Reduccion Del Consumo de Agua en Fracturamiento Hidraulico

8/13/2019 Reduccion Del Consumo de Agua en Fracturamiento Hidraulico

http://slidepdf.com/reader/full/reduccion-del-consumo-de-agua-en-fracturamiento-hidraulico 1/25

SPE 151819

Water Conservation: Reducing Freshwater Consumption by UsingProduced Water for Base Fluid in Hydraulic Fracturing –Case Histories inArgentina

Juan Bonapace and Mariano Giglio, SPE, Halliburton; Juan Moggia and Angeles Krenz, Pan American Energy

Copyright 2012, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Latin American and Caribbean Petroleum Engineering Conference held in Mexico City, Mexico, 16 –18 April 2012.

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 have not beenreviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect any position of the Society of Petroleum Engineers, itsofficers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the Society of Petroleum Engineers is prohibited. Permission toreproduce 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 SPE copyright.

AbstractIn the Golfo San Jorge basin (GSJB), in the area of Cerro Dragón, a sustained increase of activity has developed over the past

ten years. Further development of the field has been achieved with well stimulations (hydraulic-fracturing treatments),

increasing production rates.

This has resulted in increased water consumption for these operations. In an effort to ration water sources, well owners

must find alternatives to reduce the volume of fresh water consumed.

Throughout the past five years, one operator has worked to use produced water, or low-quality injection water, as a source

of water for stimulation applications. The methodology applied consisted of

Identifying different types of water produced in the field

Performing tests on the waters (physical-chemical and bacteriological)

Performing tests with commonly used fracturing fluids to identify good performance

Conducting tests during field operationsEstablishing procedures for water injection from the results obtained

Continuous monitoring, as well as the introduction of new techniques and improvements

The results that were obtained demonstrate how the operations performed have gradually increased by using this

methodology (412 of 760 fracturing treatments during 2010). In addition, the reduction in freshwater use was 54.8% of the

total amount in 2010 (5.7 of 10.5 million gal).

This work has helped identify sources of produced water in the field that are acceptable for use in fracturing stimulation

operations. A procedure has been established to help ensure a reliable fracturing fluid with good performance is always used

in these operations. Additionally, new techniques and equipment have been incorporated that contribute to continuous

improvements to the process.

IntroductionA large area called Patagonia exists in the southern part of Argentina, within which are three productive hydrocarbon basins

(Neuquina, Golfo San Jorge, and Austral). The water sources for human use and consumption in Patagonia are rivers and

lakes, which are fed by waters traveling from the mountain-chain melting sector; in the same way, this water supply is

obtained from phreatic levels. The first hydrocarbon finding in Argentina was that of the GSJB, discovered during the search

for water for human consumption.

The city of Comodoro Rivadavia was officially founded in 1901; this settlement was located 160 km from the nearest

lake, Lake Muster. The water supply for the early population was derived from natural water springs, and the search for water

(phreatic levels) started in 1903. Oil was found by chance at a depth of 540 m in December of 1907.

Currently, freshwater supply for the city is obtained from Lake Muster; this water is transferred through an aqueduct

using three pumping platforms (Cerro Negro, Valle Hermoso, and Cerro Dragón). This system supplies water to the cities of

Comodoro Rivadavia and Rada Tilly in the province of Chubut, 170 km away, and Caleta Olivia in the province of Santa

Cruz, about 220 km away.



2 SPE 151819

According to the last national census conducted in 2010, Comodoro Rivadavia and Rada Tilly have a population of

176,000 inhabitants in the province of Chubut; Caleta Olivia is inhabited by 104,000 people in the province of Santa Cruz.

Considering that 0.28 m3 is the average daily water consumption per person, the following consumption averages have been

obtained: 78,512 m3 per day or 235,536 m3 per month.

In the last few years, the well operator has increased its oil production from 7,909 m 3 per day (January 2000) to 15,565 m3

per day (January 2010), attributed to strong activity in the Cerro Dragón area (Fig. 1). This led to increased rigs, well

completions, and well workover operations. There was a sustained increase in fracturing activity associated with these

completions and interventions. Two fracturing campaigns were performed in 2000 and six in 2010 (Figs. 2 and 3).The freshwater supply for operations in the Cerro Dragón area comes from the aqueduct that provides water for the city of

Comodoro Rivadavia. Because of this, fresh water used in fracturing represents greater than 90% of the total amount of fresh

water used in completion and workover operations. Approximately 20% of the total amount of fresh water from the aqueduct

is consumed.

Because the chemicals used in fracturing operations are sensitive to certain components present in water used for

fracturing-fluid preparation, and the need for a certain degree of fracturing-gel stability, historically speaking, fresh water has

been used to help ensure the success of the operation. However, careful tests were performed in the laboratory to adjust

fracturing-fluid formulations. In an effort to help preserve natural freshwater resources, the operator decided to replace fresh

water with field-water sources from its water-flooding system.

Fig. 1—Increased oil production.



SPE 151819 3

Fig. 2—Increased rigs.

Fig. 3—Increased fracturing sets.

Reservoir GeologyThe GSJB is located in the provinces of Chubut and Santa Cruz in southern Argentina. It is irregular in shape and has a

longer east-west-oriented elongation, with its continental surface covering 65% and the remaining 35% settling down on the

continental platform. The basin genesis is of the extensional type and of the Mesozoic era. Its main filling occurred during the

rifting stages of the late Jurassic period, early Cretaceous, and is mainly of lacustrian and fluvial origin.

Source rocks were from the D-129 formation and Matasiete, with a large variety of light and heavy oils (15 to 35 API),

middle and heavy oils being the most common. The accumulated oils in the Mina el Carmen formation are lighter than thoseof the Comodoro Rivadavia formation. The traps are mostly of a combined nature (sedimentary and structural), and the

structural style is mainly found in the east area of the extensional type and in the west area of the compressional type (San

Bernardo fault through belt).

The main productive levels are the Mina el Carmen (continental, lacustrian, and fluvial environments), Comodoro

Rivadavia, and El Trébol (continental environment of the fluvial and deltaic kind) formations. Sand bodies show a 2- to 10-m

thickness, with variable porosity from 17 to 27%, diminishing in depth and with fluctuating permeability values, maintaining

50 md as an average value.

