SPE 87472 Groet 1 Scale Study: Analysis, Diagnosis & Solution Implementation

H.F. Nieuwland, SPE, BP Nederland Energie BV; I.R Collins, SPE, BP Exploration Operating Company

Copyright 2004, Society of Petroleum Engineers Inc. This paper was prepared for presentation at the 6 International Symposium on Olfield Scale held in Aberdeen, UK, 26-27 May 2004. This paper was selected for presentation by an SPE Program Committee following review of information contained in an abstract submitted by the author(s). Contents of the paper, as presented, have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material, as presented, does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Papers presented at SPE meetings are subject to publication review by Editorial Committees of the Society of Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.
th

260 240 220 200 180 160 P/Z [bara] 140 120 100 80 60 40 Measured Trend

Abstract Gas well Groet 1 (GRT1), located onshore in the Netherlands, has been experiencing unstable production for six months since Q1 2002. A down hole video survey identified salt (halite) deposition in the tubing which was subsequently removed by a gauge cutter. The problem reoccurred in Q4 2002 leading to an 8.3 MBOE loss in the quarter. A scale prediction study was conducted to analyse the problem arising with well GRT1 and to diagnose whether the scaling problem is transient or persistent. This study indicated that the scaling behaviour is persistent. Several remedial actions and control options were considered, based upon the diagnosis. All of the remedial actions and control options were compared on a risk / cost / benefit basis. The outcome of this study was that the most economical and lowest risk halite control option was bi-weekly fresh water soaks. Introduction GRT1 well was drilled in 1965 and discovered the Groet Rotliegend sandstone dry gas reservoir. Gas production from the field started in 1974 with 4 wells after flowline construction was finished. Initial reservoir pressure was 230 bara at 2100 m TVDss. Reservoir pressure dropped to 36 bara after a cumulative production of 6.1 Bnm3 (December 2003). The field is under volumetric depletion and has shown no indication of active aquifer drive, see the P/Z plot in Figure 1. GRT1 is now the only active well in the gas field and produces the remaining gas reserves. The other three production wells in the field are shut in due to water coning.

20 0 0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 4,500 5,000
3

5,500

6,000

6,500

7,000

7,500

Cumulative Production [million nm ]

Figure 1: Groet gas field P/Z plot.

Gas production from the GRT1 well has been stable since start of production in 1974. In April 2002 flow became erratic. A real time down hole video survey identified salt deposition in the tubing at two different levels, see Figure 2 and Figure 5. The liner could not be surveyed because of a salt plug in the tubing end. The salt was removed by a gauge cutter run on slickline and the well performed without any problems for another six months.

Figure 2: Scale observed in tubing GRT1 by down hole video.

The problem reoccurred in Q4 2002 leading to an 8.3 MBOE loss in the quarter. GRT1 well is the only well that produces the remaining reserves of the Groet gas field.

www.petroman.ir

SPE 87472

A scale study was set up because of the importance of this well. The scale study aims to analyze the problem with well GRT1 by: 1. determining the mechanism of salt precipitation, 2. diagnosing whether the scaling problem is transient or persistent, 3. determining the likely mass of salt that can form and its location in the well. On the basis of the diagnosis, remedial/control options can be developed to secure production and reserves. Well Test Analysis Comparison of historic well test data could indicate timing of production impairment. Four GRT1 multi-rate well tests were reviewed, dated September 1989, September 1991, September 1994 and May 1999. Figure 3 shows the derivative plot of the final build-up of the well tests. The derivate plots of the 1989 (purple), 1991 (green) and 1994 (red) well tests are alike. The derivate plot of the 1999 (blue) well test is different and shows an increase in skin.
total skin [-]

5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0 100,000 200,000 300,000 400,000 500,000 600,000 gas rate [nm3/d] 1989 1991 1994 1999

Figure 4: Groet 1 multi-rate skin plot.

