SPE 121757

Rheological Characterization of Novel, Delayed-Transition Metal

Crosslinked Fracturing Fluids: Correlation With First Field Applications
Andrey Mirakyan, Carlos Abad, Mike Parris, Yiyan Chen, and Fred Mueller, Schlumberger

Copyright 2009, Society of Petroleum Engineers

This paper was prepared for presentation at the 2009 SPE Annual Technical Conference and Exhibition held in New Orleans, Louisiana, USA, 4–7 October 2009.

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
The extent of crosslinking produced by a polymeric fracturing gel can significantly contribute to the success or failure of a
hydraulic fracturing treatment. In certain cases, excessive crosslinking while the fluid is in the tubulars can result in friction
pressures that are too high, and these may prohibit the treatment from achieving the design goals. With titanium (Ti) or
zirconium (Zr) crosslinked gels, which are prone to irreversible shear degradation, early crosslinking in the tubulars can
substantially reduce the final gel strength, even to the degree that near-wellbore proppant transport is compromised, and the
treatment may screen out. On the other hand, a Ti or Zr crosslinked gel that crosslinks substantially after exiting the
perforations may not have sufficient proppant-transport capability to complete the treatment successfully.

Varying treatment conditions such as mix-water composition and temperature, tubular shear rate and transit time, and
reservoir temperature pose difficult challenges to routinely achieving the optimal crosslinking state. Conditions such as
offshore wellbore temperature and viscous heating in smaller-diameter tubing can further complicate the task. Various
chemical means have been employed to tune the crosslinking temperature for specific treating conditions. These methods
involve addition of chemicals to control the crosslinking kinetics, such as pH buffers; adjustment of crosslinker
concentration; and competing ligands to temporarily bind the metal crosslinker. Blended crosslinkers containing a single
Group 4 (Ti, Zr) metal with different organic ligands with different binding strengths have been employed to achieve
crosslinking at two temperature ranges so that early viscosity for proppant transport can be developed. Mixed metal
crosslinkers such as aluminum (Al) and Zr have also been formulated for this same purpose.

This paper examines the disadvantages of these strategies, and describes the development and application of a high-
temperature fracturing fluid system that has overcome the issues described.

Introduction
Achieving the design fluid viscosity at in-situ conditions is critical for a number of reasons. Fracture initiation, propagation,
and resulting dimensions are strongly influenced by the viscosity. Proppant transport, also determined by fluid properties, is
necessary for fracture areal coverage and helps determine well performance.

Overcoming the various demands on a gel formed from a single organometallic crosslinker is difficult to achieve (Fig. 1).
Ideally, the crosslinking could be triggered instantaneously just after the fluid exits the perforations, avoiding high friction
pressure and degradation resulting from shear in the tubulars and perforations. Crosslink the fluid too early and the friction in
the tubulars will rise significantly and, as is shown later in the paper, a significantly crosslinked fluid cannot be sheared
without permanent damage to the polymer molecular weight, and therefore to the fluid viscosity. Low initial viscosity from
crosslinking too late may result in a narrow near-wellbore fracture that will impede proppant transport. Crosslinking a single
organometallic system either too early or too late risks premature screenout (Almond 1984; Nolte 1988; Walser 1988).

Gelling at varying temperatures can be achieved by delivering a crosslinker that has two (or more) organometallic complexes:
one that can yield limited crosslinking at a lower temperature for initial proppant transport, and one for the higher
temperatures in the fracture. This approach will manage friction pressures, but the gel will still suffer irreversible degradation
because of the strong metal bonds existing during high shear in the tubulars.
2 SPE 121757

Boron is known to produce labile chemical crosslinks that can break under shear but reform upon its cessation. If properly
formulated, the viscosity recovery from boron crosslink after shear occurs rapidly, supporting proppant transport near the
wellbore. When boron is combined with a compatible delayed organometallic in a single crosslinker package, a gel can be
produced that achieves the required balance between high-temperature and low-temperature fluid performance.