Geographical DescriptionThe Cerro Dragón field is located 85 km from the city of Comodoro Rivadavia (Fig. 4) on the west side of the GSJB in the

province of Chubut, the Argentine Republic. The area of operation consists of approximately 50 different fields, covering an

operative area close to 3,500 km2, which contains 3,800 active wells with an average depth of 2,200 m. About 50% of the



4 SPE 151819

oilfield production derives from 60 secondary recovery projects, with a total number of 650 injector wells. The Cerro Dragón

field has been under development and exploitation since 1959.

Fig. 4—Area of operation.

Fracturing FluidFracturing fluid must have the following characteristics: security, easy preparation, low fluid loss, high proppant transport

capacity (viscosity), low pipe friction, and easy removal from the reservoir. In general, water-based fluids, hydrocarbons, or

foams (nitrogen or carbon dioxide) can be used. Within these parameters, these fluids can generally achieve very good

viscosity, generally achieved by means of a gelling agent. . For water based fluids, guar-gum derivatives, such as

hydroxipropil guar (HPG) and carboximetil-hydroxipropil guar (CMHPG), are generally used. These tend to produce less

residue and have faster hydration; but, they are more sensitive to salt content and solids in the source water. Crosslinkers

generate a significant viscosity increase in the guar-polymer gels, using borate (B), as well as zirconium (Zr), titanium (Ti),

and, in smaller quantities, antimony (Sb) and aluminium (Al) as the crosslinkers.

Crosslinked fluids have a greater capacity to transport fracturing proppants than uncrosslinked fluids. This transporting

capacity is measured using a rheometer in which fluid is tested at specific temperature conditions and shear rates. In these

tests, it is possible to observe the transporting elastic regime zone and proppant settling; a minimum transport-capacity value

of 400 cp at 40 1/sec can be determined (Harris et al. 2005). This value was taken as a reference for fracturing-fluid tests

developed using the rheometer.

During the last decade, studies have concentrated on fluid-system development of gel polymers, and friction reducers that

can be used for fracturing treatments using produced water. Some of these cases have been documented by the following

authors:

Parket et al. (2003a, 2003b) and Hanes et al. (2003) present a fracturing fluid (high-performance fluid) that can be

reused after being recovered during the flowback stage.

Huang et al. (2005) documents the analysis performed with fluid (CMHPG) using produced water.

Walters et al. (2009) presents a new fracturing fluid (new biopolymer fracturing fluid) that can be used with

produced water.

Applications

Applications and Experience: 2005. Because of increased and sustained activity during the years 2000 to 2005, a need

developed for alternative sources to fresh water in the Cerro Dragón area. The first step consisted of identifying different

sources (water), water-loading platforms, fracturing-fluid analyses, and evaluations using these water sources. The tests

performed on water as well as on fracturing fluids are presented in Tables 1 and 2. It is important to point out that, during the

first stage, tests on bacteria in water were not performed because of the time required to develop these particular tests.

Initially, two temperatures of 190 and 160°F (65% of the total number of operations) were selected. This was simplified

by working with one fracturing fluid — low-guar borate medium temperature (LGB-MT) at two formulations: 30-ppt guar at

190°F and 27-ppt guar at 160°F.



SPE 151819 5

TABLE 1—WATER, PHYSICAL-CHEMICAL TESTS

Produced-Water Tests

pH Determination

Temperature determination

Calcium

Sulphites

Phosphates

Alkanity

Magnesium

Iron

Sulphates

Chlorides

Hardness

Test 1. In May 2005, two geographical areas were evaluated with water sources in Valle Hermoso and Cerro Dragón.

Valle Hermoso — Four sampling sources were identified in this area:

Source A — Water-producer well (Sample 1).

Source B — Valle Hermoso plant, without treatment (Sample 2).

Source C — Valle Hermoso plant, treated (Sample 3).

Source D — Valle Hermoso loading platform (Sample 4).

Cerro Dragón — Four sampling sources were identified in this area:

Source E — Water-producer well (Sample 5).

Source F — Cerro Dragón plant, treated (Sample 6).

Source G — ICD-13 loading platform (Sample 7).

Source H — Cañadón Grande loading platform (Sample 8).

Source A — The sample was taken from the producer well; it showed organic-material remains (moss), fetid smell, and

high temperatures. Routine tests were conducted on water with normal values for that kind of water (no bacteria

determination was performed). Fracturing-fluid tests (hydration test) showed base-gel viscosity loss, 7 cp at 511 at 1/sec at

120 minutes, denoting a typical behavior of bactericide presence.

Source B — The sample was taken at the fluid entrance of the treatment plant. This sample resulted in high hydrocarbon

content and was discarded for analysis, as well as for fracturing-fluid application.

Source C —

The sample was taken at the exit of the Valle Hermoso treatment plant. The sample showed crystallineaspects. The routine water-based tests performed remained within normal values (chlorides > 3,000 ppm). Fracturing-fluid

tests showed normal values with respect to viscosity and hydration time. Two fracturing-fluid stability tests were performed

under dynamic conditions in a rheometer (Fig. 5, Test 1 and Fig. 6, Test 6):

190°F at 120 min = viscosity 858 cp at 40 1/sec.


TABLE 2—FRACTURING FLUID TEST

Hydration gel(viscosity, cp at 511 1/sec)

Vortex time

pH Final

Lip test

Gel stability(static and dynamic,rheometer )

Transportation time capacity(viscosity = 400 cp at 40 1/sec )



6 SPE 151819

Fig. 5—Rheometer measurements using different water sources (2005).

Source D — The sample was taken at the Valle Hermoso loading platform (treated water). The water showed oil traces and

impurity, which were determined to be potentially useable. Fracturing-fluid tests showed normal values with respect to

viscosity and hydration time. A fracturing-fluid stability test was performed under dynamic conditions in a rheometer (Fig. 5,

Test 2):


Source E — The sample was taken at the producer well. It showed high hydrocarbon content; as a consequence, it was

discarded.

