- Email: [email protected]
Experimental study of friction reducer flows in microfracture
JFUE 8044
No. of Pages 8, Model 5G
30 April 2014 Fuel xxx (2014) xxx–xxx 1
Contents lists available at ScienceDirect
Fuel journal homepage: www.elsevier.com/locate/fuel 4 5
Experimental study of friction reducer flows in microfracture
3 6
Q1
Yongpeng Sun a, Qihua Wu b, Mingzhen Wei a, Baojun Bai a,⇑, Yinfa Ma b a
7 8
b
Department of Geological Sciences and Engineering, Missouri University of Science and Technology, Rolla, MO 65409, United States Department of Chemistry, Missouri University of Science and Technology, Rolla, MO 65409, United States
9 10
h i g h l i g h t s
1 2 13 14
Micro-meter sized fracture models are successfully employed in the study.
15
Flow behavior of friction reducer in microfracture is carefully investigated, and compared with that in macro-sized pipeline.
16
We analyzed the friction reducer emulsion particle size.
17
Fluid impact on shale matrix during slickwater fracturing is discussed.
18
a r t i c l e 2 3 0 3 21 22 23 24 25
i n f o
Article history: Received 4 August 2013 Received in revised form 13 April 2014 Accepted 15 April 2014 Available online xxxx
26 27 28 29 30 31 32
Keywords: Shale gas Slickwater fracturing Friction reducer Microfracture Flow behavior
a b s t r a c t Tight formations with extremely low matrix permeabilities, such as gas shale, can produce at economical rates is due to the inborn fissures and fractures introduced during hydraulic stimulation. These microfractures have much more contact area with the matrix and therefore hold the majority of the productivity potential of shale gas. Slickwater fracturing has been proved to be an effective method by which to increase the recovery of shale gas reservoirs. And friction reducer is the primary component of this fluid. However, the flow characteristics of this solution in microfractures are not clear. Micro-sized fluidic chip was used to represent the microfracture. Friction reducer solution is a shear thinning fluid. Rather than reducing flowing friction, with 0.075 vol% of this fluid flowing in a 1000 lm height, 50 lm width and 4.14 cm length microfracture, the injection pressure increased more than 50%. The impact of the solution concentration was found to be more obvious at low velocities. If a flowback additive is considered for slickwater fracturing, its performance at low velocity or low shear rate would be critical. At the same shear rate, the apparent viscosity is higher in large microfractures. At the same velocity, large microfractures display higher residual resistance factors. Through the analysis of fluid emulsion particle size and gas shale matrix pore size, this friction reducer solution will not go into the matrix pores easily, but can block the pore entrance on fracture face to prevent the fluid from leak off and help pressure build up during slickwater fracturing. Ó 2014 Published by Elsevier Ltd.
34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52
53 54
1. Introduction
55
Shale gas reservoir with extremely low matrix permeabilities are producing at economical rates. This can be attributed to the inborn fissures and introduced fractures. Due to the rock mechanical properties of gas shale, hydraulic fracturing can connect and generate these fractures, causing them to be a fracture network than a pair of main fractures. The fracture network will expose more matrix as the number of micro-sized fractures increases [1–5]. Among the various fracturing methods, slickwater fracturing has been proved to be an effective method by which to increase
56 57 58 59 60 61 62 63 64
Q2
⇑ Corresponding author. Tel.: +1 573 341 4016. E-mail address: [email protected] (B. Bai).
the recovery of shale gas reservoirs [6–8]. By adding a very small amount of chemical to the fluid (<1 vol% of the liquid volume), slickwater fracturing fluid can lower the surface pumping pressure below that achieved with the traditional cross-linked fracturing fluid. This fluid also demonstrates a relatively low viscosity, which significantly reduces the gel damage during hydraulic stimulation. In order to carry proppant in this low-viscosity fluid, high pump rates usually are required. Therefore, the friction along the pipeline could be significant. Friction reducer (FR) is one of the primary components of this fluid. Most of the common FR are polyacrylamide-based polymer, usually manufactured as water-in-oil emulsions and added to the fracturing fluids (hydration) ‘‘on the fly’’. Polymers disrupt the near-wall turbulence regeneration cycle and reduce the turbulent friction drag by directly interacting with the vortex, thereby
http://dx.doi.org/10.1016/j.fuel.2014.04.050 0016-2361/Ó 2014 Published by Elsevier Ltd.
Please cite this article in press as: Sun Y et al. Experimental study of friction reducer flows in microfracture. Fuel (2014), http://dx.doi.org/10.1016/ j.fuel.2014.04.050
65 66 67 68 69 70 71 72 73 74 75 76 77 78 79
JFUE 8044
No. of Pages 8, Model 5G
30 April 2014 2
Y. Sun et al. / Fuel xxx (2014) xxx–xxx
Nomenclature a A b D DI water dP/dL FR Fr Frr h k M q Re v x
after FR solution flows cross-sectional area (m2) before FR solution flows equivalent diameter of microfracture (m) deionized water pressure gradient (MPa/m) friction reducer resistance factor residual resistance factor fracture height (m) permeability (m2) fluid mobility (m3/(Pa s)) fluid flow rate, m3/s Reynolds number fluid velocity (m/s) boundary layer development length (m)
109
decreasing the flow friction in pipeline [9–11]. Flow loop tests in the laboratory [12–21] have addressed this phenomenon well, showing 10–82% friction reductions in the lab, compared with fresh water. During slickwater fracturing treatment, a pair of main fractures firstly is generated perpendicular to the wellbore direction. As the fluids continue to pump, more microfractures are generated near the main fractures. These microfractures have much more contact area with the shale matrix and therefore hold the majority of the productivity potential of shale gas [1,2,5,22]. However, the flow characteristics of FR solution in these microfractures are not clear. Microfluidic chips have been widely used in the area of chemistry, biology, microelectromechanical systems, etc. The flowing channel in the microfluidic chip could be manufactured from micrometer to nanometer depth. Therefore, a single straight channel in microfluidic chip with micrometer width and height would act like a microfracture. The present study investigates how the friction reducer solution flows in microfractures by employing the microfluidic chip model. The fluid flow in microfracture had been extensively examined. A commercial FR was prepared with deionized water at various concentrations. FR solution concentration effect, microfracture size effects, and residual resistance factor to water were investigated in detail. The fluid shear rates and Reynolds number in microfractures also were studied. Then the microfracture experimental results were compared with that in macro tubing. FR solution impact on fracture face, which is shale matrix, also was analyzed. The emulsion particle size in FR solution was analyzed from micrometer to nanometer scale. Then it was compared with the pore size of typical gas shale.
110
2. Experiment
111
2.1. Materials
112
A commercial friction reducer, FR, a polyacrylamide-based polymer, was used in experiment. Four concentrations, 0.025, 0.05, 0.075, and 0.1 vol% were prepared according to industry practice. Deionized (DI) water was used to prepare the FR solution. Microfluidic chip (Micronit, The Netherlands) was bonded with two pieces of glass of 145 lm and 1.1 mm thick, respectively. The channel was etched in the later one with a quarter circles of 50 lm radius on top and bottom of the fracture. Each chip contains 3 separated microfractures with 50 lm width, 1500 lm, 1000 lm, and 500 lm heights, respectively. Fig. 1 shows the microfluidic chip
80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108
113 114 115 116 117 118 119 120 121
DP
l c q gapp r
t d
pressure drop (MPa) fluid dynamic viscosity (Pa s) shear rate (s1) fluid density (kg/m3) apparent viscosity (Pa s) edge radius of the microfracture (m) fluid kinematic viscosity (m2/s) boundary-layer thickness (m)
Unit conversion 1 in. 0.0254 m 1 lm 106 m 1 md 1015 m2
with micro-sized fractures (a) and cross-sectional view of a single fracture (b). In order to calculate Reynolds number and shear rate, equivalent diameter was introduced. The area of equivalent circle is the same with the microfracture cross-sectional flowing profile. Equivalent diameter is the diameter of this circle, as listed in Table 1. Since microfractures were not of equal length, pressure gradient is used in the Results and Discussion part.
122
2.2. Equipment
130
The apparatus used in the experiment consisted of a pump, a digital pressure gauge, two non-piston accumulators, microfluidic chip inlet assemblies, and a data acquisition system, as shown in Fig. 2. A high-pressure ISCO 500D syringe pump (Teledyne Technologies, Thousand Oaks, CA) provided the fluid driving power, with a flow rate ranging from 0.001 to 204 mL/min. The digital pressure gauge (Keller, Winterthur, Switzerland) measured the microfracture inlet pressure over a pressure range of 0–3.1 MPa with an accuracy of ±0.1%. To minimize the friction in the flow line, two non-piston accumulators (Swagelok, Solon, OH) were used. Decane (Fisher Science, Waltham, MA), a nonpolar liquid that will not dissolve in water, was employed to fill the pump so that it could work as a driving fluid to push the DI water and FR solution, respectively, from the accumulators into the microfractures. A 250 lm inner diameter capillary was used to connect the 1/800
131
Fig. 1. Microfluidic chip with microfractures (a) and cross-sectional view of a single fracture (b).