Nodal Analysis Nodal analysis was performed on the four well tests to establish temperature and pressure profiles along the well bore as input for the scale prediction study. The production string of GRT1 is mainly 4 tubing. Figure 5 zooms in on the section of the completion where the salt scale was found inside the tubing during the down hole video survey. The nodal analysis software used Petroleum Experts). The well model was data measured during the well tests, see details. The data of January 2003 is measurements only.
Zoom : 2250 m MD

was Prosper (ex conditioned to the Table 1 for more based on surface

1999 DP & DERIVATIVE (BAR2/CP/M3/D)

10 -1

1991 1994

1989

10 -2

4.5 in Tbg

4 in Tbg Packer 3.5 in Tbg

10 -3 10 -2 10 -1 10 0 10 1

10 -3

Scale along tubing wall

Del ta-T (hr)

Figure 3: Groet 1 well test derivative plot.

4.5 in Tbg 7 in Csg

Figure 4 shows the multi-rate skin plot of the well tests. The data points of the 1989, 1991 and 1994 test data scatter but are along the same trend. The data points of the 1999 well test are completely off from the 1989 - 1994 trend and show an increase in skin and non-Darcy factor. From the well test data it can be concluded that there were no major changes in reservoir behavior before September 1994. The scaling most likely started to develop between September 1994 and May 1999.

3.5 in Tbg Packer 2.875 in Tbg Scale plug in tubing

Top perforations

5 in Lnr

Bottom Perforations Zoom : 2669 m MD

Figure 5: GRT1 completion schematic zoomed in.

www.petroman.ir

SPE 87472

Sample Analysis Solid Sample Analysis A solid sample recovered from the well was send to the laboratory for analysis. The first part of the solid sample was dissolved in deionized water and the soluble ion content determined by Inductively Coupled Plasma Atomic Emission Spectroscopy (ICP-AES). The dissolved sample consisted mainly (97.6%) of sodium chloride. A second part of the solid sample was sent for X-ray diffraction analysis to determine the mineral phases present and for SEM analysis including EDX surface elemental analysis. Figure 6 shows the recorded XRD pattern. As can be seen from this pattern, the sample consisted of halite (sodium chloride) with a small amount of sand. Figure 7 shows an image of the crystal. EDX analysis of the crystals confirmed halite.

Wellhead Water Sample Analysis A wellhead water sample was taken from the well and send to the laboratory for compositional analysis by ICP-AES. This compositional analysis was used to determine under what production conditions halite scale is predicted to form and to then history match the model to past and current production parameters. This allows the operational envelope within which scale is expected to form to be determined. Scale Prediction Study Scale prediction has been performed using ScaleChem 3.0 (build 11). This model is essentially a geochemical model that allows the oversaturation of mineral phases in equilibrium with brine, oil and gas to be predicted. The oversaturation is characterized by the saturation ratio, which is a number describing by how many times a mineral exceeds its thermodynamic solubility limit. When the saturation ratio equals 1, the brine is saturated with the mineral in question. Also predicted by the ScaleChem software is the maximum mass of the mineral that can form under the given conditions. It should be noted that such models are thermodynamic in nature and hence take no account of the rate at which the mineral forms. Thus a mineral can be predicted to be oversaturated but it will not form in the production timescale. This is not generally the case with halite (and even 5% oversaturated solutions will instantly deposit solid halite). Model parameters

Halite

Halite

Quartz Halite
4 10 20 30 40

Halite Quartz Halite

50 60

Halite

70

2 T heta

S c a le

Figure 6: Solid sample recorded XRD pattern.

The isochronal well test in 1999 determined the reservoir pressure as 47.6 bara. This pressure was used in the modeling along with a temperature of 87C for the reference reservoir parameters. The nodal well model was then used to calculate the flowing bottom hole pressure and temperature and the flowing wellhead temperature and pressure. These points defined the set points of the well model. The reservoir gas composition was taken back to the reservoir conditions and a dew point calculation was performed to determine the water content of the gas. It is essential to get this calculation right since there are two driving mechanisms for halite deposition: 1. halite solubility changes with temperature (reducing temperature reduces solubility); and 2. water evaporation from brine phase to satisfy the gas water requirement leading to the concentration of brines (likewise the condensation of water from the gas phase can dilute brines and hence mitigate halite formation).1,2,3 Accordingly, the resulting gas compositions were benchmarked against HYSYS (ex Aspen Technologies). The water content of the gas was calculated to be 1.48 mol% under the reference conditions. It is likely that the sample as received has been depleted in sodium and chloride ions, due to the fact that the water sample was a wellhead sample and that halite scale has been recovered from the well. An initial scoping study was performed using the reference conditions in which the formation water chemistry was reconstructed by using the composition of the solid recovered to ratio the chloride content to that of the bromide and nitrate contents based upon the

Figure 7: Image of the solid sample.

