Performance improvement of the ethylene-vinyl acetate copolymer (EVA) pour point depressant by small dosages of the polymethylsilsesquioxane (PMSQ) microsphere: An experimental study

Fuel 207 (2017) 204–213

Contents lists available at ScienceDirect

Fuel journal homepage: www.elsevier.com/locate/fuel

Full Length Article

Performance improvement of the ethylene-vinyl acetate copolymer (EVA) pour point depressant by small dosages of the polymethylsilsesquioxane (PMSQ) microsphere: An experimental study Fei Yang a,b,⇑, Bo Yao a,b, Chuanxian Li a,b, Xin Shi a,b, Guangyu Sun a,b, Xiaobin Ma c a b c

College of Pipeline and Civil Engineering, China University of Petroleum, Qingdao, Shandong 266580, PR China Shandong Provincial Key Laboratory of Oil & Gas Storage and Transportation Safety, Qingdao, Shandong 266580, PR China Pipeline Oil-transmitting Department of Qinghai Oilfield, Golmud, Qinghai 816000, PR China

h i g h l i g h t s
Small dosages of the PMSQ microsphere (10 ppm) improve the performance of EVA PPD.
The best dosage of the PMSQ microsphere is found to be around 2.5 ppm.
The EVA molecules can adsorb and concentrate on the PMSQ microsphere.
The formed EVA/PMSQ composite particles can act as the nucleation template for wax precipitation.
The formed composite particles greatly modify the precipitated wax crystals’ morphology.

a r t i c l e

i n f o

Article history: Received 21 March 2017 Received in revised form 19 June 2017 Accepted 20 June 2017

Keywords: Waxy crude oil EVA PMSQ microsphere Flow behavior Performance improvement

a b s t r a c t In a previous work, the addition of the polymethylsilsesquioxane (PMSQ) microsphere (50–400 ppm) can improve the flow behavior of waxy crude oil through the spacial hindrance effect. However, the flow improving efficiency of the neat PMSQ microsphere is not as good as the traditional polymeric pour point depressants (PPDs). In this paper, the effect of the ethylene-vinyl acetate copolymer (EVA2806) PPD together with the PMSQ microsphere (with the size around 2 mm) on the flow behavior of a typical waxy crude oil was investigated. The results show that adding 50 ppm EVA PPD can greatly improve the flow behavior of the oil. The neat PMSQ microsphere cannot improve the flow behavior of the oil at small dosages (10 ppm), but can significantly improve the performance of the EVA PPD. The gelation point, G0 , G00 , transient apparent viscosity and yield stress of the oil decrease further after adding both 50 ppm EVA and a small amount of the PMSQ microsphere (10 ppm). The best flow improving efficiency is found at 50 ppm EVA + 2.5 ppm PMSQ. The addition of the PMSQ microsphere has little influence on the WAT and precipitated wax crystal amount of the oil doped with EVA, but outstandingly changes the morphology of the precipitated wax crystals into larger and more compact flocs. The adsorption tests show that the EVA molecules can adsorb and concentrate on the PMSQ microsphere, thus causing the formation of the EVA/PMSQ composite particles. The composite particles can act as nucleation templates for the wax precipitation, resulting in larger and more compact wax microstructures and then further improving the flow behavior of the oil. The PMSQ microsphere dosage and the amount of EVA PPD adsorbed on the microsphere obviously influence the performance of the composite particle with the best performance at 50 ppm EVA + 2.5 ppm PMSQ. The findings mentioned above provide a new way to improve the performance of polymeric PPDs efficiently. Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction The precipitation of paraffin waxes brings huge challenges to the production and pipeline transportation of waxy crude oil. ⇑ Corresponding author. http://dx.doi.org/10.1016/j.fuel.2017.06.083 0016-2361/Ó 2017 Elsevier Ltd. All rights reserved.

When the temperature of waxy crude oil is below its wax appearance temperature (WAT), paraffin waxes start to precipitate from the oil due to super-saturation [1]. The precipitated wax crystals are normally irregular (plate-like or needle-like) and are liable to form a continuous three dimensional network at relatively low wax crystal concentrations (around 1 wt%) [2,3]. The formed wax

