JTMGB 2 2009 - umexpert.um.edu.my laboratorium untuk reaktivasi lapangan-x dengan injeksi kimia acidizing in esp wells: an efficient production optimization kebergantungan pada migas:

Jurnal Teknologi Minyak & Gas Bumi - Edisi 2 - 2009 1

An Alternative Approach for Estimating Water Saturation inIndonesian Carbonate Reservoirs

STUDI LABORATORIUM UNTUK REAKTIVASI LAPANGAN-X DENGANINJEKSI KIMIA

ACIDIZING IN ESP WELLS:AN EFFICIENT PRODUCTION OPTIMIZATION

Kebergantungan Pada Migas:Kebijakan Pemerintah RI Yang Cinta Kemudahan

Tracer Test Evaluation on Core Flooding

This paper presents an observation over a suggested approach for establishing water saturationmodel that is specifically designed without the need of resistivity log data. One of the main strengthof the approach is that the resulting water saturation model can be specifically established for localor specific use only. This is true since the approach can be applied using carbonate rocks that areobtained locally or from specific areas. Another important conclusion is that this approach can also

be applied for any carbonate rock classification as long as the classification can clearly groupcarbonate rocks into groups with distinctive petrophysical properties.

3

16

27

35

40Underbalanced perforation is widely known as one of the best technique to increase hydrocarbon

flow. Numerous researchers have proved that the resulting production increases is due to thesurge of hydrocarbon that eliminate permeability damage around the perforation tunnel. In order to

achieve the underbalance condition downhole, super light weight fluids needs to be formulated.The formulated fluid has lower density compared to the existing fluid in the market. In this

research, optimization formulation of super light weight completion fluids was conducted based onexperimental design software.

49

Optimization Of Super Light Weight Completion FluidsUsing Fractional Factorial Design

DAFTAR ISI

Jurnal Teknologi Minyak & Gas Bumi - Edisi 2 - 20092

EDITORIAL

Sekretariat IATMI Pusat:Patra Office Tower Lt.15, Suite 1501-A

Jl. Jend. Gatot Subroto Kav. 32-34Jakarta Selatan- 12950Telepon: 021-520.3057

Faksimili: 021-520.3057Homepage: http://www.iatmi.or.id

Email:[email protected]

KEPUTUSAN KETUA UMUM IATMI PUSATNO: 007/SK/IATMI/IV/2009

Penanggung Jawab:Bagus Sudaryanto (Pertamina EP)

Pemimpin Redaksi:P. Suryono Adisoemarta (Chevron)

Redaktur Pelaksana:Ir. Asep Kurnia Permadi, PhD (ITB)

Anggota Redaksi:Iman Nurkamal (BPMIGAS)

Tri Atmaja (Medco E&P Indonesia)Taufan Marhaendrajana (ITB)

Usman Pasarai (Lemigas)Elly M. Yusuf (Star Energy)

Peer Review:Ir. Septoratno Siregar, PhD (ITB)

Ir. Doddy Abdassah, PhD (ITB)Prof. Ir. Pudji Permadi, PhD (ITB)

Ir. Leksono Mucharam, PhD (ITB)Ir. Sudjati Rachmat, PhD (ITB)Ir. Rudi Rubiandini, PhD (ITB)

Ir. Sudarmoyo, Msc (UPN)Ir. Sugiatmo Kasmungin, PhD (Univ.Trisakti)

Sponsorship:Sunoto Murbini (Sekretariat IATMI)

Humas:Baruno Subroto (Chevron)

Dita Dwityawarman (CNOOC)

Layout Design:Ophy Irawan (IATMI)

Sirkulasi:Nursyantini (IATMI)

Abdul Manan (IATMI)

Pertama-tama ijinkan segenap pengurus IATMI-Pusat periode 2008-2010 mengucapkan Selamat HariUlang Tahun Kemerdekaan Republik Indonesia yangke 64 tahun, semoga Bangsa Indonesia bisa menjadibangsa yang besar dan sejahtera. Dalam suasanamenyambut datangnya bulan Suci Ramadhan, tidaklupa segenap jajaran pengurus IATMI periode 2008-2010 mengucapkan selamat menunaikan ibadahpuasa Ramadhan 1430H.

Pada ulang tahun yang ke-64 ini, Bangsa Indonesiamendapatkan salah satu hadiah yang sangatberharga yaitu ternyata tingkat pertumbuhanekonomi berkisar antara 3.5 % - 4.5 % dalam situasikrisis keuangan Global, dimana bangsa lainpertumbuhannya malah ke arah negative. BangsaIndonesia menempati pertumbuhan ekonomi nomortiga di asia, dibawah negara Cina dan India. Hadiahini adalah buah kerja yang luar biasa oleh semuakomponen bangsa dan menjadi suatu optimismebangsa ke depan.

Pada Saat penerbitan JTMGB yang pertama di bulanApril 2009, harga minyak NYMEX di NYSE berkisardi harga $45 - $50 /Bbl, di bulan Agustus ini sudahmulai menanjak ke harga diatas $70/bbl dengankenaikan rata-rata mendekati 50 %. Kenaikan hargaminyak ini tentunya bagi Industri Migas sangatberdampak banyak, salah satunya adanyakesempatan untuk melakukan lebih banyak riset,eksplorasi dan penerapan teknologi yang lebihmutakhir. Kegiatan ini diharapkan dapatmeningkatkan nilai lebih bagi kegiatan migas diIndonesia, terutama dalam kenaikan cadangan,kenaikan produksi migas dan biaya yang lebihefektif.

JTMGB edisi ini memuat beberapa tulisan teknisseperti : Estimating Water Saturation in IndonesianCarbonate Reservoir An efficient ProductionOptimization : Acidizing in ESP well. Termasuk jugaadanya studi-studi laboratorium, seperti ReaktivasiLapangan dengan injeksi kimia dan Tracer TestEvaluation on Core Flooding. Ada 1 paper mengenaihasil inovasi optimisasi, yaitu Optimization of SuperLight Weight Completion Fluids using FractionalFactorial design.

Jajaran team JTMGB juga mengucapkan terimakasih kepada Bapak Widjajono P yang sudahberkenan memberikan sumbangan artikel bagiIATMI mengenai arah kebijakan negara,Kebergantungan pada Migas: Kebijakan PemerintahRI yang cinta kemudahan.

Semoga tulisan-tulisan kalangan perminyakanIndonesia ini semakin menambah sumbangsih baginegara tercinta.

Salam dan Terima kasih

Redaksi


IntroductionWater saturation as one of the most

important data for estimating hydrocarbonacumulation needs to be determined reliably.However, in cases that are characterized by highlevel of heterogeneity, such as in carbonatereservoirs, transisition zones and capilary networkabove water table are influential in determining fluidsaturation throughout the reservoirs.

Capilary pressure, which is basically aninteraction between fluids and rocks, is very muchinfluenced by pore configuration, pore throatdimension, wettability, and interfacial tensions bothfluid – fluid and fluid – rock. It is understandabletherefore that carbonate reservoirs with their usualhigh level heterogeneity – meaning varried poreconfiguration and dimension – have their watersaturation distribution very much determined byvariation in capillary pressure.

Developments in conventional log analysishave demonstrated that most of water saturationmodels are dedicated to petrophysical evaluationsin sandstone reservoirs. These models usually differamong each others with regard to clay distributionand other correction-related additions on the classicArchie model. Not much has been devoted tocarbonate reservoirs, since it is commonly assumedthat carbonate rocks are clay-free and thereforeArchie model sufices. This gross simplification mayprove wrong since Archie model was actually derivedfor sands and sandstones.

An Alternative Approach for Estimating Water

Saturation in Indonesian Carbonate Reservoirs

olehBambang Widarsono , Heru Atmoko, Ridwan, KosasihPPPTMGB ”LEMIGAS”

Abstrak

This paper presents an observation over a suggested approach for establishing water saturationmodel that is specifically designed without the need of resistivity log data. One of the main strength of theapproach is that the resulting water saturation model can be specifically established for local or specificuse only. This is true since the approach can be applied using carbonate rocks that are obtained locallyor from specific areas. Another important conclusion is that this approach can also be applied for anycarbonate rock classification as long as the classification can clearly group carbonate rocks into groupswith distinctive petrophysical properties.

Key words: carbonate rocks, classification, water saturation model, capillary pressure

One important aspect that is oftenneglected, even though very well realized, aboutcarbonate reservoirs in log analysis is theirheterogeneity and complexity. Additionally,conventional log analysis that relies solely onresistivity log may record and reflect whatever fluidsaturation nearby wellbore but in case of highlyheterogeneous rocks, in which mud invasion mayvary considerably from shallow to very deep, fluidsaturation distribution may not reflect the truecondition in the reservoir. It is in this light that analternative approach – free from mud invasion effect– is needed.

Apart from complexity and mud invasionissues, it is often in Indonesia to find cases in whichno resistivity logs available (or if available, of oldvintage with its low reliability) for old wells. Thisbecomes a problem when in the wake of high oilprices old oil fields have again come into center ofattention. This requires an alternative method thatdoes not rely on availability of resistivity log data.

In relation to complexity in carbonatereservoir rocks, Lucia (1983) has classifiedcarbonate reservoir rocks into three groups basedon their hydraulic quality. Using these three classeshe generated empirical relationships between watersaturation (Sw), porosity, and heights (h) above freewater level for each class. Thi Lucia model wasgenerated using core data taken from Illinois andWest Texas – USA, which is not neccessarilly validfor Indonesian carbonate rocks. Efforts have to be

ilmiah


spent to generate ones that are valid for Indonesiancases. It is therefore the purpose of this paper –first part of two – to present results of an effort toestablish the models. This first part presentstheoretical considerations and existing models,whereas the second part later on will cover datainventory, modeling/ formulating, and validity testing.

Classification of carbonate rocksClassification of carbonate rocks is in

general based on genetical aspects that are relatedto grain size and fabric. Such classification hasbeen suggested by various workers includingDunham (1962), Choquette and Pray (1970), andmore recently Lucia (1983).

Dunham (1962) established a stratifiedclassification starting from the top by recognising‘crystalline’ and ‘non-crystalline’, dividing the ‘non-crystalline’ into ‘components bound’ andcomponents not bound’, dividing further the‘components not bound’ into ‘containing mud’ andnot, down to division of the rocks classified as‘containing mud’ into ‘mud dominated’ and ‘graindominated’. These definitions, based clearly on rockfabric, classify carbonate rocks into rocks types ofcrystalline, boundstone, grainstone, packstone,wackstone, and mudstone. See Figure 1 for theclassification.

In a manner differently, Lucia (1983)stressed the importance of rock petrophysicalproperties, especially porosity and permeability, inthe defining of rock groups. He showed that mouldicand intra-particles porosities differ significantly frominter-particle and intercrystalline porosities. In brief,Lucia’s classification for carbonate rocks includesthree groups: a) rocks with interparticle porosity, b)rocks with interparticle porosity as background mass+ vugs are mostly unconnected, and c) type (b)rocks but with connected vugs (touching vugs).Figures 2 and 3 present illustrative pictures for thethree groups. The two figures also show that Dunhamclassification serves only as classification of‘background masses’ to the vugs in the Luciaclassification.

This porosity-derived classification hascome into its relevance when it is related to rockspermeability. Lucia showed that the three groupsmay have members overlaping in permabilitymagnitudes but the three groups can bedistinguished from their three different porosity –permeability relationships. Most of type (a) rocksare characterized by relatively low permeability whilemost of type (b) rocks show higher permeability andtype (c) rocks in general show permeability higherthan permeability of the two other rock types. Thesethree distinct groups were then classified, referringto their permeability trends, as Class – 1, Class –

2, and Class – 3 representing type (c), type (b), andtype (c) rocks, repectively.

Capillary pressure and water saturationCapillary presure is defined as pressure

difference between wetting phase and non-wettingphase fluids under immiscible and static conditions.Capillary pressure reflects interactions betweenfluids and rocks, and is controlled by pore geometry,interfacial tension, and wettability. In a simple way,capillary pressure of a capillary pipe is expressedin the form of:

rPC

θσ cos2=

(1)whereó = interfacial tension,è = contact angle (related to wettability and solid-fluid interactions), andr = capillary radius (related to porosity andpermeability)

Equation (1) is also used for describing capillarypressure that prevails in the rock pores and thejustification comes from analoging the pore systemin the rock with a bundle of capillary tubes. Thetubes diameter(s) are then considered asrepresenting the pore throat sizes. Capillary pressurebehaviour of a given rock is normally measured inlaboratory, and upon its use for real reservoirapplications the capillary pressure data is convertedusing (Amyx, 1960)

( )( )lab

reslabPresP cc θσθσ

coscos)()( =

(2)

Where the subscribes of res and lab representreservoir and laboratory conditions, respectively.Table 1 presents some interfacial tension andcontact angle data normally used for the conversion.Pressure gradient for oil and water in reservoir isinfluenced by the fluid’s density difference. Thisdifference in density controls buoyancy forces andin general controls water saturation distributionabove free water level (FWL). This is described,after conversion to field unit, through

( )144

owc

hP ρρ −=

(3)

with Pc in psi, density in lbm/ft3, and height above

FWL (h) in ft.