Source F — The sample was taken at the exit of the Cerro Dragón treatment plant. It showed crystalline aspects, and water

tests showed normal values. Fracturing-fluid tests showed normal values for viscosity and hydration time. Two fracturing-

fluid stability tests were conducted under dynamic conditions in a rheometer (Fig. 5, Test 3 and Fig. 6, Test 7):





SPE 151819 7

Fig. 6— Rheometer measurements using different water sources (2005).

Source G — The sample was taken at the ICD-13 loading platform. It showed crystalline aspects; water tests showed

normal values. Fracturing-fluid tests were normal with respect to viscosity and hydration time. Two stability tests were

conducted in the rheometer (Fig. 5, Test 4 and Fig. 6, Test 8).



Source H — The sample was taken at the Cañadón Grande loading platform. Its aspect was determined to be dirtier, with

strong fetid odor and hydrocarbon traces. Routine water-based tests performed remained within normal values (chlorides >

3,000 ppm) with respect to viscosity and hydration time. A fracturing-fluid stability test was conducted under dynamic

conditions in a rheometer (Fig. 5, Test 5):


Table 3 presents results obtained from fracturing-fluid tests on hydration, base-gel viscosity, and vortex time.

TABLE 3—FRACTURING-FLUID TESTS

DataTest 130 ppt

Test 230 ppt

Test 330 ppt

Test 430 ppt

Test 530 ppt

Test 627 ppt

Test 727 ppt

Test 827 ppt

Water pH 7.48 7.77 7.35 7.88 7.98 7.48 7.48 7.94

Water temperature, °F 66.7 63.1 65.8 68.4 80.4 69.1 66.6 67.6Gel viscosity, cp at 511 at 1/sec 25 24.5 24 24 23.5 22 20 21

Gel pH 7.22 7.46 7.22 7.91 7.65 7.58 7.12 7.65

Vortex time, sec 9 20 9 12 11 22 18 17

XI gel pH 9.27 9.26 9.28 9.18 9.21 9.08 9.15 9.25

The total testing series concluded that the samples from Sources F, G, C, and H showed the best results for both

temperatures (190 and 160°F).

I niti al Test 1. In June 2005, there was an opportunity to perform well stimulations on Well PCG-870 using produced

water. Based on the recommendations obtained from tests, it was decided that water from Source G should be used. When an

attempt was made to obtain a sample, the charge pump was not on; therefore, the sample could not be taken . It was



8 SPE 151819

determined that there was a stoppage at the plant, which impacted the pumping stations. A sample from Source H was then

suggested, but the same problem occurred. Later, a sample from Source H was obtained.

Routine tests were conducted and revealed a dirty aspect with dark color, particles in suspension, and high values of

chlorides, as in previous samples. Normal hydration values were obtained with respect to gel tests; the results in the

rheometer showed a fast viscosity decrease, demonstrating that the water from Source H was not acceptable for the treatment

to be performed.

The possible causes of Source H water not being acceptable for use with a fracturing fluid are that the water remained

stagnant in the pump system for several days, or the stoppage of the plant could have influenced normal treatments performedon the water.

I niti al Test 2. In July 2005, there was anotheropportunity to perform a test on Well PCG-947. Source H was chosen for

this treatment and water and fracturing-fluid tests were conducted, obtaining normal values in the rheometer (Fig. 7, Test 10).

Finally, breaker tests were developed, evidencing the need to reformulate the breaker concentrations to achieve a viscosity

profile acceptable for the operation (Fig. 7, Test 11).

In addition, a sample of Source G was taken as an alternative. It was tested and showed normal parameters with respect to

water and gel during rheometer tests (Fig. 7, Test 12).

Fig. 7—Initial Test 2, Source H sample in 2005.

Table 4 presents results obtained from fracturing-fluid tests on hydration, pH, viscosity, and vortex-time tests.


DataTest 9 Tap Water

30 pptTest 1030 ppt

Test 1130 ppt + Breaker

Test 1230 ppt

Water pH 7.78 7.87 8.08 7.96

Water temperature, °F 62.4 61.5 62.2 63.5

Gel viscosity, cp at 511 1/sec 26 26.5 26 25.5

Gel pH 6.95 7.04 7.15 7.06

Vortex time, sec 17 17 16 18

crosslinked gel pH 9.52 9.51 9.5 9.43



SPE 151819 9

Because of a redesign of the well intervention plan (delayed), the treatment has not yet been performed; however, initial

tests were still performed for use on a future well.

I niti al Test 3. In early August 2005, an opportunity arose to use produced water on Well PVH-923. Considering the

geographical location of the well, Sources D and H were selected as possible water sources to be tested. Prejob tests were

performed with the two samples; water parameters and gel-hydration parameters tested normal. A better water quality was

suspected for Source H. As a result, the Source H fluid sample was tested in the rheometer (Fig. 8, Test 13) and Sample D

was discarded.At the time of collecting the Source H water supply, a problem in the supply system was detected, making it impossible to

load and transport water from Source H to the well. Given the time proximity to perform the treatment, it was decided to

replace the water with Source G, which had been tested as an alternative sample in the previous well (PCG-947). The water

was loaded into the fracturing fluid tank while routine tests were simultaneously conducted. Once good fluid-stability results

were obtained, the break was reformulated and tested (Fig. 8, Tests 14 and 15).

Fig. 8—Initial Test 3, Source G sample in 2005.

Table 5 presents results obtained from fracturing-fluid tests on hydration, pH, viscosity, and vortex time.


DataTest 1330 ppt

Test 1430 ppt

Test 1530 ppt + Breaker

Water pH 7.93 8.32 8.03

Water temperature, °F 60.8 84.9 71.6

Gel viscosity, cp at 511 1/sec 28 24 26

Gel pH 6.9 7.03 7.18

Vortex time, sec 12 12 11

Crosslinked gel pH 9.3 9.33 9.36

Finally, the treatment was performed within normal parameters with no evidence seen of abnormal values that could

impact fluid performance; there were only pressure variations caused by a cavitation problem in one of the pumps (Fig. 9).