Additional considerations in developing a fluid system that can routinely be delivered successfully include operational
complexity, defined by the number of chemical additives, and the tolerance of the fluids to variations of the mix-water.
Often, organometallic crosslinked fluids require a multiplicity of additives, which often results in complex formulations, and
at times error-prone pumping schedules. Mix-water compositional variations, such as water hardness and bicarbonate
content, can strongly affect the rate and extent of the organometallic crosslinking. These routine variations often require fluid
formulation changes for each treatment.

Experimental Methods and Equipment

Various laboratory methods, techniques, and equipment have been used in the development of a new shear-tolerant, high-
temperature fracturing fluid. Some of these methods are listed below:
• High-temperature viscosity was measured in Grace 5600 (Fann50 geometry) rheometers, and shear-history tests
were carried out following the ISO 13503-1 standard in a shear history loop.
• Low-shear steady viscosity was measured using cup and bob geometry in a temperature-controlled rheometer.
• Dynamic rheology was measured using cup and bob geometry in a Bohlin Gemini.

Crosslinked Fluid Shear Sensitivity

Group 4 metal crosslinked polymeric gels have been used for fracturing since the early 1980s. Initially the fluids did not
have a crosslink delay, and the performance suffered (Nolte 1988). To lessen shear sensitivity, chemical methods to delay the
crosslinking were developed, and the industry agreed upon testing standards that became API RP 39. This recommended
practice included shear tolerance in the fluid test protocols and later became ISO 13503-1.

Delayed Zr crosslinkers were developed (Almond 1984; Hodge 1987; Putzig and Smeltz 1989) that improved the shear
sensitivity. These systems were typically complexed with a ligand or, optionally, contained a delay agent such as an alpha
hydroxy carboxylic acid or a polyhedric alcohol. Other delay agents, such as sodium bicarbonate (Brannon et al. 1989), were
also introduced. However, this delay strategy involved the addition of another chemical, adding to the complexity, and could
compromise the final gel strength by permanently complexing some of the metal crosslinker.

Fig. 2 provides an example of the loss in final gel strength that can occur if the crosslinking has progressed while the fluid is
under high shear, such as that experienced during tubular transit. The crosslinker was injected into the flowing linear polymer
stream using a precision syringe pump. The sample labeled “unsheared” was taken downstream from a static mixer section in
the shear-history simulator and loaded immediately into the rheometer. The sheared sample was dynamically loaded into the
rheometer after passing through the tubing for 5 min at a shear rate of 1,350 s-1. In this case the fluid was chemically delayed
to a crosslink temperature, which was determined to be 107 degF by a microwave test. However, there was still significant
decrease in viscosity after shear. Had there been no delay in the fluid, even less viscosity would have been retained. These
results confirm the suggestion from Gidley et al. (1989) that these fracturing gels can suffer significant viscosity losses when
crosslinked too quickly in the tubulars. Delaying the crosslinking may be the best compromise when using a single
crosslinker system, but this technique has too many deficiencies to be considered optimal.

First, defining an exact point of crosslinking is a difficult task. It is defined by any of the following:
• the point where the fluid vortex closes in the blender
• the point where the gel “crowns” or forms a bulge above where the vortex existed
• the point where a small, finger-sized lip can be extended from and retracted back into a container
• the point where a broad, hanging “tongue” can be extended from a container and then retracted.

All these conditions represent some degree of crosslinking based on an observed fluid behavior.

Second, for Group 4 metal crosslinkers, the point of crosslinking is often described by a single parameter: the crosslinking
temperature. But, as for any other chemical reaction, there is a kinetic component; to appropriately describe crosslinking one
must include a time element as well. Low-rate steady-shear rheometry as a function of time can be employed to measure very
slight degrees of crosslinking, whose effects are too small to detect by the visual observations mentioned previously.
Fig. 3 shows the rate of viscosity increase in typical delayed 25-lbm/1,000 gal Zr-crosslinked carboxymethyl hydroxypropyl
guar (CMHPG) polymer gels. For these tests, the samples were placed into the temperature-controlled rheometer
immediately after introducing the crosslinker, and viscosity was measured continuously at room temperature. The
crosslinking temperatures shown in the Fig. 3 legend were determined by heating a separate sample of the fluid in a
SPE 121757 3

microwave in 5-s bursts and examining it for the first sign of “lipping” behavior. Both samples exhibit some crosslinking, as
evidenced by the increased viscosity with time, but it is much more pronounced on the sample with the lowest microwave
crosslinking temperature. This increase occurred well below the crosslinking temperature determined by the microwave
method, and under actual treatment conditions the fluid would experience high shear while in tubular transit in those early
stages.