The distribution of water saturation aboveFWL can be related to height above FWL underconditions of: 1) hydrocarbon and water pressuresare equal at FWL, 2) hydrocarbon and water haveto be continuously in contact above FWL, and 3)system is under static equilibrium.

Under the conditions presented above,water saturation decreases gradually from FWL toa level at which the water saturation reaches as lowas irreducible water saturation (Swirr) and thehydrocarbon saturation reaches its highest level (Shc= 1 - Swirr). This level is called water oil contact(WOC) for oil-water system and gas water contact(GWC) for gas-water system. The interval aboveFWL within which the gradual decrease in watersaturation, acompanied by gradual increase inhydrocarbon saturation, is called the transition zone.Thickness of this transition zone (i.e. thicknessbetween FWL and WOC/GWC) is dependent on thecapillary behaviour of the system. The transition zoneabove FWL is regarded important especially inrelatively thin reservoir with large capillary forces.In this case the whole reservoir column is withinthis transition zone and interval(s) with Sw = Swirr issimply non-existent. It is in this case that estimationof water saturation using the concept of capillaryforce is at its utmost relevance and importance.

Lucia water saturation modelFollowing Lucia’s concept of water

saturation modeling (Lucia, 1995) presented

( ) 537.881035.43 φ××=K (4)

745.1316.002219.0 −− ××= φhSw (5)

for Class 1 rocks,

( ) 38.661004.2 φ××=K (6)

44.1407.01404.0 −− ××= φhSw (7)

for Class 2 rocks, and

( ) 275.4310884.2 φ××=K (8)

21.1505.0611.0 −− ××= φhSw (9)

for Class 3 rocks.

Model for Indonesian Reservoirs

Core DataAs many as 407 limestone plugs taken

from fields in Indonesia, mostly from West Java andSouth Sumatera, were used in the study. Table 2lists the regional origins of the core samples. Datafor each sample consists of porosity, permeability,grain density, fluid saturation, visual description, andtype/ classification. Capillary pressure data for thestudy was provided by special core analysislaboratory (SCAL) data from seven wells among thefields. Table 3 presents an example of capillary datataken Sukamandi field’s plug samples.

Apart for the basic and capillary pressuredata, other core-derived data was also prepared forthe purpose of validity test on the new formula. Thedata is log suites, resistivity (a, m, and n), drill-stemtest (DST), and water analysis result taken from awell in a West Java field. Table 4 summarizes theprimary data availability for the model validity chack.Other auxilary data was also prepared consisting ofpetrographic and scanning electron microscope(SEM) data. This visual data was used for confirmingaspects such as litology model and poreconfiguration type.

Classification of the Rock SamplesIn classifying the rock samples, analyses

were performed in order to interpret the rocks’characteristics with regards to their facies, class,petrophysical properties, diagenesis, anddepositional environment. The rock samples werein general interpreted as to belong to wackestone,packstome, and grainstone classes.

Briefly, the rock samples involved in thisstudy are generally characterized by some typicalfeatures. Wackestones are usualy of fine to coarse-grained skeletal wackestones dominated by coralfragmens, planktons, echinoderms, bryozoa,milliolids, and ostracods (Figure 4). This poorly tomoderately sorted limestone is made of 40% to 65%carbonate mud. As in accordance with the Luciaclassification, this rock is classified among Class 3rocks.

Packstones of the studied samples arecharacterized by their medium to coarse skeletalmaterials consisting of larger fossils of coral,bryozoa, and echinoderms with irregular appearanceof red algae and bivalves (Figure 5). Smaller fossilsare also present represented by benthonicforaminifera such as Amohistegina sp,Lepidocyclina sp, and Miogypsina sp. This kind ofrocks are classified into Class 2 in the Luciaclassification.

The rock samples that were classified asgrainstone are characterized by their poor grain


sortation and presence of large amount of skeletalmaterial such as larger coral fragmens andforaminiferas (Figure 6). Other skeletal material thatare usually present are bryozoa, bivalves,gastropods, and other calcareous benthonicforaminiferas. These rocks are categorized amongClass 1 rocks in accordance with Luciaclassification. Some boundstones among thestudied samples are also classified into this class.

Formulation of The New ModelsSelection of capillary pressure data

First step in the formulation is to selectcapillary pressure data in accordance with the rocksamples that have been classified into Class 1,Class 2, and Class 3. As discussed previously thesethree classes are represented grainstones (andboundstones), packstones, and wackestones,respectively. Figures 7 through 10 present thecapillary pressure data for the four rock types. Allcapillary curves have been converted into reservoircondition following the method presented inWidarsono et al. (2008).

The capillary pressure curves shown inFigures 7 through 10 have apparently diffences inthe pore configuration. This is, at least, shown bythe difference in irreducible water saturation (Swirr)values. The wackestones, having the expectedlylowest permeability, are represented by a range ofSwirr values of 30% - 70%, packstones by 25% -55%, graistones by 23% - 37%, and bounstonesby 25% - 35%.

As the second step, correlations weremade between water saturation and capillarypressure averaging (power law) and normalizationamong the capillary pressure curves using the J-function (JF) approach to yield

·Class 1: 8774.371.0 −= wSJF with R2 = 0.8372

·Class 2: 3589.53606.0 −= wSJF with R2= 0.8081

·Class 3: 9542.4653.0 −= wSJF with R2= 0.968

Porosity vs. Permeability correlation

Following the classification of the rocksamples into three classes, porosity – permeabilityrelationships were determined through plottingporosity and permeability data belonging to eachclass. Figure 8 shows the plots and correlationsfor the three classes. As exhibited by the plots, nosharp correlations are visible even though there areclear correlations between porosity and permeability.

However, scatters are also obvious implying thatthere could be overlapped domains among the rocktypes. Some wackestones could be part of Class 2while some packstones could also be part of Class3 rocks. Nevertheless, porosity (f) – permeability(k) relationships for the three classes have beenfound as follows

Class 1: 2923.207021.0 φ=k (10)

Class 2: 1744.3000925.0 φ=k (11)

Class 3: 527.30000509.0 φ=k (12)

Formulation of new water saturation modelsBy combining the J-function form of

Equation (3) of

φρρ khFJ )(

144)( 21 −=

(13)where r1 and r2 are densities of heavier and lighterfluids, with Equations (1) through (3) water saturation(Sw) models - see Atmoko (2007) for completederivation – water saturation equations of

03815.025759.025472.1 −−= φhSw (14)

for Class 1 rocks,

02442.01866.055178.1 −−= φhSw (15)

for Class 2 rocks, and

03592.020185.04350.2 −−= φhSw (16)

for Class 3 rocks are derived, with h is height abovefree water level at which capillary pressure is zero.From Equations 14 through 16 it is clear that thenew water saturation models are functions of, besideporosity, height above free water level. This impliesthat the models are suitable for cases in whichreservoirs of concern are at their initial condition.The models simply estimate water saturation verticaldistribution, depending on the porosity (i.e.


permeability), up to positions above which watersaturation values are simply the irreducible watersaturation. At any other conditions that have involvedflows as the results of production operations thesemodels are simply not valid.

Model Validity TestFor validity test on the newly proposed

water saturation models, log data from two wells ina West Jawa field was used. A complete set ofstandard log data (gamma ray, spontaneouspotential, resistivity, neutron, density, acoustic, andcaliper) was available for the analysis.

In term of rock quality, the field’s structurecan be divided into two parts; the upper 30-meter ofpoor to fair quality rocks and the better and thickerlower parts with average thickness of 120 meters.Porosity estimation using the most suitable methodhas yielded average porosity values of 14% and 22%for the upper and lower parts, respectively.

Information from petrographic (thin section)and scanning electon microscope (SEM) has shownthat the upper interval is generally made of bioclasticpackstones with intercrystalline porosity, dolomitecrystalls, and secondary porosities caused bydisolution and fracturing. In general, this intervalcan be classified as having Class 2 rocks. For thelower interval no laboratory data was available butfrom the log analysis it appears that the interval ismade of rocks belonging to Class 1.

In the log analysis for the two wells (WJ-X1and WJ-X2) used in the model validity test, porositywas estimated using neutron-density log cross-plot.Considering that the lithology in general is carbonatewater saturation was estimated using Archie watersaturation model with input data as listed in Table3. Results of porosity and water saturationestimation are presented in Table 5.

Water saturation (Archie) estimates appearto be in reasonably good conformity with well testand capillary pressure data. For instance, watersaturation estimate for WJ-X1 well’s Lower zone(Class 1 rocks) is 33.15%. This zone was testedand yielded gas and water at 7 MMSCFD and 89BWPD, respectively. When this data is confrontedto potential irreducible water saturation (Swirr) rangeof 23% - 37% (Figures 9 and 10 combined) the watersaturation estimate seems to be a bit too optimistic.

However, when it is considered that Class1 rocks are characterized by ‘touching vugs’ (Figure12), which could result in very high porosity andpermeability, then the more likely Swirr values shouldtake the lower side of the range. Swirr values of 25%- 30% are likely to conform well with the well testdata.

For the Lower zone of WJ-X2 well, the watersaturation (Archie) estimate is 31.3% with

corresponding well test data of Qg = 3.5 MMSCFDwith no water production. The absence of producedwater means that the water saturation in that zoneis equal to irreducible water saturation or Swirr = Sw= 31.3%. Comparing this estimate to thecorresponding analysis of WJ-X1’s Lower zoneabove, one can speculate that even though bothLower zones are of Class 1 rocks the WJ-X1’s Lowerzone is likely to have higher permeability than theWJ-X2’s Lower zone. This is supported by the factthat the tested gas production rate of WJ-X2’s Lowerzone is only half of the WJ-X1’s tested gasproduction rate.

For the two wells’ Upper zones (Class 2rocks) the water saturation (Archie) estimates arearound 46% - 48%. No firm justification for the Swestimates, but since the Sw estimates for the Lowerzones have been well justified using the well testand Swirr data then it can safely be assumed thatthe estimates for the Upper zones are also justified.

After the application of Archie models forthe two wells is well justified estimation using thenew models, as well as the Lucia models, wereperformed. Water saturation estimation using theoriginal Lucia models (Equations 5, 7, and 9) appearto yield very optimistic (i.e. low) estimates (see Table6). For the Lower zones in the two wells the Swestimates are around 8% - 10%. Comparing to the23% - 37% irreducible water saturation range forClass 1 rocks, the Sw estimates are far too low.Similarly to the Lower zones, Sw estimates for theUpper zones may also be considered too optimistic.

Application of the newly formulated models(i.e. the ‘new’ models) has proved encouraging.Tests on Class 1 and Class 2 rocks resulted in Swestimates that are very close to the estimatesproduced using Archie model. Considering thevalidity that has been shown by the Archie model,this means that the new models have provedthemselves to be valid also, at least for Class 1 andClass 2 rocks. No validity test has been made onClass 3 rock since, understandably, no data wereof primary importance for such low flow-qualityrocks. Nevertheless, such test should also beperformed in the future whenever the needed datahas become available or been made available. Withthe establishment of the new models, an alternativeway for determining water saturation data incarbonate reservoirs has been proposed.

ConclusionsFrom the study, some primary conclusions

have been drawn:· An alternative method for estimating water

saturation in carbonate reservoirs has been validatedproposed to be used.


Table 2 Regional origins of the core samples

· Water saturation estimation for carbonate reservoirscan now be performed in cases where there is noreliable resistivity log data and absence oflaboratory-derived resistivity data.

· The new models are valid at least for use on Class1 and Class 2 rocks. Validity test on Class 3 rockshas to be performed in the future.

· Lucia models tend to produce too optimistic (i.e.too low) water saturation estimates for, at least,the reservoir rocks used in this study.

· The new models are valid for Indonesian carbonatereservoirs, even though further application and testsusing core samples from other carbonate reservoirsare suggested.

References1. Atmoko, H. et al (16 others) (2007) Estimasi

Saturasi Air Tanpa Data Resistivitas danParameter Archie pada Batuan Karbonat.Unpublished report, program code:05.04.03.0039.0044 A.

2. Amyx, J.W., Bass, Jr., D.M. & Whiting, R.L.(1960) “Petroleum Reservoir Engineering -

Table 1 Interfacial tension and contact angle data for some fluid systems.

Physical Properties.” McGraw-Hill Book Co., NewYork, pp. 610.

3. Choquette, P.W. & Pray, L.C. (1970) “GeologicNomenclature and Classification of Porosity inSedimentary Carbonates.” AAPG Bulletin, v. 54,No. 2, p. 207 – 250.

4. Dunham, R.J. (1962) “Classification of CarbonateRocks According to Depositional Texture”. In Ham,W.E., ed.: “Classification of Carbonate Rocks –A Symposium,” AAPG Memoir No. 1, p: 108 –121.

5. Lucia, F.J. (1995) “Rock-Fabric/PetrophysicalClassification of Carbonate Pore Space forReservoir Characterization.” AAPG Bulletin, V.79, No. 9, p: 1275 – 1300.

6. Lucia, F.J. (1983) “Petrophysical ParametersEstimated from Visual Descriptions of CarbonateRocks: A Field Classification of Carbonate PoreSpace.” J. of Pet. Tech., March, p: 626 – 637.