10 SPE 151819

Fig. 9—Fracturing treatment chart.

Soon after, the cleaning and reloading of fracturing fluid tanks using water from sample Source G was performed. A

sample was taken to the laboratory and tested at a lower temperature (165°F). The results of the water and gel-hydration tests

were normal (Table 6). The next step was stability tests performed in the rheometer.


DataTest 1627 ppt

Test 1730 ppt + Stabilizer

Test 1830 ppt New Guar

Water pH 8.01 7.89 7.93

Water temperature, °F 62.4 63.1 62.2

Gel viscosity, cp at 511 1/sec 20 24 25

Gel pH 7.14 7.19 7.26

Vortex time, sec 13 14 13

Crosslinked gel pH 9.35 9.37 9.41

Good results were not obtained (Fig. 10, Test 16); therefore, some modifications were implemented in the fluid to reach

the viscosity profile expected. Gel loading was increased, a temperature stabilizer was added (Fig. 10, Test 17), and another

type of gelling agent (guar gum of higher performance) was used (Fig. 10, Test 18), but the minimum acceptable values to

perform the treatment were not reached.



SPE 151819 11

Fig. 10—Additional stability tests in 2005.

Applications and Experience — 2006. In 2006, there were no tests on fracturing fluid, nor were initial field tests conducted.

However, all the information previously obtained was evaluated and a job methodology to use produced water in fracturing

fluids was developed. These items were considered:

A better water-quality source from the Cerro Dragón plant, from Sources F and G, was identified.

Other source samples showed great variability as well.

Freight and transportation systems must be clean (trucks).The main problems detected were related to bacteria generation, favorable temperature conditions, pH, and impurity

in water.

Methodology developed for produced water use with fracturing fluids:

Empty fracturing tank; do not leave gel remains.

Fill up the tank with 3,000 gal of water. Raise the pH with a sodium-hydroxide solution until a pH value equaling 10

or 11 is obtained.

Retrieve water from a clean fracturing fluid tank.

Empty the tank and clean it with water. Start collecting water and add bactericide.

Perform a prejob test.

Do not gel fracturing fluid until a plug positive result is obtained (well operational issue); minimize time with

prepared gelled fluid.

Perform prejob with gelled fluid in fracturing fluid tank.Start the operation.

Once the operation is finished, repeat.

Applications and Experience — 2007. In 2007, the goal was to perform 20 fracturing stages using this methodology. In

March, there was an unusual situation in the field regarding a freshwater supply stoppage. This meant implementing the

methodology previously stated; a total of 17 operations were then successfully conducted using this methodology.

Notwithstanding this, for intervening Well PZ-1047, the first two fracturing treatments were conducted at a higher

temperature (230°F) in which a type of fluid CMHPG-zirconate is normally used ( Fig. 11, Test 19). Because of this, the fluid

was tested with produced water from sample Source G; however, poor results were obtained (Fig. 11, Test 20). Guar-borate

for high temperatures was suggested, which is not as sensitive to high-chloride contents. A sample of water from Source C

was used for this test. Even though the fluid was formulated with a larger polymeric load and temperature stabilizer, the

result was negative (Fig. 11, Test 21).



12 SPE 151819

Fig. 11—Test for Well PZ-1047.

Because of this, it was decided to perform fracturing treatment operations at a lesser depth (lesser temperature), perform

fracturing- fluid tests to obtain good results, and wait to obtain a freshwater supply to be able to use high-temperature fluid

(CMHPG-Zr) in the deeper two operations.

In May and July of 2007, more operations were completed using produced water, with which a total number of 27

treatments were completed. This surpassed the goal proposed, amounting to 3% of the activity for the year. This amount

represents a consumption of 1,730 m

3

(457,000 gal) of produced water, achieving a reduction of 5% of the total amount offresh water used to perform 907 fracturing treatments (Fig. 12).

In August, a physical-chemical and bacteriological analysis of the water from Sample G was performed. The results are

presented in Tables 7 and 8.



SPE 151819 13

Fig. 12—Fracturing treatment stages performed from 1999 through 2010.

TABLE 7—WATER-PHYSICAL-CHEMICAL ANALYSIS

Physical-Chemical Analysis Sample, mg/L

pH 7.25

Carbonates 0

Bicarbonate 3.8

Chlorides 91

Sulphates 0.79

Calcium 4

Magnesium 1.97

Iron 0

Sodium 130.6

Potassium 0

TABLE 8—WATER-BACTERIOLOGIST ANALYSIS

Bacteriological Analysis Sample, ppm

R2A (aerobia oligotrophics) 2.90E+04

TSBC (aerobia Heterotrophics) 1.29E+05

BSR 1.00E+03

TG (bacteria anaerobics) 1.00E+01

The chemical composition of the Source G sample was similar to many others from the GSJB. This composition presents

a high content of sodium chloride with a normal pH level and the bacterial amount not being large. From the bacteriological

analysis, it can be determined that the sample showed bacteria, but in a smaller proportion than what was expected to

generate gel instability. The gels prepared with this water sample did not show meaningful modifications out of the four

bacterial groups analyzed that could develop from this kind of water.

Applications and Experience — 2008. The expectation for 2008 was to conduct 50 frac stages using this methodology.

During that year, it was decided to continue using water from Source G.

The activity that developed (Table 9) that year consisted of 72treatments, surpassing the goal proposed and reaching 8%

of the entire activity developed. Because of the geographical location of the source, the activity was concentrated in nearby

fields because of supply concerns.



14 SPE 151819

TABLE 9—2008 ACTIVITY

Fields Cañadón Grande Cerro Dragón La Madreselva Oriental Valle Hermoso Zorro

Wells 15 7 2 1 2 1

Frac stages 36 22 6 1 6 1

Avg depth, m 2090 1750 2230 1560 1540 2480

Avg temp, °F 187 163 195 231 156 208

Frac fluid LGB-MT LGB-MT LGB-MT LGB-MT LGB-MT & LT LGB-MT

Fluid volume, m3 2581 1588 347 99 298 51

The most activity during 2008 was developed in Cañadón Grande and Cerro Dragón, where 81% of the operations were

conducted with 84% of the total fluid used (Fig. 13).