Several explanations can be offered as the root cause for the reduced strength that occurs after shearing a crosslinked gel.
First, the crosslink points could be irreversibly compromised by the shear. This might be caused by a crosslink cleavage,
leaving the metal associated with only one polymer strand, which would then not contribute to the gel viscosity. A second
possibility is that the metal crosslink remains intact during shear, but the mechanical forces applied to the crosslinked
network break the polymer strands. Regardless of whether the mechanism of irreversible damage is due to effects on the
crosslink points or to the polymer backbone, it is generally accepted that the final viscosity is substantially compromised
when the gel is exposed to high shear conditions after crosslinking (Royce et al. 1984; Hodge and Baranet 1987; Roll et al.
1987). Parris et al. (2009) showed that shear-induced polymer chain degradation is responsible for the crosslinked fluids’
drop in viscosity after shear, and also for the loss of viscosity and molecular weight of the linear polymer after decrosslinking
a fully crosslinked delayed Zr-based fluid.

Results and Discussion

Fluid Development
A list of performance goals was created. These are briefly stated here:
• Develop early viscosity for proppant transport by generating a labile, shear-tolerant crosslink independent of
selected clay stabilizer.
• Design fluid system for high-temperature and high-pressure stability by selecting a Group 4 metal secondary
crosslinker.
• Eliminate the possibility of shear degradation in tubulars by deferring metal crosslink until past the perforations.
• Make the fluid robust to mix water and additive rate control by reducing crosslink sensitivity to pH.
• Improve efficiency and eliminate potential error on dilution, mixing, and transferring by designing chemicals to be
pumped as received.
• Reduce the number of additive streams and complexity by combining components and functionalities.
• Address a broad range of treatment conditions by including compatibility with various types of clay stabilizers and
concentrations.

In addition, the additives had to be shelf stable; meet stringent health, safety, and environmental requirements; and result in a
fluid that was cost competitive.

Examination of commercial products failed to find any that would deliver on all the performance goals, so a process to
formulate the additives was begun. Although many considerations are incorporated into a fluid design, only a few of the more
novel aspects are discussed here.

Titanium organometallic crosslinkers are normally formulated at relatively acidic pH, which limits the high-temperature
stability with natural polysaccharides, and were not considered for these high-temperature applications. Zr-based
crosslinkers, on the other hand, are very active in the alkaline pH region where polysaccharides are more thermally stable.

CMHPG, a lower-residue, more thermally stable polysaccharide was selected as the base polymer. Although CMHPG
responds very well to Zr-based crosslinkers in solutions of varying salt concentrations (Harry et al. 1999), the viscosity
increase from a boron-based crosslinker is dependent on ionic strength of the linear gel solution. Linear gels prepared with
KCl in typical field concentrations for clay control yield a reasonably strong crosslinked gel capable of proppant transport.
However, when organic KCl substitutes are used there is very little detectable viscosity gain. This deficiency was solved by
adding a small amount of a crosslinking aid to the KCl substitute (Parris et al. 2008).

In crosslinkable fluids, Group 4 metals must be complexed to remain in solution and to maintain crosslinking activity.
Selection of the complexing ligand, sometimes referred to as a chelant, is critical for performance. Criteria for ligand
selection are based on a number of factors, including metal binding strength, complex solubility in the crosslinker solution,
and compatibility with other components in the crosslinker solution and final fracturing fluid. The binding strength, largely
defined by the formation constant, determines the temperatures at which crosslinking is active, the rate of crosslinking, and
the compatibility of the boron and metal complex.
4 SPE 121757

Boron forms a strong, dative bond with certain ligands that have been used for Zr-based crosslinkers. This causes a metal
exchange on the ligand, which releases the Zr and complexes with the boron, resulting in an undesirable, nondelayed metal
crosslink. The National Institute of Standards and Technology (NIST) has published the database of stability constants
compiled and evaluated by Martell and Smith (2004). This reference, however, which is perhaps the most comprehensive
compilation, has limited critically reviewed data for Zr(+4) complexes; therefore, a substantial amount of laboratory
experimentation was involved in ligand selection. A complexing ligand was selected that offered several key advantages
(Mirakyan et al. 2008). First, it allowed crosslinking to proceed at desired temperatures without including delaying agents,
such as competing ligands. This ability reduces the treatment complexity by eliminating additive streams and avoids loss of
final gel strength that results when a portion of the Group 4 metal is sequestered, or permanently complexed by the
competing ligand. Also, the ligand selected has a great deal less influence on pH or crosslinking temperature. Precise pH
control in fluid systems under field conditions is challenging because of the mix-water composition and low-rate liquid
stream metering.

Fig. 4 demonstrates the reduced sensitivity to pH for the crosslinking temperature of this fluid compared with that of other
commercial crosslinkers measured with the traditional “temperature to lip” microwave method. This method has been used
for many years to estimate crosslink temperature for organometallic crosslinked fluids, and an attempt to provide a
rheological means of determining crosslinking temperature was undertaken during the development of this fluid. For fluids
crosslinked with boron and Zr, viscosity is increased by two mechanisms that can overlap with each other: a low-temperature
viscosity development mainly driven by the equilibria of boron species and polysaccharide in aqueous media that is a
function of pH; and a higher-temperature crosslink driven by the complexation of Zr metal with the polysaccharides. For
fluids with a strong lip at low mix-water temperatures it can be difficult to accurately estimate the Zr crosslinking
temperature. Dynamic rheology is a common method used to determine gellation of crosslinked polymers in several
industries (Winter 1987; Ross-Murphy 2005). In this work we used dynamic rheology to provide further insight into the
behavior of fracturing fluids with dual boron-zirconium crosslinkers. Measurements were made at a 10% strain and a single
frequency (1 Hz), with varying temperature profiles. Fig. 5 describes the method employed to extract the crosslinking
temperature from the onset of the complex viscosity, η*. Slightly different information can be extracted from the same
experiments when reporting other variables such as shear modulus, complex modulus, or tan δ. The example presents results
for three different fluids with varying amounts of sodium hydroxide, and depicts the estimated Zr crosslink temperatures. In
addition, Fig. 4 shows a comparison of the crosslinking temperature determined by the standard microwave lip test and the
method described herein. Similar results were obtained in both cases.

The next step in the fluid development process was to verify the impact of shear on the fluid performance. The behavior of
relatively successful fluids formulated with delayed Zr lactate, which had been monitored in a 70 degF shear history test
carried out according to the ISO standard (Fig. 2), were used as a reference. Fig. 6 depicts the comparison between a sheared
and an unsheared 30-lbm/1,000 gal formulation of the fluid developed in this paper. Clearly, the viscosity yield of fluids
formulated with traditional organometallic crosslinkers after extensive shearing is significantly lower than their full potential;
in contrast, the newly developed fluid was substantially insensitive to shear.

Field Implementation
The new shear-tolerant, high-temperature fracturing fluid formulations described in this paper have been implemented by
multiple operators in a variety of formations since its first introduction. At time of publication more than 100 treatments had
been successfully pumped worldwide.

South Texas has long been a center of high-temperature fracturing activity, and the area operators rely heavily on new
technology to stimulate their wells. Producing reservoirs are deep and must frequently be stimulated through long tubing or
casing strings. This wellbore geometry is complicated for two main reasons. First, treatment pump rates must be reduced to
minimize friction-pressure losses so that surface treating pressure is kept below the maximum allowable based on well
configuration or surface equipment limitations or both. In addition, conventional fracturing fluids experience high amounts of
shear degradation through small-diameter tubulars, and Zr-based fluids are particularly susceptible to premature viscosity
deterioration. The new fluid presented in this paper has allowed operators to successfully address these problems.

Fig. 7 summarizes some of the key well and treatment parameters of the first 60 fracturing operations pumped in south
Texas, which involved 11 operators in 10 counties. Well BHSTs ranged from 223 degF to 375 degF. The fluid was tested in
mix-water temperatures of 60 degF and 88 degF. Treatment pump rates were as low as 11 bbl/min and as high as 53 bbl/min
through small-diameter tubings of 2.4 in, or typical 4.5- 5.5-in casing, resulting in shear rates as low as 400 s-1 in the most
forgiving treatments and as high as 3,300 s-1 in the most stringent conditions. Formations treated included the Wilcox,
Vicksburg, Lobo, Perdido, and Edwards lime, which have measured Young’s Modulus ranging from 1.3 x 106 psi to 5.6 x 106
psi. Average job results were 98% proppant placement.
SPE 121757 5

Furthermore, these early treatments demonstrated that the fluid could be effectively pumped in a wide variety of treatment
conditions and mix-water temperatures and chemical compositions, without further customization. Fig. 8 shows a
comparison between four 40-lbm/1,000 gal CMHPG fluids prepared according to the formulation developed in this work
using different mix-water sources. In addition, Fig. 9 shows the crosslinking temperature measured as a function of pH for
fluids formulated with some of the same mix-water sources. Results show the fluid can be formulated with minor impact of
mix-water composition on the fluid rheological performance.

One application in this initial phase involved use of the fluid to fracture-stimulate a well with 310 degF BHST that had
experienced casing collapse after a previous fracturing treatment operation. Workover operations to restore communication
and production were unsuccessful, and the operator decided to sidetrack the wellbore and to cement in place a 2 3/8-in
slimhole completion. Wellbore integrity was a serious concern because of damage resulting from the workover operations. In
addition, the operator was concerned about friction-pressure losses and high leakoff rates arising from treatment diversion
into the existing fracture network. To address the operator’s concerns, a treatment with an unusually large pad volume (65%
of the total slurry) formulated with a high polymer concentration of 45 lbm/1,000 gal was pumped at a maximum rate of 12
bbl/min, placing 62,000 lbm of resin-coated bauxite at concentrations up to 8 ppa (lbm proppant added to 1 gal). Following
this treatment success the operator used the new fluid in additional slimhole applications.

Conclusions
A new shear-tolerant, high-temperature fracturing fluid has been developed. Unlike the existing delayed organometallic
crosslinked fluids, the fluid formulation is tolerant to impurities and to variations in the mix-water temperature and chemical
composition. The fluid was formulated to reduce the number of liquid additive streams to a minimum. For the majority of the
treatments to date the fluid was pumped with a polymer, a clay stabilizer (organic or inorganic), the crosslinker and
crosslinker activator, and an appropriate breaker package. For treatments at very high temperatures, an additional high-
temperature stabilizer stream was required.

Instrumental in the development of the fluid was a careful evaluation of novel and proprietary organometallic crosslinkers by
well-known benchtop processes and novel rheological and analytical test methods. Tests showed the fluid prevented
significant polymer degradation when exposed to high shear regimes, which translates into maximized high-temperature
viscosity. During the preliminary field test implementation, the fluid was pumped in a variety of conditions and environments
with minor need for customization. Additionally, the fluid proved both robust and effective.

Nomenclature

BHST = Bottomhole StaticTemperature

bpm = Barrels per minute
CMHPG = CarboxyMethylHydroxyPropylGuar
cP = Centipoise
gal = Gallons
gpm = Gallons per minute
gpt = Gallons per thousand gallons
ft = Foot or feet
HPLC = High Performance Liquid Chromatography
KCl = Potassium chloride
M = Mol per liter
MALLS = MultiAngle Laser Light Scattering
Min = Minutes
Ml = Milliliter(s)
Mw = Molecular weight
Pa s = Pascal seconds
PPA = pounds of proppand added to one gallon of clean fluid
ppt = Pounds per thousand gallons
s = Seconds
SEC = Size Exclusion Chromatography
Tan δ = Loss Tangent
Ti = Titanium
Zr= Zirconium
ºC = Temperature in Celsius
ºF = Temperature in Fahrenheit
η* = Complex Viscosity
6 SPE 121757

Acknowledgments

The authors thank Schlumberger for allowing the presentation of this paper.

Reference List

Almond, S.W. 1984. Method and Compositions for Fracturing Subterranean Formations, US Patent No. 4,477,360.

Brannon, H.D., Hodge, R.M., and England, K.W. 1989. High Temperature Guar-Based Fracturing Fluid, US Patent No.
4,801,389.

Gidley, J.L., Holditch, S.A., Nierode, D.E., and Veatch, R.W. Jr. 1989. Recent Advances in Hydraulic Fracturing.
Monograph Series, SPE, Richardson, Texas 12: 251.

Harry, D.N., Putzig, D.E., Moorhouse, R., DelPesco, T., and Jernakoff, P. 1999. Chemical Structures of Group 4 Metal
Crosslinkers for Polygalactomannans, Paper SPE 50731 presented at the SPE International Symposium on Oilfield
Chemistry, Houston, Texas, 16–19 February.

Hodge, R.M. 1987. Hydraulic Fracturing Method Using Delayed Crosslinker Composition. US Patent No. 4,657,081.

Hodge, R.M. and Baranet, S.E. 1987. Evaluation of Field Methods to Determine Crosslink Time of Fracturing Fluids. Paper
SPE 16249-MS presented at the SPE International Symposium on Oilfield Chemistry, San Antonio, Texas, 4–6 February.

ISO 13503-1. 2003. Petroleum and Natural Gas Industries–Completion Fluids and Materials–Part 1: Measurement of Viscous
Properties of Completion Fluids. Geneva, Switzerland: International Organization for Standardization.

Martell, A.E. and Smith, R.M. eds. 2004. NIST Standard Reference Database 46, NIST Critically Selected Stability
Constants of Metal Complexes. National Institute of Standards and Technology.

Mirakyan, A., Parris, M.D., Abad, C., and Chen, Y. 2008. Well Treatment with Complexed Metal Crosslinkers. US Patent
Application No. 2008/0280790 A1.

Nolte, K.G., 1988. Application of Fracture Design Based on Pressure Analysis, SPEPE 3 (1): 31–42. SPE-13393-PA.

Parris M.D., Mirakyan, A., Abad, C. Chen, Y., and Mueller F. 2009 A New Shear-Tolerant High-Temperature Fracturing
Fluid. Paper SPE 121755 presented at the SPE International Symposium on Oilfield Chemistry, Houston, Texas, 20–22
April.

Parris, M.D., Mirakyan, A., Chen, Y., and Abad, C. 2008. Well Treatment with Ionic Polymer Gels. US Patent Application
No. 2008/0280788 A1.

Putzig, D.E. and Smeltz, K.C. 1989. Cross-Linking Titanium and Zirconium Chelates and Their Use. US Patent No.
4,808,739.

Roll, D.L., Himes, R., Ewitt, D.P., and Doerksen, J. 1987. Effects of Pumping Equipment on Sand-Laden Slurries. SPEPE 2
(4): 291–296. SPE-15071-PA.

Ross-Murphy, S. B. 2005. Gelation Kinetics—Problems and Prospects, Journal of Macromolecular Science, Part B, 44 (6)
1007–1019.

Royce, T.N., Beck, L.M., and Rickards [confirm spelling], A.R. 1984. Rheological Characteristics of Adjustable Cross-
Linked Fracturing Fluids. Paper SPE 13178-MS presented at the SPE Annual Technical Conference and Exhibition, Houston,
Texas, 16–19 September.
SPE 121757 7

RP39, Recommended Practices on Measuring the Viscous Properties of a Cross-linked Water-based Fracturing Fluid, third
edition. 1998. Washington, DC: API.

Walker, M.L., Suchart, C.E., Yaritz, J.G., and Norman, L.R. 1995. Effects of Oxygen on Fracturing Fluids. Paper SPE 28978
presented at the SPE International Symposium on Oilfield Chemistry, San Antonio, Texas, 14–17 February.

Walser, D.W., 1988. Field Study of a New High-Temperature Fracturing Fluid in South Texas, SPEPE 3 (2): 187–191. SPE-
13813-PA.

Winter, H.H. 1987. Can the Gel Point of a Crosslinking Polymer Be Determined by the G’-G’’ Crossover? Polym. Eng. Sci.
27 (22): 1698–1702.

Conversion Factors
1 bbl = 0.159 m3
1 ft = 0.3048 m
1 cp = 10-3 Pa·s
T(°C ) = ( T(°F) - 32) / 1.8
1 galUS = 3.785412 10-3 m3
1 lbm = 0.4535924 kg
1 mL = 1 cm3
1 psi = 6.894757 kPa
8 SPE 121757

Crosslinking too early

„ High friction pressure
„ Shear degradation
„ Screen-out Crosslinking too late
„ Low viscosity
„ Small fracture width
„ Screen-out

What is the actual

temperature at the
perforations?

Fig. 1—The two most common sources of fluid failures in a high-temperature organometallic fracture treatment are
related to a suboptimal control of the crosslinking temperature.

800 320

700 280

600 240
-1
Viscosity, cP at 100 s

500 200 Temperature, ºF

400 160

300 120

200 80

100 40
Unsheared 55% Viscosity loss upon shear Temp(F)
0 0
0 20 40 60 80 100 120 140
Time, min
Fig. 2—Shear history simulation for a Zr lactate 30-lbm/1,000 gal CMHPG performed according to ISO shear history
simulation with fluid sheared at 1,350 s-1 for 5 min with the coil submersed in a 90 degF thermostated bath.
SPE 121757 9

400

114ºF
106ºF

300
-1
Viscosity, cP at 0.1 s

200

100

0
0 5 10 15 20
Time, min

Fig. 3—Viscosity development at 80 degF for 25-lbm/1,000 gal delayed Zr-crosslinked CMHPG. Legend values
are the crosslinking temperatures determined by a microwave “first lip” method.

175

150
Crosslink Temperature, ºF

125

100

75
Commercial Zr crosslinker. Microwave lip method
New crosslinker. Microwave lip method
New crosslinker. Dynamic rhelogy method

50
8 8.5 9 9.5 10 10.5
Fluid pH

Fig. 4— Comparison of crosslinking temperature determined by microwave method and dynamic rheology method for
fluids prepared at different pH for a commercial Zr crosslinking and the new crosslinking system.
10 SPE 121757

10 90
Complex viscosity 0 gpt NaOH 10% 84oC
Complex viscosity 2 gpt NaOH 10%
Complex viscosity 4 gpt NaOH 10% 80
Temperature

64oC 70

Temperature (oC)
η* (Pa.s) at 1 Hz

60
1
50

40

36oC 30

0.1 20
0 100 200 300 400
time (s)
Fig. 5—Time evolution of the complex viscosity, η*, measured at 1 Hz for three fluids formulated with various
concentrations of sodium hydroxide for a given temperature ramp. Crosslinking temperature for each fluid is determined as
the temperature at which the onset of the complex viscosity increase occurs.

Fig. 6—Shear history simulation of 30-lbm/1,000 gal CMHPG gel at 1350 s-1 for 5 min with coils submersed in 90 degF
water bath. Viscosity was measured continuously according to the ISO standard at 100 s-1, with shear ramps to 75, 50, and 25
s-1.
SPE 121757 11

Bottomholestatic temperature (OF) 223 298 375

Mix-water Temp (OF) 60 79 88
Injection rate (bpm) 11 30 53
Pipe size OD(in) 2.4 4.0 5.5
Shear rate (s-1) 400 1200 3300
Young’s Modulus (Millions of psi) 1.3 4.0 5.6

av
Fig. 7—Minimum (blue), average (black), and maximum (red) values for some of the key treatment parameters for the
first 60 wells stimulated with the new shear-tolerant, high-temperature fracturing fluid.

1,000

Laredo Alice
800 San Ygnacio Zapata
Viscosity, cP at 100 s-1

600

400

200

0
0 20 40 60 80 100 120 140 160 180
Time, min

Fig. 8—New shear-tolerant fluid formulated at 40 lbm/1,000 gal with four different mix-water sources used during the
early implementation in south Texas.
12 SPE 121757

175

Zapata
Laredo
150 San Ygnacio
Crosslink Temperature ( η* on-set)

125

100

75

50
8.5 9 9.5 10 10.5 11
pH
Fig. 9—Crosslinking temperature determined by dynamic rheology method as a function of final fluid pH for three
different water sources.