Please cite this article in press as: Sun Y et al. Experimental study of friction reducer flows in microfracture. Fuel (2014), http://dx.doi.org/10.1016/ j.fuel.2014.04.050
123 124 125 126 127 128 129
132 133 134 135 136 137 138 139 140 141 142 143 144 145
JFUE 8044
No. of Pages 8, Model 5G
30 April 2014 3
Y. Sun et al. / Fuel xxx (2014) xxx–xxx Table 1 Microfracture parameters. Height (lm)
Width (lm)
Length (cm)
Equivalent diameter (lm)
1500 1000 500
50 50 50
3.94 4.14 4.14
306.8 249.59 174.54
additional pressure drop. Therefore, experiments were conducted with two conditions respectively: capillary only and microfractures with same capillary connected. The pressure drop in microfracture was acquired by subtracting the pressure drop in capillary only from that in microfracture with the same capillary connected. For each experiment, time, pressure and flow rate were recorded. The data presented in this paper were all at the steady state. To clean the inner surface after each experiment, microfluidic chip was flushed with the following steps:
184
(1) 1 mol/L Nitric acid for 15 min with 25 ml. (2) DI water for 15 min with 25 ml. (3) Repeat (1) and (2) one time, and use methanol and DI water (1:1) for 15 min with 25 ml. (4) DI water for 15 min with 25 ml.
194
185 186 187 188 189 190 191 192 193
195 196 197 198 199
Fig. 2. Schematic diagram of experiment.
146 147 148 149 150 151
stainless steel tubing and the microfluidic chip. The data acquisition system was connected to the digital pressure gauge to collect the pressure data over time. A dynamic light-scattering particle size analyzer U1732 (Nanotrac, Montgomeryville, PA) was used to characterize the FR solution emulsion particle size. All experiments were carried out at room temperature.
152
2.3. Procedure
153
The pump cylinder was filled with Decane first and allowed to sit for two hours. Then, the Decane was pumped into the infill line at a low flow rate until no more gas bubbles come out. The infill line valve then was closed, and the accumulators were filled with DI water and FR solution with a syringe, respectively. To ensure that no more gas bubbles existed, the fluids were stirred with a clean glass stick. The entire flow line was checked before running the experiments to prevent any future gas bubbles or leaking. The microchannel inlet was checked to ensure that it was tight enough to hold the maximum pressure before each experiment. Based on the flow rate and the cross-sectional area of the microfracture, the fluid velocity was calculated by:
154 155 156 157 158 159 160 161 162 163 164
165 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183
v¼
q q ¼ A p r 2 =2 þ 50 ðh 2 rÞ
3. Results and discussion
200
3.1. Concentration effect
201
The following four concentrations of FR solution were studied: 0.025, 0.05, 0.075, and 0.1 vol%. Each sample was injected into a 1000 lm height microfracture, respectively. Fig. 3 illustrates the effect of the different FR solution concentrations vs. pressure gradient. High-concentration solutions display a larger pressure gradient that decreases as the fluid velocity decreases. The four lines are closing to each other at lower velocities. At the velocity of 0.1 m/s, they are almost identical. This FR solution is polyacrylamide-based polymer, so it is a nonNewtonian fluid. Eq. (2) can be used to calculate its apparent viscosity, gapp:
202
gapp
A dP ¼k q dL
ð2Þ
where k represents the permeability in m2; A is the cross-sectional area in m2; dP is the pressure drop in MPa; and dL is the microfracture length in m. Each concentration of FR solution, respectively, was used to flux the same sized microfractures. At the same velocity, in Eq. (2), K, A, and q will not change. With Darcy’s equation, apparent viscosity can be simplified to:
gapp ¼
ðdP=dLÞFR ðdP=dLÞDI water
ð3Þ
ð1Þ
where v is the fluid velocity in m/s; q is the fluid flow rate in m3/s; A is the cross-sectional area of the microfracture in m2; and h is the microfracture height in m. DI water was firstly injected into the microfracture at various velocities. FR solution was fluxed at the same velocities. Then DI water was injected again. During one experimental run, the following five fluid velocities were implemented in the microfractures: 4.2–4.6, 2.1–2.3, 1.04–1.14, 0.52–0.57, and 0.01–0.011 m/s. The pump was set to flux at the lowest velocity initially. Pressure vs. time was recorded. When the pressure was remained within 0.3% of the current reading over a 1 min period, the flow was considered reaching stable condition. Then, the next higher velocity was employed. Repeat the measurement until all velocities were tested. The pressure gauge was seated before the capillary connection. The capillary connection, which has a smaller diameter compare with that of the stainless steel tubing, would generate an
Fig. 3. Effect of FR solution concentration.
Please cite this article in press as: Sun Y et al. Experimental study of friction reducer flows in microfracture. Fuel (2014), http://dx.doi.org/10.1016/ j.fuel.2014.04.050
203 204 205 206 207 208 209 210 211 212
213 215 216 217 218 219 220 221 222
223 225
JFUE 8044
No. of Pages 8, Model 5G
30 April 2014 4
Y. Sun et al. / Fuel xxx (2014) xxx–xxx
viscosity. With the 55 mD, 269 mD, and 5120 mD cores, at the same fluid velocity, the resistance factor is higher in high-permeability rocks than that in low-permeability ones, as shown in Fig. 11. Comparing with Xanthan, FR has a much lower resistance factor, which would make it easier to flow through microfractures, thus increase fracture length. However, Xanthan with a much has larger resistance factor which can effectively reduce water fingering and result in higher sweep efficiency for a polymer flooding project to enhance oil recovery. Reynolds number gives a measure of the ratio of inertial forces to viscous forces and is used to characterize different flow regimes. It can be calculated using Eq. (4):
qVD Re ¼ l
ð4Þ 3
Fig. 4. Apparent viscosity of FR solution.
226 227 228 229 230 231 232 233 234
As displayed in Fig. 4, the apparent viscosity of FR solution is always higher than that of DI water (1 103 Pa s). This viscosity is small at high velocity (4.3 m/s) and large at low velocity (0.1 m/s). This is because FR is polyacrylamide-based polymer, and its solution usually behaviors as non-Newtonian fluid. It also indicated that the FR solution had a higher resistance in microfracture than that with water. This resistance could reduce FR penetration intro microfracture during hydraulic stimulation and thus would minimize its damage to formation.
235
3.2. Microfracture size effect
236
A 0.05 vol% FR solution was used to flux the 1500 lm, 1000 lm, and 500 lm height microfractures at five velocities, respectively. It takes microfractures with small height longer to reach a steady state than that in large height. When comparing the flow behavior of FR solution with DI water, as shown in Fig. 5, the apparent viscosity increases as the fluid velocity decreases. Large microfractures have a higher apparent viscosity than that in small ones. Xanthan, a shear thinning polymer, was found to behave in a similar way [23], as shown in Appendix A. Resistance factor (Fr) reflects the mobility ratio of water to the mobility of polymer. For the same experiment, resistance factor equals to the apparent
237 238 239 240 241 242 243 244 245 246 247
Fig. 5. Effect of microfracture size.
248 249 250 251 252 253 254 255 256 257 258 259
260 262
where Re is Reynolds number; q is the fluid density in kg/m ; D is the equivalent diameter of the microfracture cross-sectional area in m; and l is the fluid dynamic viscosity in Pa s. Re is presented vs. velocity with different sized microfracture in Fig. 6. Large microfractures have a larger Re at similar velocities. In this study, Re can reach to 890, much smaller than the transitional Re of 2300, which indicates that the experiments were under laminar flow regime.
263
3.3. Shear rate effect
271
Shear rate is the velocity gradient measured across the diameter of a fluid flow channel. It is the rate change of velocity at which one layer of fluid passes over an adjacent layer. 0.05 vol% of FR solution was used to flux the 1500 lm, 1000 lm and 500 lm height microfractures, respectively. As shown in Fig. 7 in log–log scale, with the shear rate increases, the apparent viscosity of the FR solution decreases. This indicates that FR solution is a shear thinning fluid. At the same shear rate, the fluid in larger microfracture displayed a higher apparent viscosity. However, for the same fluid, ideally the apparent viscosity should be the same at the same shear rate, where the three curves here should be on the same line. The difference may partially explained by the boundary layer theory. Based on the laminar flow, two dimensional boundary layer theory [24], the boundary layer thickness can be estimated with Eq. (5),
272
d5
rffiffiffiffiffi tx
v
rffiffiffiffiffiffiffiffiffiffi lx ¼5 qcD
ð5Þ
where d is the boundary layer thickness in m, t is the kinematic viscosity in m2/s, x represents the boundary layer development length; and c is the shear rate in s1.
Fig. 6. Reynolds number as a function of velocity.