F. Yang et al. / Fuel 207 (2017) 204–213

crystal network structure occludes large amounts of liquid oil and then seriously aggravates the rheology of waxy crude oil, which makes the pipeline transportation of waxy crude oil more difficult and dangerous [4,5]. Polymeric pour point depressants (PPDs) are often added into waxy crude oil to improve the rheology of the oil, thus ensuring the safe and economic transportation of the oil in pipelines [6]. The molecular structure of polymeric PPDs normally contains both the non-polar long alkyl chains (C18) and the polar groups. The long alkyl chains, which could be located either in the backbone or in the side chain of PPD molecules, can take part in the precipitation process of paraffin waxes through nucleation, adsorption and co-crystallization effects. The polar groups such as the ester group, maleic anhydride group, and vinyl acetate (VA) group, could control the dispersion state of PPD molecules in oil phase and interfere the growth of wax crystals [7,8]. Ethylene-vinyl acetate copolymer (EVA) is a kind of effective polymeric PPDs and has been widely used in pipelines transporting waxy crude oil [6,9–12]. The polyethylene group of EVA belongs to the long alkyl chains, while the VA group of EVA is a polar group. Much work has been done on the mechanism and performance EVA PPD so as to guide the application of EVA PPD better [6,9–12]. The results showed that the EVA PPD favors the formation of island defects on the wax crystals surface and weakens the interactions between the precipitated wax crystals [6,10,12]. Therefore, the tendency of wax crystals to interlock into a continuous network structure is impeded and the rheology of waxy crude oil is improved. Meanwhile, the VA content and the average molecular weight are the two key factors influencing the performance of EVA PPD [9,11]: the EVA containing VA content around 28 wt% usually has the best pour point depressing performance, while the optimal average molecular weight of EVA may be varied for different waxy crude oils. Nowadays, polymer/inorganic nanocomposites have been prepared in large scale and widely used in industry due to their excellent mechanical stability, thermal stability and toughness, etc [13,14]. Inspired by the advances of polymer/inorganic nanocomposites, some kinds of polymer/inorganic nanocomposite PPDs were recently developed and evaluated. He et al. [15] prepared a nanocomposite PPD by dispersing nanoclays into an EVA PPD. They found that the nanocomposite PPD could further decrease the pour point and viscosity of a waxy crude oil based on the neat EVA PPD’s performance. Yang and Norrman et al. [16,17] prepared a nanocomposite PPD by dispersing hydrophilic nanosilica into polyoctadecylacrylate (POA) PPD. They found that: (a) the prepared nanocomposite PPD can exist in oil phase as micro-sized composite particles (dozens of micron); (b) the composite particles can act as nucleation templates for wax precipitation and change the morphology of precipitated wax crystals into large spherical-like flocs, thus further improving the rheology of waxy oil; (c) the compatibility between the hydrophilic nanosilica and the organic POA PPD is poor and then the composite particles dispersed in oil phase are unstable, causing that the performance of the nanocomposite PPD decreases with time. In order to improve the compatibility between the inorganic nanoparticles and the polymeric PPDs, Yang et al. [18,19] first prepared organically modified nanoclays (abbreviated as organic nanoclays) through cationic exchange and then dispersed the organic nanoclays into POA PPD matrix. The obtained POA/organic clay nanocomposite PPD disperses well in oil phase as small composite particles (several microns) and the composite particles can also act as nucleation templates for wax precipitation. Therefore, the precipitated wax crystals’ morphology is greatly modified after the addition of the nanocomposite PPD, resulting in the further improvement of waxy crude oil rheology. In addition, the time-effectiveness of the POA/organic clay nanocomposite PPD was also greatly improved. Al-Sabagh et al. [20,21] successfully prepared the poly(methylmethacrylate)/graphene oxide (PMMA/

205

GO) and PMMA/montmorillonite (PMMA/MMT) nanocomposite PPDs by dispersing the inorganic nanosheets of GO or MMT in the organic PMMA matrix via in situ free radical polymerization. They found that the two nanocomposite PPDs disperse well in oil phase as small composite particles (several microns) and the performance of the two nanocomposite PPDs is much better than the neat PMMA PPD. They attributed the excellent performance of the two nanocomposite PPDs to the nucleation effect of the composite particles and the electrostatic repulsion between the composite particles. Based on the research work mentioned above, it is clear that the polymer/inorganic nanocomposite PPDs have become a research hotspot in petroleum industry. In order to further develop the theory of nanocomposite PPDs, it is necessary to understand if the nano and micro particles alone can improve the rheology of waxy crude oil. Polysilsesquioxane (PSQ) microsphere is a kind of organosilicone materials, which has excellent morphological and structural properties [22–24]: the PSQ microsphere is normally monodispersed sphere with the size ranging from nano to micrometer; the PSQ microsphere has specific organic-inorganic hybrid structure, which imparts the PSQ microsphere outstanding thermal stability, mechanical stability, solvent resistance, lubricity, and excellent dispersiblity in organic solvents or polymers. Recently, Yang et al. [25] synthesized the polymethylsilsesquioxane (PMSQ) microsphere with different sizes and found that: (a) the PMSQ microsphere disperses well in oil phase as single spheres; (b) adding 50–400 ppm PMSQ microsphere can greatly improve the flow behavior of waxy crude oil; (c) the best dosage and size of the PMSQ microsphere are 200 ppm and 2–5 mm, respectively; (d) the PMSQ microsphere cannot participate in wax precipitation process and change the morphology of precipitated wax crystals, but can impede the interactions of the precipitated wax crystals through the spacial hindrance effect, which inhibits the development of wax crystal network structure and then improves the flow behavior of waxy crude oil; (e) however, the flow improving efficiency of the PMSQ microsphere is not as good as the traditional polymeric PPDs. Could it be possible to obtain a better flow improving efficiency by adding both the polymeric PPDs and the PMSQ microsphere into waxy crude oil? And if so, what is the synergistic mechanism? To answer the questions mentioned above, in this paper, the effect of an EVA PPD together with a monodispersed PMSQ microsphere (with the size around 2 mm) on the flow behavior of a typical waxy crude oil is investigated. It is found that the neat PMSQ microsphere cannot improve the flow behavior of the oil at small dosages (10 ppm), but can significantly improve the performance of the EVA PPD (fixed at 50 ppm dosage). The performance improving mechanism of the PMSQ microsphere on the EVA PPD is also discussed based on the microscopic images of precipitated wax crystals and the adsorption behavior of EVA molecules onto the PMSQ microsphere.