7. Widarsono, B., Atmoko, H., Ridwan & Kosasih(2008). Estimation of Water Saturation inCarbonate Reservoirs without Resistivity LogData. Part I: Theory and Existing Model. LemigasScientific Contribution to Petroleum Science &Technology, Vol. 31, No.2, August.


Table 3 Capillary pressure data for West Java limestone samples

Table 4 Summary of data to be used in model validity test

Table 5 Results of log analysis for WJ-X1 and WJ-X2 wells

Table 6 Comparison between results of conventional log analysis (Archie), Lucia models, andthe new models


Figure 3 Lucia classification, vuggy.

Figure 1 Dunham classification.

Figure 2 Lucia classification, inter-crystalline.


Figure 4 A thin section showing an example of wackestone among the rock samples understudy

Figure 5 A thin section showing an example of packstone among the rock samples understudy


Figure 7 Capillary pressure curves for wackestone samples

Figure 6 A thin section showing an example of grainstone among the rock samples understudy


Figure 8 Capillary pressure curves for packstone samples

Figure 9 Capillary pressure curves for grainstone samples


Figure 11 Porosity – permeability correlations for the three classes

Figure 10 Capillary pressure curves for boundstone samples


Figure 12 A thin section showing vugs that are connected to form ‘touching vugs’ porosity andpermeability system


PendahuluanMeningkatnya kebutuhan terhadap energi,

melambungnya harga minyak mentah mencapaiUS$100 per barrel dan banyaknya lapangan-lapangan minyak tua di Indonesia , menjadi alasanutama mengapa peningkatan pengurasan minyaktahap lanjut (EOR) sangat perlu untuk dilakukan.Selain itu, adanya peraturan dari pemerintah melaluiPeraturan Menteri ESDM No. 1 tahun 2008 tentangPedoman Pengusahaan Pertambangan minyakbumi pada sumur tua, dan kebijakan pemerintahyang terdapat pada Pedoman dan Pola TetapPengembangan Industri Minyak dan Gas BumiNasional 2005-2020 tentang adanya paket insentifuntuk pengembangan lapangan marjinal danlapangan minyak tua (brownfield), menjadi faktorpendorong diaplikasinya EOR di Indonesia.

STUDI LABORATORIUM UNTUK REAKTIVASI

LAPANGAN-X DENGAN INJEKSI KIMIA

olehHestuti Eni, Suwartiningsih, SugihardjoPPPTMGB ”LEMIGAS”

Abstrak

Studi laboratorium untuk penentuan rancangan fluida injeksi kimia diperlukan sebelumimplementasinya di lapangan minyak. Untuk mereaktivasi suatu lapangan minyak tua (lapangan-X) telahdilakukan serangkaian studi yang meliputi beberapa tahap pekerjaan, seperti screening surfaktan,screening polimer, pencampuran surfaktan-polimer dan core flooding.

Screening surfaktan dilakukan untuk memastikan kandidat surfaktan yang digunakan cocok(compatible) dengan air formasi. Dalam penelitian ini, screening dilakukan terhadap 4 jenis surfaktan,yaitu S-F1, S-F2, S-A1, S-A2. Pada masing-masing surfaktan dilakukan serangkaian uji dengan perlakuansama. Rangkaian uji tersebut adalah uji kompatibilitas, pengukuran IFT (interfacial tension), uji kestabilanterhadap panas, uji filtrasi dan uji adsorpsi. Dari keempat jenis surfaktan, diambil 2 jenis surfaktan yangterbaik untuk dilakukan uji pencampuran dengan polimer.

Screening polimer dilakukan terhadap 3 jenis polimer, yaitu P-MA, P-MB dan P-MC. Polimerdiperlukan untuk memperbaiki mobilitas rasio selama pendesakan. Adapun jenis-jenis uji yang dilakukanadalah uji rheologi, uji kestabilan terhadap panas, uji filtrasi dan uji adsorpsi. Selain screening polimer,pada ketiga jenis polimer juga dilakukan serangkaian uji laboratorium setelah dicampur dengan 2 jenissurfaktan terbaik yang telah dilakukan pada uji sebelumnya.

Dari hasil screening terhadap surfaktan dan polimer, campuran surfaktan S-A2 dan polimer P-MCmenunjukkan performa terbaik. Oleh karena itu, core flooding dilakukan menggunakan campurankeduanya. Recovery minyak hasil core flooding menunjukkan kenaikan yang cukup signifikan, hampirmencapai 25% OOIP.

Kata kunci : kompatibilitas, IFT, thermal stability, filtrasi, rheologi, injeksi (core flooding)

Ada beberapa teknologi pengurasan tahaplanjut (EOR) yang sudah dikembangkan parapeneliti, disamping penemuan-penemuan baru yangterus berlanjut seperti injeksi mikroba dan fibro-seismik yang masih terus dikembangkan. Beberapateknologi pengurasan tahap lanjut yang sudahdikembangkan meliputi beberapa jenis, yaitu: injeksigas, injeksi panas, injeksi kimia, dan beberapakombinasi darinya seperti WAG (water alternatinggas), foam dan sebagainya.

Injeksi kimia merupakan teknologi EORyang sangat menjanjikan, terutama pada lapangan-lapangan dangkal yang tidak mungkin dilakukaninjeksi gas CO2 atau N2 karena tekanan rekahnyayang rendah. Data-data lapangan membuktikaninjeksi kimia sebagai cara efektif untuk me-recoverminyak yang masih tersisa. Hasil evaluasi penelitianlaboratorium secara mendetail juga mendukungkelayakan injeksi kimia. Apalagi, chemical yang

ilmiahilmiah


digunakan sekarang ini terbukti mampu bekerja lebihefektif pada konsentrasi 10 kali lipat lebih rendahdibanding chemical hasil penemuan terdahulu. Tentusaja ini menjadi hal yang penting karena berartichemical cost menjadi lebih rendah.

Injeksi kimia dilakukan denganmenginjeksikan chemical seperti surfaktan, polimerdan alkali baik secara sendiri, gabungan atauberkelanjutan pada sumur-sumur tua yang diyakinimasih mengandung minyak potensial. Materialtersebut menyebabkan perubahan pada interaksibatuan dengan fluida dan meningkatkan recoveryfactor meningkat pada daerah kontak reservoir.

Sebelum implementasi injeksi kimiadilaksanakan di lapangan, perlu dilakukan beberapatahap studi laboratorium. Pada penelitian ini,chemical yang digunakan adalah gabungan surfaktandan polimer. Oleh karena itu, tahapan studi yangdilakukan adalah screening surfaktan, screeningpolimer, pencampuran surfaktan dan polimer, danyang terakhir dilakukan untuk mengetahui seberapabesar kinerja chemical yang digunakan adalah coreflooding.

Sampel fluida dan batuan berasal dariLapangan-X dengan viskositas minyak 4,20 cP dantemperature reservoir 85 oC.

SCREENING SURFAKTANSurfaktan adalah senyawa organik yang

dalam molekulnya memiliki sedikitnya satu gugushidrofilik dan satu gugus hidrofobik dimana apabiladitambahkan ke suatu cairan pada konsentrasirendah, dapat merubah karakteristik teganganpermukaan dan antarmuka cairan tersebut.

Untuk meningkatkan recovery minyaksecara optimum, sejumlah uji terhadap surfaktandilakukan di laboratorium seperti uji kompatibilitas,uji pengukuran IFT, uji kestabilan terhadap panas,uji filtrasi dan uji adsorpsi sebelum implementasiinjeksi surfaktan di lapangan. Screening dilakukanpada empat jenis surfaktan, yaitu : S-F1, S-F2, S-A1 dan S-A2.

Uji KompatibilitasUji kompatibilitas merupakan uji screening

paling awal untuk mengetahui apakah suatu jenissurfaktan compatible dengan air formasi darireservoar tertentu.

Keempat jenis surfaktan tersebut dilarutkandalam air formasi dengan konsentrasi 0,1%, 0,2%dan 0,3%. Kemudian masing-masing larutandimasukkan dalam tabung, dan dilakukanpengamatan tiap waktu tertentu. Hasil lengkaptertera pada Tabel 1.

Pada larutan surfaktan S-F1 sudah terlihatwarna putih susu (milky color) dan keruh sejak haripertama dibuat. Ini mengindikasikan adanya

presipitasi yang berarti tidak kompatibel dengan airformasi Pengamatan yang hampir sama juga terlihatpada surfaktan S-A1. Sedangkan larutan surfaktanS-F2 dan S-A2 terlihat jernih dari awal sampai akhirpengamatan. Gambar 1 menunjukkan hasil ujikompatibilitas surfaktan S-F1 dan S-F2 hari ke-1.Uji Tegangan Antarmuka

Tegangan antar muka antara minyak/airdengan mikroemulsi merupakan salah satuparameter utama dalam EOR. Pengukuran nilaitegangan antarmuka menggunakan alat SpinningDrop Tensiometer pada suhu sekitar 60oC. Indikasidari kinerja surfaktan adalah menurunnya teganganantarmuka minyak-air, semakin rendah semakinbaik. Nilai IFT yang sekarang ini diyakini bagus agarsurfaktan disebut layak untuk diinjeksikan adalahsekitar 10-3 Dyne/cm.

Hasil pengukuran tegangan antarmukauntuk ketiga jenis surfaktan tertera pada Tabel 2.Berdasarkan seluruh hasil pengukuran teganganantar muka menunjukkan surfaktan S-A2 dengankonsentrasi 0.3% mempunyai harga IFT yang palingkecil yaitu sebesar 3,46.10-3 Dyne/cm.

Uji Thermal StabilityUji thermal stability dilakukan untuk

mengetahui ketahanan surfaktan terhadap panas.Surfaktan yang bagus, kinerjanya akan tetap stabiloleh pengaruh panas.

Uji ini dilakukan dengan cara memasukkanlarutan pada botol borosilikat yang tertutup rapatkemudian diletakkan pada oven pada temperaturreservoir, yaitu 85oC. Tiap waktu tertentu dilakukanpengamatan dan pengukuran IFT. Diharapkan hasilpengukuran IFT stabil yang berarti surfaktan tidakrusak oleh panas. Hasil pangamatan ditunjukkanpada Tabel 3 dan hasil pengukuran IFT pada Tabel4.

Sampai waktu 1 minggu, surfaktan S-F2dan S-A2 menunjukkan performance yang bagus,dimana tidak terjadi perubahan signifikan yangmenunjukkan tidak adanya kerusakan padasurfaktan secara kasat mata. Tapi pada minggukedua, sudah mulai tampak adanya perubahan.Pada larutan surfaktan S-F2 terbentuk sedikitendapan. Keberadaan endapan sangat tidakdiharapkan. Walaupun endapan ini jumlahnyasangat sedikit, tapi ini juga berpotensi menyumbatpori batuan jika dilakukan flooding. Sedangkanperubahan yang terjadi pada larutan surfaktan S-A2 masih bisa ditoleran karena tidak menunjukkanadanya perubahan yang signifikan.

Pada pengukuran IFT sebagai rangkaian ujithermal stability untuk semua jenis surfaktan terlihatbahwa surfaktan S-F2 dan S-A2 lebih stabildibanding yang lain. Namun demikian, seperti telahdijelaskan sebelumnya, bahwa nilai IFT yang


dianggap bagus adalah pada kisaran 10-3 Dyne/cm. Jadi, surfaktan S-F2 dianggap kurang bagus.Uji Filtrasi

Uji filtrasi dilakukan dengan melewatkan500 ml larutan surfaktan melalui membran saringukuran 0,22 mikron dengan diberi tekanan. Setiap50 ml larutan surfaktan yang yang melewati kertassaring, dicatat waktunya. Kemudian dibuat grafikvolume (ml) versus waktu (detik).

Semua larutan surfaktan, kecuali S-F2menunjukkan garis lurus (Gambar 2), yang berartilaju alir konstan yang mengindikasikan tidak adanyapenyumbatan pada saat melewati membran saring.Hasil ini harus dipenuhi agar suatu jenis surfaktandinyatakan layak untuk diinjeksikan ke dalambatuan.

Uji AdsorpsiAda 2 tipe uji adsorpsi, yaitu adsorpsi statik

dan dinamik. Sebagaimana namanya, adsorpsistatik dilakukan pada keadaan statik/diam,sedangkan adsorpsi dinamik, sebaliknya, surfaktandiinjeksikan pada core. Kemudian diukurkonsentrasinya. Jika konsentrasi setelah prosesadsorpsi berkurang banyak, maka jelas akan sangatmengurangi kinerja surfaktan dalam menurunkantegangan antarmuka minyak dan batuan.Karenaberarti chemical loss yang tinggi.

Hasil uji adsorpsi baik statik maupundinamik ditampilkan pada Tabel 5. Pada surfaktanS-F2 tidak dilakukan uji adsorpsi karena dariserangkaian uji sebelumnya, surfaktan inimenunjukkan performance yang kurang bagus.Surfaktan S-A2 menunjukkan adsorpsi terkecildibanding lainnya.

Dari semua uji yang sudah dilakukanterhadap surfaktan, surfaktan S-F1 dan S-A2dianggap lebih bagus dibanding lainnya dan padakeduanya akan dilakukan uji lebih lanjut dengandicampur polimer.