Please cite this article in press as: Sun Y et al. Experimental study of friction reducer flows in microfracture. Fuel (2014), http://dx.doi.org/10.1016/ j.fuel.2014.04.050
264 265 266 267 268 269 270
273 274 275 276 277 278 279 280 281 282 283 284 285
286 288 289 290 291
JFUE 8044
No. of Pages 8, Model 5G
30 April 2014 Y. Sun et al. / Fuel xxx (2014) xxx–xxx
5
Fig. 8. Residual resistance factor in different microfractures. Fig. 7. Shear rate effect.
311
Consider the boundary layer development length equals the microfracture length, and then the boundary layer thickness was in D1/2 relation with the microfracture equivalent diameter. The boundary layer area occupied the cross-sectional area would be in D2.5 relation with the microfracture equivalent diameter. This indicated that at the same shear rate, when the fluid is flowing in small channels, the boundary layer occupied more portion of the cross-sectional area than that in large channels, which would result in smaller flow path diameter in small channels than that in large channels. Therefore, the actual shear rate in small microfracture would be higher than calculated. For this shear thinning fluid, at the same shear rate, the apparent viscosity was lower in small microfractures, and vice versa in large ones. Xanthan flooding experiment in porous media also confirmed this phenomenon. Assuming a homogeneous rock property, porous media was simplified to capillary bundle model. Based on their permeability and porosity, capillary diameter was calculated as in Appendix B Table 2. Fig. 12 shows the shear rate vs. resistance factor, indicating that the large porous media has a larger resistance factor at the same shear rate, which is consistent with this study in Fig. 7.
312
3.4. Residual resistance factor to water
313
The chemical residual condition after stimulation is closely related to the residual fluid flowing resistance, fracturing fluid flowback, gas production rate, etc. This phenomenon can be described by residual resistance factor, Frr, as in Eq. (6):
292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310
314 315 316
317 319 320 321 322 323 324 325 326
327 329 330 331 332 333
F rr ¼
Mb ðk=uÞb ¼ M a ðk=uÞa
ð6Þ
where M is the fluid mobility in m3/(Pa s); b and a represent DI water flows before and after the FR solution, respectively. If Frr equals 1, then residual FR solution in microfracture will not be affected by the introduced fluid during hydraulic fracturing. In this study, DI water was used to flux the microfracture before and after 0.05 vol% FR solution. Hence, the viscosity is the same in Eq. (6). With Darcy’s equation, Eq. (6) can be converted into Eq. (7).
DP a F rr ¼ DP b
ð7Þ
Fig. 8 gives the residual resistance factor as a function of velocity in various sized microfractures. The residual resistance factor is low at high velocity and high at low velocity. This is due to the viscous property of FR solution. As a polymer solution, it may leave
some amount of fluid behind after flooding. For the same sized microfractures, at a high velocity or high shear rate, the fluid had high mobility and low apparent viscosity (Fig. 7); therefore, it could be flushed out easily. The exact opposite is true under the condition of low velocity or low shear rate. Furthermore, if a flowback additive is considered for slickwater fracturing, its performance at low velocity or low shear rate would be critical. Large microfractures displays larger residual resistance factor at the same velocity. This could be explained in this way. Assume the fluid can achieve a same velocity at both large and small microfractures, respectively. The shear rate in small ones would be higher than that in large ones. The sweep efficiency in small ones would be larger than that in large ones. Therefore at the same fluid velocity, the residual resistance in small microfractures is smaller than that in larger ones.
334
3.5. Data comparison with flow loop experiment
349
With 0.677 in, 0.9 in, and 1.162 in macro tubing, flow loop experiment were conducted with the 0.075 vol% FR solution [19]. At the fluid velocity around 4 m/s, the injection pressure decreased more than 40% compared with that of water. However, with the same fluid flowing at the velocity of 4.3 m/s in a 1000 lm height, 50 lm width and 4.14 cm length microfracture, the injection pressure did not decrease but rather increased more than 50%, as shown in Fig 4. This means that when FR solution flows in pipeline, it can reduce the flowing friction. When it comes to the microfracture with the size mentioned, it cannot achieve turbulent flow regime. The flowing friction increased. The fluid will not easily penetrate into the microfractures compare with DI water.
350
3.6. Analysis of FR solution emulsion particle size
362
FR was produced as water-in-oil emulsion. When pumped into the pipeline with a very high pump rate, the emulsion was reversed and the polymer was released, which swelled (hydrates). The oil phase of the FR will form diluted oil-in-water emulsion. The emulsion particle size distribution in 0.05 vol% FR solution was analyzed by a Nanotrac U1732 Particle Size Analyzer. The average results over five tests are shown in Fig. 9. The solution has two peaks. The left peak indicates that there are particles with a 0.00093 lm diameter of 3.7 vol%. The majority of the particles lie in the right peak. Their diameters are 0.0723–1.635 lm of 96.3 vol%, and the peak diameter is 0.555 lm.
363
Please cite this article in press as: Sun Y et al. Experimental study of friction reducer flows in microfracture. Fuel (2014), http://dx.doi.org/10.1016/ j.fuel.2014.04.050
335 336 337 338 339 340 341 342 343 344 345 346 347 348
351 352 353 354 355 356 357 358 359 360 361
364 365 366 367 368 369 370 371 372 373
JFUE 8044
No. of Pages 8, Model 5G
30 April 2014 6
Y. Sun et al. / Fuel xxx (2014) xxx–xxx
Fig. 9. Emulsion particle size distribution of 0.05 vol% FR solution. 374
3.7. Potential impact of FR fluid on shale matrix
375
Shale gas is stored in shale. It is composed of ‘‘free’’ compressed gas in pores and fractures [25] and adsorbed gas in organic kerogen in shale matrix [26,27]. Due to the extremely tight properties of gas shale, the oil-in-water emulsion particles in slickwater fracturing fluid may impact on the flow of fracturing fluid into the micropore, microfracture and matrix. During hydraulic fracturing, once the gas rich pores, fracture or organic kerogen were reached by fracture networks, compared with the original reservoir pressure, there would be a pressure drop. If this pressure drop is big enough, gas would be desorbed from the kerogen and start to flow through microfractures to wellbore [28]. However, at the same time, if the matrix pore throat is blocked by the emulsion particles from the introduced fluid, which will form a filter cake on the fracture face, an additional pressure drop would be required. If this pressure drop required is too high, the gas production would be impaired. Further formation damage recovery methods need considered.
376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391
The pore throat size of typical gas shale usually ranges from 0.005 to 0.1 lm [29,30], as shown in Fig. 10. When compared with the emulsion particle size distribution of the FR solution (Fig. 9), there is only a small overlap, less than 5%. From Fig. 10, only Pennsylvanian shales can reach up to 0.1 lm. Pliocene shales, source rocks, Devonian shales, and Jurassic-Cretaceous shales are ranging from 0.008 to 0.07 lm, which pore throat size hardly overlaps with emulsion particle size of FR solution. Therefore, such FR solution will not easily go into the shale matrix pores, but it may block the matrix pore entrance to build a filter cake, prevent the fluid from leak off. This filter cake on the fracture face would also help to build up pressure, once the pressure gradient is high enough, more fractures would generated.
392
4. Conclusions
405
Microfracture experiments were conducted to study the effects of concentration, microfracture size, Reynolds number, shear rate and residual resistance factor on the flow behavior of FR solution. Experimental results were then compared with that in macro tubing. FR solution emulsion particle size was analyzed and its impact on fracture face was discussed. The following can be concluded from this study:
406
(1) FR solution is a shear thinning fluid. At the same shear rate, the apparent viscosity is higher in large microchannels. (2) The impact of FR solution concentration was more obvious at low velocities. Higher concentrations of FR solution displayed a larger pressure gradient. (3) With similar velocity in flow loop experiment, the fluid flow in microfracture experiments were under laminar flow regime, instead of turbulent flow in flow loop. The same friction reducer did not decrease the injection pressure, but increased more than 50%. (4) The residual resistance factor to water is relative low for this friction reducer. Moreover, at the same velocity, large microfractures shows higher residual resistance factors.
413
Fig. 10. Pore throats sizes of different porous media [29].
Please cite this article in press as: Sun Y et al. Experimental study of friction reducer flows in microfracture. Fuel (2014), http://dx.doi.org/10.1016/ j.fuel.2014.04.050
393 394 395 396 397 398 399 400 401 402 403 404
407 408 409 410 411 412
414 415 416 417 418 419 420 421 422 423 424 425
JFUE 8044
No. of Pages 8, Model 5G
30 April 2014 7
Y. Sun et al. / Fuel xxx (2014) xxx–xxx 426 427 428 429 430
(5) Based on the analyzes of emulsion particle size and the typical gas shale pore size, this FR solution will not go into the matrix pores easily, but can block the pore entrance on fracture face to prevent the fluid from leak off, and help pressure build up during slickwater fracturing.
431 432
Acknowledgements
433
Funding for this project is provided by RPSEA through the ‘‘Ultra-Deepwater and Unconventional Natural Gas and Other Petroleum Resources’’ program authorized by the U.S. Energy Policy Act of 2005. RPSEA (www.rpsea.org) is a nonprofit corporation whose mission is to provide a stewardship role in ensuring the focused research, development and deployment of safe and environmentally responsible technology that can effectively deliver hydrocarbons from domestic resources to the citizens of the United States. RPSEA, operating as a consortium of premier U.S. energy research universities, industry, and independent research organizations, manages the program under a contract with the U.S. Department of Energy’s National Energy Technology Laboratory. The authors also wish to acknowledge for support from Baker Hughes and the China Scholarship Council. We acknowledge Minghang Chen in the Chemistry Department of Missouri S&T for conducting the distribution analysis of FR solution emulsion particle sizes. The authors also want to thank Hanzheng Wang in the Electrical & Computer Engineering Department of Missouri S&T for conducting the microfluidic chip size examination.
434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451
452
Appendix A. Resistance factor vs. velocity of Xanthan
453
459
Resistance factor is defined as the ratio of the mobility of the water to the mobility of the polymer. Under the same experimental conditions, resistance factor equals the apparent viscosity. Preview study reported that Xanthan produces a similar phenomenon, as shown in Fig. 11 [31]. With 55 mD, 269 mD, and 5120 mD cores, at the same fluid velocity, the resistance factor is high in high-permeability rocks than that in low-permeability ones.
460
Appendix B. Resistance factor vs. shear rate of Xanthan
461
With the same experiment data in Appendix A, under the assumption that the rock is homogeneous, the pore size of each sample were converted to circular capillary diameters from permeability and porosity data, as shown in Table 2. Then, the velocity was converted to shear rate. Fig. 12 shows the relationship
454 455 456 457 458
462 463 464 465
Fig. 11. Resistance factor vs. flux of 600 ppm Xanthan [31].
Table 2 Permeability to capillary diameter conversion. K (mD)
/
D (lm)
55 269 5120
0.17 0.212 0.35
10.22 20.23 68.69
Fig. 12. Resistance factor vs. shear rate of 600 ppm Xanthan.
between the shear rate and the resistance factor, indicating that at the same shear rate, the resistance factor in higher permeability sample is larger than that in low permeability sample.
466
References
469
[1] King GE. Thirty years of gas shale fracturing: what have we learned?. In: SPE annual technical conference and exhibition, Soc Pet Eng, Florence, Italy; 2010. [2] Apaydin OG, Ozkan E, Raghavan R. Effect of discontinuous microfractures on ultratight matrix permeability of a dual-porosity medium. SPE Reservoir Eval Eng 2012;15:473–85. [3] Wang FP. Production fairway: speed rails in gas shales. In: 7th Annual gas shales summit, Dallas, TX, USA; 2008. [4] Ding W, Li C, Li C, Xu C, Jiu K, Zeng W, et al. Fracture development in shale and its relationship to gas accumulation. Geosci Front 2012;3:97–105. [5] Wei C, Wang Y, Qin G, Li Q, Killough JE. Experimental and numerical studies on the micro-fractures and its significance toward production of shales: a case study. In: SPE eastern regional meeting, Soc Pet Eng, Pittsburgh, Pennsylvania, USA; 2013. [6] Grieser B, Hobbs J, Hunter J, Ables J. The rocket science behind water frac design. In: SPE production and operations symposium, Oklahoma City, Oklahoma, USA: Soc Petrol Eng Inc.; 2003. [7] Palisch T, Vincent M, Handren P. Slickwater fracturing: food for thought. SPE Prod Oper 2010;25:327–44. [8] Hassanpoor D, Hayatdavoudi A, Boukadi F. The effect of shale chemistry on gas production decline in hydraulically fractured shale gas wells. In: SPE international symposium on oilfield chemistry, Soc Pet Eng, Texas, USA: The Woodlands; 2013. [9] Ram A, Finkelstein E, Elata C. Reduction of friction in oil pipelines by polymer additives. Ind Eng Chem Process Des Dev 1967;6:309–13. [10] White CM, Mungal MG. Mechanics and prediction of turbulent drag reduction with polymer additives. Annu Rev Fluid Mech 2008;40:235–56. [11] Sun H, Stevens RF, Cutler JL, Wood B, Wheeler RS, Qu Q. A novel nondamaging friction reducer: development and successful slickwater frac applications. In: Tight gas completions conference, Soc Pet Eng, San Antonio, Texas, USA; 2010. [12] Shah SN, Kamel AHA. Investigation of flow behavior of slickwater in large straight and coiled tubing. SPE Prod Oper 2010;25:70–9. [13] Rimassa SM, Howard P, Arnold MO. Are you buying too much friction reducer because of your biocide. In: SPE hydraulic fracturing technology conference, Soc Pet Eng, Texas, USA: The Woodlands; 2009. [14] Kaufman PB, Penny GS, Paktinat J. Critical evaluation of additives used in shale slickwater fracs. In: SPE shale gas production conference, Soc Pet Eng, Fort Worth, Texas, USA; 2008. [15] S.M. Rimassa, P.R. Howard, K.A. Blow. Optimizing Fracturing Fluids From Flowback Water. in: SPE Tight Gas Completions Conference, Society of Petroleum Engineers, San Antonio, Texas, USA; 2009. [16] Zelenev AS, Gilzow GA, Kaufman PB. Fast inverting, brine and additive tolerant friction reducer for well stimulation. In: SPE international symposium on oilfield chemistry, Soc Pet Eng, Texas, USA: The Woodlands; 2009.
470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512
Please cite this article in press as: Sun Y et al. Experimental study of friction reducer flows in microfracture. Fuel (2014), http://dx.doi.org/10.1016/ j.fuel.2014.04.050
467 468
JFUE 8044
No. of Pages 8, Model 5G
30 April 2014 8 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534
Y. Sun et al. / Fuel xxx (2014) xxx–xxx
[17] Paktinat J, O’Neil BJ, Aften CW, Hurd MD. Critical evaluation of high brine tolerant additives used in shale slick water fracs. In: SPE production and operations symposium, Soc Pet Eng, Oklahoma City, Oklahoma, USA; 2011. [18] Baser B, Shenoy S, Gadiyar B, Jain S, Parlar M. An alternative method of dealing with pressure: friction reducer for water packing of long horizontal open holes in low-fracturing-gradient environments. SPE Drill Completion 2010;25: 300–8. [19] Zhou J, Sun H, Stevens RF, Qu Q, Bai B. Bridging the gap between laboratory characterization and field applications of friction reducers. In: SPE production and operations symposium, Oklahoma City, Oklahoma, USA; 2011. [20] Ferguson ML, Anderson CC, Eichelberger PB, Hallock JK, Qiu X, Roell RL. Innovative friction reducer provides improved performance and greater flexibility in recycling highly mineralized produced brines. In: SPE unconventional resources conference-USA, Soc Pet Eng, Texas, USA: The Woodlands; 2013. [21] Zhou MJ, Baltazar M, Sun H, Qu Q. Water-based environmentally preferred friction reducer in ultrahigh-TDS produced water for slickwater fracturing in shale reservoirs. In: SPE/EAGE european unconventional resources conference and exhibition, Soc Pet Eng, Vienna, Austria; 2014. [22] Odusina EO, Sondergeld CH, Rai CS. An NMR study of shale wettability. In: Canadian unconventional resources conference, Soc Pet Eng, Alberta, Canada; 2011.
[23] Seright RS, Fan T, Wavrik K, Balaban RdC. New insights into polymer rheology in porous Media. SPE J 2011. [24] Schlichting H, Gersten K. Boundary-layer theory. Gersten: Springer; 2004. [25] Ross DJ, Marc Bustin R. Impact of mass balance calculations on adsorption capacities in microporous shale gas reservoirs. Fuel 86 (2007) 2696–2706. [26] Gasparik M, Ghanizadeh A, Bertier P, Gensterblum Y, Bouw S, Krooss BM. Highpressure methane sorption isotherms of black shales from the Netherlands. Energy Fuels 2012;26:4995–5004. [27] Bai B, Elgmati M, Zhang H, Wei M. Rock characterization of Fayetteville shale gas plays. Fuel 2012. [28] Leahy-Dios A, Das M, Agarwal A, Kaminsky RD. Modeling of transport phenomena and multicomponent sorption for shale gas and coalbed methane in an unstructured grid simulator. In: SPE annual technical conference and exhibition, Soc Pet Eng, Denver, Colorado, USA; 2011. [29] Nelson PH. Pore-throat sizes in sandstones, tight sandstones, and shales. AAPG Bull 2009;93:329. [30] Clarkson CR, Solano N, Bustin R, Bustin A, Chalmers G, He L et al. Pore structure characterization of North American shale gas reservoirs using USANS/SANS, gas adsorption, and mercury intrusion. Fuel 2012. [31] Seright R, Fan T, Wavrik K, Balaban R. New insights into polymer rheology in porous media. SPE J 2011;16:35–42.
Please cite this article in press as: Sun Y et al. Experimental study of friction reducer flows in microfracture. Fuel (2014), http://dx.doi.org/10.1016/ j.fuel.2014.04.050
535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556
No. of Pages 8, Model 5G
30 April 2014 Fuel xxx (2014) xxx–xxx 1
Contents lists available at ScienceDirect
Fuel journal homepage: www.elsevier.com/locate/fuel 4 5
Experimental study of friction reducer flows in microfracture
3 6
Q1
Yongpeng Sun a, Qihua Wu b, Mingzhen Wei a, Baojun Bai a,⇑, Yinfa Ma b a
7 8
b
Department of Geological Sciences and Engineering, Missouri University of Science and Technology, Rolla, MO 65409, United States Department of Chemistry, Missouri University of Science and Technology, Rolla, MO 65409, United States
9 10
h i g h l i g h t s
1 2 13 14
Micro-meter sized fracture models are successfully employed in the study.
15
Flow behavior of friction reducer in microfracture is carefully investigated, and compared with that in macro-sized pipeline.
16
We analyzed the friction reducer emulsion particle size.
17
Fluid impact on shale matrix during slickwater fracturing is discussed.
18
a r t i c l e 2 3 0 3 21 22 23 24 25
i n f o
Article history: Received 4 August 2013 Received in revised form 13 April 2014 Accepted 15 April 2014 Available online xxxx
26 27 28 29 30 31 32
Keywords: Shale gas Slickwater fracturing Friction reducer Microfracture Flow behavior
a b s t r a c t Tight formations with extremely low matrix permeabilities, such as gas shale, can produce at economical rates is due to the inborn fissures and fractures introduced during hydraulic stimulation. These microfractures have much more contact area with the matrix and therefore hold the majority of the productivity potential of shale gas. Slickwater fracturing has been proved to be an effective method by which to increase the recovery of shale gas reservoirs. And friction reducer is the primary component of this fluid. However, the flow characteristics of this solution in microfractures are not clear. Micro-sized fluidic chip was used to represent the microfracture. Friction reducer solution is a shear thinning fluid. Rather than reducing flowing friction, with 0.075 vol% of this fluid flowing in a 1000 lm height, 50 lm width and 4.14 cm length microfracture, the injection pressure increased more than 50%. The impact of the solution concentration was found to be more obvious at low velocities. If a flowback additive is considered for slickwater fracturing, its performance at low velocity or low shear rate would be critical. At the same shear rate, the apparent viscosity is higher in large microfractures. At the same velocity, large microfractures display higher residual resistance factors. Through the analysis of fluid emulsion particle size and gas shale matrix pore size, this friction reducer solution will not go into the matrix pores easily, but can block the pore entrance on fracture face to prevent the fluid from leak off and help pressure build up during slickwater fracturing. Ó 2014 Published by Elsevier Ltd.
34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52
53 54
1. Introduction
55
Shale gas reservoir with extremely low matrix permeabilities are producing at economical rates. This can be attributed to the inborn fissures and introduced fractures. Due to the rock mechanical properties of gas shale, hydraulic fracturing can connect and generate these fractures, causing them to be a fracture network than a pair of main fractures. The fracture network will expose more matrix as the number of micro-sized fractures increases [1–5]. Among the various fracturing methods, slickwater fracturing has been proved to be an effective method by which to increase
56 57 58 59 60 61 62 63 64
Q2
⇑ Corresponding author. Tel.: +1 573 341 4016. E-mail address: [email protected] (B. Bai).
the recovery of shale gas reservoirs [6–8]. By adding a very small amount of chemical to the fluid (<1 vol% of the liquid volume), slickwater fracturing fluid can lower the surface pumping pressure below that achieved with the traditional cross-linked fracturing fluid. This fluid also demonstrates a relatively low viscosity, which significantly reduces the gel damage during hydraulic stimulation. In order to carry proppant in this low-viscosity fluid, high pump rates usually are required. Therefore, the friction along the pipeline could be significant. Friction reducer (FR) is one of the primary components of this fluid. Most of the common FR are polyacrylamide-based polymer, usually manufactured as water-in-oil emulsions and added to the fracturing fluids (hydration) ‘‘on the fly’’. Polymers disrupt the near-wall turbulence regeneration cycle and reduce the turbulent friction drag by directly interacting with the vortex, thereby
http://dx.doi.org/10.1016/j.fuel.2014.04.050 0016-2361/Ó 2014 Published by Elsevier Ltd.
Please cite this article in press as: Sun Y et al. Experimental study of friction reducer flows in microfracture. Fuel (2014), http://dx.doi.org/10.1016/ j.fuel.2014.04.050
65 66 67 68 69 70 71 72 73 74 75 76 77 78 79
JFUE 8044
No. of Pages 8, Model 5G
30 April 2014 2
Y. Sun et al. / Fuel xxx (2014) xxx–xxx
Nomenclature a A b D DI water dP/dL FR Fr Frr h k M q Re v x
after FR solution flows cross-sectional area (m2) before FR solution flows equivalent diameter of microfracture (m) deionized water pressure gradient (MPa/m) friction reducer resistance factor residual resistance factor fracture height (m) permeability (m2) fluid mobility (m3/(Pa s)) fluid flow rate, m3/s Reynolds number fluid velocity (m/s) boundary layer development length (m)
109
decreasing the flow friction in pipeline [9–11]. Flow loop tests in the laboratory [12–21] have addressed this phenomenon well, showing 10–82% friction reductions in the lab, compared with fresh water. During slickwater fracturing treatment, a pair of main fractures firstly is generated perpendicular to the wellbore direction. As the fluids continue to pump, more microfractures are generated near the main fractures. These microfractures have much more contact area with the shale matrix and therefore hold the majority of the productivity potential of shale gas [1,2,5,22]. However, the flow characteristics of FR solution in these microfractures are not clear. Microfluidic chips have been widely used in the area of chemistry, biology, microelectromechanical systems, etc. The flowing channel in the microfluidic chip could be manufactured from micrometer to nanometer depth. Therefore, a single straight channel in microfluidic chip with micrometer width and height would act like a microfracture. The present study investigates how the friction reducer solution flows in microfractures by employing the microfluidic chip model. The fluid flow in microfracture had been extensively examined. A commercial FR was prepared with deionized water at various concentrations. FR solution concentration effect, microfracture size effects, and residual resistance factor to water were investigated in detail. The fluid shear rates and Reynolds number in microfractures also were studied. Then the microfracture experimental results were compared with that in macro tubing. FR solution impact on fracture face, which is shale matrix, also was analyzed. The emulsion particle size in FR solution was analyzed from micrometer to nanometer scale. Then it was compared with the pore size of typical gas shale.
110
2. Experiment
111
2.1. Materials
112
A commercial friction reducer, FR, a polyacrylamide-based polymer, was used in experiment. Four concentrations, 0.025, 0.05, 0.075, and 0.1 vol% were prepared according to industry practice. Deionized (DI) water was used to prepare the FR solution. Microfluidic chip (Micronit, The Netherlands) was bonded with two pieces of glass of 145 lm and 1.1 mm thick, respectively. The channel was etched in the later one with a quarter circles of 50 lm radius on top and bottom of the fracture. Each chip contains 3 separated microfractures with 50 lm width, 1500 lm, 1000 lm, and 500 lm heights, respectively. Fig. 1 shows the microfluidic chip
80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108
113 114 115 116 117 118 119 120 121
DP
l c q gapp r
t d
pressure drop (MPa) fluid dynamic viscosity (Pa s) shear rate (s1) fluid density (kg/m3) apparent viscosity (Pa s) edge radius of the microfracture (m) fluid kinematic viscosity (m2/s) boundary-layer thickness (m)
Unit conversion 1 in. 0.0254 m 1 lm 106 m 1 md 1015 m2
with micro-sized fractures (a) and cross-sectional view of a single fracture (b). In order to calculate Reynolds number and shear rate, equivalent diameter was introduced. The area of equivalent circle is the same with the microfracture cross-sectional flowing profile. Equivalent diameter is the diameter of this circle, as listed in Table 1. Since microfractures were not of equal length, pressure gradient is used in the Results and Discussion part.
122
2.2. Equipment
130
The apparatus used in the experiment consisted of a pump, a digital pressure gauge, two non-piston accumulators, microfluidic chip inlet assemblies, and a data acquisition system, as shown in Fig. 2. A high-pressure ISCO 500D syringe pump (Teledyne Technologies, Thousand Oaks, CA) provided the fluid driving power, with a flow rate ranging from 0.001 to 204 mL/min. The digital pressure gauge (Keller, Winterthur, Switzerland) measured the microfracture inlet pressure over a pressure range of 0–3.1 MPa with an accuracy of ±0.1%. To minimize the friction in the flow line, two non-piston accumulators (Swagelok, Solon, OH) were used. Decane (Fisher Science, Waltham, MA), a nonpolar liquid that will not dissolve in water, was employed to fill the pump so that it could work as a driving fluid to push the DI water and FR solution, respectively, from the accumulators into the microfractures. A 250 lm inner diameter capillary was used to connect the 1/800
131
Fig. 1. Microfluidic chip with microfractures (a) and cross-sectional view of a single fracture (b).
Please cite this article in press as: Sun Y et al. Experimental study of friction reducer flows in microfracture. Fuel (2014), http://dx.doi.org/10.1016/ j.fuel.2014.04.050
123 124 125 126 127 128 129
132 133 134 135 136 137 138 139 140 141 142 143 144 145
JFUE 8044
No. of Pages 8, Model 5G
30 April 2014 3
Y. Sun et al. / Fuel xxx (2014) xxx–xxx Table 1 Microfracture parameters. Height (lm)
Width (lm)
Length (cm)
Equivalent diameter (lm)
1500 1000 500
50 50 50
3.94 4.14 4.14
306.8 249.59 174.54
additional pressure drop. Therefore, experiments were conducted with two conditions respectively: capillary only and microfractures with same capillary connected. The pressure drop in microfracture was acquired by subtracting the pressure drop in capillary only from that in microfracture with the same capillary connected. For each experiment, time, pressure and flow rate were recorded. The data presented in this paper were all at the steady state. To clean the inner surface after each experiment, microfluidic chip was flushed with the following steps:
184
(1) 1 mol/L Nitric acid for 15 min with 25 ml. (2) DI water for 15 min with 25 ml. (3) Repeat (1) and (2) one time, and use methanol and DI water (1:1) for 15 min with 25 ml. (4) DI water for 15 min with 25 ml.
194
185 186 187 188 189 190 191 192 193
195 196 197 198 199
Fig. 2. Schematic diagram of experiment.
146 147 148 149 150 151
stainless steel tubing and the microfluidic chip. The data acquisition system was connected to the digital pressure gauge to collect the pressure data over time. A dynamic light-scattering particle size analyzer U1732 (Nanotrac, Montgomeryville, PA) was used to characterize the FR solution emulsion particle size. All experiments were carried out at room temperature.
152
2.3. Procedure
153
The pump cylinder was filled with Decane first and allowed to sit for two hours. Then, the Decane was pumped into the infill line at a low flow rate until no more gas bubbles come out. The infill line valve then was closed, and the accumulators were filled with DI water and FR solution with a syringe, respectively. To ensure that no more gas bubbles existed, the fluids were stirred with a clean glass stick. The entire flow line was checked before running the experiments to prevent any future gas bubbles or leaking. The microchannel inlet was checked to ensure that it was tight enough to hold the maximum pressure before each experiment. Based on the flow rate and the cross-sectional area of the microfracture, the fluid velocity was calculated by:
154 155 156 157 158 159 160 161 162 163 164
165 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183
v¼
q q ¼ A p r 2 =2 þ 50 ðh 2 rÞ
3. Results and discussion
200
3.1. Concentration effect
201
The following four concentrations of FR solution were studied: 0.025, 0.05, 0.075, and 0.1 vol%. Each sample was injected into a 1000 lm height microfracture, respectively. Fig. 3 illustrates the effect of the different FR solution concentrations vs. pressure gradient. High-concentration solutions display a larger pressure gradient that decreases as the fluid velocity decreases. The four lines are closing to each other at lower velocities. At the velocity of 0.1 m/s, they are almost identical. This FR solution is polyacrylamide-based polymer, so it is a nonNewtonian fluid. Eq. (2) can be used to calculate its apparent viscosity, gapp:
202
gapp
A dP ¼k q dL
ð2Þ
where k represents the permeability in m2; A is the cross-sectional area in m2; dP is the pressure drop in MPa; and dL is the microfracture length in m. Each concentration of FR solution, respectively, was used to flux the same sized microfractures. At the same velocity, in Eq. (2), K, A, and q will not change. With Darcy’s equation, apparent viscosity can be simplified to:
gapp ¼
ðdP=dLÞFR ðdP=dLÞDI water
ð3Þ
ð1Þ
where v is the fluid velocity in m/s; q is the fluid flow rate in m3/s; A is the cross-sectional area of the microfracture in m2; and h is the microfracture height in m. DI water was firstly injected into the microfracture at various velocities. FR solution was fluxed at the same velocities. Then DI water was injected again. During one experimental run, the following five fluid velocities were implemented in the microfractures: 4.2–4.6, 2.1–2.3, 1.04–1.14, 0.52–0.57, and 0.01–0.011 m/s. The pump was set to flux at the lowest velocity initially. Pressure vs. time was recorded. When the pressure was remained within 0.3% of the current reading over a 1 min period, the flow was considered reaching stable condition. Then, the next higher velocity was employed. Repeat the measurement until all velocities were tested. The pressure gauge was seated before the capillary connection. The capillary connection, which has a smaller diameter compare with that of the stainless steel tubing, would generate an
Fig. 3. Effect of FR solution concentration.
Please cite this article in press as: Sun Y et al. Experimental study of friction reducer flows in microfracture. Fuel (2014), http://dx.doi.org/10.1016/ j.fuel.2014.04.050
203 204 205 206 207 208 209 210 211 212
213 215 216 217 218 219 220 221 222
223 225
JFUE 8044
No. of Pages 8, Model 5G
30 April 2014 4
Y. Sun et al. / Fuel xxx (2014) xxx–xxx
viscosity. With the 55 mD, 269 mD, and 5120 mD cores, at the same fluid velocity, the resistance factor is higher in high-permeability rocks than that in low-permeability ones, as shown in Fig. 11. Comparing with Xanthan, FR has a much lower resistance factor, which would make it easier to flow through microfractures, thus increase fracture length. However, Xanthan with a much has larger resistance factor which can effectively reduce water fingering and result in higher sweep efficiency for a polymer flooding project to enhance oil recovery. Reynolds number gives a measure of the ratio of inertial forces to viscous forces and is used to characterize different flow regimes. It can be calculated using Eq. (4):
qVD Re ¼ l
ð4Þ 3
Fig. 4. Apparent viscosity of FR solution.
226 227 228 229 230 231 232 233 234
As displayed in Fig. 4, the apparent viscosity of FR solution is always higher than that of DI water (1 103 Pa s). This viscosity is small at high velocity (4.3 m/s) and large at low velocity (0.1 m/s). This is because FR is polyacrylamide-based polymer, and its solution usually behaviors as non-Newtonian fluid. It also indicated that the FR solution had a higher resistance in microfracture than that with water. This resistance could reduce FR penetration intro microfracture during hydraulic stimulation and thus would minimize its damage to formation.
235
3.2. Microfracture size effect
236
A 0.05 vol% FR solution was used to flux the 1500 lm, 1000 lm, and 500 lm height microfractures at five velocities, respectively. It takes microfractures with small height longer to reach a steady state than that in large height. When comparing the flow behavior of FR solution with DI water, as shown in Fig. 5, the apparent viscosity increases as the fluid velocity decreases. Large microfractures have a higher apparent viscosity than that in small ones. Xanthan, a shear thinning polymer, was found to behave in a similar way [23], as shown in Appendix A. Resistance factor (Fr) reflects the mobility ratio of water to the mobility of polymer. For the same experiment, resistance factor equals to the apparent
237 238 239 240 241 242 243 244 245 246 247
Fig. 5. Effect of microfracture size.
248 249 250 251 252 253 254 255 256 257 258 259
260 262
where Re is Reynolds number; q is the fluid density in kg/m ; D is the equivalent diameter of the microfracture cross-sectional area in m; and l is the fluid dynamic viscosity in Pa s. Re is presented vs. velocity with different sized microfracture in Fig. 6. Large microfractures have a larger Re at similar velocities. In this study, Re can reach to 890, much smaller than the transitional Re of 2300, which indicates that the experiments were under laminar flow regime.
263
3.3. Shear rate effect
271
Shear rate is the velocity gradient measured across the diameter of a fluid flow channel. It is the rate change of velocity at which one layer of fluid passes over an adjacent layer. 0.05 vol% of FR solution was used to flux the 1500 lm, 1000 lm and 500 lm height microfractures, respectively. As shown in Fig. 7 in log–log scale, with the shear rate increases, the apparent viscosity of the FR solution decreases. This indicates that FR solution is a shear thinning fluid. At the same shear rate, the fluid in larger microfracture displayed a higher apparent viscosity. However, for the same fluid, ideally the apparent viscosity should be the same at the same shear rate, where the three curves here should be on the same line. The difference may partially explained by the boundary layer theory. Based on the laminar flow, two dimensional boundary layer theory [24], the boundary layer thickness can be estimated with Eq. (5),
272
d5
rffiffiffiffiffi tx
v
rffiffiffiffiffiffiffiffiffiffi lx ¼5 qcD
ð5Þ
where d is the boundary layer thickness in m, t is the kinematic viscosity in m2/s, x represents the boundary layer development length; and c is the shear rate in s1.
Fig. 6. Reynolds number as a function of velocity.