Fig. 13—Main activity of Cañadón Grande and Cerro Dragón,

Finally, a reduction of 13% of the fresh water used for the performance of the 891 fracturing treatments was reached that

year. This represents 4,964 m3 of produced water consumption (1.3 millon gal) (Fig. 12).

Applications and Experience —

2009. During the beginning of 2009, the objective was to perform 90 frac stages using thismethodology; at the same time, some improvements were implemented.

Started using a preblender unit in real-time.

Searches were conducted for new water sources and loading platforms.

Preblender Uni t. A preblender unit is designed to supply additives and liquid or powdered gel in quantities accurately

measured in a blender unit. The preblender unit allows the development of up to 100 bbl/min operations (Yeager et al. 1998;

Weinstein et al. 2009).

The preblender unit used for operations has a storage tank with room for 50 liquid gel concentrate (LGC) barrels, plus a

hydration tank for 75 job volume barrels. The LGC is composed of a base fluid (diesel) and a dry power gelling agent so that

the latter can be kept in suspension; LGC contained 4 lbm of a high-viscosity, rapid-yielding, and intermediate-residue guar

gum per gal of concentrate (Fig. 14).

ANTIC LINALFUNES

ANTICLINAL

GRANDE SUR

HUETEL

BAYO

CHULENGO

MESETACATORCE

ESCORIAL

ORIENTALOESTE ORIENTAL

LA MADRESELVASUR

LA MADRESELVAVALLEMARTIN

TRES PICOS

ZORRO

RESERO

CAÑADON

GRANDE

CERRO D RAGON

CERROTORTUGA

CAÑADONPEDRO

ELGATOVALLE HERMOSO

VALLE HERMOSONORTE

LAS FLORES

RIO CHICO

ELTRIANGULO

PADRE CORTIEL CONDOR

PAMPA

PAMPASUR

SAN AGUSTINLA PIEDRA

EL KIKEN

MARIANA

SourceWater



SPE 151819 15

Fig. 14—

Preblender unit at operation.

This unit was limited to minimum fluid and proppant volume operations averaging 55 m3 water produced and 210,000

lbm of proppant. The goals were to evaluate the operation’s performance, as well as reduce operation time, which were

obtained because of the application of the blender methodology for fracturing fluid when using produced water (Fig. 15).

Fig. 15—Typical fracturing treatment chart for small treatment.



16 SPE 151819

This unit removed certain steps in the methodology that would have generated additional time. On average, two operative

hours were saved from each of these treatments.

The methodology for produced water use with the preblender unit was as follows:

Filled up water in fracturing fluid tank and add bactericide.

Performed prejob test with LGC.

Waited for plug positive result (well operational issue).

Performed operation.

Once operation was over, repeat.

This unit implementation was used on 5% of the total number of operations performed using produced water.

New Water Sources. The main objective was to identify other sources as well as evaluate a larger area within the Cerro

Dragón oil field.

For Stage 4, new sources were evaluated as follows:

Source I – Escorial area (Sample 1).

Source J – Oriental area (Sample 2).

Source K – Zorro area (Sample 3).

Source L – Anticlinal Grande area (Sample 4).

Water samples showed a crystalline aspect, in general, without color or a particular odor. Routine tests (physical-

chemical) performed were within normal values for field waters (Table 10), and tests for bacteria determination were notconducted.

TABLE 10—PHYSICAL-CHEMICAL TEST

AnalysisSource IEscorial

Source JOriental

Source KZorro

Source L Anticlinal Grande

pH 6.8 7.4 7.2 7.1

SS in 10 min, mg/L 0 0 0 0

SS in 2 min, mg/L 0 0 0 0

TSS, mg/L 10 0.1 0.1 200

Dry residue 105°C, mg/L 9100 6200 4900 6600

Carbonate, mg/L 1 1 1 1

Bicarbonate, mg/L 80 121 134 201

Chloride, mg/L 8940 5502 4126 5158

Sulfates, mg/L 22 24 25 10

Nitrite, mg/L 80 40 30 60

Nitrate, mg/L 0.5 0.1 0.1 0.1

Amony, mg/L 43 9 0.01 5

Calcium, mg/L 1082 601 400 721

Magnesium, mg/L 1 1 1 1

Iron, mg/L 2 1 1 1

Fracturing-fluid tests (hydration, viscosity, vortex time, and pH) showed normal results, without out-of-range values.

Finally, tests were performed in the rheometer to observe the viscosity-profile behavior. The tests were standardized at 180°F

(LGB-MT) for samples of Sources I, J, and K (Fig. 16, Tests 22, 23, and 24); for Sample L, the tests were conducted at

140°F for a LGB-LT fluid (Fig. 16, Test 25).



SPE 151819 17

Fig. 16—Tests performed in 2009 for Samples I through L.

The results showed the sources from the Zorro and Oriental fields were acceptable for use; on the contrary, sources from

the Escorial and Anticlinal Campamento fields were not.

Activity. The activity in 2009 totaled 191 treatments, surpassing the amount proposed. This comprised 24% of the entire

activity for the year. The use of this methodology was implemented on nine fields in the geographical proximity of the water

sources. Initially, the methodology was implemented in nearby fields, such as the Cañadón Grande and Cerro Dragón. Later,its use was extended to La Madreselva, Resero, and Valle Hermoso. Detailed information is presented in Table 11.

TABLE 11—2009 ACTIVITY

FieldsCañadónGrande

CerroDragón Huron

LaMadreselva

LasFlores Resero

RioChico

V.Hermoso Zorro

Wells 14 9 1 8 5 13 6 15 4

Frac stages 24 21 2 22 13 27 20 49 13

Avg depth, m 1998 1750 650 1935 1440 2300 1180 1630 2250

Avg temp, °F 181 167 106 178 150 198 136 160 195

Frac fluid LGB-MT& LT

LGB-MT LGB-MT

LGB-LT LGB-MT &LT

LGB-MT

LGB-MT& LT

LGB-MT& LT

LGB-MT

Fluid volume,m

3

1576 1441 125 1698 1037 1703 1156 3815 625

The greatest amount of activity was concentrated within these five fields, where 75% of the operations, using 78% of the

total fluid, were carried out (Fig. 17).



