SPE-174118-MS

Water Management: What We Have Learned and What We Need to Consider for Developing a Shale Play in Argentina Juan Carlos Bonapace, Halliburton; Facundo Alric, Adrian Angeloni and Luciano Zangari, Total Austral

Copyright 2015, Society of Petroleum Engineers This paper was prepared for presentation at the SPE Latin American and Caribbean Health, Safety, Environment and Sustainability Conference held in Bogotá, Colombia, 7–8 July 2015. 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 been reviewed 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, its officers, 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 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 SPE copyright.

Abstract Hydraulic fracturing has been active in Argentina since the 1960s. The first jobs were performed using oil-based fluids.

Throughout the years, new water-based fluids were introduced to replace alcohol-based fluids and foams based on reservoir

requirements, economics, and safety and environmental issues. Currently, more than 95% of hydraulic fractures performed in

the country are made using aqueous-based fluids.

Recently, exploration and development of resource shale plays, such as the Vaca Muerta, have begun. To achieve

commercial production, this type of reservoir must be stimulated by hydraulic fracturing using large volumes of water. From

2009 to present, various exploration techniques have been performed in different shales, such as Los Molles, Vaca Muerta,

Agrio (Neuquén Basin), Cacheuta (Cuyo Basin), and the D-129 (Golfo San Jorge Basin).

This paper discusses aspects of water logistics necessary during the well completion phase, fracture treatment designs

applied in Vaca Muerta, and laboratory studies performed on flowback and produced waters to help evaluate the potential for

water reuse.

The focus is on three different phases of water cycles for these projects.

• Water sources and stimulation: information for vertical and horizontal wells based on physical-chemical

characteristics of various freshwater for stimulation, volume of water used, type of fracture treatment, and

fracture fluid and additives used.

• Logistics: evolution of different water storage and transport options used for shale projects on single or multiple

well pads.

• Reuse of flowback and produced water: laboratory tests on different flowback and produced water and/or blends

(freshwater-flowback-water), treated and untreated including:

‒ Physico-chemical characteristic of water (flowback and produced) from different wells.

‒ Formation sensitivity tests with different water sources and usage possibilities.

‒ Fracture fluids, conventional borate fluids, and a new low-residue CMHPG-metal formulated fluid using no

traditional water treated and untreated with high total dissolved solids (TDS).

Introduction Well stimulation using hydraulic fracturing has been widely used for producing oil and gas reservoirs in Argentina since the

1960s. This stimulation technique has been applied in the five hydrocarbon producing basins shown in Fig. 1, as well as in a

variety of formations and types of reservoirs, such as conventional, tight, and more recently in shale (hydrocarbon source

rock). The hydraulic fractures created in Argentina present a variety of conditions and challenges related to depth (from 300

to 4,500 m), bottomhole temperature (BHT) (100 to 300°F), reservoir pressure (from subnormal to overpressure), formation

permeability (high, medium, low, and ultralow perm), multilayer reservoirs, and multitarget wells.

Throughout the years, there have been noticeable changes to the types of treatment and wells in which fracturing fluids

are used, from oil-based systems, alcohol-water mixtures, and foams, to water-based fluids currently used. The steady

increase in drilling activity and, therefore, well completion and stimulation has led to increased water consumption; thus,

alternatives have been sought to help minimize this impact in certain basins. Bonapace et al. (2012) documents the use of

2 SPE-174118-MS

produced water for use in a fracturing fluid in the Golfo San Jorge Basin, managing to replace 55% of freshwater

consumption.

Early work (hydraulic fractures) to develop Argentina’s shale basins was conducted during 2010. The majority of

exploration and development has been in the Vaca Muerta formation, but work has also been assessed in other formations,

such as Los Molles, Cacheuta, D-129 and Agrio more recently. Experience gained related to water management in these shale

plays during the completion of more than 40 wells (>200 hydraulic fractures) by different operators is presented.

Furthermore, laboratory studies were conducted on treated and untreated flowback waters and their assessment for use as

fracturing fluid water is presented.

Fig. 1—Map of five hydrocarbon producing basins discussed.

Water Sources and Stimulation Currently, Argentina’s largest shale reservoir development is the Neuquén Basin in the Vaca Muerta formation; however,

development has been performed in other plays, such as training in the Los Molles, Agrio, some stimulation in the Cuyo

Basin (Cacheuta formation), and in the Golfo San Jorge Basin (D-129 formation). These basins have a history of

conventional reservoir development and corresponding stimulation techniques (primarily hydraulic fracturing). Thus, water

sources normally used for these developments (conventional reservoirs) are the same as those used during the early stages of

exploration and subsequent development of shale reservoirs. Some particularities in terms of water type have been observed

in such exploration wells for other plays. In the Los Molles formation, a mixture of fresh water (85%) and produced water

(15%) was used because of the large volume of water necessary for hydraulic fracturing of a 10 stage horizontal well. For the

completion in the D-129 formation, the operator decided to use 100% produced water (low salinity < 10,000 TDS) to run all

the stimulation treatments (five fracture stages).

Given the economic potential of the Vaca Muerta play, the focus lies on this reservoir. Considering its vast extension in

the Neuquén Basin, the Vaca Muerta comprises many fields exploited at levels located below or above the Vaca Muerta by

several operators. This helps because some fields and operators have existing surface facilities; this is favorable during the

exploration and development phases of this shale in relation to the logistics of water.

The primary sources of water in the Neuquén Basin used to develop these hydrocarbon resources are rivers (Neuquén,

Limay, Colorado), lakes, or reservoirs (Cerro Colorado, Pellegrini), or groundwater sources, such as wells with low salinity

(< 5,000 TDS). These types of wells for water supply need a permit from regulatory authority and produced water is not

suitable for human consumption or farmlanding.

