o opimise he inerface beween hose wo crucial processes. Because of our alliance, we were able o work wih KBR o opimise he feed. You canno opimise unis individually and expec o maximise he refinery s overall performance. 1. Overwhelming complexiy: We idenified processes ha would achieve our objecives. Bu when i came o licensing he echnologies from differen suppliers he complexiy was jus overwhelming. Bes pracices are key. A sraegic licensor should be experienced a co-ordinaing he inerfaces and will supply unis ha leverage he same design sandards. 2. Poor soluion inegraion: To ge he bes performance ou of our new uni, we have o compromise on oher processes downsream. This is so frusraing. Opimising he overall performance of he refinery requires wide-ranging experise across muliple disciplines. There is no single ideal process line-up; ha is coningen on he available crudes, he desired produc slae and he Shell Global Soluions has muliple sraegic echnical alliances wih specialis echnology providers, some of which dae back decades. In addiion, we leverage relaionships wih our affiliaes such as Crierion and our in-house rading eams (for crude supply opimisaion and produc offake deals). All of hese share a common fundamenal objecive: o help us work wih cusomers o enhance he value ha we deliver o hem. Dave Clark General Manager, Process Licensing Shell Global Soluions Inernaional BV AVOID THESE THREE COMMON REFINERY PROJECT PROBLEMS boundary condiions in erms of emissions. Wihin ha box, a sraegic licensor wih operaional experience should be able o unlock value. 3. Inabiliy o secure funding: I know ha he projec will deliver on is objecives. Bu making he case o invesors and heir banks is anoher maer; hey are no so enhusiasic and ha is refleced in he erms ha hey are offering. Having a sraegic licensor on board ha has operaional and echnical experience and a rack record of acing as boh an adviser and a projec manager on major energy projecs around he world ends o make a difference. I provides reassurance ha he developmen will be delivered on schedule, wihin budge and o world-class benchmarks

7 Capialising on opporuniy US refinery improves low-grade kerosene-processing performance wih new Crierion caalys Refinery profiabiliy may be improved by operaing wih a cheaper crude oil die, hough, in pracice, his is harder han i sounds. Cheaper crudes are cheaper for a reason: hey are generally heavier, sourer and more conaminaed han refiners would normally consider ideal. Consequenly, convering hem ino premium-grade producs a high yields and wihou undermining plan reliabiliy presens a considerable challenge. I is clearly possible, however. One Norh American refiner, for insance, has successfully processed large volumes of difficul, discouned crude oil o reurn an impressive produc slae of clean fuels ha mee he highes inernaional specificaions. This refiner s success is parly due o he close aenion i pays o is hydroprocessing asses, he skilful way ha i operaes hem and is abiliy o respond posiively o challenges as hey arise. An example of his is he way ha he refiner, wih help from Crierion, ackled he acue drop in performance of one Deailed performance sudies indicae ha i should be possible o run for 24 monhs wih high volume fracions of difficul kerosene feeds. of is wo kerosene hydroreaers, ulimaely urning he problem around and generaing business benefis in erms of increased cycle lengh and enhanced operaional flexibiliy. The sie hydroreas sraigh-run kerosene a relaively low pressure (<500 psi) in wo unis. The firs of hese ends o receive higher-grade kerosene o make je fuel, wih is igh colour and colour reenion specificaions. I is mainly hydrodenirogenaion (HDN) and sauraion aking place in his reacor. The second uni is assigned he ougher kerosene and i generally makes a ULSD-specificaion produc for blending purposes. HDS is he prime objecive here. Colour is no so imporan for his produc; is value lies in is abiliy o improve he ULSD pool in erms of volailiy, graviy and viscosiy. Unforunaely, he pracice of using he second uni o rea low-grade kerosene ran ino difficulies. This was soon afer The refinery has gained he flexibiliy o swing beween ULSD and je fuel producion, depending on he prevailing produc marke economics. he refinery had begun o impor more opporuniy and synheic crudes. The kerosene disilled from hese synheic crudes had higher sulphur and nirogen levels and challenging API graviies. All his kerosene was direced o he second uni, which had jus been loaded wih fresh caalys, Crierion s proven ASCENT DC The impac of he feed change was fel almos immediaely. The refinery had o increase he operaing emperaure of he uni by 8 11 C o mee he ULSD specificaion. And he caalys deacivaion rae ripled. The refinery echnologiss ried o manage his by esablishing a weighed average bed