SCREENING POLIMERSelain surfaktan, polimer juga diperlukan

untuk proses chemical flooding, yaitu sebagai fluidapendorong minyak untuk memperbaiki mobilitasrasio. Pada kajian ini, beberapa uji screening sepertirheologi polimer, uji thermal stability, uji filtrasi danuji adsorpsi dilakukan terhadap 3 (tiga) jenis polimer,yaitu P-MA, P-MB dan P-MC.

Rheologi PolimerUji rheologi polimer bertujuan untuk menentukankonsentrasi polimer yang optimal untuk coreflooding, yaitu konsentrasi dimana viskositas larutanpolimer sedikit lebih tinggi di atas minyak , sehinggaproses pendesakan minyak menjadi efektif.

Injeksi larutan polimer sebagai salah satu metodeEOR dimaksudkan untuk menghindari fingering yangkadang terjadi pada injeksi air. Fingering terjadikarena viskositas air sebagai fluida pendesak lebihrendah dibanding fluida minyak yang didesak.Efektifitas penyapuan dapat ditingkatkan denganpenambahan polimer ke dalam air injeksi agarmobilitas air injeksi mengecil.

Pada ketiga jenis polimer dibuat larutanmasing-masing dengan konsentrasi 800, 1000,1500, dan 2500 ppm. Pengukuran viskositasdilakukan pada suhu reservoar, yaitu 85oC dan shearrate 7 s-1 . Diasumsikan fluida dalam reservoirmengalir pada shear rate tersebut.

Hasil pengukuran viskositas ditampilkanpada Tabel 6. Pertimbangan dalam menentukankonsentrasi optimum polimer selain harus lebihtinggi dari viskositas minyak, juga harus diperhatikancost dari polimer itu sendiri, yang tentunya semakintinggi konsentrasi, semakin tinggi pula cost polimer.

Uji Thermal StabilitySalah satu faktor yang mempengaruhi

efektivitas larutan polimer selama mengalir dalammedia berpori adalah degradasi polimer, yaituterputusnya rantai molekul polimer menjadi unit-unit yang lebih kecil. Ini merupakan fenomenakehilangan berat molekul polimer yang tentunyaakan menyebabkan pengurangan viskositas.

Salah satu jenis degradasi polimer adalahdegradasi thermal yang disebabkan oleh suhu yangtinggi. Oleh sebab itu, dalam kajian ini, perludilakukan uji thermal stability. Uji ini dilakukan untukmengetahui kestabilan viskositas larutan polimer,jika dipanaskan sampai pada suhu tertentu, dalamhal ini suhu reservoar.

Untuk uji ini, diambil data harga viskositaspolimer sebelum dan sesudah dipanaskan (Tabel7). Terlihat bahwa polimer P-MA dan P-MC cukupstabil dengan penurunan viskositas sekitar 5%, yangberarti masih masuk dalam batas toleransi, yaitu20%.

Uji FiltrasiPenurunan harga permeabilitas batuan

dapat terjadi selama uji core flooding denganmenggunakan larutan polimer. Untuk mengantisipasikejadian tersebut, dilakukan uji filtrasi. Uji dilakukandengan menggunakan kertas saring (membran)berukuran 3 mikron, yaitu dengan cara mencatatwaktu yang diperlukan untuk melewatkan sejumlahfluida melalui kertas saring tersebut.

Gambar 3 adalah plot volume terhadapwaktu hasil uji filtrasi ketiga jenis polimer. SurfaktanP-MB membentuk garis melengkung yangmengindikasikan adanya penyumbatan pada kertassaring sehingga laju alir menjadi tidak konstan.


Uji AdsorpsiSebagaimana surfaktan, pada polimer juga

dilakukan uji adsorpsi untuk mengetahui seberapajauh chemical loss pada saat berinteraksi denganbatuan. Tentunya diharapkan hasilnya adalahserendah mungkin. Pada Tabel 8 terlihat bahwaadsorpsi yang terjadi pada polimer P-MA dan P-MCcukup rendah untuk kedua jenis batuan.

PENCAMPURAN SURFAKTAN-POLIMER(SP)

Untuk mengoptimalkan recovery minyak,polimer ditambahkan pada surfaktan. 2 jenissurfaktan yang lolos screening adalah S-F1 dan S-A2. Pada keduanya akan dilakukan serangkaian ujidengan pencampuran dengan 3 jenis polimer.Surfaktan masing-masing dengan konsentrasi 0,3%dan polimer masing-masing 1000 ppm.

Uji KompatibilitasHasil Uji kompatibilitas campuran surfaktan-

polimer ( SP ) ditunjukkan pada Tabel 9. Padapencampuran semua jenis polimer dengan surfaktanS-F1, terbentuk larutan milky (warna susu) danterbentuk gumpalan di atas. Sedangkanpencampuran polimer dengan surfaktan S-A2menunjukkan warna jernih dari hari pertama sampaipengamatan berakhir ( Gambar 4 ).

Uji Tegangan AntarmukaPada dasarnya, pencampuran dengan

polimer tidak mempengaruhi hasil pengukuran IFT.Hasil menunjukkan hampir semua sistemmempunyai nilai IFT 10-3 Dyne/cm, kecuali laruransurfaktan S-F1 dicampur dengan polimer P-MA (Tabel10).

Uji RheologiTabel 11 memperlihatkan viskositas larutan

polimer setelah dicampur dengan surfaktan. Secaraumum, viskositas polimer akan menurun setelahdicampur dengan surfaktan.

Uji Thermal StabilityUji dilakukan dengan mengukur IFT dan

viskositas secara berkala terhadap campuransurfaktan dan polimer yang dipanaskan pada suhureservoir.Selain itu, juga dilakukan pengamatanterhadap perubahan yang terjadi pada larutan (Tabel12). Tabel 13 menunjukkan hasil pengukuran IFTsampai 2 bulan . Terlihat bahwa IFT cukup stabilterhadap panas. Sedangkan viskositas, untuk keduajenis surfaktan dan polimer P-MA dan P-MCpenurunannya relatif kecil, dibawah 10%, namuntidak demikian dengan polimer P-MB (Tabel 14).

Gambar 5 adalah contoh larutan dalam uji thermalstability.

Uji FiltrasiHasil uji filtrasi ditunjukkan pada Gambar 6.

Hampir semuanya membentuk garis lurus, kecualisampuran surfaktan S-F1 dan P-MB yang cenderungmelengkung.

CORE FLOODINGDari semua uji screening, campuran

surfaktan S-A2 dan polimer P-MC menunjukkanperformance terbaik, karenanya, campuran inidipersiapkan untuk diinjeksikan pada batuan dengankonsentrasi surfaktan 0,3% dan polimer 1000 ppmsebagai slug utama. Selain itu, untuk lebihmengoptimalkan penyapuan, setelah slug utamajuga diinjeksikan polimer P-MC 800 ppm. Pemilihankonsentrasi tersebut karena mempunyai viskositasrelatif di atas slug utama sebagai fluida yang akandidorong.

Uji core flood dilakukan pada native coredengan volume main slug 0,3 PV dan volume polimerpendorong 0,2 PV. Recovery factor dari minyak yangdihasilkan adalah sekitar 23.5%. Hasil secaralengkap ditunjukkan pada Gambar 7.

KESIMPULANBerdasarkan kajian yang telah dilakukan

maka dapat disimpulkan :1. Dari 4 jenis surfaktan yang telah discreening,

surfaktan S-A2 dengan konsentrasi 0,3%menunjukkan performance terbaik pada hampirsemua uji screening.

2. Campuran surfaktan S-A2 0,3% dan Polimer P-MC 1000 ppm menghasilkan IFT dan viskositasyang optimal, karenya terpilih sebagai slug utamauntuk uji core flooding.

3. Untuk lebih mengoptimalkan penyapuan, setelahslug utama juga diinjeksikan polimer P-MC 800ppm yang mempunyai viskositas relatif di atasslug utama sebagai fluida yang akan didorong

4. Recovery factor hasil core flooding dengan 0,3PV slug utama dan 0,2 PV polimer pendorongmenggunakan native core hampir mencapai 25%.

DAFTAR PUSTAKA1. www.migas.go.id2. www.oil-chem.com3. Don W.Green, G. Paul Willhite, “Enhanced Oil

Recovery” SPE Textbook Series Vol.6, 19984. API RP 63 , “Recommended Practice for

Evaluation of Polymer Used in EOR Operations”1st Edition, 1990


Tabel 1. Hasil Uji Kompatibiltas Surfaktan

Tabel 2. Hasil Pengukuran Tegangan Antarmuka Surfaktan

Table 3. Hasil Uji Ketahanan terhadap Panas pada Surfaktan


Tabel 4. Hasil Pengukuran IFT pada Uji Ketahanan terhadap Panas

Tabel 5. Hasil uji Adsorpsi surfaktan

Tabel 6. Viskositas polimer pada suhu reservoir dan shear rate 7 s-1

Tabel 7. Uji Thermal Stability polimer


Tabel 8. Hasil Uji Adsorpsi Polimer

Tabel 9. Hasil Uji Kompatibilitas Surfaktan – Polimer ( SP )

Tabel 10. Hasil Pengukuran Tegangan Antarmuka Surfaktan – Polimer ( SP )

Tabel 11. Hasil Pengukuran Viskositas Surfaktan – Polimer ( SP )


Tabel 12. Hasil Pengamatan Uji Thermal Stability Surfaktan – Polimer ( SP )

Tabel 13. Hasil Pengukuran IFT pada Uji Thermal Stability Surfaktan – Polimer ( SP )

Tabel 14. Hasil Pengukuran Viskositas pada Uji Thermal StabilitySurfaktan – Polimer ( SP )

Gambar 1. Hasil Uji Kompatibilitas Surfaktan S-F1 dan S-F2 Hari Ke-1


Gambar 3. Uji Filtrasi Polimer

Gambar 4. Uji Kompatibilitas Surfaktan-Polimer

Gambar 2. Uji Filtrasi surfaktan


Gambar 5. Pengamatan uji thermal stability campuran Surfaktan – Polimer ( SP )

Gambar 6. Uji Filtrasi campuran Surfaktan – Polimer ( SP )

Gambar 7. Hasil Core Flood pada Native Core



IntroductionKaji-Semoga Field is in Rimau Block, South

Sumatera, Indonesia. Oil was found and productioncommenced in 1997.

There are around 350 wells, which consistof oil producers and water injection wells. Most wellsare gas lifted and deviated in a cluster system. Eachcluster consists of 3 to 15 wells, either producersor injectors. .

This field produces oil from 3 formations:Telisa Sandstone (TLS), Baturaja Carbonate (BRF),and Talang Akar Sandstone (TAF). Most oil isproduced from BRF, since it has the largesthydrocarbon reserve among all. Its porosity rangesfrom 13 to 20% and matrix permeability ranges from20-150 md. Its productivity index (PI) is between0.5 to 8 bfpd/psi. The depth is between 3000 to 4000ft. The oil is relatively light, 38 degrees of API.Average field properties are: 82% water cut and GOR700 scf/stb. Currently the BRF pressure is about800 psig.

The water-flood project has been carried intothis formation in order to maintain the pressure and

ACIDIZING IN ESP WELLS:

AN EFFICIENT PRODUCTION OPTIMIZATION

olehNoke Fajar Prakoso, Dendy WicaksanaMedco E&P Indonesia

Abstrak

Many electric submersible pumps (ESPs) are being installed in Kaji Semoga Field, South Sumatra,Indonesia (operator: Medco E&P Indonesia) to change the artificial lift method from gas lift. This is beingimplemented in accordance with the water-flood project and is also due to decreasing produced gas. Untilnow, 15 ESPs have been installed and are running. After ESP installation, three wells suffered influximpairment, so oil production did not meet the target. Therefore stimulations were needed to restore theinflux. In this case, the stimulations were in the form of acidizing to Baturaja carbonate formation.

Acidizing without pulling ESP out of hole has been done in three ESP wells in Kaji Semoga Field.15% HCl was pumped thru annulus to solve the formation damage problem, which was suspected to havebeen caused by workover fluid quality and also loss circulation material (LCM). The results were good::influx improved, the oil gain was more than 400 BOPD, and also costs of up to USD 35,000 per job weresaved, since these jobs did not need the use of a rig. As per November 2008 (3 months), the three ESPsare still running in good condition. This shows that the acidizing job without pulling the ESPs out of holedoes not have any negative effect on the pumps since the techniques are implemented.

This paper presents well performance analysis, acid risk analysis, acid recipe, acidizing technique,and cost justification. These three success stories can be implemented as best practice in other similarcases.

to sweep the oil. It has been observed that theassociated gas is declining while the water cut isincreasing. Medco E&P Indonesia has thereforeconducted studies on artificial lifts to replaceexisting gas lift, since make-up gas for gas liftcompressors has declined and gas lift cannot createmaximum drawdown for the current conditions. Thestudies recommend changing from gas lift to electricsubmersible pump (ESP). More than 50% of gaslifted wells will be changed to ESP wells. At themoment15 ESPs are running.

ESP Installation, Influx Impairment andRejuvenation

In 2006, Medco started to convert 15 gas lifted wellsto ESP, with different optimum pump ranges.Unfortunately, three of them, KS-38, KS-57 and KS-63, suffered influx impairment.