2. Experimental 2.1. Materials All the chemicals (including the EVA PPD) used here were purchased from Sigma-Aldrich Co., Ltd and used as received. The PPD used here is EVA2806, which has the best performance on the crude oil used in this work (performances of some other PPDs can be seen in the Table S1 of the support information). The VA content and the melting index of the EVA2806 are 28 wt% and 6, respectively. The PMSQ microsphere used here was synthesized based on the method mentioned in a previous paper [25]. As seen in Fig. 1, the synthesized PMSQ microsphere is monodispersed and

206

F. Yang et al. / Fuel 207 (2017) 204–213

Fraction / %

15

10

5

0 1

10

(c)

d / µm

Fig. 1. Microscopic image (a), TEM image (b) and particle size distribution (c) of the PMSQ microsphere dispersed in n-dodecane (0.5 wt%).

disperses well in n-dodecane as single spheres. The average size of the PMSQ microsphere is around 2 lm. Before usage, the EVA PPD together with the PMSQ microsphere was first dissolved/dispersed in n-dodecane. The EVA concentration in n-dodecane was fixed at 5 wt%, while the concentration of PMSQ microsphere was zero, 0.05 wt%, 0.1 wt%, 0.25 wt%, 0.5 wt% and 1 wt%, respectively. Then, the solution/dispersion containing both EVA PPD and PMSQ microsphere was added into waxy crude oil. The oil sample used here was kindly provided by Qinghai oilfield in China. As shown in Table 1, the wax content of the oil sample is relatively high (24.6 wt%) but the resins and asphaltenes contents are relatively small, resulting in high WAT (39 °C) and pour point (31 °C) of the oil. As seen in Fig. 2, the oil sample has a wide carbon number distribution range (C11–C32), with the peak carbon number around C21. It is clear that the oil sample is a typical waxy crude oil. 2.2. Methods Fig. 2. The carbon number distribution of n-alkanes in the Qinghai waxy crude oil.

2.2.1. Pour point tests of the undoped/doped waxy crude oils The pour points of the undoped/doped waxy crude oils were measured on the basis of the method given in the Chinese Standard SY/T 0541-2009 [18,19]. The undoped/doped waxy crude oils sealed in glass bottles were first preheated at 60 °C for 30 min and then transferred into the pre-heated glass test tube. Next, under a fixed cooling rate of 0.5 °C/min, the oil sample in the test

tube was cooled to the temperatures that were slightly higher than the presumptive pour point and the surface movement of the oil sample was checked by tilting the oil in the test tube. The pour point was defined as the highest temperature at which the oil sample in the tilted test tube did not move for at least 5 s.

Table 1 Physical properties of the Qinghai waxy crude oil.

*

Wax (wt%)

Resin (wt%)

Asphaltene (wt%)

WAT (°C)

Pour point* (°C)

Initial boiling point (°C)

q20 (g/cm3)

24.6

8.72

0.83

39

31

82

0.8547

The preheat temperature in the pour point test was fixed at 60 °C.

F. Yang et al. / Fuel 207 (2017) 204–213

2.2.2. Flow behavior tests of the undoped/doped waxy crude oils In the flow behavior tests, the dosage of EVA PPD was fixed at 50 ppm, and the corresponding dosage of PMSQ microsphere was zero, 0.5 ppm, 1 ppm, 2.5 ppm, 5 ppm and 10 ppm, respectively. The undoped/doped waxy crude oils sealed in glass bottles were first preheated at 60 °C for 30 min and then used for the following tests. A rheometer (AR-G2, TA Instruments, USA) equipped with a coaxial cylinder system (a standard cup having a diameter of 30 mm, configured with a DIN Rotor having a diameter of 28 mm) was used to evaluate the effect of the EVA PPD together with the PMSQ microsphere on the flow behavior of the waxy crude oil. Viscoelastic test during cooling. The development of the viscoelastic parameters (the elastic modulus G0 , the viscous modulus G00 and the loss angle d) for the undoped/doped waxy crude oils during cooling was measured by temperature sweeping test under oscillatory mode. The applied amplitude is 0.0005, which is small enough not to disturb the wax crystal network development. The cooling rate and oscillatory frequency were fixed at 0.5 °C/min and 1 Hz. The gelation point of the waxy crude oil, at which G0 begins to become larger than G00 or d begins to become smaller than 45°, could also be obtained through the viscoelastic test. Transient flow curves test. The undoped/doped waxy crude oils were cooled quiescently under the cooling rate of 0.5 °C/min from 60 °C to 15 °C and then maintained isothermally at 15 °C for 30 min. Subsequently, transient flow curves of the oil samples were measured at 15 °C by shear rate ramping from 10 to 200 s1 within 10 min. The variation of the transient apparent viscosity with shear rate was recorded. Yield behavior test. The yield stress is the minimum stress necessary to flow a material, and this stress should not be dependent on the experiment [26]. However, it is hard to determine the real yield stress of gelled waxy crude oil by experimental test. When the yield stress of gelled waxy crude oil is measured, the yield stress varies with the experimental method. Therefore, the yield stress test of gelled waxy crude oil is conditioned, and many methods such as shear stress ramping, shear rate ramping and oscillatory ramping, have been used to determine the yield stress of gelled waxy crude oil [27–30]. In this paper, the yield behavior (yield stress and yield strain) of the undoped/doped waxy crude oils at 15 °C was measured by shear rate ramping from 0 to 1 s1 within 5 min. 2.2.3. DSC test of the undoped/doped waxy crude oils A DSC calorimeter (821e, Mettler-Toledo Co., Switzerland) was used to analyze the exothermic character of undoped/doped waxy crude oils in the temperature scanning range of 85  20 °C. The cooling rate was fixed at 10 °C/min. Based on the exothermic curves, WAT and the amounts of precipitated wax crystals at different temperatures were calculated and recorded [1–3,6–8,15– 19]. 2.2.4. Microscopic observation of the undoped/doped waxy crude oils A microscope (BX51, Olympus Co., Japan) fitted with an automatic thermal stage was used to observe the precipitated wax crystals in undoped/doped waxy crude oils. The oil samples sealed in glass bottles were first preheated at 60 °C for 20 min and then one droplet of the oil samples was transferred to a glass slide covered by a coverslip. The loaded crude oil was cooled from 50 °C to 15 °C on the thermal stage under the cooling rate of 0.5 °C/min. The microstructure of the oil samples at 15 °C was photographed and recorded. 2.2.5. Adsorption behavior of the EVA PPD on PMSQ microsphere The EVA PPD was first dissolved in n-dodecane at 50 °C to prepare a 5 wt% EVA solution. Then, a small amount of the PMSQ