The conclusion from solid sample analysis is that halite scale has been deposited in the tubing.

www.petroman.ir

SPE 87472

ratios of the known solubilities of sodium chloride, sodium bromide and sodium nitrate at reservoir conditions. This meant that the chloride content had to be increased from 61,954 mg/l to 129,071 mg/l and the sodium content prorated accordingly. However, no down hole halite was predicted for this brine chemistry using the January 2003 conditions. A sensitivity study was performed by increasing the sodium chloride content of the brine and performing scale predictions to determine at what brine composition halite scale was formed. This was found to occur when the brine was saturated with halite under static reservoir conditions (increasing chloride to ca. 196,000 mg/l). Therefore the protocol was adopted of ensuring that the formation brine was saturated in halite as starting point for each set of calculations. There is the possibility of calcium carbonate formation because of the presence of both bicarbonate ions and calcium ions in the brine. However, invariably the bicarbonate content of samples taken from the wellhead is incorrect owing to the out-gassing of CO2 from the sample. The recommended procedure to counteract this problem is to equilibrate the brine with calcite under reservoir conditions (including the dissolved gas) and to adjust the brine composition until the calcite saturation ratio is 1. The reason for this is that the brine has been in equilibrium with calcite minerals in the reservoir over geological time and so the brine must be in equilibrium. This protocol has been employed in the scale predictions. Table 2 gives the as measured brine composition from the analysis of the wellhead sample and the equilibrated brine composition used in the predictions. Having determined the behavior of the brine under the initial reference conditions, a full suite of simulations were performed for fixed conditions using the protocol established above (saturate the gas with water under the set reservoir conditions, saturate the brine with calcite and halite under the same conditions and then produce the fluids using the production conditions listed in Table 1). Scale Prediction Results No scales are predicted for the reservoir equilibrated formation waters generated for the 1989, 1991 and 1994 well and reservoir conditions. For the 1999 conditions, a very mild calcite scaling tendency is predicted at the perforations (and most probably in the near well region owing to the large draw down in this region) for water rates of 80 BWPD and greater. This is illustrated by the graph in Figure 8.

1.2 Calcite Scaling Tendency @ perforations

12

10

0.6

0.4

0.2

Calcite Scaling Tendency Mass [mg/l]

0 0 100 200 300 Water Rate [BWPD] 400 500

0 600

Figure 8. Calcite scale predicted at perforations for well GRT1 in May 1999.

The thermodynamic equilibrium for calcite occurs at a scaling tendency of 1. This shows that the scaling tendencies are very low. In general, scale would not be expected to form at such low scaling tendencies. However, unique circumstances in individual wells can lead to scale formation (this is especially true in gas wells where high velocities can promote scale formation). If scale is deposited, the deposited masses of calcite scale will be low in the near well region. Even this low mass of scale could cause significant flow restriction and generate additional skin in the well. At the very least, the predictions do highlight that a calcite scaling tendency was starting to develop in 1999. The scale predictions performed under the January 2003 conditions showed both a halite and calcite scaling tendency. The graph in Figure 9 shows the development of the calcite scaling tendency along the well bore with water cut.
Calcite Scaling Tendency 0.0 0 0.5 1.0 1.5 2.0 2.5 3.0

500

1000 MD [m]

8 bwpd 10 bwpd 100 bwpd

1500

300 bwpd

2000

2500

Figure 9: Calcite scaling tendency as function of depth and water rate for Jan 2003 conditions

The data shows that there is a calcite scaling tendency at the perforations (and in the near well region owing to the draw down) for all water rates. Higher in the well the scaling tendency is, generally, too low to cause scale formation. The scaling tendencies are again very low but nevertheless higher than for 1999. The masses are all below 0.25 kg per day and at

www.petroman.ir

Calcite Mass [mg/l]