18 SPE 151819

Fig. 17—Five fields where operations were performed 2009.

Finally, a reduction of 35% of the total amount of fresh water used was reached for 840 fracturing treatments during this

year. This meant a total consumption of 13,176 m3 (3.5 million gal) of produced water (Fig. 12).

Applications and Experience – 2010. During the beginning of 2010, the objective was to perform 50% of the total number of

treatments using this methodology. The same course of action was desired for 2009:

Increase the use of preblender unit

Continue searching for new water sources and loading platforms.

Preblender Uni t. Because of the good results obtained, this unit continued to be used. Its application was increased and

its use was also adopted for larger-volume fluid and proppant operations (average frac stage, 153 m3 water produced and

120,500 lbm of proppant) (Fig. 18).

ANTICLINALFUNES

ANTICLINALGRANDE SUR

HUETEL

BAYO

CHULENGO

MESETACATORCE

ESCORIAL

ORIENTALOESTE ORIENTAL

LA MADRESELVA

SUR

LA MADRESELVAVALLEMARTIN

TRES PICOS

ZORRO

RESERO

CAÑADON

GRANDE

CERRO DRAGON

CERROTORTUGA

CAÑADONPEDRO

ELGATOVALLE HERMOSO

VALLE HERMOSONORTE

LAS FLORES

RIO CHICO

ELTRIANGULO


PAMPA

PAMPASUR

SAN AGUSTINLA PIEDRA

EL KIKEN

MARIANA

Source Water



SPE 151819 19

Fig. 18—Fracturing treatment chart when using produced water in 2010.

This unit was used in 31 operations in 2010, which comprised 8% of the entire activity. On the other hand, the total

amount of fluid used in these operations represents 8% of the produced water used in the period analyzed.

New Water Sources. Continuing with this alternative, there was a search for water sources in nearby areas. A search of

the La Madreselva area led to the discovery of four new sources:

Source M – Producer well (Sample 1).

Source N – Transfer manifold (Sample 2).

Source O – Transfer pump (Sample 3).

Source P – CD-LM transfer (Sample 4).

Water samples showed good general results, without color or particular odors; routine tests (physical-chemical)

performed were within the normal values for field waters (Table 12). Bacteria determination tests were not conducted.



20 SPE 151819

TABLE 12—PHYSICAL-CHEMICAL WATER TEST

AnalysisSource M

(S1)Source N

(S2)Source O

(S3)Source P

(S4)

pH 7.8 7.4 7.6 7.4

SS in 10 min, mg/L 0 0 0 0

SS in 2 min, mg/L 0 0.2 0 0

TSS, mg/L 0.1 0.1 < 0.1 < 0.1

Dry residue 105°C, mg/L 3400 3600 3000 3400Carbonate, mg/L 1 1 < 1 < 1

Bicarbonate, mg/L 295 309 188 456

Chloride, mg/L 3095 2751 1685 1857

Sulfite, mg/L 37 43 61 14

Nitrite, mg/L 50 90 < 10 < 10

Nitrate, mg/L 0.1 0.1 0.1 0.7

Amony, mg/L 9 2 3 13

Calcium, mg/L 200 160 160 160

Magnesium, mg/L 1 1 < 12 < 1

Iron, mg/L 1 1 2 1

Tests on frac fluid (hydration, viscosity, vortex time, and pH) showed normal values, denoting no out-of-range values.

Finally, rheometer tests were performed to determine the viscosity profile behavior. The tests were conducted at 180°F

(LGB-MT) for the four samples (Fig. 19, Tests 26, 27, 28, and 29).

Only the fluid formulated with the Source O sample met the requirements to be used (Fig. 19, Test 28).

Fig. 19—Viscosity profile-behavior tests.

Practice of Using Produced Water Interrupted. During May 2010, a screenout increase of 9% (based on a monthly

average) was recorded. A quick review of the fluid compositions used on the treatments that had ascreenout revealed poor

fluid properties with the produced water used. A considerable loss of viscosity was observed on tests performed in the

rheometer. Use of produced water was immediately stopped and only fresh water was used.

In Fig. 20, three viscosity profiles can be observed from the fluid prepared with fresh water (Test 30: tap water) and the

same fluid prepared with produced water from Source G (Test 31), with a difference in viscosity of > 1,000 cp between both



SPE 151819 21

profiles being highly noticeable. Additionally, the fluid viscosity profile formulated with a higher polymer load (Test 32) can

be observed, with an improvement between 500 to 600 cp.

Fig. 20—Viscosity profile-behavior tests.

After approximately a 20-day produced-water-use interruption, activity resumed using a 35-ppt polymer load fluid,

mainly for wells with temperatures higher than 180°F, as well as in field sector sources LM-LMS (Source O). Additionally, it

was advisable to continue most preblender unit jobs.

Activity. The activity at that time consisted of 412 treatments, which represents 54% of the total number of treatments. This

practice was spread out over 18 fields according to geographical proximity to loading platforms.

The largest development was located at Cañadon Grande, Cerro Dragón, La Madreselva, La Madreselva Sur, Oriental,

Resero, Rio Chico, Valle Hermoso, and Zorro, where 90% of the operations were performed using 92% of the total amount

of water used (Fig. 21). Table 13 presents more information with respect to activity in 2010.

TABLE 13—ACTIVITY 2010

FieldsCañadónGrande

CerroDragón

LaMadreselva

LaMadreselva

Sur Oriental ReseroRio

ChicoV.

Hermoso Zorro Other

Wells 13 14 8 15 4 26 4 31 18 13

Frac

stages

23 25 25 50 21 76 14 95 52 31

Avg depth,m

1842 1884 2044 2058 1905 2065 890 1586 2025 1468

Avg temp,°F

170 175 178 193 176 190 120 157 183 148

Frac fluid LGB-MT LGB-MT LGB-MT LGB-MT LGB-MT

LGB-MT

LBG-LT

LGB-MT &LT

LGB-MT

LGB-MT& LT

Fluidvolume,m

3

967 1086 1020 2646 698 3742 562 6480 1738 928

Fresh-water reduction in consumption for 2010 (412 fracturing treatments) was 55%; this means 21,805 m 3 produced

water was used (5.76 million gal) (Fig. 12).