Please cite this article in press as: Sun Y et al. Experimental study of friction reducer flows in microfracture. Fuel (2014), http://dx.doi.org/10.1016/ j.fuel.2014.04.050
264 265 266 267 268 269 270
273 274 275 276 277 278 279 280 281 282 283 284 285
286 288 289 290 291
JFUE 8044
No. of Pages 8, Model 5G
30 April 2014 Y. Sun et al. / Fuel xxx (2014) xxx–xxx
5
Fig. 8. Residual resistance factor in different microfractures. Fig. 7. Shear rate effect.
311
Consider the boundary layer development length equals the microfracture length, and then the boundary layer thickness was in D1/2 relation with the microfracture equivalent diameter. The boundary layer area occupied the cross-sectional area would be in D2.5 relation with the microfracture equivalent diameter. This indicated that at the same shear rate, when the fluid is flowing in small channels, the boundary layer occupied more portion of the cross-sectional area than that in large channels, which would result in smaller flow path diameter in small channels than that in large channels. Therefore, the actual shear rate in small microfracture would be higher than calculated. For this shear thinning fluid, at the same shear rate, the apparent viscosity was lower in small microfractures, and vice versa in large ones. Xanthan flooding experiment in porous media also confirmed this phenomenon. Assuming a homogeneous rock property, porous media was simplified to capillary bundle model. Based on their permeability and porosity, capillary diameter was calculated as in Appendix B Table 2. Fig. 12 shows the shear rate vs. resistance factor, indicating that the large porous media has a larger resistance factor at the same shear rate, which is consistent with this study in Fig. 7.
312
3.4. Residual resistance factor to water
313
The chemical residual condition after stimulation is closely related to the residual fluid flowing resistance, fracturing fluid flowback, gas production rate, etc. This phenomenon can be described by residual resistance factor, Frr, as in Eq. (6):
292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310
314 315 316
317 319 320 321 322 323 324 325 326
327 329 330 331 332 333
F rr ¼
Mb ðk=uÞb ¼ M a ðk=uÞa
ð6Þ
where M is the fluid mobility in m3/(Pa s); b and a represent DI water flows before and after the FR solution, respectively. If Frr equals 1, then residual FR solution in microfracture will not be affected by the introduced fluid during hydraulic fracturing. In this study, DI water was used to flux the microfracture before and after 0.05 vol% FR solution. Hence, the viscosity is the same in Eq. (6). With Darcy’s equation, Eq. (6) can be converted into Eq. (7).
DP a F rr ¼ DP b
ð7Þ
Fig. 8 gives the residual resistance factor as a function of velocity in various sized microfractures. The residual resistance factor is low at high velocity and high at low velocity. This is due to the viscous property of FR solution. As a polymer solution, it may leave
some amount of fluid behind after flooding. For the same sized microfractures, at a high velocity or high shear rate, the fluid had high mobility and low apparent viscosity (Fig. 7); therefore, it could be flushed out easily. The exact opposite is true under the condition of low velocity or low shear rate. Furthermore, if a flowback additive is considered for slickwater fracturing, its performance at low velocity or low shear rate would be critical. Large microfractures displays larger residual resistance factor at the same velocity. This could be explained in this way. Assume the fluid can achieve a same velocity at both large and small microfractures, respectively. The shear rate in small ones would be higher than that in large ones. The sweep efficiency in small ones would be larger than that in large ones. Therefore at the same fluid velocity, the residual resistance in small microfractures is smaller than that in larger ones.
334
3.5. Data comparison with flow loop experiment
349
With 0.677 in, 0.9 in, and 1.162 in macro tubing, flow loop experiment were conducted with the 0.075 vol% FR solution [19]. At the fluid velocity around 4 m/s, the injection pressure decreased more than 40% compared with that of water. However, with the same fluid flowing at the velocity of 4.3 m/s in a 1000 lm height, 50 lm width and 4.14 cm length microfracture, the injection pressure did not decrease but rather increased more than 50%, as shown in Fig 4. This means that when FR solution flows in pipeline, it can reduce the flowing friction. When it comes to the microfracture with the size mentioned, it cannot achieve turbulent flow regime. The flowing friction increased. The fluid will not easily penetrate into the microfractures compare with DI water.
350
3.6. Analysis of FR solution emulsion particle size
362
FR was produced as water-in-oil emulsion. When pumped into the pipeline with a very high pump rate, the emulsion was reversed and the polymer was released, which swelled (hydrates). The oil phase of the FR will form diluted oil-in-water emulsion. The emulsion particle size distribution in 0.05 vol% FR solution was analyzed by a Nanotrac U1732 Particle Size Analyzer. The average results over five tests are shown in Fig. 9. The solution has two peaks. The left peak indicates that there are particles with a 0.00093 lm diameter of 3.7 vol%. The majority of the particles lie in the right peak. Their diameters are 0.0723–1.635 lm of 96.3 vol%, and the peak diameter is 0.555 lm.
363
Please cite this article in press as: Sun Y et al. Experimental study of friction reducer flows in microfracture. Fuel (2014), http://dx.doi.org/10.1016/ j.fuel.2014.04.050
335 336 337 338 339 340 341 342 343 344 345 346 347 348
351 352 353 354 355 356 357 358 359 360 361
364 365 366 367 368 369 370 371 372 373
JFUE 8044
No. of Pages 8, Model 5G
30 April 2014 6
Y. Sun et al. / Fuel xxx (2014) xxx–xxx
Fig. 9. Emulsion particle size distribution of 0.05 vol% FR solution. 374
3.7. Potential impact of FR fluid on shale matrix
375
Shale gas is stored in shale. It is composed of ‘‘free’’ compressed gas in pores and fractures [25] and adsorbed gas in organic kerogen in shale matrix [26,27]. Due to the extremely tight properties of gas shale, the oil-in-water emulsion particles in slickwater fracturing fluid may impact on the flow of fracturing fluid into the micropore, microfracture and matrix. During hydraulic fracturing, once the gas rich pores, fracture or organic kerogen were reached by fracture networks, compared with the original reservoir pressure, there would be a pressure drop. If this pressure drop is big enough, gas would be desorbed from the kerogen and start to flow through microfractures to wellbore [28]. However, at the same time, if the matrix pore throat is blocked by the emulsion particles from the introduced fluid, which will form a filter cake on the fracture face, an additional pressure drop would be required. If this pressure drop required is too high, the gas production would be impaired. Further formation damage recovery methods need considered.
376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391
The pore throat size of typical gas shale usually ranges from 0.005 to 0.1 lm [29,30], as shown in Fig. 10. When compared with the emulsion particle size distribution of the FR solution (Fig. 9), there is only a small overlap, less than 5%. From Fig. 10, only Pennsylvanian shales can reach up to 0.1 lm. Pliocene shales, source rocks, Devonian shales, and Jurassic-Cretaceous shales are ranging from 0.008 to 0.07 lm, which pore throat size hardly overlaps with emulsion particle size of FR solution. Therefore, such FR solution will not easily go into the shale matrix pores, but it may block the matrix pore entrance to build a filter cake, prevent the fluid from leak off. This filter cake on the fracture face would also help to build up pressure, once the pressure gradient is high enough, more fractures would generated.
392
4. Conclusions
405
Microfracture experiments were conducted to study the effects of concentration, microfracture size, Reynolds number, shear rate and residual resistance factor on the flow behavior of FR solution. Experimental results were then compared with that in macro tubing. FR solution emulsion particle size was analyzed and its impact on fracture face was discussed. The following can be concluded from this study:
406
(1) FR solution is a shear thinning fluid. At the same shear rate, the apparent viscosity is higher in large microchannels. (2) The impact of FR solution concentration was more obvious at low velocities. Higher concentrations of FR solution displayed a larger pressure gradient. (3) With similar velocity in flow loop experiment, the fluid flow in microfracture experiments were under laminar flow regime, instead of turbulent flow in flow loop. The same friction reducer did not decrease the injection pressure, but increased more than 50%. (4) The residual resistance factor to water is relative low for this friction reducer. Moreover, at the same velocity, large microfractures shows higher residual resistance factors.
413
Fig. 10. Pore throats sizes of different porous media [29].
Please cite this article in press as: Sun Y et al. Experimental study of friction reducer flows in microfracture. Fuel (2014), http://dx.doi.org/10.1016/ j.fuel.2014.04.050
393 394 395 396 397 398 399 400 401 402 403 404
407 408 409 410 411 412
414 415 416 417 418 419 420 421 422 423 424 425
JFUE 8044
No. of Pages 8, Model 5G
30 April 2014 7
Y. Sun et al. / Fuel xxx (2014) xxx–xxx 426 427 428 429 430
(5) Based on the analyzes of emulsion particle size and the typical gas shale pore size, this FR solution will not go into the matrix pores easily, but can block the pore entrance on fracture face to prevent the fluid from leak off, and help pressure build up during slickwater fracturing.