SPE-174118-MS 3

Physical-Chemical Analysis. Table 1 presents a summary of different water sources that have been used during stimulation

in the Vaca Muerta. It presents the primary feature chemical of these sources, which have been identified according to

operator, field, and nature (surface or underground). Additionally, the first column presents the requirements for fresh water

to be used as the base element of a fracturing fluid (according to service company standards).

Fresh Water

Field Water Requirements

Sample 1 Sample 2 Sample 3 Sample 4 Sample 5 A B#a B#b B#c C

Water Source Type River River River River River Well Well Well Well River

Specific gravity 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000

pH 6 to 8 7.71 8.11 8.03 8.38 7.91 9.14 8.73 8.76 8.71 7.63

Resistivity (ohms-cm) 59.551 3.525 1.012 12.234 37.386 3.331 2.450 2.235 2.618 31.434

Temperature (°C) 15 to 40 24.4 21.0 20.7 19.1 20.9 20.6 19.9 19.6 20.4 20.7

Carbonate (mg/L) < 600 0 0 0 0 0 0 7.2 9.6 14.4 0

Bicarbonate (mg/L) < 600 67.1 158.7 268.5 97.6 134.2 278.2 244.1 249.0 258.7 68.3

Chloride (mg/L) < 30,000 2.0 620.2 80.0 36.0 24.0 348.1 428.2 348.1 372.2 24.0

Sulfate (mg/L) < 500 7.5 50.0 85.0 36.3 32.5 475.0 650.0 875.0 625.0 12.5

Calcium (mg/L) 50 to 250 1.6 34.7 83.4 32.1 28.1 8.0 1.6 0.0 0.0 19.2

Magnesium (mg/L) 10 to 100 2.0 8.1 17.5 14.6 7.3 1.0 1.0 1.0 1.0 6.8

Barium (mg/L) 0 0 0 0 0 0 0 0 0 0

Strontium (mg/L) 0.02 1.3 1.2 1.0 0.02 n/a n/a n/a n/a n/a

Total Iron (mg/L) 1 to 20 0.12 0.20 0.10 0.00 0.56 0.17 0.07 0.12 0.05 0.32

Aluminum (mg/L) 0.002 0.002 0.002 0.002 0.002 0.020 0.020 0.020 0.020 0.020

Boron (mg/L) 0 to 20 0.0 0.2 10.9 0.6 0.02 n/a n/a n/a n/a n/a

Potassium (mg/L) 100 to 500 0.0 2.6 13.3 11.5 1.5 0.0 0.0 0.0 0.0 0.0

Sodium (mg/L) 2,000 to 5,000

24.7 427.7 51.7 1.6 34.2 546.5 682.1 743.4 646.7 12.3

TDS (mg/L) < 50,000 105 1,302 599 230 262 1,657 2,014 2,226 1,918 11,603

TSS (mg/L) < 50 2.5 5.6 0.3 0.8 1.1 30.0 16.0 12.0 7.5 143.6

Table 1—Summary of various sources of water used during stimulation in the Vaca Muerta.

As Table 1 depicts, the underground water sources (wells) are higher than surface sources (rivers) in terms of pH, TDS,

total suspended solids (TSS), chlorides, sulphates, bicarbonates, and sodium. However, all referenced water sources meet the

requirements established to be used in fracturing fluids. Five samples of water from the Neuquén River (different points of

location) are included as references and comparison.

Types of Treatment. The most common hydraulic fracturing treatments performed in different plays in Argentina were

hybrid fracturing designs. Fig. 2A presents average water volumes per stage (m3) for various shales in Argentina and types of

reservoir fluids. Additionally, Fig. 2B presents corresponding percentages according to the fluid system used, distribution of

slickwater (SW), linear gel (LG), and crosslinked gel (XL). In general, it can be observed that the greatest volumes of water

(m3) by stage correspond to reservoirs of gas and wet-gas (Fig. 2A red and yellow bars), being hybrid treatment designs for

both SW-LG (Los Molles) and SW-LG-XL (Vaca Muerta).

4 SPE-174118-MS

Fig. 2—(A) average water volume per stage (m3); (B) percentage according to fracturing fluid type.

A statistical analysis regarding the type of hydraulic fracture in Vaca Muerta was performed for six fields, A, B, C, D, E,

and F. Fig. 3 shows more detail for the treatments performed in these wells based on reservoir fluid composition. A total of

13 wells and more than 65 hydraulic fractures were analyzed. In general, average water volume per stage (Fig. 3A) varied

according the fluid reservoir; for oil wells, average water volume was 1300 m3; it was 1850 m

3 for wet gas; and 2180 m

3 for

gas wells. Fig. 3B presents the fracturing fluid distribution for each. The fracturing treatments were primarily hybrid SW-XL,

where in some cases used SW-LG-XL; the percentages of most systems ranged from 30 (oil) to 84% (gas) for slickwater and

70 (oil) to 10% (gas) for crosslinked systems. LG was commonly used as a contingency in the transition from SW to XL.

Small percentages were used (less 10%) based on the pressure response during the treatment and reservoir conditions.

Normally, the completion of a Vaca Muerta well has involved a total water volume of ~ 6500 m3 for vertical wells and

~14500 m3 for horizontal wells.

Fig. 3—(A) average water volume per stage (m3); (B) percentage based on fracturing fluid type.

SPE-174118-MS 5

Types of Systems. These systems used fresh water and contained chemical additives to provide various functions:

• SW: contained friction reducer and friction reducer breaker.

• LG: contained gelling agent, buffer, and breaker.

• XL: consisted of buffer, gelling agent, crosslinker, and breaker (usually used a 20-lbm/1000 gal XL guar-borate

fluid.

• Additionally, each of the fluid systems also typically contained a biocide, clay inhibition, and surfactant additives.

Specific surfactants were selected according to reservoir fluid type (oil, wet gas, and gas).

Yang et al. (2013) and Patel et al. (2014) presented a detailed analysis of the evolution of fracturing fluid designs,

fracturing fluids, water consumption, chemical additives, and proppant used for different basins in the US. This information

is valuable for identifying tendencies or changes in these points in other US shale plays.