KS-38ESP was installed in KS-38 on March 12,

2008. Everything was as normal during installation,

ilmiah


but after the well was put on production the rate didnot meet the target. On the contrary, the rate waslower than when gas lift was used (see Figure 1).From pump intake pressure (PIP) and rate dataconverted to inflow performance relationship (IPR,see Figure 2), it was suspected that there was influximpairment. It may have been caused by unfilteredkilling fluid in ESP installation job and incompatiblekilling fluid to formation fluid. Therefore acid washwas conducted in this well. The result was verysatisfying, since after acidizing the oil productionincreased 472 BOPD (see Figure 1).

KS-57ESP was installed in KS-57 on May 14,

2008. This well produces oil from 2 perforationintervals in BRF (flow unit 1 and 2). Duringinstallation loss circulation occurred, so the LCMwas pumped into the formation to stop losscirculation. But after it was produced, the same thinghappened. Production dropped (see Figure 3). It wassuspected that this formation damage happenedbecause LCM did not work properly in the formation,since killing fluid loss circulation happened at onewell. After mechanical well service was completedand the well was produced, the rate did not meetthe design rate or pump range performance becauseof influx impairment. It was suspected the sealingof LCM which invaded the formation did not breakperfectly and Influx impairment happened as shownin its IPR (see Figure 4). Then it was acidized. Afteracid wash the rate was better than before, but stilllower than when it was gas lifted. The damage maystill exist in the formation.

KS-63ESP was installed in KS-63 on May 10,

2008. Everything was as normal during installation,but after the well was put on production, the ratewas lower than when it was gas lifted (see Figure5).From its IPR as shown in Figure 6, it is knownthat there was influx impairment, and therefore acidwash was conducted. After acid wash the rate wasbetter than before, but still lower than when gas liftwas used. The damage may still exist in theformation.

EconomicsThere were two options for performing the

acidizing job: to follow the common procedure at ahigh cost, or to try the lower cost alternative. Thecommon procedure at a high cost is that prior tothe acidizing job the ESP must be pulled out of thehole, while the lower cost alternative means acidizingwithout pulling ESP out of hole at all. After severaldiscussions and careful consideration, Medco

decided to do rig-less acidizing in order to studyand to find the most efficient way to do acidizing inESP wells. Cost savings per job reached USD35,000. A cost comparison between both options ispresented in Table 1.

Acid RecipeThe rig-less acid washing in the ESP wells

used 15% HCl, which is similar to common acidwashing in this formation. Table 2 shows volumecalculation of acid & formation water to be injectedthru annulus.

Risk Analysis and Acidizing TechniquesFormation damage in carbonate could be

solved by acid stimulation. The main problem of acidin ESP well was damaging pump stages (impeller& diffuser), but this problem was mitigated byminimizing acid exposure time to ESP down holeequipment. Some negative effects which may haveoccurred as the impact of contacting with acid were:1. Electrical short at down-hole cable.2. Damage on bag type protector.3. Damage on other parts, such as seals, bearings,

impellers & diffusers.

To minimize the risks, a sequence of jobsshould be performed in order to achieve the objective(see Figure 7):1. Prepare acid pumping equipment.

a. Acid pump combob. Water tankc. Square tank for return fluid

2. Ensure that the ESP is running.3. Hold a pre-job safety/operations meeting and

review job procedures and safety issues.4. Rig up acid pump and connect to annulus.5. Pressure test to maximum wellhead working

pressure (2000 psig) and hold for 10 minutes.6. Bleed off pressure to 0 psig and open the flow

line.7. Ensure the ESP is running8. Open casing valve, kill well thru annulus with 78

bbls (filled up volume of 5½” casing, 3280 ft deepof well) 8.4 ppg of salt water. Pump in killing fluiduntil it is circulated to the surface to ensurewellbore is fulfilled with killing fluid.

9. Shut off ESP, close tubing valve at surface to trapkilling fluid in the annulus and tubing, then performinjectivity test thru annulus.

10. Prepare 125 gallons (3 bbls) of 15% acid, opencasing valve, spot 3 bbls 15% acid thru annulus,followed by 49 bbls 8.4 ppg of formation water(pump pressure = 500 psig and 2 bpm rate ofinjection)


11. Stop pumping killing fluid thru annulus while 15%acid placement in formation (soak 15% acid for30 minutes)

12. Open tubing valve and start up ESP to flush outacid mixture from the annulus until all acid comesout to the surface and cleans up well.

13. Measure the pH of return fluid from tubing. Oncethe acid is returned, switch return line to squaretank

14. Keep monitoring the return pH. Once all the acidhas already returned, indicated by normal returnpH, switch return back to production line.

15. Close annulus valve.16. Neutralize the spent acid with caustic soda.17. Rig down acidizing equipment18. End job.

The question is: how reliable are the pumpstages after exposure to acid, and what is thepump’s acid tolerance level? The standard pumpstage material supplied is ni-resist alloy. The vendorspecified there should be a limit of 4 hours maximumexposure with 15% acid threshold. Based on thisallowance, it was decided to soak acid thru theannulus to the formation without needing a rig. Thisis because the pump does not need a lengthyexposure to acid, and our technique achieve lessthan 15 minutes of pump exposure, and acidplacement for 30 minutes into formation. Then, theacid is pumped out thru tubing to the surface andthe well is cleaned up.

Conclusion and RecommendationAfter reviewing the results of this job, the

authors have reached the following conclusions:1.Rig-less acidizing in KS-38 ESP well was

successful. This technique can be implementedin other wells.

2.Rig-less acidizing in KS-57 and KS-63 wells canbe considered successful because of influximprovements as shown on their IPR, althoughthey have not been fully recovered. The authorsrecommend that rig-less acidizing should berepeated with a larger acid volume, especially inKS-57.

AcknowledgementThe authors would like to thank Mr. Anwas

Tanuwijaya in addition to the Petroleum EngineeringDivision and Rimau Asset of Medco E&P Indonesia.

References1. Allen, O.Thomas and Roberts, Alan P.:

“Production Operations 2”, OGCI and PetroskillsPublications, Tulsa, OK (Chapter 5 and 7)

2. Brown.K.E.: “The Technology of Artificial LiftMethods”, Petroleum Publishing Co., Tulsa, OK(1980, vol 2A)

3. Ojukwu, K.I.: “Through Pump Stimulation: A CostEffective Alternative”, Society of PetroleumEngineers Inc (2008)

Figure 1 Production Performance KS-38



Figure 2 IPR Curve KS-38 (B)


Figure 4 IPR Curve KS-57



Table 1 Cost Comparison

Table 2 Acid Recipe and Acid Volume Calculation

Figure 6 IPR Curve KS-63


Figure 7 Job Sequences of Rigless Acidizing in ESP Wells



Resensi Buku :

Kebergantungan Pada Migas :Kebijakan Pemerintah RI Yang Cinta Kemudahan

Dari Buku Berjudul “Migas dan Energi di Indonesia: Permasalahan dan Analisis Kebijakan”tulisan Widjajono Partowidagdo

Penerbit : Development Studies Foundation, Juni 2009 (430 halaman)*)

Menarik bahwa Prof.Widjajono di buku terbarunya ini, dalam mengambarkan kebijakan energidi Indonesia yang sangat ‘pragmatis’, ‘enak’ dan ‘luwes’, beliau hanya mengutip kata-kata KepalaSekolah Dumbledore dalam buku “Harry Potter’ (2007) :

“Someday, you will have to choose between what is right and what is easy” yang artinya : suatuhari nanti kau harus memilih di antara apa yang benar dan apa yang mudah…

Selain kutipan kata-kata Yasadipura (kakek Ranggawarsita) dalam buku pujangga Jawa klasiktsb. (1873):

“Waniya ing gampang, wediya ing pakewuh,sabarang nora tumeka.”, yang artinya: sukailahkemudahan, takutilah kesulitan, insya’Allah tidak ada yang diperoleh.

Persoalan energi dan bangsa tidak bisa hanya diselesaikan oleh Pemerintah sendiri, apalagipemerintah terpilih saat ini yang dipersepsikan sangat-sangat hati-hati dalammemutuskan segala sesuatu. Negara yang baik membutuhkan adilnya Pemimpin, amalnyaPengusaha, ilmunya Akademisi (Ulama) serta kesabaran, kemandirian dan keperdulianMasyarakat.

populer


Migas dan energi penting untukpembangunan di Indonesia, sehinggapermasalahannya perlu dipahami oleh parapemangku kepentingan (stakeholders)-nya yaituPemerintah, Pengusaha, Akademisi dan Anggotamasyarakat lain. Buku Widjajono tampaknyaberusaha untuk menjelaskan agar para pemangkukepentingan lebih memahami permasalahan dananalisis kebijakan migas dan energi di Indonesiasehingga diharapkan bisa dipilih kebijakan energiyang lebih tepat dan penanganan permasalahanenergi dengan lebih baik. Untuk lebih memahamipermasalahan, ditambahkan teori-teori pendukungbaik dari tekno-ekonomi maupun analisis kebijakan.

Dalam pandangan Widjajono (2009), adabeberapa anggapan yang keliru mengenai migas danenergi, diantaranya:1. Indonesia adalah Negara yang kaya minyak,

padahal tidak. Kita lebih banyak memiliki energilain seperti batubara, gas, CBM (Coal BedMethane), panas bumi, air, BBN (Bahan BakarNabati) dan sebagainya,

2. harga BBM (Bahan Bakar Minyak) harus murahsekali tanpa berpikir bahwa hal ini menyebabkanterkurasnya dana Pemerintah untuk subsidi hargaBBM, ketergantungan kita kepada BBM yangberkelanjutan serta kepada impor minyak danBBM yang makin lama makin besar serta makinsulitnya energi lain berkembang,

3. investor akan datang dengan sendirinya tanpaperlu kita bersikap terlalu ‘ramah’ dan usahakanmemberikan iklim investasi yang lebih baik,

4. peningkatan kemampuan Nasional akan terjadidengan sendirinya tanpa keberpihakanPemerintah.

Menurut Prof. Widjajono yang juga anggotaDewan Energi Nasional ini, kita tidak perlumemberlakukan harga pasar apabila harga BBMtinggi. Yang perlu diberlakukan adalah sebaiknyakita tidak menerapkan harga BBM yang lebih rendahdari biaya (ditambah keuntungan) pengembanganenergi lain. Misal BBN (Bahan Bakar Nabati, untukminyak jarak dan ketela sebesar Rp 5500/liter) atautidak menerapkan harga listrik lebih rendah dariharga rata-rata listrik panasbumi (sebesar 8 sendollar AS /kwh) dengan demikian BBN danpanasbumi akan berkembang.

Kita tidak perlu mengikuti harga BBM dipasar bebas atau mengenakan pajak untuk BBMsehingga harganya bisa diatas dua kali lipat hargapasar seperti berlaku dikebanyakan negara-negaralain. Walaupun demikian, kita tidak bisa mengikutiharga BBM murah di negara-negara yang produksiminyaknya berlimpah. Perlu disadari bahwawalaupun kita menghasilkan minyak 953 ribu barelper hari dan mengekspor 370 ribu barel per hari,kita juga mengimpor minyak mentah sebesar 318ribu barel/hari dan BBM sebesar 411 ribu barel/hari

(Ditjen Migas, 2007). Cadangan terbukti minyak kitahanya 4,4 milyar barel atau 0,4 % dunia. Sebagainet importer minyak yang tidak memiliki cadanganterbukti minyak yang banyak, kita tidak bijaksanaapabila mengikuti harga BBM murah di Negara-negara yang produksi minyaknya melimpah.

Apabila kita harus menerapkan harga politik(subsidi), maka seharusnya kita tidak hanyamensubsidi harga BBM baik untuk non listrik danlistrik tetapi juga harus mensubsidi harga energialternatif. Dengan demikian energi alternatif akanberkembang sehingga kita bisa mengurangi imporminyak dan BBM. Apalagi kebutuhan energi kitamakin lama makin besar dengan peningkatankesejahteraan.

Untuk mereka yang kurang mampu perludiberikan subsidi untuk pemakaian LPG (LiquefiedPetroleum Gas) dan gas kota untuk memasak,transportasi umum (di kota-kota besar sebaiknyamenggunakan BBG) dan listrik 450 watt untukpemakaian tertentu (kalau boros maka’pemborosannya’ – berdasarkan ukuran intensitaskonsumsi energi - tidak perlu disubsidi).Pengurangan subsidi harga akan meningkatkandana untuk subsidi langsung (kesejahteraan rakyat)dan depletion premium.

Menurut Widjajono, mengundang investoradalah ibarat mengundang pelanggan untukmengunjungi rumah makan. Rumah makan hanyaakan laku apabila pelanggannya tahu, makanannyaenak, harganya bersaing, pelayanannya danlingkungannya baik. Perlu informasi yang baik,sistem fiskal dan regulasi yang menarik, birokrasidan birokrat yang mendukung, koordinasi antarsektor dan antar pusat-daerah untuk menarikinvestor baik asing maupun domestik di sektorenergi.