207

microsphere (0.05 wt%, 0.1 wt%, 0.25 wt%, 0.5 wt% and 1 wt%, respectively) was added into the EVA solution. The solution/dispersion containing both EVA and PSMQ microsphere was stirred for 2 h and then kept statically for another 6 h at 50 °C. After that, the microstructure of PMSQ microspheres dispersed in the solution/dispersion was observed through a microscope (BX51, Olympus Ltd., Japan) and a TEM (JEM-100CX II, JEOL Ltd., Japan), respectively. Finally, the solution/dispersion was centrifuged under 10,000 r/min. The obtained sediments were vacuum dried overnight, then tested by TGA (TA-60WS, SHIMADZU Co., Japan) under oxygen atmosphere and the mass loss percentage (at 600 °C) were remarked as fc. As a comparison, the TGA test for the neat PMSQ microsphere was also conducted and the mass loss percentage was remarked as fn. Compared with the neat PMSQ microsphere, the amount of the EVA PPD adsorbed on the PMSQ microsphere, fads could be calculated as follows:

f ads ¼

mEVA f  fn ¼ c  100% mPMSQ 1  fc

ð1Þ

3. Results and discussions 3.1. Flow behavior of the undoped/doped waxy crude oils 3.1.1. Effect of the neat PMSQ microsphere on the flow behavior of waxy crude oil As reported in a previous paper [25], a small dosage of the PMSQ microsphere (50 ppm) can improve the flow behavior of waxy crude oil, and the improving efficiency of the PMSQ microsphere is greatly influenced by the size and dosage of the microsphere (the best size and dosage are 2–5 lm and 200 ppm, respectively). In this paper, the effect of the neat PMSQ microsphere (with the size around 2 mm) on the transient apparent viscosity of Qinghai waxy crude oil at 15 °C was studied in a smaller dosage range (0–50 ppm). As seen in Fig. S1, the transient apparent viscosity of the oil decreased by about 8% upon the addition of 50 ppm PMSQ microsphere. We also consider here that the PMSQ microsphere can inhibit the interactions of the precipitated waxy crystals through the spacial hindrance effect [25], thus improving the flow behavior of the oil. When the PMSQ microsphere dosage decreases to 20 ppm, the viscosity reducing rate decreases to 2.5%, meaning that decreasing the dosage weakens the spacial hindrance effect of the PMSQ microsphere. As the PMSQ microsphere dosage decreases to 10 ppm, the transient apparent viscosity of the oil is nearly unchanged. Therefore, it could be concluded that the PMSQ microsphere (2 mm) has little influence on the flow behavior of the waxy crude oil when its dosage is less than or equal to 10 ppm. 3.1.2. Viscoelastic development of the undoped/doped waxy crude oils during cooling The viscoelastic test during cooling is an effective way to monitor the structural development of waxy crude oil. The viscoelastic parameters such as G0 , G00 , and d, are often used to describe the structural state of waxy crude oil [2,4,27–29]. The viscoelasticitytemperature curves of the undoped/doped Qinghai waxy crude oils are shown in Fig. 3. At temperatures higher than or around the WAT (39 °C for the undoped oil and 37 °C for the doped oil, see Fig. 6), the amount of precipitated wax crystals are little and the oil behaves as a pure viscous fluid with the value of G00 much higher than G0 and d approaching 90°. With the further decrease of oil temperature, the amount of precipitated wax crystals is enough to influence the flow behavior of the oil and then the oil shows viscoelasticity: at temperatures higher than the gelation point, the G0 is smaller than G00 while the d is larger than 45°, meaning that vis-

208

F. Yang et al. / Fuel 207 (2017) 204–213

Fig. 3. Viscoelastic development of the waxy crude oil undoped/doped with EVA and EVA/PMSQ during cooling.

cous response of the oil is dominant; at temperatures lower than the gelation point, the G0 is larger than G00 but the d is smaller than 45°, meaning that elastic response of the oil is dominant. It is clear that the waxy crude oil transits from a sol to a gel at the gelation point. According to Fig. 3, the G0 , G00 at 15 °C and the gelation point of the undoped/doped Qinghai waxy crude oils are obtained and listed in Table 2. The gelation point of the undoped oil is 33.7 °C, and the G0 , G00 at 15 °C are 137,600 Pa and 7844 Pa, respectively. The high values of G0 and G00 indicate a strong gel structure of the

undoped oil at 15 °C. After adding 50 ppm EVA, the gelation point of the oil decreases to 29.4 °C, and the G0 , G00 at 15 °C decrease dramatically to 11,840 Pa and 2483 Pa. The lower values of G0 and G00 indicate that adding EVA PPD greatly weakens the gel structure of the oil formed at 15 °C. The values of G0 , G00 at 15 °C and the gelation point of the oil decrease further after the addition of both EVA PPD and PMSQ microsphere, indicating that small dosages of the PMSQ microsphere can further enhance the ability of the EVA PPD to weaken the crude oil gel structure. Adding 50 ppm EVA and 0.5 ppm PMSQ microsphere could decrease the gelation point

209

F. Yang et al. / Fuel 207 (2017) 204–213 Table 2 The pour point, gelation point, G0 , G00 , yield stress and yield strain of the waxy crude oil undoped/doped with EVA and EVA/PMSQ.