0.8

SPE 87472

low water rates are very small indeed (but are an order of magnitude higher than for the 1999 predictions); however, given that the scaling is predicted to occur at the perforations and around the near well bore, the amounts could be sufficient to cause flow restrictions and increase skin especially if the scale has been depositing for a number of years. Whether calcite scale is contributing to the skin damage in well GRT1 depends upon the current and historical water rates and upon the unique conditions present in the well. However, calcite scale damaging the near well and perforation region cannot be ruled out as a contributory factor. The scale prediction study has also confirmed a large halite formation tendency. A graph of halite scaling tendency plotted against measured depth for different water rates is shown in Figure 10.
Halite Scaling Tendency 0.0 0 0.5 1.0 1.5 2.0 2.5 3.0

Halite mass [mg/l] 0 0 100,000 200,000 300,000 400,000 500,000

500

1000 MD [m]

8 bwpd 10 bwpd 100 bwpd

1500

300 bwpd

2000

2500

Figure 11: Halite mass as a function of depth and water rates for Jan 2003 conditions.

500

1000 MD [m]

8 bwpd 10 bwpd 100 bwpd

1500

300 bwpd

2000

2500

Figure 10: Halite scaling tendency as function of depth and water rate for Jan 2003 conditions.

Figure 11 shows that the halite problem is purely water related, i.e., occurs on the well cutting measurable water, and is not driven solely by dehydration of the formation water (clearly this is an important mechanism at low water rates leading to higher masses of scale predicted than at higher water rates)4. The consequences of this are that the halite formation is a pernicious problem and will not go away as the water cut increases. Given that halite is always predicted at the perforations irrespective of the water rate, it seems reasonable to assume that halite scaling is contributing to the skin damage in well GRT1. The masses predicted are significant. (However, halite scale prediction is currently inaccurate and the predicted masses of halite should only be taken as indicative. The magnitude of these numbers does, however, suggest that the halite problem is significant.) Conclusions Scale Prediction Study

The halite scaling tendencies are very high at the perforations for low water rates. However even for significantly higher water rates, the scaling tendencies are still greater than 1; this means unequivocally that halite scale will form (unlike for calcite where scaling tendencies between 1 and 2 fall into a grey area owing to the kinetic control of the precipitation of this scale). The associated masses are also high (Figure 11). Figure 10 shows an increased halite scaling at 2350 m MD. This agrees with the depth in the tubing at which scale was found during the down hole camera survey, see Figure 5. This shows that there is good agreement between the scale model and the field observations.

Whilst this study confirms that between 1999 and the present day, well GRT1 has developed a persistent halite scaling problem, the lack of predicted halite scale prior to 2002 suggests that halite scale is not responsible for the skin that developed between 1994 and 1999. The fact that low calcite scaling tendencies have been predicted for 1999 onwards, suggests that potentially low levels of calcite scale could be contributing to the skin damage. Potential Production Improvement Figure 12 shows the inflow performance relationship (IPR) for the GRT1 well. The green curve represents the 1999 inflow performance relationship. The blue curve represents the 1994 inflow performance relationship. The difference between the curves is caused by the skin factors and non-Darcy factors. The red curve represents the tubing performance curve at 11.6 bara tubing head pressure. This figure shows that the production increase is 20,000 nm3/d when the original IPR (1994 curve) is re-installed by a well intervention. This rate gain needs to justify a well intervention to remedy the scaling problem with well GRT1.

www.petroman.ir

SPE 87472

60 IPR 1994 50 rate increase = 20,000 nm3/d Pressure [bara] 40 IPR 1999 Tubing Curve k = 11 mD h = 185 m r = 0.0752 m IPR old (1994): S = 0.8 D = 4.1e-6 [1/(nm3/d)] IPR new (1999): S = 2.1 D = 13.1e-6 [1/(nm3/d)]