Peningkatan Kemampuan Nasional akanterjadi apabila terdapat keperpihakan pemerintahmisalnya untuk kontrak-kontrak yang sudah habismaka pengelolaannya diutamakan untukperusahaan nasional dengan mempertimbangkanprogram kerja, kemampuan teknis dan keuangan.Hal lain yang perlu dilakukan adalah pinjaman daribank nasional untuk membiayai kegiatan produksinasional dengan kehati-hatian.

Untuk memperbaiki iklim investasi danmeningkatkan kemampuan nasional di bidangenergi perlu digunakan dana depletion premium darienergi tak terbarukan yang untuk migas diperkirakansekitar 10 persen dari equity to be split (revenuedikurangi recoverable cost). Dana tersebutdigunakan untuk meningkatkan kualitas informasibagi penawaran konsesi-konsesi baru, untukmeningkatkan kemampuan sumber daya manusiadan penelitian, untuk mempersiapkan infrastrukturpendukung migas serta untuk pengembangan energiterbarukan dan energi perdesaan. Peningkatan


kemampuan nasional di bidang energi justru akanlebih menarik investor asing.

Sejalan dengan pendapat Ketua UmumKadin MS Hidayat sebelumnya, dalam diskusi“Upaya Meningkatkan Produksi Migas dengan TetapMemberdayakan Produk dan Jasa Dalam Negeri”di Jakarta, Selasa (26/4/2009, perubahan paradigmakita menurut Prof.Widjajono adalah antara lainmigas bukanlah sumber utama APBN, perlunyadiversifikasi energi, keharusan melakukankonservasi, dan efisiensi dalam pengelolaan BBM.“Migas dan terutama Energi Terbarukan janganlahdibebani kewajiban untuk memikul peran sebagaisumber utama APBN. Sumber utama APBN adalahpajak,” kata Hidayat. Migas diprioritaskan bukanuntuk menghasilkan devisa secara cepat (dan karenaitu mengapa dibenarkan untuk terus diekspor dalambentuk “mentah” seperti selama ini), tetapi lebihsebagai bahan bakar untuk menggerakkan ekonominasional dan mendorong pertumbuhan sektor-sektorekonomi yang berkontribusi pada pertumbuhanpajak. Dengan mengubah paradigma seperti itumaka kita tidak perlu “tergesa-gesa” menyerahkanpekerjaan hulu migas dan energi terbarukan kepadaperusahaan asing demi untuk mengejar devisa.Sebaliknya , Indonesia dapat lebih memberikanperhatian kepada pengembangan pengusahanasional. Menurut Widjajono, Indonesia masihmembutuhkan investasi asing terutama untukeksplorasi apalagi di laut dalam. Untuk itu perluperbaikan iklim investasi dengan penataan birokrasidengan regulasi dan birokrat serta koordinasi yanglebih baik.

Kalau melihat laporan dari DepartemenKeuangan, utang kita setiap tahun naik, sekarangsudah mencapai Rp 1695 trilyun. Penghasilanmigas sekitar 40 % dari APBN (Rp 1000 trilyun),jadi sangat disayangkan kalau penghasilan migashabis untuk bayar utang dengan subsidi BBM danlistrik. Untuk itu perlu diubah paradigmapengelolaan migas bukan semata-mata untukAPBN tapi harus mempunyai nilai tambahdengan ’multiplier effect’ yang luas danmenyentuh sektor real. Salah satu contohnyagas untuk kepentingan domestik seperti untuk gaskota, BBG untuk transportasi, Industri Pertanian(Pupuk) dan Petrokimia, pengolahan / kilang minyakmentah selain bisa menghasilkan BBM jugamemproduksikan LPG (4%) dan Petrokimia.

Prof.Widjajono cukup dalam mengajak kitamemahami permasalahan migas, mulai dariDinamika, Cadangan dan Produksi, EvaluasiKeekonomian dan Resiko, Investasi dan Biaya,Sejarah, Hukum, Fiskal dan Kerangka Kerja Huluserta Daya Saing Dunia. Kita perlu belajar dari masalalu, masukan dari para pemangku kepentingan danbelajar dari bangsa lain sehingga dibahas juga :\Belajar dari kasus LNG Tangguh, Pentingkah

Menjadi Anggota OPEC?, Belajar dari Pak Ibnu,Musibah Lusi (Lumpur Sidoardjo), Masukan dariPemangku Kepentingan dan Pakar serta Belajar dariPetronas/Malaysia, Bekas Murid Pertamina. Dalammembahas Kebijakan Migas serta pembahasantentang Peningkatan Produksi, Investasi danKemampuan Nasional Hulu Migas serta KontrakKerja Sama, Institusi dan Iklim Investasi Migas diIndonesia dan Pertamina dan Elnusa Baru,Widjajono cukup terinci juga sampai penyajianperhitungan dan rumus-rumus teknoekonomi takterlewatkan pula.

Tentang Energi Terbarukan danPembangunan Berkelanjutan

Dalam bab-bab energi terbarukan danberkelanjutan (renewable & sustainable energies)Prof.Widjajono membahas Kebijakan Energinasional tanpa melupakan fakta bahwa cadangan(tepatnya : potensi, karena cadangan panas bumiini benar-benar terbarukan dan berkelanjutan,insya’allah takkan habis) nasional kita sangatmelimpah, bahkan untuk ukuran dunia. Dalampembahasan tentang Harga Energi, Panasbumi,CBM, BBN, LPG, Memperkuat Ketahanan Energidan Pangan, BBN dan Pengentasan Kemiskinan,para pembaca diantarkan kepada konsepKemandirian Energi berdasarkan Potensi EnergiAlternatif kita. Karena energi berhubungan eratdengan pembangunan maka dibahas aspekPemahaman Pembangunan, Bangsa yang Tolerandan Ramah (cenderung kurang disiplin, yangmenjurus pada kompromi dalam hal Etika Bisnis)dan Praktek Pembangunan Berkelanjutan yangbelum berdasarkan pemahaman yang baik.

Karena terus-menerus BBM mendapatkansubsidi, energi alternatif menjadi sulit bersaing. Halitu terbukti pada energi panas bumi, yang selama20 tahun terakhir tidak mampu berkembang danmenyaingi BBM, karena secara keekonomian selalukalah oleh BBM yang disubsidi.

Disamping rendahnya harga BBM (insentifuntuk pemakaian BBM), tidak adanya insentif untukpengembangan energi alternatif menyebabkan krisisenergi dewasa ini. Sebagai contoh, PLN dulu selaluberdalih “least cost” kalau mau membeli gas ataupanasbumi, sehingga mereka selalumembandingkannya (net back) dengan batubara.Akibatnya, pengembangan gas (lapangan kecil danmenengah) dan panasbumi macet karena hanyamau dibeli dibawah 5 sen per KWh, sedangkanbiayanya diatas 7 sen per KWh. Padahal, akibattidak cukupnya pasokan listrik PLN maka untukmenutupinya PLN menggunakan BBM yangsekarang biayanya 15-30 sen per KWh (tergantungharga minyak). Hal ini yang menyebabkan kenapa


Tabel 1 Potensi Energi Nasional 2006

subsidi BBM untuk listrik pada APBN_P 2008sebesar 60,3 trilyun rupiah.

Contoh lain adalah bagaimana jarak(biodiesel) dan ketela (bioethanol) bisa berkembangapabila harga BBM masih Rp 4.500 / liter. Untukmemproduksikan satu liter BBN dibutuhkan palingtidak 5 kg jarak atau ketela yang jika petani memintaharga Rp800,-/kg sehingga bahan mentahnya sajasudah Rp 4.000 per liter BBM, belum proses dantransportasinya. Usaha biodiesel dan bioethanolbaru menguntungkan apabila dijual Rp 5.500 perliter. Insentif untuk pengembangan energi alternatiflebih bermanfaat daripada subsidi harga BBM yangterlalu mahal dan menyebabkan pemakaian BBMyang boros.

Perlu dicatat bahwa cadangan terbuktiminyak Indonesia (4,4 milyar barel pada akhir 2007)hanya 0,4% cadangan terbukti minyak dunia,cadangan gas 1,7% cadangan terbukti gas dunia,cadangan terbukti batubara 0,5% cadangan terbuktibatubara dunia dan potensi panasbumi Indonesiadiperkirakan 40% potensi panasbumi dunia.1

Potensi energi nasional 2006 diperlihatkan padaTabel 1.

Nasib serupa juga dialami sumber energialternatif lain yang masih dalam tarafpengembangan, seperti BBN, matahari, angin, danmikrohidro. Sumber-sumber alternatif itu tak akanbisa sejajar atau meningkat mengejar penggunaanBBM, selama kebijakan pemerintah masih berpihakpada BBM.

Kini sebetulnya saat yang tepat untukmemindahkan subsidi dari BBM ke energi alternatif.Dengan demikian, dalam kurun 5-10 tahun ke depan,ketahanan energi nasional dapat terjamin, karenaenergi alternatif yang melimpah ruah di bumiNusantara dapat berperan dengan nyata.

Apabila pemerintah memiliki keinginan yangkuat menata energi nasional di masa yang akandatang, seyogyanya BBM dilepaskan dari subsidi,dan mengikuti harga pasar yang berlaku. Biayasubsidi yang sudah ada di APBN selanjutnyadialihkan untuk mensubsidi energi alternatif (inganpotensi Panas Bumi kita 40% atau terbesar di duniasedang pemanfaatannya baru 0,4%), sehinggabisnis dan industri energi alternatif tumbuh dengansendirinya karena secara keekonomian menjadiatraktif. Dampak lanjutannya, kesadaranmasyarakat sedikit demi sedikit akan terpola untukmengurangi energi dari BBM, beralih ke energialternatif.

Mengacu studi JICA 2007, harga panasbumiuntuk 55 MW dengan IRR 15%, apabila dikenakanbea masuk, PPN impor dan PPh impor adalah 8,68sen dolar AS per kWh. Apabila bea masukdihapuskan dan PPN impor dan PPh imporditanggung pemerintah adalah 8,09 sen dolar ASper kWh. Apabila pemerintah mengandalkanbatubara dan gas, untuk memenuhi kenaikanpermintaan listrik sebesar 5% atau lebih di masadepan serta menggantikan pemakaian BBM tahun

2007 sebesar 23.571 GWh atau 23,6 x 109 kWhdengan panasbumi maka pemerintah dapatmenghemat 23,6 x 109 kWh x US$(0,26-0,08) /kWhx Rp. 9300 /US$ atau Rp 39,5 trilyun per tahun.Untuk itu dibutuhkan penambahan pembangkit listrikpanasbumi hanya sekitar 3000 MW.

Menurut Prof.Widjajono, masyarakat saatini sebaiknya dibuat lebih banyak belajar hematbelanja energi dan memahami kenyataan bahwaharga bensin premium terbaik adalah tetap di kisaranRp 5.500 per liter. Karena, bersamaan dengan itu,energi alternatif yang melimpah ruah di Nusantaraini akan termanfaatkan.


Kita juga jangan membiarkan PLN untukmelanjutkan praktek monopsoninya danmelanjutkan impor BBMnya dengan harga tinggi(15-30 sen dolar AS / Kwh) demi puluhan pembangkitkonvensionalnya, sambil tetap melanjutkanpemaksaan harga beli listrik murah (kurang dari 6sen dolar AS/Kwh) dari Pembangkit Tenaga ListrikPanas Bumi lokal (PLTP, yang mengoperasikanenergi terbarukan, berkelanjutan, bersih lingkungan,mendapat hadiah lingkungan internasional, meraihPROPER ’Gold’ dari Kementrian KLH dsb.).

Seperti kata Yasadipura yang dikutipcucunya: pujangga Ranggawarsita, jika kita

lanjutkan sikap suka kemudahan, takut kesulitan,insya’allah takkan ada yang diperoleh, demi tanahair kita tercinta....

(Penulis : Prijo Hutomo, praktisi di industri energialternatif yang tengah belajar dan berusaha kerasmemahami praktek-praktek manajemenkeberlanjutan energi di Indonesia; dan Prof.ZulkifliHusin, Ketua Program Doktor Ekonomi Bidang’Sustainable Development’ di Universitas Trisakti,Jakarta, Indonesia).


IntroductionIn the upstream oil and gas industry,

perforation is referred to small holes made at theportion of the production casing which connects tothe production zone. This connection provide pathfor hydrocarbon to flow from the reservoir up to thesurface. It has been proven in the literatures thatduring perforation operation, rock around theperforation tunnel will be squeezed and compactedas the perforation bullets enter the formation(Behrmann et al., 1996; Jilani et al. 2002; Smith etal 1997). This process can result “low-permeabilitycrushed zone” as the primary cause of wellboredamage. In order to minimize this damage,underbalanced perforating should be considered (Janet al. 2007; King et al. 1986).