emperaure (WABT) budge and reducing he hroughpu of he uni. However, despie he refiner s bes effors, he cycle lengh was curailed o12 monhs afer is sar. Given he likelihood ha synheic crudes would coninue o feaure srongly in he crude die hrough he course of he nex cycle, he refinery approached Crierion for help in devising he bes way of maximising he cycle lengh, as Kevin Carlson, one of Crierion s business developmen managers, explains. A eam of hydroprocessing specialiss from he refinery and Crierion was assembled o examine he composiion and properies of he feed and o formulae he bes operaing sraegy for he hydroreaing uni and he crude disillaion uni used o process he synheic crude. As par of his exercise, Crierion eam members proposed swiching o a new-generaion CoMo alumina caalys, DC-2618, based on Crierion s CENTERA echnology plaform. The new caalys had already demonsraed ousanding HDS and HDN aciviy and class-leading sabiliy in oher low- o medium-pressure ULSD unis running difficul feeds. Given he uncerainy surrounding he qualiy of he feed o he uni, he eam was unable o predic accuraely he effec of he new operaing sraegy and caalys on cycle lengh. I was hoped, however, ha he changes would add a leas 4 monhs o he 12 monhs achieved previously. The uni was loaded wih CENTERA DC-2618 and resared wih 100% kerosene from synheic crude, which had been held in sorage during he shudown. The sar-of-run WABT was as prediced. Afer his poor-qualiy kerosene had been processed, he WABT fell by 6 C, and i became clear from he subsequen rae of emperaure rise ha he caalys was reaining is aciviy beer han is predecessor, even hough he uni coninued o be fed kerosene of variable qualiy. Daa obained during he firs monhs of operaion led he eam o revise is esimae of he cycle lengh, says Carlson. Technologiss a he refinery began o believe i would be possible o reach he 24-monh mark, which is double he lengh of he las run. The enhanced performance of he uni has been pu down o he caalys s ousanding HDN aciviy. While levels of 60 70% were ypical during he previous run, analysis of daa from he uni shows ha he CENTERA DC-2618 caalys is achieving an HDN level of 90 95%. This means here is less nirogen inhibiion of he caalys; consequenly, he required HDS level can be achieved a reduced emperaures. This is highly significan: i has improved he colour and colour sabiliy of he produc from he second uni o he exen ha i may be used as a je fuel componen. Deailed performance sudies by he eam, aking ino accoun feed qualiy and various operaional scenarios, indicae ha i should be possible o run for 24 monhs wih high volume fracions of difficul kerosene feeds, Carlson says. A he same ime, he refinery has gained he flexibiliy o swing beween ULSD and je fuel producion, depending on he prevailing produc marke economics. The refinery has a srong record of using is hydroprocessing asses o process less favoured crudes and achieve good margins. Therefore, i will no be a surprise o learn ha he company is working wih Crierion o increase he cycle lengh beyond he projeced 24 monhs. I is also looking a oher feed and produc possibiliies ha have opened up as a resul of he improved performance of he second uni wih Crierion s new CENTERA caalys

8 echniques, kerosene blending requires large volumes of kerosene, and addiivaion has a limied effec (i is more effecive a addressing he cold filer plugging and pour poins). Following an economic evaluaion, Berlyaev s eam chose o include a dewaxing caalys (Crierion s SDD-800) in he reacor configuraion. Subsanial analysis work helped Ukranafa o decide wheher o implemen single-sage or second-sage dewaxing (see boxed ex on page 16). Evenually, he single sage opion was chosen, as i suied he refiner s objecives beer. UKRTATNAFTA S ULSD JOURNEY How he refiner me is cleaner fuels mandae and enhanced compeiiveness When he firs ankers loaded wih Euro 4 qualiy diesel and gasoline rolled away from PJSC Ukranafa s depo in January 2011, i marked he end of a long, demanding business journey for he echnologiss a he refinery in Kremenchug, Ukraine. They sared o consider how he 18.6-M/y refinery, which is he larges in Ukraine, could mee he Euro 4 (50-ppm sulphur conen) and Euro 5 (10-ppm sulphur conen) clean fuels specificaions back in Valeriy Berlyaev, Chief Engineer, PJSC Ukranafa, explains why hey were ahead of he game. The refinery s fuure compeiiveness was a sake, he says. We knew ha meeing hose sandards would become a prerequisie for selling o inernaional markes and ha i would improve our posiion in he Ukrainian marke. However, adds Berlyaev, we had a long way o go because we were producing 1,500-ppm-sulphur-conen diesel a he ime. Wih he projec playing ou agains ongoing geopoliical issues in Ukraine and, laer, he global economic crisis of 2008, i proved o be a remendously