possible formation damage and the low pressure reservoir that needs to produce back the squeezed fluids. Another option would be to operate the well outside of the halite and calcite scaling regions. Using the January 2003 data (Table 1), this would mean increasing the bottom hole pressure to 34 bara to eliminate both calcite and halite at the perforations (and in the near well region). This would prevent calcite entirely but would not control halite other than at the perforations (i.e., halite would still form higher up the well and would obviously have the potential to cause restrictions). This could be achieved by a stimulation job, e.g. fracture, and would increase production by some 100,000 nm3/d. This would mean to reduce the skin from 2 to 5, which is a challenging stimulation with no guarantee that it can be realized. Alternatively, calcite can be eliminated by decreasing the bottom hole flowing pressure to 23 bara. This, however, does not prevent halite formation and in fact increases the halite propensity at the perforations. The reduction in bottom hole flowing pressure can be achieved by replacing the current tubing by a 5 tubing and would increase production by 100,000 nm3/d. A tubing change out is costly. Increasing the tubing size will bring the tubing unloading rate closer to the well IPR curve, hence increasing exposure to liquid loading problems. The bottom line is that the well can potentially be operated in such a way as to ensure that there is no chance of calcite forming but that halite cannot be fully prevented; however, the location of halite formation can be controlled to a degree to prevent it forming at the perforations and in the near well region. This is a consequence of the fact that if a brine is oversaturated in halite, the salt will always drop out and in GRT1's case we do not have the flexibility to control the conditions such that the salt forms topside rather than down hole. This is consistent with previous studies of similar problems1. Comparison of Options and Selection All considered options for calcite and halite scale control were discussed briefly in the previous section and are summarized in Table 3. The first four columns in Table 3 show the areas tubing, perforations or near well bore, effected by the calcite or halite scale. A red block means effected by the scale. Thereafter, columns five to eight show the suggested remedial and control options, and the area these options are effective. Columns nine to twelve show the cost, risk, 12-month incremental production and well work cost. The best remedial control option for scale in the GRT1 well based on a risk / cost / benefit evaluation is bi-weekly (period is based on experience) fresh water soaks.

30

20

10

0 0 100 200 300 400 500 Gas Rate [knm3/d]

Figure 12: Groet 1 inflow performance curve.

Scale Remedial / Control Options Calcite scale can be removed by acidisation with HCl (12%) or organic acids. In order to remove near well damage, the acid would have to be squeezed into the reservoir (ca. 12 inches) and so damage testing and compatibility testing would need to be performed (especially given that the package will contain corrosion inhibitors). By sampling the acid returns the presence of calcite could be determined and the mass removed calculated. If calcite is found, the scaling tendency is too low to warrant active inhibition and the best control strategy would be periodic removal with acid. This technique is for the GRT1 well seen as high risk because of possible formation damage, low pressure reservoir that needs to produce back the squeezed fluids, and uncertainty about masses of calcite present in formation. If large masses of halite scale were deposited in the well then batch washing with fresh water would be a highly inefficient control strategy due to the frequency with which the treatment would need to be repeated. Nevertheless, for the GRT1 well it would be a very cheap and low risk option. Chemical mitigation is a better solution to control halite scale. Chemical inhibitors for halite do exist and have been applied on a continuous basis and as squeeze chemicals5,6,7. There have also been attempts to develop controlled release versions of the products for placement in the rat hole of wells. If a down hole injection line can be deployed, then chemical can be injected to control the halite; however, injection of chemicals down to the perforations is difficult and costly and doesnt provide protection to the near well bore in the case of well GRT1, if halite scale is responsible for or contributing to the skin damage then it is essential to prevent halite from depositing in the perforations and near well region. This is not realized by chemical injection down hole. Additionally, a complete tubing change out would be required to install a chemical injection string without jeopardizing the function of the down hole safety valve in the well. This adds costs and high risk to the control option. It would be sensible, therefore, to determine the feasibility of squeezing well GRT1. As mentioned before, squeezing is seen as high risk because of