*

Crude oil sample

Pour point/°C

Gelation point/°C

G0 */ Pa

G00 */Pa

Yield stress*/Pa

Yield stain*

Undoped 50 ppm EVA 50 ppm EVA + 0.5 ppm PMSQ 50 ppm EVA + 1 ppm PMSQ 50 ppm EVA + 2.5 ppm PMSQ 50 ppm EVA + 5 ppm PMSQ 50 ppm EVA + 10 ppm PMSQ

31 19 18 18 18 18 19

33.7 29.4 27.6 26.9 27.4 27.9 28.4

137600 11840 8593 5515 4192 4703 6802

7844 2483 1683 1295 975.3 1064 1490

1357 144.7 116.8 77.43 68.91 85.33 95.95

0.016 0.025 0.030 0.031 0.031 0.027 0.029

The G0 , G00 , yield stress and yield strain were all tested at 15 °C.

Fig. 4. Transient flow curves of the waxy crude oil undoped/doped with EVA and EVA/PMSQ at 15 °C.

Fig. 5. Yield behavior of the waxy crude oil undoped (a)/doped with EVA and EVA/PMSQ (b) at 15 °C.

Fig. 6. DSC curves (a) and precipitated wax crystals’ amount (b) of the waxy crude oil undoped/doped with 50 ppm EVA and 50 ppm EVA + 2.5 ppm PMSQ.

210

F. Yang et al. / Fuel 207 (2017) 204–213

to 27.6 °C and the G0 , G00 at 15 °C to 8593 Pa and 1683 Pa, respectively. The gelation point of the oil is suppressed to the lowest (26.9 °C) when 50 ppm EVA and 1 ppm PMSQ microsphere are added. Upon adding 50 ppm EVA and 2.5 ppm PMSQ, the G0 , G00 at 15 °C are decreased to the minimum values of 4192 Pa and 975.3 Pa, respectively. When the dosage of the PMSQ microsphere increases further (5 ppm), however, the performance improvement of the EVA PPD by the addition of PMSQ microsphere is weakened, leading to the gradual recovery of the G0 , G00 and gelation point of the oil with increasing PMSQ microsphere dosage. It is clear that the 50 ppm EVA + 2.5 ppm PMSQ has the strongest ability to inhibit the viscoelastic development of the waxy crude oil. The pour points of the undoped/doped Qinghai waxy crude oils are also listed in Table 2. As seen in this table, adding 50 ppm EVA decreases the pout point of the oil from 31 °C to 19 °C, while adding 50 ppm EVA + 0.5–10 ppm PMSQ microsphere could only further decreases the pour point by 1 °C. It is clear that the gelation point depressing ability and the pour point depressing ability of the same additive varies, the mechanism of which has been discussed in detail in a previous work [19]. 3.1.3. Transient flow curves and yield behavior of the undoped/doped waxy crude oils The transient flow curves of the undoped/doped Qinghai waxy crude oils at 15 °C are demonstrated in Fig. 4. It is obvious that all the undoped/doped oils are non-Newtonian fluids and show clear shear-thinning behavior. Adding the EVA PPD greatly decreases the transient apparent viscosity of the oil. For example, at the fixed shear rate of 50 s1, the transient apparent viscosity of the oil decreases from the initial 2368 mPas to 227.4 mPas after adding 50 ppm EVA PPD. Small dosages of the PMSQ microsphere can further improve the efficiency of EVA PPD in reducing the oil viscosity. For example, at the fixed shear rate of 50 s1, the transient apparent viscosity of the oil doped with 50 ppm EVA + 0.5 ppm, 1 ppm, 2.5 ppm, 5 ppm and 10 ppm PMSQ decreases to 182.3, 124.4, 123.3, 129.1 and 139.2 mPas, respectively, and the viscosity reducing rates compared with the neat EVA are 19.83%, 45.29%, 45.78%, 43.23% and 38.79%, respectively. It is clear that the viscosity reducing efficiency of the EVA + PMSQ reaches the maximum at 2.5 ppm dosage of the PMSQ microsphere. The yield behaviors of the undoped/doped Qinghai waxy crude oils at 15 °C are shown in Fig. 5, from which the yield stress and yield strain of the oil are obtained and listed in Table 2. The yield stress and yield strain of the undoped oil is 1357 Pa and 0.016, respectively. The addition of 50 ppm EVA decreases the yield stress dramatically to 144.7 Pa but increases the yield strain to 0.025. The lower value of yield stress and the higher value of yield strain indicate that after the addition of EVA PPD, the gel strength and brittleness of the wax crystal network structure formed at 15 °C are greatly weakened. Small dosages of the PMSQ microsphere facilitate the further decrease of the yield stress and the further increase of the yield strain of the oil doped with 50 ppm EVA. The yield strains of the oils doped with EVA + PMSQ remain almost the same around 0.03, meaning that the PMSQ microsphere dosage has little influence on the brittleness of the wax crystal network structure. However, the PMSQ microsphere dosage greatly affects the yield stress of the oils doped with EVA + PMSQ. Adding 50 ppm EVA + 0.5 ppm PMSQ slightly decreases the yield stress to 116.8 Pa. The yield stress decreases further to 77.43 Pa after the addition of 50 ppm EVA + 1 ppm PMSQ, and then to 68.91 Pa upon the addition of 50 ppm EVA + 2.5 ppm PMSQ. With the further increase of the PMSQ microsphere dosage, the yield stress rises up to 85.33 Pa at 50 ppm EVA + 5 ppm PMSQ and 95.95 Pa at 50 ppm EVA + 10 ppm PMSQ, both of which are still lower than the yield stress of the oil doped with neat 50 ppm EVA. It is clear that the