difficul journey. Meeing he specificaions involved overcoming major challenges and consrains, he explains. For example, he lack of major capial precluded one of he more obvious soluions: he insallaion of a new diesel hydroreaer. Insead, Berlyaev s eam eleced o revamp he exising hydroreaer, which feaured wo rains in parallel, wih wo reacors per rain. In addiion, o avoid disrupion o producion, i was necessary for he revamp eam o execue he projec in phases ha coincided wih scheduled shudowns. During he phase 1 revamp, under he direc echnical supervision of specialiss from Shell Global Soluions, Ukranafa replaced he inernals in all four reacors wih sae-of-he-ar inernals for improved caalys uilisaion. In addiion, Crierion inroduced a nex-generaion, more-acive caalys: ASCENT DC The resuls were impressive: he changes faciliaed he refinery s producion of Euro 3 diesel (350-ppm sulphur). Even he Euro 4 specificaions could be me if a ligher feedsock was used. Euro 4 and 5 specificaions were he arge of he phase 2 revamp. However, his ime, sulphur was no he only challenge. The cold flow properies, such as he cloud, pour and cold filer plugging poins, of Ukranafa s middle disillaes posed a major challenge. The financial incenive was clear. In he pas, he cold flow properies were enhanced by kerosene blending or end-poin reducion, which reduced he value or he size of he refinery s middle disillae pool, or by addiivaion, which can be expensive. Ukranafa s principal challenge was he cloud poin: his was much higher han he specificaion for winer diesel, whereas he cold filer plugging and pour poins were close o he specificaions. Shell Global Soluions experience shows ha he cloud poin is very difficul o reduce economically wihou dewaxing echnology. For he oher In addiion, o mee he sulphur specificaion, subsanially more reacor volume was required. Analysis showed ha a new reacor needed adding o each rain and, crucially, he eam was able o minimise he cos for his. Kremenchug is an enormous refinery ha was buil during he Sovie era and has a lo of heavy-duy spare capaciy available. The eam was able o idenify suiable redundan reacors in he faciliy s warehouses. These, however, were no designed for a rickle-flow process. Much higher performance was required o mee he sulphur specificaions, and so several reacor modificaions were made. For insance, Shell Global Soluions reacor inernals were inroduced, including high dispersion (HD) rays for highly uniform vapour liquid disribuion and excellen hermal disribuion, and filer rays o preven foulans from enering he caalys beds. These reacor inernals can produce 30 50% aciviy gains hrough improved caalys uilisaion and exra volume for loading caalys. In addiion, an even higher aciviy caalys was used: Crierion s ASCENT DN Again, he revamp resuls were sriking. No only was he hydroreaer able o produce Euro 4 and Euro 5 diesel, he cloud poin also fell specacularly o 31 C, subsanially lower han he 15 C arge. Alhough his demonsraes he dewaxing caalys s very high aciviy, i also indicaed ha exra dewaxing aciviy was occurring, which Crierion s echnologiss soon opimised o bring he cloud poin closer o he arge. Anaoliy Dyachenko, Chief Process Engineer, PJSC Ukranafa, reflecs ha Crierion s relaionship wih Shell Global Soluions has helped o solve issues like his hroughou he projec, and has coninued ino he operaional phase. For insance, he says, if we are forced o change feedsock, hey are on hand o help us forecas oucomes and, if necessary, o make process adjusmens ha will help o avoid undesirable consequences

9 Delivering value Ukraine s domesic markes became harmonised wih he Euro 4 and 5 sandards in January 2011, jus as he phase 2 revamp concluded. Sergey Koshelyuk, Depuy Chairman of he Board, PJSC Ukranafa, says he projec has been key o he company s ongoing compeiiveness. These projecs have helped us o mainain and srenghen our marke posiion, he says. The compleion of phase 1 in Ocober 2009 mean ha we could sar o produce baches of Euro 4 diesel. And, when we had compleed phase 2 in January 2011, we could produce Euro 5 diesel and he full range of winer diesel grades, including Arcic diesel, a compeiive prices. Couresy of PJSC Ukranafa Enhanced performance: Ukranafa replaced he inernals in all four reacors wih sae-of-he-ar inernals for improved caalys uilisaion. DIESEL DEWAXING Diesel dewaxing is emerging as a powerful echnique for meeing produc specificaions and creaing addiional margin, wries Lauren Huve, Dewaxing and Base Oil Technology Manager, Shell Global Soluions Inernaional BV. To avoid wax compounds crysallising ou and causing engine problems, he cold flow properies of diesel fuels o be markeed in cold climaes mus mee sringen specificaions. Dewaxing can be used o remove he wax o mee he required values for he cloud, pour and cold filer