Underbalaced perforating is a perforationconducted when wellbore pressure is lower than thereservoir pressure (King et al. 1986). It is generallyregarded as one of the best methods of creatingopen and undamaged perforations. The benefits ofunderbalanced perforation extend far beyond aproducer’s ability to utilize previously hard-to-harvestoil and gas. In addition, it also helps to minimizedamage to under-pressurized hydrocarbon pools orformations prone to lost circulation; increaseddrilling rates have been observed with certain rocktypes; bit life may be extended; reduced chipholdown occurs, allowing the bit to continuously cut

Optimization Of Super Light Weight Completion

Fluids Using Fractional Factorial Design

olehMunawar Khalil, Badrul Mohamed JanDepartment of Chemical Engineering, Faculty of Engineering,University of Malaya, 50603, Kuala Lumpur, Malaysia

Abstrak

Underbalanced perforation is widely known as one of the best technique to increase hydrocarbonflow. Numerous researchers have proved that the resulting production increases is due to the surge ofhydrocarbon that eliminate permeability damage around the perforation tunnel. In order to achieve theunderbalance condition downhole, super light weight fluids needs to be formulated. The formulated fluidhas lower density compared to the existing fluid in the market. In this research, optimization formulationof super light weight completion fluids was conducted based on experimental design software. There are4 optimized factors being optimized namely the amount of homogenizing agent (A), amount of additive(B), mixing speed (C) and mixing time (D). Model equation was proposed with regard to the mentionedfactors. The results shows that the optimized condition of the formulation is 4% w/w of homogenizingagent, 10% w/w of additive and mixed at 14276.11 rpm during 1 hour. The desirability of the optimizedcondition is 0.710.

new rock which is then swept away by the drillingfluid; and, oftentimes, well stimulation can beeliminated, further protecting damage-proneformations (Wong and Arco 2003).

It has been reported that an underbalancedperforation condition could be achieved byformulating a stable fluid with non-damagingchemical properties that would have a significantlylower density which called as completion fluid (Janet al. 2007; Wong and Arco 2003). Jan et al. in 2003had successfully formulate a super light weightcompletion fluid which consist of Shell Sarapar 147synthetic oil [Shell MDS (M)], 3MTM Glass Bubblesas a density reducing agent and an appropriaterheology control agent. The measured density ofthe formulated completion fluids is around 5.0 ppg.

The measured value is much lower than theexisting completion fluids in the market which isaround 7.57 ppg. Field test has also shown theincrease of production rate after perforation job wasconducted with fluids based on the optimizedformulation from laboratory (Jan et al. 2007).

In order to optimize the formulation, it isnecessary to perform critical parametersoptimization. Such parameters include the amountof rheology controlling agent, amount of additive,speed of mixing, and mixing time. Fractional factorialexperimental design was used to evaluate thesignificance the variables, as well as theirinteractions (Morgan 1991).

ilmiahilmiah


It is well known that the number ofexperimental runs increase geometrically with thenumber of model parameters (factors). For example,a model with 8 factors would need 28 = 256experimental runs in order to investigate all thesingle and interaction between them (Montgomery1997). From the mathematical and statistical pointof view, it is not necessary to conduct completeparameter variation in order to optimize the factorssince not all of the parameter interactions haveappreciable effects on the model results (Akgüngör& Yildiz 2007). Thus, only the fraction of wholeexperimental runs in factorial design is performedfor the sensitivity analysis.

Materials and Methods2.1 Materials

To formulate the super light weightcompletion fluids, synthetic oil based completionfluids were used. The completion fluids are ShellSarapar 147 synthetic oil [Shell MDS (M)]. 3MTM

Glass Bubbles (HGS) were used as a densityreducing agent. Homogenizing agent was used asrheology controlling agent in order to suspend theglass bubbles in homogenous slurry, and additivewas used to increase the stability. For thedetermination of density, a 25 ml picnometer wasused. The rheological measurements wereconducted by a HAAKE VT 550 shear ratecontrolled-viscometer (Gebruder Haake GmbH,Karlsruhe, Germany). For the slurry mixing,disperser T25 (IKA LABORTECHNIK, Germany) wasused.

2.2 Formulation of super light weight completionfluids

To formulate super light weight completionfluids, firstly, 65 % w/w of completion fluids wasmixed manually with additive. The amount of additivewas optimized from 1 % w/w until 10 % w/w.Secondly, 35 % w/w of glass bubbles was mixedwith the homogenizing agent. The amount ofhomogenizing agent was optimized from 1 % w/wuntil 4 % w/w. The solution of the completion fluidsand additive was placed under homogenizer, andthe powder mixture (glass bubbles and homogenizingagent) was added slowly into the solution. The slurrywas agitated at various speeds and mixing time.Finally, other variables of response such as density,plastic viscosity, and stability were also determined.

2.3 Experimental DesignSince a full factorial design is introduced to

the experimental design with k factors, there wouldbe 2k experimental runs. However, as we discusearlier, it is not necessary to investigate the whole2k experimental runs. In a two level half fractional

factorial design experimental, each factor wasassigned as two levels: low (-1.000) and high (1.000).If k factor are considered, there will be 2k - 1

measurements in order to perform the analysis. Inorder to gain the accuracy, at least one replicatemust be conducted. Into the bargain, it is necessaryto conduct at least one central point measurementin order to test for quadratic term within the low-to-high range (Abdul-wahab & Abdo 2007). Therefore,the total number of test is given as:

N = r2-k- – 1 + C (1)

Where N represents the total number of test, r isthe number of replicates, and C represents thenumber of center-point measurements. The aims ofthis optimization are to get the minimum density ofthe slurry in order to perform underbalancedperforation condition, to gain the stability of the fluid,and to minimize the viscosity.

In this research 4 critical factors offormulating super light weight completion fluids, k= 5, i.e. A (amount of homogenizing agent), B(amount of additive), C (speed of mixing), and D(time of mixing) were investigated. The coded levelfor these factors is given in table 1. Since two testsare made for each combination (r = 2) to estimatefor the pure error, and three center-pointmeasurement are conducted in the analysis (C = 3)to estimate curvature in the model, it was resultedin a total N = 19 experimental measurements basedon equation 1. Table 2 represents the experimentaldesign using fractional factorial design which usedin this research, the extra experimentalmeasurements (center-point measurements) werecollected at mid-level (coded as zero). Experimentswere randomized to maximize the effects ofunexplained variability in the observed responses,due to extraneous factors.

The statistical software was used in orderto model building, experimental design, and dataanalysis. The effect of each factors and theirinteractions were considered to three parameters,i.e. density of the slurry, plastic viscosity, andstability.

Results and Discussions3.1 Response Analysis

The software was used in the statisticalanalysis investigation. Using this software, half-normal plot in which the ranks of the absolute valueof various effects were measured. Figure 1 showsthe half normal-normal plot for density as theresponse variable. Based on figure 1, the factor thatlie along the line are negligible and the rest of thefactors and their cross interaction are significant. Infigure 1, it can be seen that speed of mixing (C),time of mixing (D), the interaction between the


amount of homogenizing agent and amount ofadditive (AB), and the interaction between amountof homogenizing agent and time of mixing (AD) aresignificant to the density of slurry. The effects of thefactor are calculated by averaging the responses ofeach factor at the plus levels and subtracting theaverage at the minus levels for the same factor(Abdul-Wahab & Abdo 2007).

Similar analysis for the significant factorsand their interaction was also conducted to otherresponses, i.e. plastic viscosity and stability. Figure2 and 3 show the half-normal plot for those tworesponses respectively. Based on figure 2, there are6 terms as the significant factor for the effects, i.e.speed of mixing (C), the interaction between amountof homogenizing agent and time of mixing (AD), theinteraction between amount of homogenizing agentand speed of mixing (AC), time of mixing (D),amount of homogenizing agent (A), and amount ofadditive (B). Afterwards, based on figure 3 thesignificant terms are amount of homogenizing agent(A), amount of additive (B), the interaction betweenamount of homogenizing agent and additive (AB),the interaction between amount of homogenizingagent and speed of mixing (AC), the interactionbetween amount of homogenizing agent and timeof mixing (AD), and speed of mixing (C).

Before the conclusion from the analysis ofvariance was adopted, the adequacy of theunderlying model should be checked by examinationof residual. The normal probability plot is a graphicaltechnique for assessing whether or not a data setis approximately normally distributed. Figure 4presents a normal probability plot of the residual fordensity, plastic viscosity, and stability respectively.

Figure 4(a) and 4(b) indicate that the modelwas adequate and it is also normally distributed forthose three responses. However based on figure4(c), the model for stability does not seem to benormally distributed. It is proved by the data doesn’tresemble to the straight line. Hence, it is necessaryto check another transformation that could havebeen applied using Box-Cox plot.

3.2 Box-Cox TransformationNormal plot of residual for stability indicates

that the model doesn’t seem to be normallydistributed. Hence, it is necessary to conduct theBox-Cox transformation check. Figure 5 shows theBox-Cox plot for stability. It is recommended to usedsquare root transformation (Lambda = 0.5) with k =0.2.

3.3 Residual AnalysisIn order to check on the equality of variance,

residual versus predicted plot was investigated.Figure 7 shows the residual versus predicted plot

for density. From this figure, it seems that theequality of variance is not violated. The sameindication also obtained at the rest of responses(plastic viscosity and stability).

3.4 Regression AnalysisUsing the software, it is possible to develop

the model equation in order to predict the outcomedepending on the level for each factor. This equationcalled the coded equation since the +1 and -1 valuesare used to represent high and low levels,respectively. In inverse, if the actual values are used,there would be actual equation. Equation 2, 3, and4 show the coded equation for density, plasticviscosity, and stability respectively. R2 value for thosemodel was very high, i.e. 0.9589 for density (equation2), 0.9727 for plastic viscosity (equation 3), and0.9920 for stability (equation 4).

Density = 4.68 + 0.046 * A - 0.027 * B + 0.29 * C + 0.17 * D+ 0.15 * A * B - 8.556E-003 * A * C - 0.10 * A * D (2)

plastic viscosity = 487.38 + 57.16 * A - 51.53 * B - 240.51* C - 61.37 * D - 14.06 * A * B - 61.88 * A * C + 119.41 * A * D (3)

0.2stability + = 1.64 + 1.03 * A + 0.81 * B + 0.10 * C

+0.049 * D + 0.65 * A * B + 0.26 * A * C - 0.11 * A * D (4)

3.5 Optimal DesignIn order to get the best condition for

formulating super light completion fluids, a fractionalfactorial experimental design tests wereinvestigated. Figure 8 shows ramps of various asthe solution to the optimization job. It was obtainedthat 4% w/w of homogenizing agent, 10% w/w ofadditive and mixed at 14276.11 rpm during 1 houras the best condition with the desirability = 0.710.The software also provides the optimal design indifferent desirability factors. Table 3 presents thealternative solution for optimal design at otherdesirability.

ConclusionIn order to get underbalanced condition in

the borehole during perforation job, low density ofthe slurry, low viscosity and higher stability shouldbe important. A fractional factorial design wasdevelop for optimizing super light completion fluids.The result have shown that speed of mixing (C),time of mixing (D), the interaction between theamount of homogenizing agent and amount ofadditive (AB), and the interaction between amountof homogenizing agent and time of mixing (AD) are


significant to the density of slurry. Then, for theplastic viscosity, the significant terms are speed ofmixing (C), the interaction between amount ofhomogenizing agent and time of mixing (AD), theinteraction between amount of homogenizing agentand speed of mixing (AC), time of mixing (D),amount of homogenizing agent (A), and amount ofadditive (B). and finally, for stability, the terms areamount of homogenizing agent (A), amount ofadditive (B), the interaction between amount ofhomogenizing agent and additive (AB), theinteraction between amount of homogenizing agentand speed of mixing (AC), the interaction betweenamount of homogenizing agent and time of mixing(AD), and speed of mixing (C).

The model was tested for its adequacy usingnormal plot residual and found that the assumptionof normality and independency are not violatedexcept for stability. Then, Box-Cox transformationhave suggested square root transformation (Lambda= 0.5) with k = 0.2. R2 value was very high, thatsuggesting that model accounted for most of thevariability. The optimal solution with specifieddesirability was calculated using the software andpresented. The selected solution is 4% w/w ofhomogenizing agent, 10% w/w of additive and mixedat 14276.11 rpm during 1 hour as the best conditionwith the desirability = 0.710.

References1. Abdul-Wahab, S.A., Abdo, J. 2007. Optimization

of multistage flash desalination process by usinga two-level factorial design. Applied ThermalEngineering. 27, 413-421.

2. Akgüngör, A.P., Yildiz, O. 2007. Sensitivityanalysis of an accident prediction model by thefractional factorial method. Accident Analysis andPrevention. 39, 63-68.

3. Behrmann, L.A., Huber, K., McDonald, B., Couet,B., Dees, J., Folse, R., Handren, P., Schmidt,J., Snider, P., Quo Vadis, Extreme Overbalance?.Oilfield Review 8, no. 3 (Autumn 1996): 18-33.

4. Jan, B.M., Rae, G., Noor, I., Suhadi, A.N.,Devadass, M., 2007. Increasing production bymaximizing underbalance during perforation.Paper SPE 108423 presented at the 2007International Oil Conference and Exhibition,Veracruz, Mexico, 27–30.

5. Jilani, S.Z., Menouar H., Al-Majed A.A., KhanM.A., 2002. Effect of overbalance pressure onformation damage. J. Pet. Sci. Eng. 36, 97-109.

6. King, G.E., Anderson, A.R., Singham, M.D. 1986.A field study of underbalance pressure necessaryto obtain clean perforating using tubing-conveyedperforating. Paper SPE 14321 presented at the1986 SPE Annual Technical Conference andExhibition, Las Vegas, 22-25 September.