Logistics During the past five years, there has been substantial progress related to water management and logistics for sustainable

development of Argentina’s shale plays. A variety of water storage systems and methods for transferring water (trucks, pipe

systems) have been used in the Neuquén Basin associated with operations, primarily in the de Vaca Muerta. Various models

of water management were used by numerous operators and depended on the stage or phase of activity (exploration or pilot

phase), type of completion (vertical, horizontal, or recompletion wells), surface facilities existing in the fields, geographical

location, and proximity to available water sources. Water storage systems have been primarily mobile fracture tanks (80 m3)

(Fig. 4B), circular tanks (1,000 to 5,500 m3) (Fig. 4A), and lined pits (15,000 to 35,000 m

3) (Figs. 4C and D). The

movement of water has been performed primarily by trucks, and some operators have built transfer systems using piping

(tubing or aluminum pipe) and centrifugal pumps.

Currently, the most common storage systems used are mobile fracture tanks and circular tanks. Pits utilization are

restricted for environmental restrictions. A greater number of suppliers of fracture tanks have been incorporated in the last

years, and also, the incorporation of new technologies in circular tanks (easy assembly, portables, and greater capabilities).

Fig. 4—(A) circular tank (6000 m3); (B) fracture tanks (6000 m

3); (C) small pit (15000 m

3); (D) large pit (35000 m

3).

Following is a discussion of varying cases of water management plans and logistics developed for vertical wells during

the initial stage of exploration, as well as cases for horizontal wells under development.

First Vertical Well The first well in Vaca Muerta was drilled in A Field, which has a very good infrastructure attributed to development wells in

tight formations. The completion of the well consisted of four hydraulic fractures stages, requiring of 7600 m3 of total water.

6 SPE-174118-MS

Alternatives were evaluated for the logistic and water management; finally, it was decided to install a transfer system through

pipes and a water storage location close to the well to be stimulated (Fig. 5).

A water well (groundwater with low salinity) was used for source water located in the same field with production flow of

1200 m3/D. A circular tank for water storage of up to 1000 m

3 was located within the vicinity. This well was 7.5 km away

from the well to be stimulated. The operator performed the laying of a 4 in. pipe from the water well to the water storage

location, in which there were installed two circular tanks of 2000 m3 capacity (Fig. 6A). Water was pumped from the well to

the location using centrifugal pumps. The service company provided a water transfer system (Figs. 6B and 6C) between the

storage location and the stimulation wellsite, 350 m of distance, consisting of two lines of 8-in. aluminum pipe and two

centrifugal pumps with a pumping capacity of 40 to 67 bbl/min. At the location of the vertical well, mobile fracture tanks

were placed (Figs. 6D and 6E) with a total storage capacity of 2000 m3. Before starting the stimulation, a water volume of

6000 m3 was available (frac tank and water location tanks were completely filled; Fig. 5); additionally, as a backup, recharge

water was onsite by means of trucks.

Fig. 5—Water management plan for first vertical well completion.

Fig. 6—Water management plan for first vertical well completion.

Vertical and Horizontal Well The second vertical well in Vaca Muerta was drilled in Field B; as the first well, this field has a very good infrastructure. The

completion consisted of four hydraulic fracture stages, requiring of 8000 m3 of water. For the stimulation of this well, it was

decided to drill a new water well (groundwater and non-potable water) close to the wellsite and apply the same water

management strategy. The water well (Fig. 7A) finally was drilled at 300 m from the wellsite and had a flow capacity of

2,000 m3/D (natural production). To complete the vertical well, a mobile fracture tank was installed in the water well location

to store water and a centrifugal pump. The water storage was transferred from the frac tank to the wellsite using one line of

aluminum pipe (8 in.) and a centrifugal pump. In the wellsite, the water was stored in three circular tanks (each had a capacity

of 1000 m3) and 15 mobile frac tanks (Figs. 7B and C), a total storage capacity of 4500 m

3 was available at the location.

SPE-174118-MS 7

Fig. 7— Water management plan for second vertical well completion.

The first horizontal well completion in Field B consisted of six fracture stages, consuming a total volume of water of

13000 m3. It was decided to use part of the infrastructure built for the vertical for the horizontal well drilled 1.5 km from the

water well (Fig. 8). The water plan involved taking water from the water well and transferring and storing it in the location of

vertical well (water location). Based on the amount of water needed, it was necessary to install a new circular tank with more

capacity (total storage capacity 3200 m3). Then, the water was transferred to the wellsite using a centrifugal pump and

aluminum pipe. The horizontal well had total water capacity storage of 5000 m3 (circular tank and mobile fracture tanks). For

this job, the service company was in charge of water logistics and it was necessary to transfer a portion of the water in real

time.

Fig. 8—Water management plan for horizontal well completion.

Use and Reuse of Flowback, Produced, and Treated Water This section analyzes the possibility of using water nontraditionally (flowback and produced), either treated or untreated.

Various alternatives uses include (a) flowback and produced water untreated; (b) mixing dilution water (flowback-produced)

untreated with fresh water; (c) and treated water.

Laboratory studies were conducted to evaluate different alternatives for the use of these nontraditional waters as suitable

for fracturing fluids, primarily for use in a XL gel system. The tests performed included:

• Detailed water (physical-chemical) analysis.

• Clay swelling and inhibition testing.

• Evaluation and development of XL fluids, proppant transport capacity.

• Damage by gel residue and TSS.

8 SPE-174118-MS

Physical-Chemical Analysis. Table 2 presents a summary of different flowback and produced waters labeled by different

fields in the Vaca Muerta formation. The same primary physico-chemical variables can be observed for these waters.