431 432
Acknowledgements
433
Funding for this project is provided by RPSEA through the ‘‘Ultra-Deepwater and Unconventional Natural Gas and Other Petroleum Resources’’ program authorized by the U.S. Energy Policy Act of 2005. RPSEA (www.rpsea.org) is a nonprofit corporation whose mission is to provide a stewardship role in ensuring the focused research, development and deployment of safe and environmentally responsible technology that can effectively deliver hydrocarbons from domestic resources to the citizens of the United States. RPSEA, operating as a consortium of premier U.S. energy research universities, industry, and independent research organizations, manages the program under a contract with the U.S. Department of Energy’s National Energy Technology Laboratory. The authors also wish to acknowledge for support from Baker Hughes and the China Scholarship Council. We acknowledge Minghang Chen in the Chemistry Department of Missouri S&T for conducting the distribution analysis of FR solution emulsion particle sizes. The authors also want to thank Hanzheng Wang in the Electrical & Computer Engineering Department of Missouri S&T for conducting the microfluidic chip size examination.
434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451
452
Appendix A. Resistance factor vs. velocity of Xanthan
453
459
Resistance factor is defined as the ratio of the mobility of the water to the mobility of the polymer. Under the same experimental conditions, resistance factor equals the apparent viscosity. Preview study reported that Xanthan produces a similar phenomenon, as shown in Fig. 11 [31]. With 55 mD, 269 mD, and 5120 mD cores, at the same fluid velocity, the resistance factor is high in high-permeability rocks than that in low-permeability ones.
460
Appendix B. Resistance factor vs. shear rate of Xanthan
461
With the same experiment data in Appendix A, under the assumption that the rock is homogeneous, the pore size of each sample were converted to circular capillary diameters from permeability and porosity data, as shown in Table 2. Then, the velocity was converted to shear rate. Fig. 12 shows the relationship
454 455 456 457 458
462 463 464 465
Fig. 11. Resistance factor vs. flux of 600 ppm Xanthan [31].
Table 2 Permeability to capillary diameter conversion. K (mD)
/
D (lm)
55 269 5120
0.17 0.212 0.35
10.22 20.23 68.69
Fig. 12. Resistance factor vs. shear rate of 600 ppm Xanthan.
between the shear rate and the resistance factor, indicating that at the same shear rate, the resistance factor in higher permeability sample is larger than that in low permeability sample.
466
References
469
[1] King GE. Thirty years of gas shale fracturing: what have we learned?. In: SPE annual technical conference and exhibition, Soc Pet Eng, Florence, Italy; 2010. [2] Apaydin OG, Ozkan E, Raghavan R. Effect of discontinuous microfractures on ultratight matrix permeability of a dual-porosity medium. SPE Reservoir Eval Eng 2012;15:473–85. [3] Wang FP. Production fairway: speed rails in gas shales. In: 7th Annual gas shales summit, Dallas, TX, USA; 2008. [4] Ding W, Li C, Li C, Xu C, Jiu K, Zeng W, et al. Fracture development in shale and its relationship to gas accumulation. Geosci Front 2012;3:97–105. [5] Wei C, Wang Y, Qin G, Li Q, Killough JE. Experimental and numerical studies on the micro-fractures and its significance toward production of shales: a case study. In: SPE eastern regional meeting, Soc Pet Eng, Pittsburgh, Pennsylvania, USA; 2013. [6] Grieser B, Hobbs J, Hunter J, Ables J. The rocket science behind water frac design. In: SPE production and operations symposium, Oklahoma City, Oklahoma, USA: Soc Petrol Eng Inc.; 2003. [7] Palisch T, Vincent M, Handren P. Slickwater fracturing: food for thought. SPE Prod Oper 2010;25:327–44. [8] Hassanpoor D, Hayatdavoudi A, Boukadi F. The effect of shale chemistry on gas production decline in hydraulically fractured shale gas wells. In: SPE international symposium on oilfield chemistry, Soc Pet Eng, Texas, USA: The Woodlands; 2013. [9] Ram A, Finkelstein E, Elata C. Reduction of friction in oil pipelines by polymer additives. Ind Eng Chem Process Des Dev 1967;6:309–13. [10] White CM, Mungal MG. Mechanics and prediction of turbulent drag reduction with polymer additives. Annu Rev Fluid Mech 2008;40:235–56. [11] Sun H, Stevens RF, Cutler JL, Wood B, Wheeler RS, Qu Q. A novel nondamaging friction reducer: development and successful slickwater frac applications. In: Tight gas completions conference, Soc Pet Eng, San Antonio, Texas, USA; 2010. [12] Shah SN, Kamel AHA. Investigation of flow behavior of slickwater in large straight and coiled tubing. SPE Prod Oper 2010;25:70–9. [13] Rimassa SM, Howard P, Arnold MO. Are you buying too much friction reducer because of your biocide. In: SPE hydraulic fracturing technology conference, Soc Pet Eng, Texas, USA: The Woodlands; 2009. [14] Kaufman PB, Penny GS, Paktinat J. Critical evaluation of additives used in shale slickwater fracs. In: SPE shale gas production conference, Soc Pet Eng, Fort Worth, Texas, USA; 2008. [15] S.M. Rimassa, P.R. Howard, K.A. Blow. Optimizing Fracturing Fluids From Flowback Water. in: SPE Tight Gas Completions Conference, Society of Petroleum Engineers, San Antonio, Texas, USA; 2009. [16] Zelenev AS, Gilzow GA, Kaufman PB. Fast inverting, brine and additive tolerant friction reducer for well stimulation. In: SPE international symposium on oilfield chemistry, Soc Pet Eng, Texas, USA: The Woodlands; 2009.
470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512
Please cite this article in press as: Sun Y et al. Experimental study of friction reducer flows in microfracture. Fuel (2014), http://dx.doi.org/10.1016/ j.fuel.2014.04.050
467 468
JFUE 8044
No. of Pages 8, Model 5G
30 April 2014 8 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534
Y. Sun et al. / Fuel xxx (2014) xxx–xxx
[17] Paktinat J, O’Neil BJ, Aften CW, Hurd MD. Critical evaluation of high brine tolerant additives used in shale slick water fracs. In: SPE production and operations symposium, Soc Pet Eng, Oklahoma City, Oklahoma, USA; 2011. [18] Baser B, Shenoy S, Gadiyar B, Jain S, Parlar M. An alternative method of dealing with pressure: friction reducer for water packing of long horizontal open holes in low-fracturing-gradient environments. SPE Drill Completion 2010;25: 300–8. [19] Zhou J, Sun H, Stevens RF, Qu Q, Bai B. Bridging the gap between laboratory characterization and field applications of friction reducers. In: SPE production and operations symposium, Oklahoma City, Oklahoma, USA; 2011. [20] Ferguson ML, Anderson CC, Eichelberger PB, Hallock JK, Qiu X, Roell RL. Innovative friction reducer provides improved performance and greater flexibility in recycling highly mineralized produced brines. In: SPE unconventional resources conference-USA, Soc Pet Eng, Texas, USA: The Woodlands; 2013. [21] Zhou MJ, Baltazar M, Sun H, Qu Q. Water-based environmentally preferred friction reducer in ultrahigh-TDS produced water for slickwater fracturing in shale reservoirs. In: SPE/EAGE european unconventional resources conference and exhibition, Soc Pet Eng, Vienna, Austria; 2014. [22] Odusina EO, Sondergeld CH, Rai CS. An NMR study of shale wettability. In: Canadian unconventional resources conference, Soc Pet Eng, Alberta, Canada; 2011.
[23] Seright RS, Fan T, Wavrik K, Balaban RdC. New insights into polymer rheology in porous Media. SPE J 2011. [24] Schlichting H, Gersten K. Boundary-layer theory. Gersten: Springer; 2004. [25] Ross DJ, Marc Bustin R. Impact of mass balance calculations on adsorption capacities in microporous shale gas reservoirs. Fuel 86 (2007) 2696–2706. [26] Gasparik M, Ghanizadeh A, Bertier P, Gensterblum Y, Bouw S, Krooss BM. Highpressure methane sorption isotherms of black shales from the Netherlands. Energy Fuels 2012;26:4995–5004. [27] Bai B, Elgmati M, Zhang H, Wei M. Rock characterization of Fayetteville shale gas plays. Fuel 2012. [28] Leahy-Dios A, Das M, Agarwal A, Kaminsky RD. Modeling of transport phenomena and multicomponent sorption for shale gas and coalbed methane in an unstructured grid simulator. In: SPE annual technical conference and exhibition, Soc Pet Eng, Denver, Colorado, USA; 2011. [29] Nelson PH. Pore-throat sizes in sandstones, tight sandstones, and shales. AAPG Bull 2009;93:329. [30] Clarkson CR, Solano N, Bustin R, Bustin A, Chalmers G, He L et al. Pore structure characterization of North American shale gas reservoirs using USANS/SANS, gas adsorption, and mercury intrusion. Fuel 2012. [31] Seright R, Fan T, Wavrik K, Balaban R. New insights into polymer rheology in porous media. SPE J 2011;16:35–42.
Please cite this article in press as: Sun Y et al. Experimental study of friction reducer flows in microfracture. Fuel (2014), http://dx.doi.org/10.1016/ j.fuel.2014.04.050
535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556