22 SPE 151819

Fig. 21—Areas of produced water 2010.

Future ActionsIn the last few years, the operator and service company have worked mainly on identification and selection of water sources

and their use to meet certain requirements. This has allowed modification of fracturing-fluid formulations used in the field on

a daily basis. During that time, water sources that did not have a good performance with the fracturing fluid were not

selected.

To accomplish all of the operator’s objectives using these water sources in fracturing operations, it is necessary to propose

different alternatives to be evaluated and applied, some examples of which are listed in the next sections.

Fit Water Sources. A water-source control was conducted, which presented a salinity increase in the west sector of the area;

two sectors with useful viable sources were identified:

Cerro Dragón (CD) and Cañadón Grande (CG) fields.

La Madreselva (LM) and La Madreselva Sur (LMS) fields have a less quality degree, though still fit for use.

High-volume loading platforms are being built in both sectors (CD-CG and LM-LMS), which will start operating at the beginning of 2012 (Fig. 22). By means of these new volume loading capibilities, the operator will be able to comply with

100% of the supply of the operations that use low-salinity waters. This fact has required an investment of approximately USD

500,000.

ANTICLINAL

FUNES

ANTICLINAL

GRANDE SUR

HUETEL

BAYO

CHULENGO

MESETA

CATORCE

ESCORIAL

ORIENTAL

OESTE ORIENTAL

LA MADRESELVA

SUR

LA MADRESELVAVALLE

MARTIN

TRES PICOS

ZORRO

RESERO

CAÑADON

GRANDE

CERRO D RAGON

CERRO

TORTUGA

CAÑADON

PEDRO

EL

GATOVALLE HERMOSO

VALLE HERMOSO

NORTE

LAS FLORES

RIO CHICO

EL

TRIANGULO


PAMPA

PAMPA

SUR

SAN AGUSTIN

LA PIEDRA

EL KIKEN

MARIANA

JORGE

ANTICLINAL

GRANDE

SourceWater

SourceWater



SPE 151819 23

Fig. 22—Loading platforms.

Unfit Water Sources. The sources that currently do not comply with requirements have been evaluated for other uses. Other

kinds of treatments are under evaluation.

Current uses: low-quality water sources are being used in fluid completion, mainly perforation muds.

Alternatives:o Water blends — unfit water sources and good-quality fresh water diluted in different percentages are being

evaluated so that fracturing fluids more sensitive to salinity levels can be tested. This tends to be a frequent

practice in the industry.

o Water treatment — a water-treatment settling plant is being evaluated with ultrafiltration membranes that

allows removing salt from water and using it in highly demanding treatments (energy-generating plants). It

has been purported to use this treatment method on produced waters for critical operations. One of the main

limits is that produced waters tend to contain hydrocarbon traces, which is highly damaging for

membranes.

Freshwater Sources (Unfit for Human Consumption). These waters are fit for animal consumption, but not for humans.

This water source belongs to the Patagoniano aquifer, which contains excessively high-salinity levels for human

consumption. There is an upstreaming plan to obtain these waters for specific uses (more sensitive frac fluids, perforation

muds used on pipe guides).

Other Technologies. The recent boom in the US of nonconventional reservoir exploitation, in which hydraulic stimulation is

essential with the use of large water volumes, has generated a significant technology development associated to use,

handling, and reuse of this source, as well as fracturing fluids.

The following is a set of technologies that could be applied at different stages of the process; these could also be used

with different kinds of water qualities.

Ultraviolet (UV) light — this process uses UV light energy instead of chemicals to control bacteria. The fluids are

pumped through UV light sources where the UV light passes through the liquid media (water) and the bacteria’s cell

wall. Exposure to UV light disrupts the organism’s DNA , rendering it unable to reproduce. Water turbidity can have

a negative effect on disinfection results; in some cases, it is necessary to use a prefilter when there are high turbidity

values. This system is set on a mobile piece of equipment, which offers a working system of up to 100 bbl/min, the



24 SPE 151819

same has been used with different types of fluids, such as slickwater, water frac, and crossliker fluid (Gloe et al.

2009; Crane et al. 2011).

Electrocoagulation — this system uses an electrical process that has the capacity to destabilize and coagulate

suspended colloidal matter in water. When contaminated water passes through the electrocoagulation cells, the

anodic process releases positively charged ions that bind onto the negatively charged colloidal particles in water,

resulting in coagulation. This technology is used for treating well flowback water for reuse in subsequent operations.

Among its benefits, it reduces 99% of the total suspended solids (TSS), coagulates particles smaller than a micron,

reduces turbidity of at least 10NTU, removes divalent and heavy metals, and breaks emulsions. This system is set ona mobile piece of equipment that has a working system of up to 20 bbl/min (Rodvelt et al. 2011).

ConclusionsThe following conclusions are a result of this work.

Water Source. A large number of water sources within the field have been tested over a six year period. The following has

been determined.

Waters in general present a salinity increase in the west sector of the Cerro Dragón field.

Currently, some water sources have been tested in four geographical areas within the Cerro Dragón field (CD-CG;

VH; LM-LMS and Z-O-R).

Until now, two sectors, CD-CG fields and LM-LMS (lesser quality), have been localized, where said sources are

being used.

Sources have been identified in the Zorro and Oriental fields that could be considered to start an initial plan in thearea.

The best water source identified was Source G (loading platform ICD-13, Cerro Dragón).

These water sources over time have shown important variations in their properties; therefore, it is advisable to take

periodic samples and perform tests with fracturing fluids.

The main problems detected were associated with the presence of bacteria, identifying favorable temperature

conditions, pH, and impurities in water.

Bacteriological-test results revealed the presence of bacteria in smaller proportions than were thought to cause fluid

instability; however, viscosity-loss effects were observed.