yield stress of the oil doped with EVA + PMSQ achieves the smallest value at 2.5 ppm dosage of the PMSQ microsphere. 3.2. DSC curves and precipitated wax crystals’ amount of the undoped/doped waxy crude oils The DSC curves and the precipitated wax crystals’ amount of the undoped/doped Qinghai waxy crude oils are exhibited in Fig. 6a and b. The WAT of the oil slightly decreases from the initial value of 39 °C to 37 °C after adding 50 ppm EVA. As reported in previous works [6–8,18], EVA PPD can effectively increasing the critical nucleation radius as well as the nucleation potential barrier of wax crystals, promote the effective (but not equilibrium) solubilization of wax molecules in the oil phase, thereby inhibiting wax precipitation (that is, decreasing the WAT of oil). The addition of 2.5 ppm PMSQ cannot further decrease the WAT of the oil doped with 50 ppm EVA. Based on the data in Fig. 8c, the EVA PPDs adsorbed on PMSQ microsphere are relatively small (about 5– 6 ppm), most of the EVA PPDs is still free molecules. Therefore, adding 2.5 ppm PMSQ microsphere has little influence on WAT. Meanwhile, the precipitated wax crystals’ amount of the oil is nearly unchanged after adding the EVA or the EVA/PMSQ. It could be concluded that a small dosage of the PMSQ microsphere (2.5 ppm) has little influence on the WAT and the precipitated wax crystals’ amount of the oil doped with EVA PPD. 3.3. Microstructure of the undoped/doped waxy crude oils Microstructures of the undoped/doped Qinghai waxy crude oils at 15 °C are illustrated in Fig. 7. As seen in Fig. 7a, the precipitated wax crystals in the undoped oil have small size and arrange haphazardly with large amounts. Adding 50 ppm EVA greatly changes the morphology of precipitated wax crystals into large wax flocs (see Fig. 7b). After adding 50 ppm EVA + 2.5 ppm PMSQ microsphere (see Fig. 7c), the precipitated wax flocs in the oil phase become larger and more compact. 3.4. Adsorption behavior of the EVA PPD on PMSQ microsphere As seen in Fig. 8a and b, EVA molecules could adsorb and concentrate on the PMSQ microspheres in oil phase. The surface of the PMSQ microsphere contains not only the nonpolar methyl group but also the polar hydroxyl group [22–24]. We deduce that the EVA molecules adsorb on the PMSQ microsphere through the polar attraction between the VA group and the hydroxyl group. Therefore, the PMSQ microsphere, as an inner core, is coated with the EVA molecules on its surface, resulting in the larger sizes and blurred outlines. As shown in Fig. 8c, after heated to 600 °C under oxygen atmosphere, the mass loss percentage of the neat PMSQ microspheres (fn) is 9.98 wt%, meaning that the original organic content of the PMSQ microsphere is relatively small. When 0.05 wt%, 0.1 wt% and 0.25 wt% PMSQ microspheres are added, the mass loss percentages of the EVA-adsorbed PMSQ microspheres (fc) are 58.43 wt%, 57.21 wt% and 58.61 wt%, respectively. Correspondingly, the amounts of the EVA PPD adsorbed on the PMSQ microsphere (fads) are calculated through Eq. (1) as 116.56 wt%, 110.37 wt% and 117.69 wt%, respectively. The values of the fads are almost the same at 0.05–0.25 wt% PMSQ concentration, corresponding to about 5–6 ppm EVA molecules adsorbed on the PMSQ microsphere. The amount of adsorbed EVA molecules on the PMSQ microsphere is directly related to the surface property of the microsphere. PMSQ microsphere is a kind of organosilicone materials with large amounts of methyl group chemically bonded on its surface. Therefore, PMSQ microsphere has good oil dispersibility and could disperse stably in oil phase, which result in that the

F. Yang et al. / Fuel 207 (2017) 204–213

211

Fig. 7. Microstructure of the waxy crude oil undoped/doped with EVA and EVA/PMSQ at 15 °C: undoped (a); doped with 50 ppm EVA (b); doped with 50 ppm EVA + 2.5 ppm PMSQ (c).

amount of adsorbed EVA molecules on the microsphere would not be large. We consider that the EVA molecules can fully adsorb on the PMSQ microsphere at the microsphere concentrations 0.25 wt%. Therefore, the amount of the EVA PPD adsorbed on the PMSQ microsphere is relatively high (around 115 wt%) and is not influenced by the microsphere concentration. Meanwhile, with the further increasing of the PMSQ microsphere concentration to 0.5 wt% and 1 wt%, the fc shows an obviously decreasing trend to 40.03 wt% and 25.99 wt%, respectively, resulting in the decrease of fads to 50.23 wt% and 21.74 wt%, respectively. It is clear that the amount of the EVA PPD adsorbed on the PMSQ microsphere decreases greatly with further increasing the microsphere concentration. It was found in adsorption experiments that the viscosity of the solution/dispersion containing both EVA and PMSQ microsphere at 50 °C increases obviously with increasing the PMSQ microsphere concentration at the concentrations 0.5 wt% (see Fig. S2 in support information). We consider that the increased viscosity of solution/dispersion greatly inhibits the adsorption of EVA molecules on PMSQ microsphere. Therefore, the value of fads decreases with the further increase of PMSQ microsphere concentration at the PMSQ concentrations 0.5 wt% (see Fig. 8c). 3.5. Performance improving mechanism of the PMSQ microsphere on the EVA PPD First of all, it should be noticed that the dosage range of the PMSQ microsphere studied in this paper is so small (10 ppm) that the PMSQ microsphere alone has little influence on the flow behavior of Qinghai waxy crude oil (see Fig. S1). According to the results in Fig. 8, EVA molecules can adsorb and concentrate on the PMSQ microsphere, thus forming the EVA/PMSQ composite particles. Different from the spacial hindrance effect of the neat PMSQ microsphere on the flow behavior of waxy crude oil [25], the EVA/ PMSQ composite particles can effectively act as the nucleation templates for the wax precipitation (see Fig. 9), resulting in the formation of larger and more compact wax flocs (see Fig. 7c). On the

one hand, the larger size of the wax flocs greatly reduces the wax crystal/oil interface area and then weakens the interactions among the precipitated wax crystals, which favors the flow improvement of the oil. On the other, the more compact microstructure of the wax flocs will enable the wax flocs to release more liquid oils previously occluded in the flocs into oil phase, which also promotes the rheological improvement of the oil. Therefore, small dosages of the PMSQ microsphere (10 ppm) improve the performance of the EVA PPD further. Meanwhile, the performance of the EVA/PMSQ composite particle is obviously influenced by the PMSQ microsphere dosage and the amount of EVA PPD adsorbed on the microsphere. At the PMSQ microsphere dosage 2.5 ppm, the amount of EVA PPD adsorbed on the microsphere is almost unchanged (around 115 wt%, see Fig. 8c). The increasing concentration of the PMSQ microsphere provides more nucleation sites for the wax precipitation and thus favors the performance improvement of the composite particle. At the PMSQ microsphere dosage 5 ppm, although the increasing concentration of the PMSQ microsphere provides more nucleation sites, the amount of EVA PPD adsorbed on the microsphere is decreased with the further increase of the microsphere concentration (see Fig. 8c), which inhibits the performance improvement of the composite particle. We consider that the amount of EVA PPD adsorbed on the microsphere is dominant in controlling the performance of the composite particle (at the PMSQ microsphere dosage 5 ppm). Therefore, the best performance improving of the PMSQ microsphere on the EVA is found at the addition of 2.5 ppm PMSQ. In addition, the composite particle should increase the WAT of Qinghai waxy crude oil through nucleation effect. However, the WAT changes little after adding 50 ppm EVA + 2.5 ppm PMSQ (see Fig. 6a). It is normally considered that the nucleation effect should increase the WAT of crude oil, that is, the EVA/PMSQ composite particles should increase the WAT. Based on the data in Fig. 8c, the EVA PPDs adsorbed on PMSQ microsphere are relatively small (about 5–6 ppm), most of the EVA PPDs is still free molecules. Therefore, the EVA/PMSQ composite particles contribute little to the increase of WAT.

212

F. Yang et al. / Fuel 207 (2017) 204–213

Fig. 8. Microscopic images (a) and TEM image (b) of the solution/dispersion containing both EVA PPD and PMSQ microspheres; amount of the EVA PPD adsorbed onto PMSQ microspheres (c).

Fig. 9. Schematic diagram of the performance improving mechanism of the PMSQ microsphere on the EVA PPD.

F. Yang et al. / Fuel 207 (2017) 204–213

4. Conclusions The monodispersed PMSQ microsphere (2 mm) alone has little influence on the flow behavior of Qinghai waxy crude oil when its dosage is less than or equal to 10 ppm. Compared with the same concentration of EVA2806, small dosages of PMSQ microsphere further inhibited the gelation process and weakened the gel structure strength of the crude oil: the gelation point, G0 , G00 , transient apparent viscosity and yield stress at 15 °C of the crude oil decreased to a lower value. When 50 ppm EVA + 1 ppm PMSQ microspheres were added, the gelation point was suppressed to the lowest from 33.7 °C to 26.9 °C; when 50 ppm EVA + 2.5 ppm PMSQ microspheres were added, the G0 , G00 , transient apparent viscosity and yield stress at 15 °C reduced to the minimum value. Based on the adsorption behavior of EVA on the PMSQ microsphere in oil phase, EVA PPDs can adsorb and concentrate on the PMSQ microsphere to form EVA/PMSQ composite particles. When the PMSQ microsphere concentration is relatively low (0.25 wt%), the EVA molecules could fully adsorb on the microsphere and attain the saturate state. When the PMSQ microsphere concentration is relatively high (0.5 wt%), however, the EVA adsorption on the microsphere could no longer attain the saturate state (unsaturated). Associated with the microstructure of waxy crude oil, it is deduced that the EVA/PMSQ composite particles can effectively act as the nucleation templates for the wax precipitation, resulting in larger and more compact wax microstructures. This kind of wax crystals microstructure reduced the solid-liquid interfacial areas, inhibiting the overlap of the wax crystals; and the compact microstructure occludes fewer liquid oils, favoring rheological beneficiation of the crude oil. Meanwhile, the performance of the EVA/ PMSQ composite particle is obviously influenced by the dosage of the PMSQ microsphere and the amount of adsorbed EVA PPD. Acknowledgments This work was financially supported by National Natural Science Foundation of China (51204202), Natural Science Foundation of Shandong Province of China (ZR2016EEM22), the Fundamental Research Funds for the Central Universities-China (17CX06019) and the Fundamental Research Funds for the Central Universities-China (14CX02210A, 15CX06072A). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.fuel.2017.06.083. References [1] Kok MV, Létoffé J-M, Claudy P, Martin D, Garcin M, Volle J-L. Comparison of wax appearance temperatures of crude oils by differential scanning calorimetry, thermomicroscopy and viscometry. Fuel 1996;75(7):787–90. [2] Yang F, Li C, Wang D. Studies on the Structural characteristics of gelled waxy crude oils based on scaling model. Energy Fuels 2013;27(3):1307–13. [3] Yang F, Li C, Li C, Wang D. Scaling of structural characteristics of gelled model waxy oils. Energy Fuels 2013;27(7):3718–24. [4] Wardhaugh LT, Boger DV. Flow characteristics of waxy crude oils: Application to pipeline design. AIChE J 1991;37(6):871–85.