Flowback and Produced Water

Field/Well G#1 A#1 B#1 C#1 D#1 E#1 F#1 B#2 D#2

Water Type PROD FB FB FB FB FB FB FB FB

Specific gravity (SG) 1.136 1.074 1.123 1.143 1.123 1.156 1.099 1.110 1.068

pH 6.48 6.74 5.06 5.25 4.65 4.82 5.59 4.50 6.31

Resistivity (ohms-cm)

0.026 0.067 0.030 0.023 0.024 0.035 0.049 0.040 0.051

Temp (°C) 20.2 26.0 23.0 24.0 24.0 20.7 20.5 20.8 20.0

Carbonate (mg/L) 0 0 0 0 0 0.0 0.0 0.0 0

Bicarbonate (mg/L) 146.4 1,196.0 131.8 107.4 0.0 61.0 329.5 61.0 500.4

Chloride (mg/L) 118,546.8 67,026.5 106,041.9 131,051.8 135,051.8 148,058.5 87,034.4 92,536.5 63,525.1

Sulfate (mg/L) 0.0 10.0 262.5 137.5 100.0 0.0 233.3 265.0 400.0

Calcium (mg/L) 21,643.0 7,134.2 23,406.7 17,955.8 30,781.4 35,671.2 18,036.0 27,655.2 5,210.4

Magnesium (mg/L) 2,140.2 1,702.4 3,988.5 2,723.8 4,669.4 2,432.0 2,918.4 1,216.0 8,755.2

Barium (mg/L) 800 800 0 0 0 1,275 2.5 0 0

Strontium (mg/L) 2,078.0 n/a 2,120.0 4,210.0 3,170.0 2,900.0 385.0 1,000.0 740.0

Total Iron (mg/L) 21.25 575.00 243.75 6.50 150.00 68.00 98.00 196.25 185.00

Aluminum (mg/L) 0.020 0.020 0.020 0.020 0.020 0.002 0.020 0.002 0.500

Boron (mg/L) 29.8 24.2 10.4 17.2 24.2 15.5 29.2 12.6 5.0

Potassium (mg/L) 2,750.0 250.0 998.0 2,130.0 1,700.0 2,905.0 504.0 1,250.0 562.5

Sodium (mg/L) 45,234.5 32,225.5 34,489.0 59,261.3 40,819.0 47,526.9 29,913.7 24,832.7 18,447.2

TDS (mg/L) 190,562 110,920 171,682 217,584 212,982 237,998 139,070 149,713 97,586

TSS (mg/L) 714.5 163.0 310.4 235.6 240.0 120.0 517.2 194.0 551.7

Table 2—Summary of flowback and produced water from various fields and wells.

Flowback and produced waters have similar characteristics, which differentiate them from freshwater. In general, these

waters have higher values of specific gravity (SG), lower pH values (being slightly acidic), higher levels of TDS and TSS, as

well as significant values of Ca, Mg, Na, K, Fe, B, and Ba concentrations. Olsen et al. (2013) presented flowback water

compositions from some North American shale plays, which can be evaluated comparatively with the information shown in

Table 2 from the Vaca Muerta formation.

Moreover, Table 3 shows the physical and chemical results for four samples of flowback and produced water, which have

been treated by four different water treatment companies. The water treatment methods performed by these different

companies were Treatment Methods I, II, and III corresponding to treatments of chemical coagulation, flocculation, and

separation; the IV Treatment method consists of electrocoagulation, pH adjustment, weir tank separation, and multimedia

filtration.

SPE-174118-MS 9

Treated Water

Field/Well A#1T B#5T B#4T G#1T

Water type FB FB FB PRO

Treated method I II III IV

Specific gravity (SG) 1.060 1.094 1.070 1.125

pH 7.84 5.87 7.32 9.12

Resistivity (ohms-cm) 0.075 0.049 0.047 n/a

Temperature (°C) 19.5 21.1 18.1 n/a

Carbonate (mg/L) 0 0 0 66.5

Bicarbonate (mg/L) 219.7 170.9 244.1 0.0

Chloride (mg/L) 59,523.5 85,033.6 61,524.3 104,687.0

Sulfate (mg/L) 0.0 325.0 6,375.0 5.0

Calcium (mg/L) 6,332.6 14,909.8 3,206.4 155.0

Magnesium (mg/L) 729.0 1,167.4 1,945.6 857.0

Barium (mg/L) 110 0 0 874

Strontium (mg/L) 1,400.0 1,080.00 177.00 1,846.0

Total Iron (mg/L) 0.45 11.00 2.60 1.32

Aluminum (mg/L) 0.020 0.002 0.002 0.920

Boron (mg/L) 12.0 13.7 8.2 22.8

Potassium (mg/L) 16.0 1,945.0 253.1 2,066.0

Sodium (mg/L) 29,984.4 34,054.7 35,389.8 47,182.0

TDS (mg/L) 96,916 137,617 108,940 172,097

TSS (mg/L) 4.4 34.6 4.3 10.1

Table 3—Physical and chemical results for four samples of flowback and produced water.

These waters generally had the same concentration of TDS ions of flowback produced waters; it is important to observe

certain indicators that varied amongst the treatments. In general, because these waters had measured pH values ranging from

slightly acidic to neutral to slightly alkaline, reduction to the amount of iron and TSS is clearly visible, while the content of

TDS and salts remain high.

For a better understanding of the action of the treatments applied to these waters, one can compare the results of the

samples corresponding to fields and Wells A, B, and G to Tables 2 and 3.

Action as Clay Inhibitor. The goal of this section is to evaluate flowback, produced, and treated waters in relation to their

inhibition properties on formation clays (cuttings). High salt content (TDS) made it very unlikely to use certain clay

stabilizing additives. Capillary suction time (CST) testing was performed to determine this property (Ramurthy et al. 2011).

During the exploration phase of the Vaca Muerta in these six fields, a routine CST was performed for each well in a new

field; the intention was to understand the degree of water sensitivity and select the better clay stabilizer and concentration. All

of the tests were performed using deionized water (DI).

The following clay stabilizers have been tested and used in these fields:

• Quaternary ammonium salt (liquid), initial wells.

• Inorganic salt—KCl (solid), in some cases (Garcia et al. 2013).

• New ultralow-molecular-weight cationic organic polymer (liquid) recently applied to replace the quaternary

ammonium salts (Weaver et al. 2011; Garcia et al. 2013).

For all of the fields, the base line was performed using a Quaternary ammonium salt (1.4 gal/1,000 gal), and it was

identified that some fields were more sensitive to the aqueous phases as A, C, D, and F, but not more sensitive than B and E

(Fig. 9A, yellow bars). Two further lines are referenced, which indicate the degree of sensitivity (no sensitivity CST ratio =

0.5 dotted line black and extremely sensitive CST ratio = 50 red dotted line). For reference, a green solid line was placed for

the CST ratio = 5, which was considered an acceptable value for the Vaca Muerta formation. There is a clear correspondence

to the high values of CST ratio (Fig. 9A, yellow bars) and the clay percentage (Fig. 9B, green bars).

Tests performed using 1% KCl (Fig. 9A, green bars) for C and D fields show values below CST = 5. For Field C, it was

necessary to increase up to 2% to achieve these values. It is important to mention that, for the completion of the first wells in

Fields C and D, it was necessary to use KCl as clay stabilizer.

10 SPE-174118-MS

A new ultralow-molecular-weight cationic organic polymer was tested for different fields and concentrations (Fig. 9A,

blue bars). In general, it was found that optimal concentrations were between 0.5 gal/1,000 gal (B) and 1.5 gal/1,000 gal (F).

All of these were below CST = 5. For the most sensitive field (C), it was necessary to increase the concentration up to 5.0

gal/1,000 gal (Fig. 9A, purple bars) to obtain the desired value. After completing all of these tests, it was decided to

implement the use of a new clay stabilizer in all of the fields.

Fig. 9—(A) CST for different clay stabilizers and fields; (B) mineralogy for each field.

Non-Traditional Water Evaluation. This test was performed on the formation material and drill cuttings (Field A), to

evaluate the inhibitory action of different water sources, such as mixed (flowback-fresh), produced, or flowback and treated

flowback. Table 5 illustrates a breakdown of the waters that were tested and the results that are presented in Fig. 10A.

Type Field/Well Percentage

(%) TDS

(mg/L) Clay Stabilizer

DI — 100 0 1.4 gpt

Freshwater B#a — 2,014 No

Blend B#a + G#1 50:50 96,288 No

Produced G#1 100 190,562 No

Treated B#4T 100 108,940 No

*Clay stabilizer (quaternary ammonium salt), blend: fresh water + produced, treated: flowback treated

Table 5—Breakdown of the water.

SPE-174118-MS 11

Fig. 10—(A) Vaca Muerta CST for different waters; (B) mineralogy for each sample.

The observed results show that, in general, for any of these options, nontraditional water (produced, blend, or treated)

CST ratio values were below five (Fig. 10A, green line), denoting a very good inhibition power that does not require

additional clay inhibition additives. Tipton (2014) documented the application of recycled water and treated water as a brine

added to the base fluid, achieving a concentration of 1% KCl (clay stabilizer) in the operations of the Woodford Shale.

Fracturing Fluid. Recently, there have been studies on the development of fracturing fluids using flowback or produced

water, treated or untreated, with high TDS values. The authors LeBas et al. (2013), Monreal et al. (2014), and Kakadjian et al.

(2013) reported primarily using CMHPG-Zr or alternative systems. Haghshenas and Nasr-El-Din (2014) documented good

results with a guar-borate system; however, there was a need to address some ions (calcium and sodium), bringing them to

acceptable levels for proper fluid performance.

Referring to the XL fluid tests performed used in current Vaca Muerta fracturing treatments, the base fluid used was fresh

water and the system used a guar-borate of 20 lbm/1,000 gal. Given the nature of the hybrid treatment and the pumping of

large volumes of SW or linear gel before a XL gel, it has effects of cooling, which makes the XL systems subject to

background temperatures of approximately 120°F and spacing out of approximately 30 to 45 minutes.

A survey methodology raised two lines of work, the first uses a mixture of water dilution (fresh water and flowback

untreated) and the second works on samples of only treated flowback water.

Blend of Water (Flowback Untreated). The first test performed used the fracturing fluid normally used (20 lbm/1,000 gal

guar-borate) using a mixture of waters. The results were not satisfactory because gel hydration and crosslinking problems

were observed, which resulted in unstable fluids at surface and bottomhole conditions. Moreover, generating filaments,

flocculants, and precipitates by increasing the pH of the system (usually borate systems works in high pH) were observed.

In Fig. 11A, evidence flocculated particles was visible as the flowback water pH (alkaline pH range for borate system)

was changed. In the bottom of the image, one can observe precipitates (Fig. 11B), crystal structures, which are evidenced by

the visible eye (Fig. 11C), compared to the stereomicroscope 40 times (Fig. 11D). Situations of this type, i. e. the generation

of precipitates, can lead to problems with the performance of the fracturing fluid and cause potential damage to the proppant

pack in the fractures, which has been documented by several authors (Monreal et al. 2014; Haghshenas and Nasr-El-Din

2014; Fedorov 2014).

Because of this, development of alternative fluids that meet the requirements for use in the fracture treatment was

necessary. The first objective was to develop a fluid that could be used with mixtures of water (fresh-flowback water

untreated) to minimize the effects described.

After several tests with different fracturing fluids, a CMHPG XL system was developed using a metal crosslinker, a low

polymer loading of 20 lbm/1,000 gal, and low pH. The same was tested for mixtures of 50:50 fresh water with various

flowback waters from different fields (Table 6). These tests, conducted without added breaker, showed repeatability in terms

of the results with slight adjustments of the system. In Fig. 12, the guar-borate (20 lbm/1,000 gal) fluid made with freshwater

was indicated for routine use in the operations of the Vaca Muerta, (Test 8). All of the tests were performed using a FANN-

12 SPE-174118-MS

50 viscometer equipped with a R1/B2 rotor-bob geometry, 1 hour by simulating BHT (120°F) and a constant shear stress of

40 1/sec.

Fig. 11—Evidence that the flocculants generated as the flowback water pH (alkaline range pH for borate system) was changed.

Test No. Water Type* Field / Well Percentage

(%) TDS

(mg/L)

1 Blend (fresh + flowback) B#a + A#1 50:50 56467

2 Blend (fresh + flowback) B#a + B#1 50:50 86848

3 Blend (fresh + flowback) B#a + C#1 50:50 109799

4 Blend (fresh + flowback) B#a + D#1 50:50 107498

5 Blend (fresh + flowback) B#a + E#1 50:50 120006

6 Blend (fresh + flowback) B#a + F#1 50:50 70542

8 Fresh water (guar.borate) B#a 100 2014

*For more details about water, refer to Tables 1 and 2; TDS = final value for the blend.

Table 6—Tested for mixtures of 50:50 fresh water with various flowback waters from different fields and wells.

SPE-174118-MS 13

Fig. 12—Guar-borate (20 lbm/1,000 gal) fluid made with freshwater was indicated for routine use in the operations of the Vaca Muerta (Test 8).

Treated Water (Flowback). According to the previously obtained results, it was decided to move forward and evaluate the

fluid with 100% treated flowback water. Slight changes in pH adjustment and concentration of the crosslinker were made to

the fluid.

This fluid was tested with the sources of water treated using Methods I, II, and III (Table 7). Refer to Table 3 for more

details about the treated water.

Test No. Water Type/Treated Field/Well Percentage

(%) TDS

(mg/L)

Treated I Flowback/treated I A#1T 100 96,916

Treated II Flowback/treated II B#5T 100 137,617

Treated III Flowback/treated III B#4T 100 108,940

Table 7—Fluid tested with the sources of water treated using Methods I, II, and III.

Fig. 13 presents the results for each of these tests. Comparatively, the reduced viscosity profile was observed between the

fluids formulated with blends of water and formulated with 100% of flowback treated water. The latter, however, had

viscosity values in the order of 300 to 400 cp (40 1/sec) for times of 30 to 45 minutes.

14 SPE-174118-MS

Fig. 13—Test results.

Proppant Transport Capacity. The next step was to analyze the transport capacity in which two types of tests were

conducted; the first was a conditional static (settling test) and the second was under dynamic conditions (slurry viscometer,

SV) to evaluate the transport behavior of the fluid.

Settling Test. For this first test, two sets of XL gel were evaluated formulated according to Table 8. The test was performed

at 120°F for approximately 8 hours (480 minutes) using a 20/40 light-weight ceramic (LWC) and a proppant concentration 3

lbm/gal (without the addition of breakers). In Figs. 14 and 15, one can observe the behavior of each fluid. Overall, the guar-

borate system, currently used in operations in the Vaca Muerta, after the first two hours lost its carrying capacity. The

CMHPG-Zr developed system exhibited proppant suspension capacity for 8 hours. Very similar responses are documented

related to these tests by Monreal et al. (2014), Kakadjian et al. (2013), and Haghshenas and Nasr-El-Din (2014).

System Water Type Fields/Wells Percentage

(%) TDS

(mg/L)

20-ppt Guar.borate Fresh water B#a 100 2,014

20-ppt CMHPG-Zr Blend B#a+F#1 50:50 70,542

20-ppt CMHPG-Zr Treated B#4T 100 108,940

Table 8—Formulations of two sets of XL gel tested.

SPE-174118-MS 15

Fig. 14—Behavior of fluids observed by means of testing.

Fig. 15—Behavior of fluids observed by means of testing.

Dynamic Test-SV. This test was performed only for the new CMHPG-Zr fluid developed formulated with 100% treated

water. It was performed at 120°F with 10-lbm/gal LWC proppant 20/40 for approximately 180 minutes with the addition of

breakers (oxidant and an activator). For this test, a slurry viscometer was used, which was developed to evaluate the transport

ability of the fracturing fluids and proppant using dynamic testing conditions (Harris et al. 2005).

In Fig. 16, one can see the test; on the graph of slurry viscosity, three periods of time can be identified, which are detailed

below. An initial period, which goes from 0 minutes to 125 minutes, is called the elastic transport region. The same one has a

very good ability to transport fluid viscosity values in the order of 180 cp at 40 1/sec. Next, the settlement process begins

(onset of proppant setting) and is identified extending over a period of time ranging up to 160 minutes and corresponding to

the region of viscous settlement. This action of the breaker is weakening the structure-based fracture fluid. At the midpoint of

this period, the maximum transport capacity of the system, which corresponds to a value of 90 cp at 40 1/sec, is determined.

16 SPE-174118-MS

Finally, between 160 to 190 minutes, only one proppant decanted with values of approximately 60 cp at 40 1/sec, which

shows the lack of system structure by the action of the breaker.

Fig. 16—Dynamic testing results.

This test indicated that, for low polymer fluid formulations using 100% nontraditional water treated, the fluid

demonstrated good proppant transport capacity, despite having low viscosity values at 40 1/sec. Harris et al. (2005) mentions

that systems (metal XL gel) with low polymer loading can carry proppant perfectly for long periods of time with viscosity

values (clean fluid) very low, unlike the borate systems.

Damage. This analysis was performed on a column fluid recovery test (Fig.17A), which consists of preparing a test column

with 70/140-mesh white sand pack, determining pore volume, passing three pore volumes initially, and then proceeding to

flow the different solution to be tested for a time period of 10 minutes. Then, the percentage of displaced fluid through this

pack for the testing duration was determined. The solutions were different types of water and broken fracture fluid (Fig.

17B).

The objective of this test was to evaluate the potential for generating damage in a sand pack by the action of different

types of water (TSS) and fluid systems broken (gel residue). Table 9 presents information about the samples tested.

System Water Type Field Percentage

(%) TSS

(mg/L)

DI-W DI — 100 0

Treated-W Treated B#4T 100 4

Blend-W Blend DI + G#1 66.6:33.3 238

Produced-W Produced G#1 100 714

FrW XL guar.borate Fresh water B#a 100 16

B XL CMHPG-Zr Blend B#a + F#1 50:50 267

T XL CMHPG-Zr Treated B#4T 100 4

*All XL gel systems are 20 ppt B = blend water T = treated water

Table 9—Information on the samples tested.

SPE-174118-MS 17

Fig. 17—(A) column fluid recovery test; (B) broken XL gel system.

The first test performed was to finally identify the differences arising from the use of waters with varying content of TSS

(DI, flowback, produced, blends of water all untreated, and flowback treated water) (Fig. 18A). This test showed a linear

response for the four samples and, as the amount of TSS increased, the percentage of displaced fluid decreased.

The second test (Fig. 18B) was conducted to evaluate the potential for damage created by different systems, broken XL

gel; this test presented a different trend in response. The systems were prepared with the same polymer loading (20 lbm/1,000

gal), but with different types of water (Table 9). Then, the same breaker concentration was added with all the broken systems

having a viscosity of 3 cp using a Fann Model 35 viscometer. The system was traditionally evaluated and used in operations

in the Vaca Muerta (FRW XL guar-borate); an alternative system was developed with either blend-mixed water (B XL

CMHPG-Zr) or treated water (T XL CMHPG-Zr).

Fig. 18—(A) TSS water effect; (B) TSS and XL gel effect.

Fig. 19 presents the value at the end of the test in bars (final time = 10 minutes); refer to the left axes. For each water

type, one can markedly observe a difference for the four samples. The percentage reduction in fluid displaced with reference

to DI water (red dots, refer to right axes) reached a maximum value of 36% when using untreated produced water (highest

value of TSS = 714 mg/L). Ye et al. (2013) presents comparative results on the effect of reducing the permeability of a

proppant pack by using flowback-production treated water and not treated water.

18 SPE-174118-MS

Fig. 19—Values post-testing.

For the XL gel system, one can see the difference for XL gel vs. the various water systems. The values obtained for the

guar-borate fluid (fresh water) and the fluid CMHPG-Zr (blend water) are very similar (FrW XL gaur-borate and B XL

CMHPG-Zr), varying in such systems only in terms of TSS content and the waste water of different types of gelling agents

(guar and CMHPG). Clearly, for the same range of reduction (93 to 95%), a greater effect of gelling agent residue was

observed in the guar-borate system and an increased incidence of TSS for CMHPG-Zr was observed. Finally, for a low TSS

water content (treated) and the CMHPG-Zr system, the best value (68%) was observed. These results show the importance of

the removal of TSS in the flowback or produced water as well as the use of gelling agent with low percentages of residue

(CMHPG-Zr). In Fig. 17B, one can see differences in residue for CMHPG-Zr between blend water and treated water; the

difference is only in relation to the amount of TSS in each water type (Table 9).

Conclusions In general, hybrid (SW-LG-XL) fracturing treatment designs dominate the fluid type used in Argentina’s shale plays. The

largest volumes of fluid phase correspond to the treatments performed in the Los Molles formation (3000 m3), being that

Vaca Muerta is approximately 1500 m3, varying based on the different reservoir fluids (1300 m

3 oil, wet-gas 1850 m

3, and

gas 2200 m3).

The water storage systems that have been primarily used are mobile fracture and circular tanks, which are usually located

at the wellsite being stimulated. Most water is delivered by transport trucks. An integrated water management plan was

developed to support wells completions in Vaca Muerta, using sources of water close to the wellsite, pipeline transfer

systems, and an important storage system, all of this to reduce costs and simplify logistics. For a large-scale development an

integral water management plan has to be performed, taking into account rivers, lakes, water wells, and synergies with other

regional operator companies.

Nontraditional water sources analyzed (flowback-produced) for the Vaca Muerta revealed substantial TDS, TSS, Ca, Mg,

Na, K, Fe, and B concentrations. It has been observed that, for such waters that have been treated using different methods

(service companies), important reduction in terms of TSS and Fe content has been made.

The use of these waters in the fracture treatments concluded that there is no need to use additional clay stabilizers because

of the water’s inhibition properties.

There has been a significant decrease to the displaced fluid (10 to 36%) as the content of TSS increases (4 to 714 mg/L) in

the nontraditional waters. This clearly demonstrates the necessity of treating such waters by considerably reducing the

content of TSS, which can negatively impact the fracture conductivity of the proppant pack.

A new fracturing fluid, with low polymer loading and low pH, was developed in the laboratory. It can be formulated with

these waters, whether blend mixing or 100% (nontraditional treated water). This system presents the advantages of working

in a range of pH in which there are not precipitates that generate reduced conductivity in the fracture pack. It generates a

good carrying capacity, as per settling test and slurry viscometer testing. Additionally, it was observed during test fluid

recovery that a lower percentage (68%) in fluid reduction, offset from the fracturing fluid guar-borate (93%) that is currently

used, proved the fluid much cleaner, despite being formulated with nontraditional treated water.

SPE-174118-MS 19

Treatment and reuse of non-traditional water for future fracturing treatments greatly mitigates the issue of fresh water

requirements for shale wells and reduces volumes to be injected in disposal wells. Water reuse is a key factor for sustainable

shale developments.

Nomenclature CMHPG = carboxymethylhydroxypropyl guar

CST = capillary suction time

DI = deionized water

FB = flowback

LG = lineal gel

LWC = lighweight ceramic

ppt = pound per thousand gallons

SV = slurry viscometer

SW = slickwater

TDS = total dissolved solids

TSS = total suspended solids

XL = crosslinked gel

XRD = x-ray diffraction

Acknowledgments The authors thank Total and Halliburton for the permission to publish this paper. Special thanks are extended to the

Argentina Unconventional Reservoir Solution Team for support as well as to Maximiliano Coronel and Dario Tello for their

unconditional work and support in the lab. Aerial pictures are courtesy of Gonzalo Arribere (HYDROFRAC, an El Fortin

Constructions division).

References Bonapace, J.C., Giglio, M.R., Moggia, J.M. et al. 2012. Water Conservation: Reducing Freshwater Consumption by Using Produced Water

for Base Fluid in Hydraulic Fracturing-Case Histories in Argentina. Presented at the SPE Latin America and Caribbean Petroleum

Engineering Conference, Mexico City, Mexico, 16–18 April. SPE-151819-MS. http://dx.doi.org/10.2118/151819-MS.

Fedorov, A., Carrasquilla, J., and Cox, A. 2014. Avoiding Damage Associated to Produced Water Use in Hydraulic Fracturing. Presented

at the SPE International Symposium and Exhibition on Formation Damage Control, Lafayette, Louisiana, 26–28 February. SPE-

168193. http://dx.doi.org/10.2118/168193-MS.

Garcia, M.N., Sorenson, F., Bonapace, J.C. et al. 2013. Vaca Muerta Shale Reservoir Characterization and Description: The Starting Point

for Development of a Shale Play with Very Good Possibilities for a Successful Project. Presented at the Unconventional Resources

Technology Conference, Denver, Colorado, 12–14 August. SPE-168666-MS. http://dx.doi.org/10.1190/URTEC2013-090.

Haghshenas, A. and Nasr-El-Din, H.A. 2014. Effect of Dissolved Solids on Reuse of Produced Water in Hydraulic Fracturing Jobs.

Presented at the SPE Latin America and Caribbean Petroleum Engineering Conference, Maracaibo, Venezuela, 21–23 May. SPE-

169408-MS. http://dx.doi.org/10.2118/169408-MS.

Harris, P.C., Morgan, R.G., and Heath, S.J. 2005. Measurement of Proppant Transport of Frac Fluids. Presented at the SPE Annual

Technical Conference and Exhibition, Dallas, Texas, 9–12 October. SPE-95287-MS. http://dx.doi.org/10.2118/95287-MS.

Kakadjian, S., Thompson, J., Torres, R., et al. 2013. Stable Fracturing Fluids from Waste Water. Presented at the SPE Unconventional

Resources Conference Canada, Calgary, Alberta, Canada, 5–7 November. SPE-167175-MS. http://dx.doi.org/10.2118/167175-MS.

LeBas, R.A., Shahan, T.W., Lord, P. et al. 2013. Development and Use of High-TDS Recycled Produced Water for Crosslinked-Gel-Based

Hydraulic Fracturing. Presented at the SPE Hydraulic Fracturing Technology Conference, The Woodlands, Texas, 4–6 February.

SPE-163824-MS. http://dx.doi.org/10.2118/163824-MS.

Monreal, G.H., N’Guessan, O., and McElfresh, P. 2014. A New Generation of Efficient Fracturing Fluids for High Total Dissolved Solids

Applications. Presented at the SPE International Symposium and Exhibition on Formation Damage Control, Lafayette, Louisiana, 26–

28 February. SPE-168192-MS. http://dx.doi.org/10.2118/168192-MS.

Olsen, D.K., Weitner, M., Olson, D.C. et al. 2013. Smart Water Management as part of Supply Chain Logistics for Source Rock

Development. Presented at the SPE Middle East Intelligent Energy Conference and Exhibition, Manama, Bahrain, 28–30 October.

SPE-167454-MS. http://dx.doi.org/10.2118/167454-MS.

Patel, P.S., Robart, C.J., Ruegamer, M. et al. 2014. Analysis of US Hydraulic Fracturing Fluid System and Proppant Trends. Presented at

the SPE Hydraulic Fracturing Technology Conference, The Woodlands, Texas, 4–6 February. SPE-168645-MS.

http://dx.doi.org/10.2118/168645-MS.

Ramurthy, M., Barree, R.D., Kundert, D.P. et al. 2011. Surface Area vs Conductivity Type Fracture Treatments in Shale Reservoirs.

Presented at the SPE Hydraulic Fracturing Technology Conference, The Woodlands, Texas, 24–26 January. SPE-140169-MS.

http://dx.doi.org/10.2118/140169-MS.

Tipton, D.S. 2014. Mid-Continent Water Management for Stimulation Operations. Presented at the SPE Hydraulic Fracturing Technology

Conference, The Woodlands, Texas, 4–6 February. SPE-168593-MS. http://dx.doi.org/10.2118/168593-MS.

Weaver, J.D., Nguyen, P.D., and Loghry, R. 2011. Stabilizing Fracture Faces in Water-Sensitive Shale Formations. Presented at the SPE

Eastern Regional Meeting, Colombus, Ohio, 17–19 August. SPE-149218-MS. http://dx.doi.org/10.2118/149218-MS.

Yang, Y., Robart, C., and Ruegamer, M. 2013. Analysis of U.S. Hydraulic Fracturing Design Trends. Presented at the SPE Hydraulic

Fracturing Technology Conference, The Woodlands, Texas, 4–6 February. SPE 163875-MS. http://dx.doi.org/10.2118/163875-MS.

20 SPE-174118-MS

Ye, X., Tonmukayakul, N., Lord, P. et al. 2013. Effects of Total Suspended Solids on Permeability of Proppant Pack. Presented at the SPE

European Formation Damage Conference & Exhibition, Noordwijk, The Netherlands, 5–7 June. SPE-165085-MS.

http://dx.doi.org/10.2118/165085-MS.