7. Montgomery, D.C., 1997. Design and analysis ofexperiments fifth edition. John Wiley & Sons,Inc: New York.

8. Morgan, E., 1991. Chemometrics: experimentaldesign. John Wiley & Sons, Inc: London.

9. Smith, P.S., Behrmann, L.A. Yang,W., 1997.Improvements in perforating performance in highcompressive strength rock. Paper SPE 38141,presented at the SPE European FormationDamage Conference, The Hague, TheNetherlands, June 2-3.

10. Wong, A, Arco, MJ., 2001. Use of hollow glassbubble as a density reducing agent for drilling.Paper No. 2001-31 presented at CODE/CAODCdrilling conference, Calgary, Alberta Canada, 23-24 October.

11. Wong, A., Arco, M.J., 2003. A new technologyfor reducing the density of drilling fluids. 3MSpecialty Materials 98-0212-2679-4.


Figure 2 Half-normal plot of plastic viscosity.

Figure 1 Half-normal plot of density


Figure 3 Half-normal plot of stability.

Figure 4 Normal plot of residual for the responses; (a) for density, (b) for plastic viscosity, (c) forstability.


Figure 5 Box-Cox plot for stability.

Figure 6 Normal plot of for stability after Box Cox transformation.


Figure 7 Residual versus predicted plot for density.

Desirability = 0.710

Figure 8 Ramps for various factors.


Table 2. Experimental design used in fractional factorial design studies by using forindependent variables and three dependent variables

Table 1. Code levels for independent factors used in the experimental design

Table 3. Optimal Solution


IntroductionSuccess of secondary and tertiary oil

recovery projects targeting the remaining oil inmature or partially depleted reservoirs stronglydepends on adequate description of reservoirheterogeneity. Process that are well-understood ina laboratory environment and properly designed forthe reservoir may fail when implemented in thereservoir because of inadequate description ofreservoir. Reports such failures are numerous inpetroleum engineering literature. A number ofprocedures exist that can be used beforeimplementation of an EOR process in attempt todescribe the reservoir geology. One of theseprocedures is tracer test.

Tracer technology plays an important rolein improving the reservoir characterization before theapplication of EOR methods by providing qualitativeinformation on reservoir compartmentalization,preferential flow paths to improve understanding offluid movement in the reservoir, stratification, andheterogeneities distribution. Basically, tracer ischemicals that can be added to fluids in smallconcentrations and used to follows their movementwithout affecting their physical properties. It alsocan be used as an effective tool to detect andestimate of remaining oil saturation using twodifferent types of tracers. The two tracer types aredifferentiated into the conservative (ideal) tracer and

Tracer Test Evaluation on Core Flooding

olehUtomo Pratama, Usman Pasarai, SugihardjoPPPTMGB ”LEMIGAS”

Abstrak

The implemented and success or failure of an EOR process are always affected by reservoirgeologic heterogeneities which causes fluids movement in reservoir is very non uniform. Factors of thistype, unless properly identified and understood before the start of EOR process, will likely cause a projectfailure.

Core flooding as the model of small scale of fluids movements in reservoir undergoes similarcircumstances. Tracer test was performed prior oilflood to characterize the core. On this experiment,lithium solution was selected as tracer solution to be then injected into core at constant rate, 4 ft/day.Afterwards, the effluent collected by Gilson sample collector in each tube for further d its concentrationusing Atomic Absorption Spectrometry (AAS).

Lithium concentration reported in some extent and subsequently analyzed by employing methodof temporal moments. This method provides numerical calculation to estimate effective core pore volume(PV). Weighing method made also use to compare the PV with aforementioned method and the resultswere comparable. Furthermore, response curves were able to determine core heterogeneities and set up.

ilmiah

the partitioning tracer. The ideal tracers do not havesolubility in other substances, at this case is oil orby definition they do not interact with the rock orother fluid phases present.. Thus when ideals areinjected in reservoir, they will flow only in the waterphase adopting the velocity of this phase. Someexamples that can be used as ideal tracer are iso-propyl alcohol (IPA), bromide and lithium.

In contrast, partitioning tracers are solublein liquid hydrocarbon as well as water or gas phases.The molecules of the partitioning tracers are movingback and forth between the water and oil phase,because they have high partition coefficient (absorbinto the rock) which is determine the tracer solubilityin other phases. Consequently, when a pulse ofaqueous solution containing a suite of partitioningtracers is injected into an oil reservoir, the tracerwill continuously partition into and out of the oilphase contacted by aqueous solution (injected gas).Hence the molecules of partitioning tracers areflowing with the water velocity when they are in thewater phase and oil velocity when they are in oilphase6. Thus, partitioning tracers propagate moreslowly in an oil reservoir than conservative tracers.This retardation of the partitioning tracer is analogueto chromatographic separation where thismechanism is utilized to estimate oil saturation inthe reservoir. Several examples from this type oftracers are n-butanol, rhodamine, and propanol

ilmiah


Tracer test is necessary to be performed incore flooding experiment when we are deeplyconcern with core properties and validation seeking.Besides, describes the heterogeneity within coreincluding its connectivity, another advantage is leakindicator in core set up. Moreover, we could obtaineffective pore volume swept by the tracer.

This paper will describe the results of tracertest on several core flooding succinctly includinghow to use the method of temporal moment analysisfrom tracer response curve data to estimate effectivecore pore volume. A quick look observation onresponse curve to reflect core heterogeneity andexperimental set up was also presented concisely.

Tracer Interpretations data methodA host of tracer analysis methods consider

the temporal behavior of tracers. The methods wereoriginally developed for closed reactor vessel, buthave been applied to more general conditions of openboundaries, characterization fractured media undercontinuous reinjection and estimates flow geometry.The methods have rigorous mathematical basis. Themethods and application mentioned above are allbased on analysis of tracer residence times. Themean residence time or first temporal moment, isthe most useful single property derived from tracertest, although other properties have been used aswell. Levenspiel10 shows the total pore volume sweptby the tracer can be determined from its residencetime.

The method of temporal moments is a verysimple, fast and robust method to estimate sweptpore volume and remaining oil saturation. Asexplained earlier, to calculate remaining oil saturationwe need two different types of tracer where the idealtracer behaves as the reference tracer and thepartitioning tracer as the partitioned one. Becauseof the presence of oil, partitioning tracers are retardedcompound to the non-partitioning. However, thispaper aims merely on effective pore volumeestimate.

Effective pore volume can be estimated usingone tracer that swept by tracer. The pore volumewas determined from tracer mean residence timewhich required steps are summarized as follows:· Normalize the tracer history· Extrapolate the history to late time· Calculate mean residence time and swept volume

Although this method provides someadvantages but it is necessary to inform that momentanalysis is a general method, and one that suffersfrom few limitations. Assumed conditions essentiallystate that the flow field is steady and tracers moveswith bulk fluid flow such that the information obtainedfrom the analysis is general applicable. Theseconditions can be stated as follows:

1. Steady state injection and extraction2. The tracer is ideal and conservative

Experiment ProcedureSeveral tracer tests were conducted by using

high pressurized core cell with low dead volume.The experiment used both standard and nativesandstone core that condition set to reservoir state(table 1). The core was prepared to contain residualoil saturation using standard procedure by alteringflow of brine and crude oil. The experimentconfiguration is illustrated in figure 1.

When residual oil saturation had beenestablished, core was flooded with synthetic brineat constant flow rate, 4 ft/day. During this period, asmall volume of tracer was injected into the flowingwater phase close to the core inlet. On thisexperiment Lithium (Li), which dissolved in LiClsolution, was selected as conventional tracer. Thesolution had been designed 100 ppm Li+ ion.Waterflood resumed at stable flow rate until the tracerchemical was eluted through the core plug, thusthe majority of the injected tracer mass could berecovered.

The Aliquots of 5 ml of the effluent werecontinuously collected by Gilson sample collectorfraction close to the outlet of the backpressureregulator valve. Then each sample fractions wereanalyzed for its content of Li by atomic absorptionspectrometry (AAS).

ResultsEach sample concentration was plotted

versus its volume, as depicted in figure 2. Thecalculation of effective swept pore volume will bedetermined from temporal moment method. The firststep is normalizing the concentration history bydividing measured Li+ concentration to total Li+injected concentration.

Tracer response curves should be completein terms of outflow measured concentration in orderto estimate effective pore volume precisely, becausemuch of the information is contained in the tails ofthe response curve. Unfortunately, the tracerresponse curve is often incomplete either due todilution of the tracer concentration below detectablelimit of apparatus or some other reasons. Thereforeto overcome this difficulty is with extrapolating thehistory to long time. The tracer response curve canbe extrapolated with an exponential function providedthe derivation of the test is sufficient to establishthis decline.

The first moment of the tracer response curveswas obtained by dividing the data into two parts.The first part represents the data from zero to thetime tb where time becomes exponential, and the


second covers the exponential part in which it goesfrom tb to infinity.

After time tb, the tracer response is assumedto follow an exponential decline given by:

−−

= att

b

b

eCC (1)4

Where 1/a is slope of the straight line when thetracer response curves are plotted in semi-log scaleand Cb is the tracer concentration at time tb whencurve becomes exponential. The extrapolated curveis seen in figure 3.

The mean residence time, or first temporalmoment, of tracer is determined directly from thenormalized and extrapolated tracer history as:

∫

∫−

−

+

++=

bb

bb

t at

t

bat

eabCdt

ateabtCdt

t

0

0 2 )1(*

(2)4

The constants a, b and tb are determinedby curve-fitting late time tracer data in spreadsheet.Pore volume estimates follow directly from the meanresidence time as describe from the equation below:

*tqMmV injinj

p =

(3)4

Then each data was tabulated and plottedagainst cumulative volume as shown in figure 4. Theeffective pore volume calculation from temporalmoments and weighing method is 118.14 cc and115.23 cc respectively. There is a fair differencebetween the estimation of effective pore volumeusing temporal moments and weighing method.

Core heterogeneity is reflected from twopeaks which formed in tracer response curve. Thesepeaks mean the core was stratified into differentflow paths (Fig. 5). Another case depicted in Figure6 shows scattered and varying noise in responsecurve that may indicate the leakage occured duringthe flooding due to imperfect core set up particularlyin core holder sleeve.

Conclusions1. Effective pore volume estimates from temporal

moments and weighing method are 118.14 cc and115.23 cc respectively

2. Tracer test provides helpful tools to improve corecharacterization by providing qualitativeinformation on preferential flow, stratification, coreconnectivity, heterogeneities distribution andapparatus set up.

3. Method of First Temporal Moment Analysis issimpler and faster to interpret tracer responsecurve.

4. Tracer test possible to be used as initialassessment to core before proceeding coreflooding. Hence, it will save times and cost inexperimental.

NomenclatureC = tracer concentration, ppmq = flow rate, cc/minM = total tracer injected, cc

References1. Tang, J.S.,”extended brigham model for remaining

oil saturation measurement by partitioning tracertest”, SPE 84874.

2. Sinha, R., K. Asakawa K, G.A. Pope, and K.Sepehnoori,”simulation of natural and partitioninginterwell tracer test to calculate saturation andswept volume in oil reservoirs”, SPE 89458.

3. Illiasov, P.A., A.D. Gupta, and D.W. Vasco, “Field-Scale Characterization of Permeability andSaturation Distribution Using Partitioning TracerTests: The Ranger Field, Texas”, SPE 71320.

4. Shook, G. M., J. Hope Forsmann, “TracerInterpretation Using Temporal Moments on aSpreadsheet” Idaho National Laboratorydocuments,(2005)

5. Asakawa, K., “a generalized analysis ofpartitioning interwell tracer test”, dissertation,university of texas, (2005).

6. Chatzichristos, C.,Ø. Dugstad, A. Haugan, J.Sagen, J. Muller,”Application of PartitioningTracers for Remaining Oil Saturation Estimation:An Experimental and Numerical Study”, SPE59369.

7. Jin, M., R.E. Jackson, G.A. Pope, S.Talfinder,”development of partitioning tracer testfor characterization of non-aqueous phase liquidcontaminated aquifers”, SPE 39293.

8. Abidin, Z., “Teknologi Perunut Untuk ManajemenReservoir Minyak bumi (EOR)”, Pusat TeknologiAplikasi Isotop dan radiasi, BATAN.

9. Bailey, R.E., and L.B. Curtis, “enhanced oilrecovery”, national petroleum council, (1984).

10. Levenspiel, O., @Chemical ReactionEngineering@, 2nd edition, New York: John Wileyand Sons, Chapter 9, (1972).


Figure 1. Core flooding scheme for Tracer Test

Table 1. General Experimental Data


Figure2. Tracer Response Curve

Figure 3. Li Extrapolated Curve


Figure 4. Pore Volume Estimation Curve

Figure 5. Core Heterogeneity Response Curve


Figure 6. Leakage on Core Holder Set up Response Curve

Documents

JTMGB 2 2009 - umexpert.um.edu.my laboratorium untuk reaktivasi lapangan-x dengan injeksi kimia acidizing in esp wells: an efficient production optimization kebergantungan pada migas: