Copyright 2014, Society of Petroleum Engineers This paper was prepared for presentation at the SPE Saudi Arabia Section Annual Technical Symposium and Exhibition held in Al -Khobar, Saudi Arabia, 21-24 April 2014. 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

Multistage acid fracture treatments are utilized in low-permeability carbonate reservoirs (permeability <10 md) to stimulate

the formation by creating highly conductive fractures in the formation and bypassing near wellbore damage. The fracture is

generated at high pressures that are required to break the rock open while using a viscous pad. The fracture is then kept open

by adding gelled or emulsified acid to create uneven etches on the surface of the fracture.

Pre-job acid fracturing treatment fluids' reaction and compatibility analysis in the laboratory are crucial as the operational

success is highly dependent on its chemicals’ reactions. The key problem with acid fracturing treatments is the difficulty in

appraising the actual downhole reactions and performance of the treatment chemicals within the heterogeneous rock. This

problem can be resolved when flow back fluids and the chemical ions are analyzed to understand the reactions that occurred

down hole. Also, since acid fracture treatments require pumping large volume of fluids, flowing back the entire fluids becomes

a challenge due to the low reservoir permeability and the associated reservoir rock capillary pressure effects.

This paper will discuss the pre-fracture treatment evaluation based on laboratory experiments - core flood, rock dissolving

capacity, and fluid compatibility in addition to comparing the expected chemical ion returns with the actual ions observed in

the flow-back fluids.

The results of this flow-back fluid analysis showed a recovery of 17% of the chemicals pumped during the treatment with a

stabilized production rate of 3 MBOPD. Further water analysis indicated the presence of 25-30% formation water while the

critical ions analyzed showed the effectiveness of the corrosion inhibitor package, acid system dissolving capacity, and

crosslinker fluid recovery. It is expected that this paper will provide a learning process for optimizing future multistage acid

fracture treatment in Saudi Arabia.

Introduction

LF Formation is the deepest of the upper Jurassic limestone reservoirs in the K-field. Underlying the two other carbonate

reservoirs, LF has been appraised as a relatively lower-quality formation with RQI ~ 0.13/FZI ~0.73 and a net to gross ratio of

~0.62. This reservoir can also be described as being heterogeneous and clayey limestone/dolomite with permeability range of 1

– 15mD and porosity range of 5 – 20% (Tiab and Donaldson 2004). The LF-reservoir is a 120 ft thick reservoir at a depth of

6,900 ft and BHST of 150 oF. Mineralogy of the core samples from this reservoir is mainly calcite with some dolomite and

ankerite. Table 1 shows the minerals in formation rock with their chemical composition

No improvement in productivity was notice with several attempts to stimulate wells in this reservoir. This is due to the tight

formation and low reservoir quality. The decision to conduct and evaluate a multistage acid fracturing operation in this

reservoir was considered highly important to assess the value of this completion design for developing this field. A multilateral

well with three laterals was drilled with a plan to create seven fractures in the mother-bore. The candidate well selected for the

multi-stage acid fracturing is a tri-lateral oil producer. The main wellbore is completed with the multi-stage frac (MSF)

completion, while the other two laterals are in open hole. The two other laterals were isolated by a blanking pipe installed and

connected between upper production packer and lower MSF completion tie-back receptacle. The pipe is used to ensure

isolation during the job and was removed after the operation was completed. In regards to the completion, both upper and

lower completion were equipped with VM-95 to withstand the pressure of the operation. All equipment exposed to the

fracturing operation in the wellbore was ensured to have a pressure rating of 10,000 psi. Also, it was ensured that sour service

piping was used to avoid any corrosion from possible H2S suspected to be in the crude from initial analysis. The completion

SPE-172246-MS

Optimization and Post-Job Analysis of the First Successful Oil Field Multistage Acid Fracture Treatment in Saudi Arabia

Tariq Almubarak, Mohammed Bataweel, Majid Rafie, Rifat Said, Hussain Al-Ibrahim, Mohammad Al-Hajri, Peter Osode, Abdullah Al-Rustum, Omar Aldajani, Saudi Aramco

SPE-172246-MS 2

tubular mainly contained Fe, Mn and Cr. The tubular were chosen to resist the expected sour environment of H2S from

corroding the tubular while producing the well. The Cr material in the tubular reacts with oxygen on surface and forms an

impermeable layer of Cr2O3. This layer protects the tubular from direct contact with H2S as shown in the equation below:

However, this layer is very sensitive to HCl and will be removed if it was in direct contact with it. A good corrosion

inhibitor package was used in this job in order to protect this type of chrome tubular from the multistage acid fracture

treatment chemicals. The Fe, Mn and Cr ion analysis in the flow back water will help us identify if there was any corrosion to

the tubular while conducting the multistage acid fracture job on this well.

The Pumping schedule to fracture this well uses crossed-linked gel to initiate the fracture followed with emulsified acid to

achieve deeper penetration in to the formation. This is followed with crossed-linked gel to maintain the bottom hole pressure,

control the leak-off rate and create a viscous fingering effect which alter the acid path in the fracture and ensure irregular

etched patterns along the fracture face. This is followed by gelled acid to break the crossed-link pad and create a different

etching path on the fracture face concluded with diversion pill. Fig. 1 illustrate sequence of pumping schedule used to acid

fracture this well. The same sequence of fluids will be repeated two times before injecting the final stage with closed fracture

acid (CFA). This acid stage is pumped below fracturing pressure to allow acid to deepen and widen existing channels, enhance

fracture conductivity and minimize fracture closure effect (Bartko et al. 2003).

Emulsified acid is a key stage in the sequence of fluids injected to create the fracture. It is a retarded acid system prepared

by acid-in-diesel emulsions (Fig. 2). Diesel acts as a diffusion barrier between the acid and the rock. de Roziere et al. (1994)

studied the effective diffusion coefficients in the emulsified acid and found it to be 3 order of magnitude lower than plain acid.

Due to the lower reaction rate the live acid can penetrate deeper in to the formation creating more effective wormholes.

Broaddus et al. (1968) showed that emulsified acid provided excellent etching and better fracture conductivity than regular

acid. They found that better fracture conductivity can be achieved by injecting different acid solutions with different degrees of

retardation. They proposed that the most retarded acid should be injected first.

Bartko et al. (2003) studied the impact of acid type and formation lithology on fracture performance by evaluating field

production data, flow back analysis and lab work. They found that emulsified acid shows superior performance in calcite and

the more challenging dolomitic rocks. Fracture conductivity is achieved in acid fracturing by generating non-uniform rock

dissolution at the fracture face. The degree of heterogeneity on the fracture face due to physical and chemical properties of the

formation rock will influence the reaction rate and enhance differential etching (Anderson and Fredrickson 1989). Another

factor that can affect the etching pattern is the difference in permeability and porosity by variation in acid leak-off rates

(Anderson and Fredrickson 1989). Viscous fingering by different acids injected inside the fracture pre-flushed with cross-

linked spacer will help to further enhance etching (Fig.1).

Researchers reported chemical analysis of flow back samples after acid treatments in sandstone (Gdanski and Peavy.1989;

Almond et al. 1990; Shuchart 1995 and Nasr-El-Din et al. 1996). Nasr-El-Din et al. (1999) and Mohammed et al (1999)

examined acid return samples from carbonate reservoirs. Chemical analysis of the flow back samples collected during well

production is used to assess the performance of the acid fracturing treatment. During acid placement and soaking in the

wellbore, acid will interact with formation rock and down-hole hardware. Analyses of concentration of key ions in the flow

back samples, collected during wellbore cleanup, are carried out to identify the source of these ions (formation or hardware).

In carbonate reservoirs high concentrations of calcium and magnesium is expected to exist in flow back samples as a product

of reaction of acid with formation rock. The following equations show the reactions between HCl/ Calcite and HCl/Dolomite:

From these equations we can clearly indicate that Ca, Mg and Cl ions are going to be in our flow back water analysis. Other

information that can be gained using the flow back analysis is the integrity of the wellbore hardware and if any sever corrosion

to metal components happened because of interaction with acid solutions. This can be assessed by monitoring the key ions

(total iron, chrome, nickel, molybdenum) that exist in the metal alloys used to build the downhole tubular. Third important

information that can be obtained from this analysis is the flow back of live acid. This is important to protect downhole

hardware from sever corrosion due to the presence of live acid in the flow back. Also, this will help in optimizing the soaking

time and over-flush volumes to allow acid to fully interact with rock till acid is totally spent. The presence of live acid can be

detected using pH and acid concentrations.

The objectives of this study were to: (1) evaluate the impact of breakers and acid stages on the viscosity of the cross-linked

gel. (2) assess the emulsified acid / rock interaction and its propagation in tight cores (3) evaluate acid stimulation treatments

based on the chemical analysis of the return fluids (4) investigate the amount of corrosion in the wellbore hardware.

CaCl2 + CO2+ H2O 2HCl + CaCO3

CaCl2 + MgCl2 CO2+ H2O 4HCl + CaMg(CO3)2

4Cr + 3O2 2Cr2O3

SPE-172246-MS 3

Procedure and Experimental Work

CEA Stability

One of the main chemical systems in this multistage acid fracture operation is the emulsified acid. The emulsified acid has

certain properties that help us generate efficient reaction rates and deep etching that are very critical in this job. The emulsified

acid consists of acid, diesel and emulsifier. The most important criteria is to keep the emulsified acid stable for the amount of

time needed to create etches and wormholes in the formation rock. In addition, it is very important for the emulsified acid to be

able to break down to its original components for an easy flow back operation. Tests were conducted by mixing the emulsified

acid then adding chunks of the reservoir's core sample to observe the reaction under reservoir conditions. The goal was to

observe the time it took the emulsified acid to separate into diesel and acid. In addition, it is critical to check if any portion of

the emulsified acid was stabilized after the reaction due to the fines in the reservoir and therefore would hinder the flow back

operation.

Viscosity Measurements

The HPHT viscometer was used to measure the apparent viscosity of the fracturing fluid samples under different shear

rates and constant pressure at reservoir temperature. The HPHT viscometer utilizes standard R1/B5 bob and rotor which

require a sample volume of 52 cm3. The viscometer uses sliding carbon block for dry heating and a temperature sensor is

mounted on the stator/bob to measure the sample’s temperature. A pressure of 1,000 psi was applied to minimize evaporation

of the sample, and to keep all generated gases in liquid state.

Viscosity measurements were performed under different shear rates to simulate the flow of the fracturing fluid through

production tubular, perforations and inside the created fracture. The goal of this test was to measure the time needed for the

crosslinked fluid to break so it can flow back easily during the flow back operation.

Gel Stability

The crosslinker system is a key component in this multistage acid fracturing operation. The crosslinker builds high

viscosity and breaks open the rock. In addition, the crosslinker fluid keeps the fracture open while maintaining high viscosity

to enable the acids and chemicals to flow inside and react with the rock. Furthermore, it enable the fluids to viscously finger

through and thereby randomly etching the rock which results in high conductivities that help with production. Moreover, the

high viscosity helps with controlling the acid leak off thereby generating deeper etching and fractures. The critical factor

though is to remove the viscous fluid after the operation is completed and to unplug the crosslinker filter cake from the

permeable reservoir areas. We always add breakers to crosslinked fluids in order to lower the crosslinker's viscosity and flow it

back easily. However, the recovery is always ranging between 30-50% and we always have speculations about how much fluid

was broken. As a backup analysis, tests were run to see the interactions between the viscous crosslinker and the different

chemicals pumped in the fracture. The tests were conducted under reservoir conditions. The goal of this test was to observe

how long it would take the fracture fluid to break down after being exposed to all the chemicals in addition to the internal

breaker.

X-Ray Diffraction (XRD)

X-Ray Diffraction (XRD) was used on core samples to gain knowledge about the mineralogy of the reservoir rock. The

samples were crushed to fine powder using a mill. The clay size fractions of the sample are separated and dried using air on a

glass slide. The air dried glass slide was glycolated in a desiccator containing ethylene glycol at 60° C in the oven. The core

samples were analyzed by X-Ray Powder Diffraction (XRD). The identification of the crystalline phases were analyzed.

Subsequent semi-quantification of XRD data was done using the Rietveld Refinement method. The clay-size material (<2

microns equivalent spherical diameter) was separated from the larger size particles by sedimentation techniques.

Environmental Scanning Electron Microscope (ESEM)

Micro-structural characterizations, in terms of porosity, pore size, presence of clay and foreign materials in the pores are

important in understanding the behavior of reservoirs. In addition such investigation will help in selecting the right acidizing

treatment for the formation in order to increase productivity.

In this test the Environmental Scanning Electron Microscope (ESEM) analytical techniques with integrated ultra-thin window

energy dispersive X-ray detector were utilized to perform comprehensive microstructural characterizations of the core samples

in this reservoir.

The ESEM/EDS data are required in order to identify the minerals in the core, any materials blocking the pore space, type

of cementing materials and also the elemental compositions of the core plug samples.

In addition, ESEM assesses any formation damages that may have been caused by the chemical treatments and subsequently

resulted in a reduction in productivity.

The primary goal in this test was to identify the main components of the reservoir core samples and correlate them with the

XRD results in order to have a better understanding of the acid/rock reactions during this multistage acid fracturing job.

In addition, a second set of cores were previously exposed to a damaging drilling mud filtrate and were tested in the (ESEM)

in order to show the magnitude of the clay swelling damage on the core pore throats.

SPE-172246-MS 4

Rock Solubility

Rock solubility is very important to know in the multistage acid fracture operation because the whole mechanism of this

stimulation treatment relies on acid reactions. We ran several tests by dissolving core plugs in the different HCl acid recipes

(EA, GA, SA) and concentrations being used on site. The goal of this test was to measure the amount of rock that will be

dissolved before flowing back the fluids. From this test we can decide whether the operation time would be sufficient to spend

the acid in order to protect the tubular from corrosion while flowing back. This test was conducted under reservoir condition

and graphs were generated to illustrate the solubility capabilities of the acid system with the reservoir rock.

Core Flood

A core flood apparatus was designed and built to simulate fluid flow in porous media in the reservoir. Positive

displacement Equipped with a programmable controller was used to deliver fluids at constant flow rates at variable speeds up

to 200 cm3/min and pressure up to 10,000 psi. The pump is connected to two accumulators to deliver brine or chemical

solutions. Accumulators with floating pistons rated up to 3,000 psi and 250o F were used to store and deliver fluids. A set of

valves were used to control the injected fluid into the core sample. The core-holder can accommodate a core plug with

diameter of 1.5 inches and a length up to 3 inches. Pressure transducers were used to measure the pressure drop across the

core. A back pressure regulator was used to control the flowing pressure downstream of the core. A second back pressure

regulator was also used to control the confining pressures on the core plug. A convection oven was used to provide

temperature controlled environment. A data acquisition system was used to collect data from the pressure transducers.

Below is the procedure used to prepare the core for the core flooding experiments:

1. Core samples were dried overnight in a 100 °C oven

2. Core samples were loaded into the core holder and confining pressure was applied

3. 6% KCl brine was injected in the core to establish ~ 100% water saturation

4. Base permeability to water were measured using different flow rates (0.5 and 1 cm3/min)

5. Emulsified acid was pumped to create worm holes in the core

CT scan

X-Ray Computer Tomography (CT scan) was used on the cores after running the core flood experiments. By applying the

X-Ray on the core we were able to determine the density distribution and therefore track the wormhole propagation created

after flooding the core with HCl acid recipes. In addition, this type of test would help in understanding the distribution of

minerals in the rock and how the acid will react. From this analysis we can anticipate the type of ions we expect from the

reactions in our flow back water analysis. The results were then processed to achieve a two color scale in order to magnify the

wormhole propagation in the core.

Flow back Analysis

Fluid samples were collected and concentrations of key ions were analyzed. Calcium, Magnesium, Iron, Chrome, Nickel,

and Molybdenum concentrations were measured using Inductively Coupled argon Plasma emission spectroscopy (ICP).

Chloride ions were measured by titration with 0.1N Silver Nitrate solution using a auto-titrator. To measure pH, a pH meter

was used.

Results and Discussion

Emulsified Acid Stability

The emulsified acid stability test results are illustrated in Fig. 3. The results indicate that the emulsified acid is going to

break into 3 layers after 75 minutes at room temperature following the reaction with the core. The main two layers are diesel

and HCl acid. However, the third layer is a portion of the emulsified acid that has not separated completely. This third layer

could happen due to fines stabilizing the emulsion and could cause some reactions in unwanted areas of the

completion/reservoir.

The same test has been conducted under reservoir condition and the results show a clear 2 layer of diesel and spent acid

after incubation time of 30 minutes. These two layers can easily flow back without issues during the flow back operation. The

Ca and Mg are the major ions that are going to be available in the acid portion of the emulsified acid after it separates. The

separation of acid from the emulsion increases corrosion rate to the tubulars if the acid was not totally spent due to short

soaking time.

Viscosity of Crosslinked Pad

Fig. 4 shows the effect of the internal breaker concentration on the viscosity of the crosslinked gel. The test was run for the

following concentrations of internal breaker 0.1, 0.25 and 0.5 gpt. The results indicate that increasing the concentration of the

SPE-172246-MS 5

internal breaker reduces the gel breaking time from 90 to 15 minutes. This also resulted in reducing the gel viscosity to less

than 500 cP at reservoir temperature. The flow back operation was planned to start after 1 week giving the breakers enough

time to react and break the crosslinked gel for fracture cleanup.

Since the acid stage was injected after each crosslinked pad, the effect of mixing the acid system with crosslikned gel was

examined. Fig. 5 shows the effect of mixing crosslinked pad (without breaker) with emulsified acid at 1:1 ratio. The results

indicate that the emulsified acid was able to break down the viscosity and allowed easier cleanup of the fracture. Many field

operations have trouble flowing back the crosslinked fluid because it doesn’t break completely even though it had adequate

internal breaker concentrations. In addition, these results are comforting and show a positive impact since it breaks the

crosslinked fluid considering the tough to remove filter cake generated on the fracture face. These results will definitely help

in flowing back the fracture fluid more easily.

Mineralogical and Rock Analysis

The XRD results showed that the samples consisted of carbonate minerals (calcite and ankerite) with minor quantities of

clay minerals (kaolinite, I-S and illite), iron sulfide (pyrite) and sand (quartz) in some of the samples. Table 2 gives the bulk

mineralogical composition of 7 core plugs from this reservoir. The data showed that calcite was the most dominant mineral in

the samples with wt.% ranges between 68 and 99. The Second dominant mineral is ankerite and dolomite with 1 – 18 wt. %.

Clay minerals were detected in some of the cores and reached up to 14 wt. % in the extreme case and no clay exist in some

core plugs. Total clays of mixed layers illite-smectite, illite and kaolinite were also detected in all these samples. Miner

quantity of potassium iron sulfide was also founded in the samples.

Fig. 6 & 7 shows the environmental scanning electron microscope ESEM images confirming the presence of CaCO3

mainly in the reservoir core sample. From this analysis we can safely say that CaCO3 will be the main component to react with

HCl and we expect large amount of calcite reaction product ions in the flow back water analysis. In addition, Fig. 7 shows the

distribution of clays on the pore throat in the reservoir core sample. This indicates the importance of using a clay control agent

in this multistage acid fracture job in order to protect the pore throats from plugging, thus, blocking the flow back and

production.

Rock / Acid Interaction

Results of rock solubility vs. time are shown in Fig. 8. The results indicate that the emulsified acid was able to dissolve

around 90% of the core in 1 hours with no precipitation. After 2 hours the graph starts to plateau which indicates that no

further reaction is happening and the acid was spent. This above result is used to design the field operation and recommend the

minimum required soaking time for acid to spent and prevent flow back of live acid. These results indicate that we are

expecting Ca, Mg from the calcite and dolomite reactions in the flow back analysis. Moreover, according to the XRD analysis

Mn and Fe are also expected due to the presence of ankerite in the core sample.

Fig. 9 shows pressure drop as function of injected pore volumes of brine and emulsified acid in the core flooding

experiment using core samples from the target reservoir. The figure initially shows the pressure drop resulting from pumping

KCl. Then, the figure shows the pressure drop resulting from pumping the emulsified acid. The emulsified acid was more

viscous that KCl so it resulted in a higher pressure drop while keeping the rate parameter constant. The results also indicate

that the emulsified acid required was less than 0.5 pore volume to generate a wormhole and break through the core.

The CT scan results are shown in Fig. 10. The results indicate that the minerals are distributed all over the core with no

specific pattern. Therefore, the core is a heterogeneous sample of the reservoir. The worm hole generated from the emulsified

acid is favorably dissolving the calcite minerals compared to the dolomite and ankerite. This will promote differential etching

phenomenon due to mineralogical difference which is a favorable attribute for effective fracture conductivity. Second

observation indicates no face dissolution as can be shown in Fig. 10.

Flow back Analysis

According to rock mineralogy by XRD and ESEM for the core samples from LF formation, we expect the acid-rock

interactions to yield Ca, Mg mainly with some Fe and Mn. Table 3 shows a summary for the ions in the mixing and formation

water. In addition, Table 4 shows the expected ions in the flow back analysis.

Dilution effect by mixing of formation brine with stimulation mixing water

To calculate the mixing coefficient (A) that result from mixing formation brine and the water used in mixing the

stimulation fluid, the following equation was used:

( ) Where: CIFlo= Concentration of ions in the flow back water

CIFor= Concentration of ions in the formation water

CIMix= Concentration of ions in the mixing water

SPE-172246-MS 6

The key ions to apply the above calculation are ions that exist in formation brine or mixing water and are not a byproduct of

treatment fluid reactions with the formation rock/wellbore materials. Since the formation water was the only source for

strontium, it was used to calculate the dilution coefficient for the other ions. Fig. 11 shows dilution coefficient as function of

time that was calculated from the Sr concentrations.

Reaction Products in Flow back Samples

Fig. 12 shows the calcium concentration in the flow back water as a function of time. The amount of Ca in the flow back

water is way higher than the Ca in both the mixing water and the formation water. This is an indication that the reaction

between the HCl acid and the calcite in the formation rock has occurred. Fig. 12 shows the ratio of Ca in the flow back water

that came from the formation brine and the reaction of HCl with the formation. The results indicate that we have around

10,000 mg/L calcium in the flow back water coming from the formation water and 40,000 mg/L as a result of the HCl acid

reacting with the reservoir rock. This large amount of calcium from the reaction was expected according to our dissolving test

results in the lab section.

Fig. 13 shows the magnesium concentration in the flow back water analysis in comparison to mixing water and formation

water Mg amounts. The amount of Mg in the flow back water is higher than the Mg in the mixing water but lower than the Mg

in the formation water. This can be an indication that the reaction between the HCl acid and the dolomite in the formation rock

has occurred. Applying the dilution factor to find the calculated Mg concentration can be used to differentiate the source of

magnesium ions. Fig .13 shows the ratio of Mg in the flow back water that came from the formation brine and the reaction of

HCl with the formation rock in the reservoir. The results indicate that we have around 800 mg/L magnesium in the flow back

water coming from the formation water and 2,200 mg/L as a result of the HCl acid reacting with the reservoir rock. This large

amount of Magnesium from the reaction was expected according to our dissolving test results in the lab section.

Fig. 14 shows a trend that indicates that both Ca and Mg are generating from the same source due to the similar slope at

late time. In this case our source is the formation rock. It was also noticed that the trend is decreasing towards the end of the

flow back operation which indicates less reaction with the formation rock.

Fig. 15 shows the pH value as function of time in the flow back analysis in comparison to mixing water pH values. The pH

value in the flow back water is 5 compared to a pH value of 7 in the mixing water. The reason behind that is the reaction of

HCl acid with the formation rock. The reaction equation yields H2O and CO2 and therefore forms carbonic acid which lowers

the pH values of spent HCl acid to around 4.5-5 pH. This is critical to know because at this pH the acid is fully spent and ready

for flow back. In addition, knowing this pH value indicates that the flow back water is going to be acidic and will requires a

good corrosion inhibitor package in order to protect our tubular during the flow back operation.

Corrosion Products in Flow back Samples

Fig. 16 shows the Iron concentration in the flow back analysis as function of time. The amount of Fe in the flow back water

is higher than both the Fe in the mixing water and the formation water. Fe ions in the flow back samples, can be generated

from two sources; reaction byproduct of ankerite with acid or corrosion to wellbore tubulars. To confirm the presence of

corrosion, analysis of Mn and Cr ions was required. Fig. 16 shows the ratio of Fe in the flow back water that resulted from the

formation water, the reaction of HCl with the formation ankerite rocks or the corrosion of the tubular due to the reaction with

HCl. The results indicate that we have around 100 mg/L iron in the flow back water coming from the formation water and 300

mg/L as a result of the HCl acid reacting with the ankerite in the reservoir rock or causing minor corrosion to the tubular. This

small amount of Fe is an indication of minor corrosion in the tubular.

Fig. 17 shows the Mn concentration in the flow back water analysis. The flow back water analysis showed some returns of

Mn ions. The ions concentrations were very low. This could be an indication of minimal corrosion. In addition, Mn ions could

be coming from the ankerite's reaction with HCl. To confirm the presence of corrosion the Cr, Ni, and Mo analysis was

conducted. Fig. 18 shows that Fe and Mn ions following the same trend and increasing at the very end of the flow back

operation. This is an indication that both Fe and Mn are originating from the same source. However, this source is not similar

to the calcium’ and magnesium’s source.

Analysis of Cr, Ni, and Mo in the flow back samples was conducted to confirm the presence of corrosion. These analyzed

ions are part of the construction material of tubular. Fig. 19 shows concentration of Cr, Ni, and Mo as function of flowing

time. There was no indication of these ions in the flow back samples. They were all below the detection limit of 1 mg/L. This

finding eliminates the possibility of corrosion in the completion hardware.

Formation Water Ions

Fig. 20 shows the Strontium concentration as function of time. The amount of Sr in the flow back water is higher than the

mixing water but lower than the formation brine. This is indicating that formation brine is part of the flowing back fluid.

Fig. 21 and 22 show Potassium and Chloride concentrations as a function of flow back time. The presence of K and Cl ions

is usually helpful in confirming the % of formation water in the flow back water. However, fluids containing KCl and HCl

were pumped during this job and that will mislead the formation water % calculation.

H2CO3 H2O + CO2

SPE-172246-MS 7

Fig. 23 shows the Sodium concentration in the flow back water analysis in comparison to mixing water and formation water

Na amounts. The amount of Na in the flow back water is a clear indication of formation water in our flow back water. Sodium

was one of the ions that were not included in our pumped chemicals.

The dilution factor is then calculated by dividing the flow back water Na amount by the formation water Na amount at

different times of flow back. This yields a 25-30% of formation water in the flow back water at late time which corresponds to

the Sr analysis done above.

Flow back Rates

The main objective of flow back is to ensure maximum treatment fluid returns and successful operation. The capability of flow

is a major indicator of success. However, in this particular job and field designs for flow require artificial lift for pressure

support. The well was design to flow using an electric submersible pump and expected to flow at a rate of 2.0 MBOD. The

results from the flow back observed following treatment were very promising. The well was able to flow for 8 hrs before the

rate was deemed too low for treatment fluid recovery. Following that, the well was artificially lifted using nitrogen pumped

through coiled tubing at the bottom. The flow back showed that we were producing at an average rate of around 3.5 MBOD as

seen in Fig. 24. In addition, we were able to recover 17% of the treated chemicals pumped as shown in Table 3. Further water

analysis indicated the presence of 25-30% formation water while the critical ions analyses showed the effectiveness of the

corrosion inhibitor package, acid system dissolving capacity, and crosslinker fluid recovery Also, it is expected that for future

jobs, optimization will be conducted in regards to increasing the percentage of treatment fluid recovered.

Conclusions

Based on the results from experimental work and flow back analysis the following conclusions can be drawn:

1) Viscosity of the cross-linked gel was reduced by the internal breaker and interaction with injected acid stages.

2) Emulsified acid was efficient in generating wormholes with less than one pore volume of injection in the core

flooding experiment conducted on cores from LF formation.

3) Propagation of wormhole in the core samples is sensitive to rock mineralogy as shown in the CT scan images due to

different reaction rate of acid with different carbonate minerals.

4) Flowed back acid is fully spent due to extended soaking time and proper displacement.

5) No indication of corrosion to downhole completion by the flow back analysis results.

SPE-172246-MS 8

Acknowledgements The authors would like to thank the management of Saudi Aramco and Halliburton for permission to publish this paper.

References Al-Anazi, H.A., Nasr-El-Din, H.A, and Mohamed, S.K. 1998. Stimulation of Tight Carbonate Reservoirs Using Acid-in-

Deisel Emulsions: Field Application. Paper SPE 39418 presented at the International Symposium on Formation

Damage Control, Lafayette, La, 18-19 February.

Almond, S.W., Brady, J.L. and Unerdown, D.R. 1989. Return Fluid Analysis From the Sadlerochit Formation, Prudhoe Bay,

Alaska: Field Study – Part 1. Paper SPE 18223 presented at SPE Annual Technical Conference and Exhibition,

Houston,Tx.

Anderson, M.S., and Fredrickson, S.E. 1989. Dynamic Etching Tests Aid Fracture Acidizing Treatment Design. SPE

Production Engineering 4 (4): 443-449. SPE 16452.

Brakto, K.M., Nasr-El-Din, H.A., Rahim, Z. et al. 2003. Acid Fracturing of a Gas Carbonate Reservoir: The Impact of Acid

Type and Lithology on Fracture Half Length and Width. Paper SPE 84130 presented at SPE Annual Technical

Conference and Exhibition, Denver, Co, 5-8 October.

Broaddus, G.C., Knox, J.A., and Fredrickson, S.E. 1968. Dynamic Etching Tests and Their Use in Planning Acid Treatments.

Paper SPE 2362 presented at the SPE Oklahoma Regional Meeting, Stillwater, Oklahoma, 25 October.

De Rozieres, J., Chang, F.F, and Sullivan, R.B. 1994. Measuring Diffusion Coefficients in Acid Fracturing Fluids and Their

Application to Gelled and Emulsified Acids. Paper SPE 28552 presented at SPE Annual Technical Conference and

Exhibition, New Orleans, 25-28 September. DOI:10.2118/28552-MS.

Gdanski, R. 2001. Modeling Acid Returns Profiles After HF Acidizing Treatments. Paper SPE 65035 presented at the

International Symposium on Oilfield Chemistry, Houston, Tx, 13-16 February.

Mohamed, S.K., Nasr-El-Din, H.A. and Al-Furaidan, Y.A. 1999. Acid Stimulation of Power Water Injectors and Saltwater

Disposal Wells in a Carbonate Reservoir in Saudi Arabia: Laboratory Testing and Field Results. Paper SPE 56533

presented at the SPE Annual Technical Meeting, Houston, Tx, 3-6 October.

Nasr-El-Din, H.A., Al-Anazi, H.A. and Mohamed, S.K. 1999. Stimulation of Water Disposal Wells Using Acid-In Diesel

Emulsion: Case Histories. Paper SPE 50739 presented at the International Symposium on Oilfield Chemistry,

Houston, Tx, 16-19 February.

Nasr-El-Din, H.A., Al-Driweesh, S., Al-Muntashiri, G.A. et al. 2003. Acid Fracturing HT/HP Gas Wells Using a Novel

Surfactant Based Fluid System. Paper SPE 84516 presented at the SPE Annual Technical Conference and Exhibition,

Denver, Co, 5-8 October.

Nasr-El-Din, H.A., Rosser, H.R. and Hopkins, J.A. 1996. Simulation of Injection Water Supply Wells in Central Arabia. Paper

SPE 36181 presented at the 7th

Abu Dhabi International Petroleum Exhibition and Conference, Abu Dhabi, UAE, 13-

16 October.

Rahim, Z., Al-Anazi, H.A., and Al-Kanaan, A. 2013. Selecting Optimal Fracture Fluids, Breaker System, and Proppant Type

for Successful Hydraulic Fracturing and Enhanced Gas Production – Case Studies. Paper SPE 163976 presented at

the SPE Unconventional Gas Conference and Exhibition, Muskat, Oman, 28-30 January.

Shuchart, C.E. 1995. HF Acidizing Returns Analyses Provide Understanding of HF Reactions. Paper SPE 30099 presented at

the SPE Formation Damage Symposium, The Hague, Netherlands, 15-16 May.

Taylor, K.C. and Nasr-El-Din, H.A 2000. Flowback Analysis of Acid Stimulation of Seawater Injection Wells: Case Histories.

Paper SPE 58736 presented at the International Symposium on Formation Damage Control, Lafayette, La, 18-19

February.

SPE-172246-MS 9

TABLE-1 CHEMICAL COMPOSITION OF MINERALS

Calcite Ankerite Dolomite Illite Koalanite Quartz Pyrite

CaCO3 Ca(Fe,Mg,Mn)(CO3)2 CaMg(CO3)2 (K,H3O)(Al,Mg,Fe)2(Si,Al)4O10[(OH)2,(H2O)] Al2Si2O5(OH)4 SiO4 FeS2

TABLE-2 XRD RESULTS

Plug Calcite Ankerite Illite + IS Kaolinite Quartz Pyrite Grain

Density

# wt% wt% wt% wt% wt% wt% g/cm3

Well-88

2 68 18 7 3 3 1 2.75

49 82 11 3 2 1 1 2.75

Well-37

813H* 89.7 9 0.2 0.8 0.3 - 2.75

815H 98.9 0.9 - 0.2 - - 2.75

829H 97.3 1.2 0.2 1.2 0.1 - 2.75

803H 96 4 Trace Trace - - 2.75

837H 98 2 Trace Trace - - 2.75

*Microline - Trace

TABLE-1 COMPARISON BETWEEN MIXING WATER AND FORMATION WATER

Parameter Ca Mg Cl Fe Mn K Na Sr SO4

Mixing water 217 55 271 5 0 15 171 3 622

Formation water 37000 6700 122500 30 0 1000 34000 1400 480

TABLE-2 MAIN IONS IN FLOW BACK WATER

Parameter Ca Mg Cl Fe Mn K Na Sr SO4

Calcite/Ankerite

Chemicals/Mixing water

Formation water

Tubular

SPE-172246-MS 10

TABLE-3 PUMPED CHEMICALS

Parameter Stage

1 Stage

2 Stage

3 Stage

4 Stage

5 Stage

6 Stage

7 Treated water 13,896 0 11,138 10,206 10,212 10,221 9,341

Pad 19,192 0 18,566 19,161 18,981 19,116 19,204

CEA 17,247 0 17,807 17,985 17,985 17,239 17,851

CSA 14,590 0 17,145 17,516 26,981 18,175 18,050

CFA 6,199 0 4,249 4,932 5,002 4,875 4,506

Diverter 4,070 0 3,964 4,035 4,044 4,113 3,830

Pumped volume, gal 75,194 0 72,869 73,835 83,205 73,739 72,782

Total, gal 451,624

Recovered water, gal 77,584

Recovered, % 17%

Figure 1–Chemicals pumped in order in each stage and viscous fingering by the acid in the fracture

Figure 2–Emulsified acid properties and components expected in the flow back

SPE-172246-MS 11

Figure 3–Emulsified acid reaction stability test in the lab

Figure 4–Viscosity measurements of the low temperature instant crosslinker with different breaker concentrations

Figure 5–Crosslinker stability test when mixing with emulsified acid and no internal breaker

SPE-172246-MS 12

Figure 6–ESEM for the core sample taken from the formation rock indicates abundant calcite

Figure 7–ESEM shows possible damage caused by clay swelling indicating the importance of a clay control chemical and ESEM

pore throat picture showing the possible clay damage due to not using clay control agent, thus hindering the flow back

SPE-172246-MS 13

Figure 8–Emulsified acid dissolving capacity and solubility test

Figure 9–Emulsified acid core flooding test shows that the emulsified acid’s worm hole needed less than 1 pore volume to break

through

SPE-172246-MS 14

Figure 10–Core sample CT scan after flooding it with emulsified acid to characterize the wormhole

Figure 11–Ratio of formation water in the flow back water using sodium and strontium ions as base comparison

SPE-172246-MS 15

Figure 12–Portion of calcium ions generating from the formation rock reactions or the formation water mixing with the flow back

water

Figure 13–Portion of magnesium ions generating from the formation rock reactions or the formation water mixing with the flow

back water

SPE-172246-MS 16

Figure 14–Calcium and Magnesium trends are similar and decreasing towards the end time indicating that they are generating from

the same source which is the formation rock reactions

Figure 15–Flow back water pH analysis indicates presence of spent acid only

SPE-172246-MS 17

Figure 16–Portion of total iron ions generating from the ankerite reactions/corrosion or the formation water mixing with the flow

back water

Figure 17–Flow back water manganese ion analysis confirms indicates minimal corrosion

SPE-172246-MS 18

Figure 18–Iron and manganese trends are similar at the end but different from the Ca and Mg trends indicating that they are

generating from a different source

Figure 19–Cr, Ni and Mo concentration are very low which indicates that corrosion of the tubular didn't occur during this

multistage acid fracture job

SPE-172246-MS 19

Figure 20–Flow back water strontium ion analysis confirms possibility of formation water in the flow back

Figure 21–Flow back water potassium ion analysis indicates possibility of formation water in the flow back

SPE-172246-MS 20

Figure 22–Flow back water chloride ion analysis indicates possibility of formation water in the flow back

Figure 23–Flow back water sodium ion analysis indicates possibility of formation water in the flow back

SPE-172246-MS 21

Figure 24–Oil and water flow back rates during cleanout