Practice and Methodology. Over time, the following items were achieved in relation to the developed methodology.

It was determined that a water-charge and transportation system (trucks) must be well-identified and must not have

contaminating substances.

A methodology for the use of fracturing-fluid produced water was developed and used.The use of a preblending unit for this practice was implemented to complement and improve logistics, achieving a

two-hour time savings per treatment, on average.

This preblending unit was initially used in small-volume operations (40 m 3); however, in 2011, it was used on

operations with large volumes (150 m3).

Fracturing Fluids and Scope. The following items were achieved in relation to fracturing fluids and scope.

Four fracturing fluids were tested (LGB-MT, LGB-LT, LCMHPGZr-HT, and GB-HT), obtaining good results only

from LGB-MT fluid, and proportionally less with LGB-LT fluid.

From 2005 to present, water-use percentage has increased on a yearly basis, and an important reduction in fresh

water consumption has been achieved, reaching values of 55% and 21,805 m3 (5.76 million gal) (Fig. 12).

A larger area than anticipated was involved in the use of this practice within the Cerro Dragón field (initially two

fields, currently more than 18 fields) (Fig. 21).

It will be necessary to conduct tests with fracturing fluids (LGB-LT, LCMHPGZr-HT, and GB-HT) using otherviable alternatives, such as water-blend use (fresh-produced) and fresh water unfit for human consumption

(Patagoniano aquifer).



SPE 151819 25

Nomenclature

bbl/min = barrel per minute

CD = Cerro Dragón

CG = Cañadón Grande

DNA = desoxy ribonucleic acid

GB-HT = guar borate-high temperature

GSJB = Golfo San Jorge Basin LCMHPGZr-HT = low CMHPG zirconate-high temperature

LGB-LT = low guar borate-low temperature

LGB-MT = low guar borate-medium temperature

LGC = liquid gel concentrates

LM = La Madreselva

LMS = La Madreselva Sur

NTU = nephelometric turbidity units

O = Oriental

ppm = parts per million

R = Resero

SS = settleable solids

TSS = total suspended solid

UV = ultraviolet

VH = Valle Hermoso

Z = Zorro

AcknowledgementsThe authors thank PAE and Halliburton for permission to publish this paper and for all the support provided during its

development.

ReferencesCrane, B., Neal, G., and Warren, W. 2011. Improved Process Provides More Effective Ultraviolet Light Disinfection of Fracturing Fluids.

Paper SPE 142217 presented at the SPE Americas E&P Health, Safety, and Environment Conference, Houston, Texas, USA, 21 – 23March. doi: 10.2118/142217-MS.

Gloe, L., Neal. G., and Kleinwolterink, K.. 2009. Ultraviolet Light Disinfection of Fracturing Fluids. Paper SPE 126851 presented at theSPE International Health, Safety, and Environment Conference, Rio de Janeiro, Brazil, 12 – 14 April. doi: 10.2118/126851-MS.

Harnes, R., Parker, M., Slabaugh, B., Weaver, J., and Walters, H. 2003. Analytical Methods for Maintaining Quality Assurance ofRecycled Fracturing Fluids. Paper SPE 80221 presented at the SPE International Symposium on Oilfield Chemistry Houston, Texas,

USA, 5 – 8 February. doi: 10.2118/80221-MS.Harris, P. and Walters, H. 2000. Real Time Control of Low Polymer Fracturing Fluids. Paper SPE 63238 presented at the SPE Annual

Technical Meeting, Dallas, Texas, USA, 1 – 4 October. doi: 10.2118/63238-MS.Huang, F., Gundewar, F., Steed, D., and Loughridge, B. 2005. Feasibility of Using Produced Water for Crosslinked Gel-Based Hydraulic

Fracturing. Paper SPE 94320 presented at the SPE Production and Operations Symposium, Oklahoma City, Oklahoma, USA, 17 – 19April. doi: 10.2118/94320-MS.

Parker, M., Slabaugh, B., Walters, H., Weaver, J., and Hanes, R.M. 2003a. New Hydraulic Fracturing Fluid Technology Reduces ImpactOn The Environment Trough Recovery And Reuse. Paper SPE 80590 presented at the SPE/EPA/DOE Exploration and ProductionEnvironmental Conference, San Antonio, Texas, USA, 10 – 12 March. doi: 10.2118/80590-MS.

Parker, M., Slabaugh, B., Walters, H., Hart, T., Walsh, H., Haley, W., Harper, J., and Weaver, J. 2003b. New Hydraulic Fracturing-FluidTechnology Increases Production in the Barnett Shale and Reduces Impact on the Environment. Paper SPE 80912 presented at theSPE Production and Operations Symposium, Oklahoma City, Oklahoma, USA, 23 – 25 March. doi: 10.2118/80912-MS.Rodvelt, G.,Yeager, V., and Hyatt, M. 2011. Case History: Challenges Using Ultraviolet Light to Control Bacteria in Marcellus Completions.

Paper SPE 149445 presented at the SPE Eastern Regional Meeting, Columbus, Ohio, USA, 17 – 19 April. doi: 10.2118/149445-MS.Walters, H., Stegent, N., and Harris, P. 2009. New Frac Fluid Provides Excellent Proppant Transport and High Conductivity. Paper SPE

119380 presented at the SPE Hydraulic Fracturing Technology Conference, The Woodlands, Texas, USA, 19 – 21 January. doi:10.2118/119380-MS.

Weinstein, J., Phillippi, M., and Walters, H. 2009. Dry-Polymer Blending Eliminates Need for Hydrocarbon Carrier Fluids. Paper SPE121002 presented at the SPE Americas E&P Environmental & Safety Conference, San Antonio, Texas, USA, 23 – 25 March. doi:10.2118/121002-MS.

Yeager, R. and Bailey, D. 1988. Diesel-Based Gel Concentrate Improves Rocky Mountain Region Fracture Treatments. Paper SPE 17535

presented at the SPE Rocky Mountain Regional Meeting, Casper, Wyoming, USA, 11 – 13 May. doi: 10.2118/17535-MS.

Documents

Reduccion Del Consumo de Agua en Fracturamiento Hidraulico