plugging poins. This can ofen be a beer economic opion han he alernaives, which include addiivaion and kerosene blending. Shell Global Soluions, in associaion wih Crierion, has esablished an enviable rack record of working wih cusomers o design and implemen value-adding dewaxing projecs ha mee heir objecives. There are wo main process configuraions for caalyic dewaxing. In single-sage dewaxing, he dewaxing bed is par of he main hydroreaing secion, and a base meal Ukranafa has now compleed his par of is journey o ULSD producion, and he newly insalled echnology coninues o mee is performance guaranees, bu Berlyaev is already hinking abou new projecs. The boom of he barrel is our nex focus area, he says. Because he oules for heavy fuel oil are shrinking, i is vial ha we maximise our conversion levels. Again, we may need capial-consrained soluions and I am sure ha we will coninue o work wih Shell Global Soluions and Crierion o evaluae our opions. dewaxing caalys ha can wihsand he severe operaing condiions is used. This configuraion can provide a low-cos and flexible soluion. Wih second-sage dewaxing, a dedicaed second-sage reacor downsream of he hydroreaing reacor provides a cleaner environmen (low sulphur and nirogen) so ha a high-aciviy noble meal caalys can be used. The invesmen coss may be higher wih his configuraion, bu produc qualiies and yields can be maximised. Diesel dewaxing has helped refiners around he world o unlock subsanial value. I has helped hem o maximise heir yield of high-value disillaes. The addiion of kerosene o he diesel oil pool is a common soluion for meeing cold flow specificaions; however, unlike diesel dewaxing, i can derac from he boom line. Oher refiners have repored enhancing heir margins hrough he use of diesel dewaxing, as i avoids he need for expensive addiives. In addiion, i can faciliae he use of cheaper feedsocks ha have a higher cloud or pour poin, or more wax. A STEP CHANGE IN HYDROCRACKER PERFORMANCE How Shell Scoford refinery rose o he dual challenge of shifing o a more difficul feed while also increasing capaciy Increasing hydrocracking capaciy can have a profound effec on a refinery s economics. Operaors ha are ineresed in he possible exen of a hydrocracker revamp are advised o look a Shell s Scoford refinery s ScoMods projec, says Wayne Munserman, ScoMods Process Design Lead, Shell Canada Ld. This was a remarkable iniiaive in which he projec eam unlocked a performance sep change in response o compelling business objecives. This was by no means he firs debolenecking exercise on he Scoford hydrocrackers. There had been a series of successful revamps before his one ha had aken he capaciy o 170% of design, so you could be forgiven for hinking ha we had exhaused all of our opions, says Munserman. Bu, because he echnology keeps evolving, we were able o increase he capaciy furher o 189% of design. The hydrocrackers were originally designed o process a relaively ligh and swee synheic gas oil feed: hydroreaed coker gas oil impored from Canadian synheic crude producers Syncrude and Suncor. However, when Shell s Ahabasca Oil Sands Projec came online in 2002, he refinery needed o find a way o process sour, heavy VGO from he Ahabasca Oil Sands upgrader. I also needed o do his a an increased rae o mach he upgrader s oupu. This was a subsanial challenge, reflecs Munserman, who was on he revamp eam. The new hydrocracker feed was much heavier and more sour han he previous one. I also had much higher levels of nirogen and a much higher oal acid number (TAN). See Table 1. The Ahabasca Oil Sands Projec exracs biumen from oil sands in Albera, Canada, and supplies sour, heavy VGO o Scoford s wo hydrocrackers, which are he refinery s principal conversion unis. These are highconversion, wo-sage unis (each wih wo reacors in series, wih recycle o he second reacor) wih fully independen high-pressure reacion secions and a common fracionaion secion. As well as meeing he feed and capaciy consrains, he echnologiss charged wih responsibiliy for he revamp also had o ensure ha he hydrocracker could achieve a minimum cycle lengh of 18 monhs while designing for opimal gasoline/diesel flexibiliy