www.petroman.ir

SPE 87472

Conclusion There is a small tendency for calcite scaling in the near well bore of well GRT1 that started prior to halite scaling. There is a large and proven tendency for halite scaling in well GRT1. The halite scaling is persistent and thus will come back after removal. The most efficient and cost effective control strategy to operate well GRT1 is bi-weekly fresh water washes. It is unfortunately not to be able to proactively treat or manage a scale problem and prevent it from occurring. However, an assurance process has been performed to confirm that the well is operated in the most efficient and cost effective way. Nomenclature GRT1 = well Groet 1 BOE = barrel of oil equivalent TVDss = true vertical depth sub sea SEM = Scanning Electron Microscopy EDX = Electron Diffraction X Ray Analysis XRD = X-ray diffraction BWPD = barrels of water per day IPR = inflow performance relationship Conversion Factors 1 nm3 = 0.00642 BOE References 1 Kleinitz, W., Koeler, M., and Dietzsch, G., The Precipitation of Salt in Gas Producing Wells, paper 68953 presented at the SPE European Formation Damage Conference, The Hage, The Netherlands, 21-22 May 2001. 2. Jasinski, R.; Sablerolle, W.; Computer modelling HTHP salt and scale formation; Expl & Prod Newsletter, July 1996 3. Jasinski, R.; Frigo, D.; The Modelling and Prediction of Halite Scale, Paper presented at the IBC Advances in Solving Oilfield Scaling Conference, Aberdeen, Scotland, Jan 1996 4. Couture, F.; Jomaa, W.; and Puiggali, J.-R.; Relative permeability relations: A key factor for a drying model; Transport. in Porous Media, 23;1996 5. Cooke, E. G.; The effect of additives on the crystal form of sodium chloride 4th. Intl. Salzsymposium, 259 267, 1974 6. Frigo, D. M., Jackson, L. A., Doran, S. M., and Trompert, R. A., Chemical Inhibition of Halite Scaling in Topsides Systems, paper 60191 presented at second SPE International Symposium on Oilfield Scaling, Aberdeen, Scotland, 26-27 Jan, 2000 7. Kirk, J. W., and Dobbs, J. B., A Protocol to Inhibit the Formation of Natrium Chloride Salt Blocks, paper 74662 presented at the SPE Oilfield Scale Symposium, Aberdeen, Scotland, 30-31 Jan, 2002

www.petroman.ir

SPE 87472

TABLE 1 SCALE PREDICTION MODEL PARAMETERS Jan-03 Reservoir Pressure [bara] Reservoir Temperature [C] Bottom hole Flowing Pressure [bara] Bottom hole Flowing Temperature [C] Wellhead Flowing Pressure [bara] Wellhead Flowing Temperature [C] Gas Rate [nm3/d] 36.8 87 26 86 11.6 54 200,000 May-99 47.6 87 38 86 24.8 56 242,000 Sep-94 65.6 87 59.2 86 44.9 44 280,570 Sep-91 71.1 87 64.1 86 48 52 309,400 Sep-89 84.9 87 75.4 86 48.3 54.4 500,100

TABLE 2: GRT1 WATER COMPOSITION

Wellhead Sample Concentration, mg/L Equilibrated Water Chemistry 32,572 121,930 622 502 3,657 2,989 945 762 59 48 <3 1 129 104 16 3 1 61,954 195,900 161 130 0 99 262 46 37 36 29 34 27 3 2 100,333 322,727

Species sodium potassium calcium magnesium ferrous iron barium strontium zinc hydrogen chloride sulfate bisulfide bicarbonate bromide nitrate acetate propionate Mass of Solute (TDS)

Ratio 3.74 0.81 0.82 0.81 0.81 0.81 0.20 3.16 0.81 2.64 0.81 0.81 0.80 0.80 3.22

TABLE 3: SUMMARY OF GRT1 SCALE CONTROL OPTIONS / COST / RISK / BENEFIT Damaged Area Controlled Area Cost Production Wellwork cost Scale Remedial / Control Option Risk TBG PRF NW TBG PRF NW [kEuro] [MBOE/yr] [Euro/BOE] Calcite Acid squeeze 180 high 47 3.83 Halite Gauge cutter 5 low 0 Halite Fresh water soak 1 low 0 Halite Chemical inhibitor squeeze 180 high 47 3.83 Halite Chemical inhibitor downhole injection 500 high 0 Halite Chemical inhibitor controlled release downhole 50 low 0 Halite Stimulation 300 high 235 1.28 Halite Tubing change out 500 high 235 2.13
TBG = Tubing PRF = Perforations NW = Near Wellbore Area controled by the technique Area possibly controled by the technique Area not controled by the technique

www.petroman.ir