213

[5] Visintin RFG, Lapasin R, Vignati E, D’Antona P, Lockhart TP. Rheological behavior and structural interpretation of waxy crude oil gels. Langmuir 2005;21(14):6240–9. [6] Yang F, Zhao Y, Sjöblom J, Li C, Paso K. Polymeric wax inhibitors and pour point depressants for waxy crude oils: a critical review. J Disper Sci Technol 2015;36 (2):213–25. [7] Yao B, Wang L, Yang F, Li C, Zhao Y. Effect of vinyl-acetate moiety molar fraction on the performance of poly (octadecyl acrylate-vinyl acetate) pour point depressants: experiments and mesoscopic dynamics simulation. Energy Fuels 2016;31(1):448–57. [8] Yang F, Xiao Z, Yao B, Li C, Wang L, Shi X, et al. Influences of different functional groups on the performance of polyoctadecyl acrylate pour point depressant. Petrol Sci Technol 2016;34(20):1712–9. [9] Machado ALC, Lucas EF, González G. Poly(ethylene-co-vinyl acetate) (EVA) as wax inhibitor of a Brazilian crude oil: oil viscosity, pour point and phase behavior of organic solutions. J Petrol Sci Eng 2001;32(2–4):159–65. [10] Wu C, Zhang J-L, Li W, Wu N. Molecular dynamics simulation guiding the improvement of EVA-type pour point depressant. Fuel 2005;84(16):2039–47. [11] Jin W, Jing J, Wu H, Yang L, Li Y, Shu X. Study on the inherent factors affecting the modification effect of EVA on waxy crude oils and the mechanism of pour point depression. J Disper Sci Technol 2014;35(10):1434–41. [12] Oliveira LMSL, Rita CP, Melo IC, Ribeiro YLL, Reis LG, Dias JCM, et al. Evaluation of the correlation between wax type and structure/behavior of the pour point depressant. Fuel Process Technol 2016;149:268–74. [13] Kango S, Kalia S, Celli A, Njuguna J, Habibi Y, Kumar R. Surface modification of inorganic nanoparticles for development of organic–inorganic nanocomposites—a review. Prog Polym Sci 2013;38(8):1232–61. [14] Kalia S, Haldorai Y. Organic-inorganic hybrid nanomaterials. Switzerland: Springer; 2015. [15] He C, Ding Y, Chen J, Wang F, Gao C, Zhang S, et al. Influence of the nano-hybrid pour point depressant on flow properties of waxy crude oil. Fuel 2016;167:40–8. [16] Yang F, Paso K, Norrman J, Li C, Oschmann H, Sjöblom J. Hydrophilic nanoparticles facilitate wax inhibition. Energy Fuels 2015;29(3):1368–74. [17] Norrman J, Solberg A, Sjöblom J, Paso K. Nanoparticles for waxy crudes: effect of polymer coverage and the effect on wax crystallization. Energy Fuels 2016;30(6):5108–14. [18] Yao B, Li C, Yang F, Sjöblom J, Zhang Y, Norrman J, et al. Organically modified nanoclay facilitates pour point depressing activity of poly(octadecyl acrylate). Fuel 2016;166:96–105. [19] Yao B, Li C, Yang F, Zhang Y, Xiao Z, Sun G. Structural properties of gelled Changqing waxy crude oil benefitted with nanocomposite pour point depressant. Fuel 2016;184:544–54. [20] Al-Sabagh AM, Betiha MA, Osman DI, Hashim AI, El-Sukkary MM, Mahmoud T. Preparation and evaluation of poly(methyl methacrylate)-graphene oxide nanohybrid polymers as pour point depressants and flow improvers for waxy crude oil. Energy Fuels 2016;30(9):7610–21. [21] Al-Sabagh AM, Betiha MA, Osman DI, Hashim AI, El-Sukkary MM, Mahmoud T. A new covalent strategy for functionalized montmorillonite–poly (methyl methacrylate) for improving the flowability of crude oil. RSC Adv 2016;6 (111):109460–72. [22] Lee Y-G, Park J-H, Oh C, Oh S-G, Kim YC. Preparation of highly monodispersed hybrid silica spheres using a one-step sol-gel reaction in aqueous solution. Langmuir 2007;23(22):10875–8. [23] Sankaraiah S, Lee JM, Kim JH. Preparation and characterization of surfacefunctionalized polysilsesquioxane hard spheres in aqueous medium. Macromolecules 2008;41(16):6195–204. [24] Ma W, Zhang D, Duan Y, Wang H. Highly monodisperse polysilsesquioxane spheres: synthesis and application in cotton fabrics. J Colloid Interf Sci 2013;392(1):194–200. [25] Yang F, Yao B, Li C, Sun G, Ma X. Oil dispersible polymethylsilsesquioxane (PMSQ) microspheres improve the flow behavior of waxy crude oil through spacial hindrance effect. Fuel 2017;199:4–13. [26] Barnes HA. The yield stress—a review or ‘pamsa qei’—everything flows? J NonNewtonian Fluid Mech 1999;81(1–2):133–78. [27] Wardhaugh LT, Boger DV. The measurement and description of the yielding behavior of waxy crude oil. J Rheol 1991;35(6):1121–56. [28] Chang C, Boger DV, Nguyen QD. The yielding of waxy crude oils. Ind Eng Chem Res 1998;37(4):1551–9. [29] da Silva JAL, Coutinho JAP. Dynamic rheological analysis of the gelation behaviour of waxy crude oils. Rheol Aata 2004;43(5):433–41. [30] Kané M, Djabourov M, Volle J-L. Rheology and structure of waxy crude oils in quiescent and under shearing conditions. Fuel 2004;83(11–12):1591–605.