10 Previous feed Feed rae (% of original design) New feed Feed origin Synheic hydroreaed coker gas oil Sour VGO ex Ahabasca biumen Feed qualiy: Specific graviy Sulphur (w%) Nirogen (ppmw) TAN (mg KOH/g) 0.5 (max.) 4.5 Table 1: The new oil-sands-derived feedsock is subsanially more difficul o process han he former feedsock. These echnical challenges were me wih a combinaion of caalys, process and engineering soluions hrough he combined effors of Shell Canada, Shell Global Soluions and Crierion. We faced numerous echnical issues relaed o boh he changes in he feed qualiy and he increased inake, says Munserman. For insance, he increase in feed sulphur was dramaic. The new feed conained more han 10 imes as much sulphur as he old one, which creaed several poenial processing difficulies. The projec eam oped o inroduce an inegraed highpressure amine absorber upsream of he recycle gas compressor o remove hydrogen sulphide from he recycle gas. This reduced he poenial pressure drop across he reacor circui and lowered he power requiremens for he recycle gas compressor, which enabled he exising compressor o be reained wih only a moderae upgrade o he urbine o provide he incremenal power requiremens. Careful aenion had o be paid o engineering he modificaions o he exising cold, high-pressure separaor (CHPS) owing o he increased flow and he waer separaion issues his would inroduce. To minimise fouling, several measures were aken, including filraion, opional hydrogen injecion ino he feed and inermien waer washing for chlorides. Because of he new feed s disinciveness, he choice of caalys was a very imporan decision. I needed o provide he required yield paern, have high sabiliy and minimise ligh The challenge for he hydrocracker was clear: a significanly higher inake of a sourer, heavier feedsock ha also conained higher levels of nirogen and had a higher TAN. gas producion and hydrogen consumpion. The projec eams conduced exensive process research and caalys esing wih he new feedsock o provide daa ha was crucial for he decision-making process. Afer a horough comparison, he choice finally fell on he applicaion of a se of Crierion and Zeolys caalyss ha delivered he bes combinaion of aciviy, sabiliy, yields and produc properies. The lead reacors, originally loaded only wih hydroreaing caalys, were convered o a sacked bed design wih wo differen caalyss: reaing caalys in he op hree beds and a nirogen-oleran cracking caalys in he boom bed. A full load of cracking caalys was insalled in he second reacors. As par of he previous debolenecking exercises, he caalys uilisaion of he reacors was maximised by replacing he original inerbed disribuion rays wih Shell HD rays. These proven inernals provide excellen gas liquid disribuion over he enire caalys bed and permi increased caalys loading by eliminaing he need for iner disribuion rings. High performance Following he hydrocracker s complee conversion o he new feed and operaion, es runs confirmed ha all he revamp objecives had been me. The produc yields and properies all me or exceeded he arge values, Munserman elaboraes. And, even a he new expanded capaciy and wih he more difficul feed, we achieved a 24-monh cycle lengh for he iniial operaing cycle. THE SCOTFORD HYDROCRACKER S CAPACITY EVOLUTION Even before he ScoMods revamp discussed in he main aricle, Shell Scoford s echnologiss had esablished a long radiion of finding ways o squeeze more value ou of he hydrocracker. Since i was commissioned in 1984, hey had worked wih Shell Global Soluions and Crierion on several programmes ha, by 2005, had creaed a oal 89% capaciy increase (Figure 2). The magniude of he capaciy increase is sriking, bu wha is ineresing is ha many of he improvemens made, especially hose implemened before 2002, are he resul of relaively Capaciy (% of original design capaciy) Year Figure 2: Scoford has a hisory of coninuously increasing is capaciy. The ScoMods projec (see main aricle) ook he capaciy o 189% of design in 2005 (echnical consrains prevened furher increases wihou major capial expendiure). The magniude of he capaciy increase is sriking, bu wha is ineresing is ha many of he improvemens made, especially hose implemened before 2002, are he resul of relaively minor and evoluionary equipmen modificaions. minor and evoluionary equipmen modificaions, says Brian Denniss, Business Opporuniy Manager, Scoford Manufacuring, Shell Canada Ld. These include, for insance, upgrading he feed pump sysem and feed piping, and modifying he make-up gas compressors. In addiion, Shell HD rays were inroduced in some secions of he reacors, and specialised inernals were fied in he high-pressure separaor. Some second reacor beds were combined o increase he overall caalys volume, and an improved filraion sysem was insalled in he feed sysem. A real-ime opimiser o run he hydrocracker on he basis of economics and scheduling arges was also insiued. A each and every urnaround, we look for invesmen opporuniies o improve our capabiliies in some way. Over he years, his has helped us o increase capaciy, handle a more difficul feedsock, save energy and improve he profiabiliy of he uni, explains Denniss. Shell Scoford s approach is herefore aligned wih he Shell Global Soluions Penagon Model. The cash generaed by hese operaional improvemen projecs (which are Penagon I iniiaives) helped o fund a larger revamp ha achieved a sep change in operaional and financial performance (a Penagon II projec). The resul: a hriving, paceseing enerprise