J. Cent. South Univ. (2016) 23: 2720-2727

DOI: 10.1007/s11771-016-3333-4

Performance comparison for oil-water heat transfer of circumferential overlap trisection helical baffle heat exchanger

WANG Wei-han(王伟晗w)7 )¹, CHENG Dao-lai(程道来w)7 )¹, LIU Tao(刘涛l)U T)², LIU Ying-hao(刘颖昊)²

1. College of Urban Construction and Safety Engineering, Shanghai Institute of Technology, Shanghai 201418, China;

2. Energy and Environment Research Institute, Baoshan Iron & Steel Co., Ltd, Shanghai 201900, China

Central South University Press and Springer-Verlag Berlin Heidelberg 2016

Abstract: The performance tests were conducted on oil–water heat transfer in circumferential overlap trisection helical baffle heat exchangers with incline angles of 12°, 16°, 20°, 24° and 28°, and compared with a segmental baffle heat exchanger. The results show that the shell side heat transfer coefficient h_o and pressure drop △p_o both increase while the comprehensive index h_o/△p_o decreases with the increase of the mass flow rate of all schemes. And the shell side heat transfer coefficient, pressure drop and the comprehensive index h_o/△p_o decrease with the increase of the baffle incline angle at a certain mass flow rate. The average values of shell side heat transfer coefficient and the comprehensive index h_o/△p_o of the 12° helical baffled scheme are above 50% higher than those of the segmental one correspondingly, while the pressure drop value is very close and the ratios of the average values are about 1.664 and 1.596, respectively. The shell-side Nusselt number Nu_o and the comprehensive index Nu_o·Eu_zo^-1 increase with the increase of Reynolds number of the shell side axial in all schemes, and the results also demonstrate that the small incline angled helical scheme has better comprehensive performance.

Key words: performance experiments; helical baffled heat exchangers; circumferential overlap of baffles; incline angle of baffle; heat transfer enhancement

1 Introduction

The segmental baffled heat exchangers (sgSTHXs) are traditional and mature shell-and-tube heat exchangers (STHXs) which are widely used in various industrial fields, such as petroleum refining, power generation and chemical process. However, they have some drawbacks including stagnant zones, higher pressure drops, and a propensity to induce vibration and fouling. Based on these reasons, YOU et al [1-2] designed the STHX with flower baffles (fbSTHX) and with trefoil-hole baffles (thbSTHX), and the results verified that they are of good thermo hydraulic performance and less liable to foul. On the other hand, some researchers improved the performance of sgSTHX by replacing segmental baffles with the helical ones. LUTCHA and NEMCANSKY [3] used planar surface quadrant sector baffles to replace the curved surface ones to achieve the spiral flow in the helical baffled heat exchangers (HSTHX). But the adjacent baffles can form the V gap, which can cause the leakage flow and affect the performance of the heat exchanger. Many researchers paid attention to the phenomenon and put forward some improved measures, which were focused on the optimum incline angle of the sector baffles and the baffle shapes and connection configurations. It is clear that with the increasing velocity, the smaller the incline angles of the sector baffles are, the higher the heat transfer coefficient and the pressure drop are. LUTCHA and NEMCANSKY [3] indicated that the optimum incline angle of the sector baffles is 40° in the chart of heat transfer coefficient h_o versus pressure drop △p_o. However, JAFARI NASR and SHAFEGHAT [4] obtained a fluctuated curve of heat transfer coefficient versus incline angle. STEHLIK et al [5-7] thought that the optimal helical angle for comprehensive performance of helical baffle heat exchanger was 40°, and proposed to use axial-overlap method to reduce the thread pitch and the leakage flow from the V gap of the adjacent baffles. WANG et al [8] tried to block the V gap with plates, but the experimental results indicated that the comprehensive index of heat transfer coefficient was lower than that of the non-block scheme and the pressure drop increased greatly. WANG et al [9-11] used continuous helical baffles to replace non-continuous helical baffles, and the experimental results showed that the heat transfer performance of the continuous helical baffle heat exchanger was better than the segmental baffle heat exchanger, and the greater the flow velocity is, the more obvious the advantage is. However, compared with the non-continuous helical baffle (such as quadrant and trisection baffles) heat exchanger, the shortcoming of the continuous helical baffle heat exchanger is higher manufacturing difficulty and the higher cost. JI et al [12-13] adopted numerical simulation to prove the staggered overlap helical baffle and continuous helical baffle can effectively improve the comprehensive performance of the heat exchanger. SONG et al [14] designed a kind of baffle structure which widened two straight edges of fan-shaped baffles to form the structure of overlap helical baffles to prevent the short circuit flow, and the experiments showed that the heat transfer performance of this baffle structure was obviously improved. LIN et al [15] applied HTRI software to simulate the best helical angle of helical baffle heat exchanger, and considered that if the shell diameter was certain the optimal helical angle was also certain and the change of the shell side flow rate had no effect on the optimal helical angle. But the simulation result needs to be verified by the plenty of experiments. CHEN et al [16-19] proposed the trisection helical baffled heat exchangers design, especially the circumferential overlap trisection helical baffled shell- and-tube heat exchanger (cothSTHX), which was more suitable to the equilateral triangular tube layouts and had better thermo-hydraulic performances in physical structure. The design of circumferential overlapped of adjacent baffles can minimize the short-cut leakage loss at triangular conjunction notches and improve the rigidity of the heat exchanger tube bundle as well, which were proved by numerical simulation and experimental research. WANG et al [20-21] used fold baffles blocked the triangle leakage zone near the joint point between two plain baffles. In the testing scope the overall heat transfer coefficient increased 7.9%-9.7%, while the shell side pressure drop increased 2.9%-8.0%, and the thermal performance factor TEF enhanced by 28.4-30.7%. DU et al [22-23] put forward a kind of sextant sector helical baffles heat exchanger with six pieces of quadrant sector baffles in each cycle. It also used circumferential overlap to reduce the leakage flow from the V gap, and the study suggested that the comprehensive heat transfer performance was best when the helical angle was 40°.

Compared with the above-mentioned folding baffle and sextant sector baffle schemes, the circumferential overlap trisection helical baffle design was liable to be used in practice as minimal number of parts and simpler structure. In addition, as a new heat exchanger must meet the demand of heat transfer firstly, the larger incline angle design for the pursuit of the best heat transfer coefficient of unit pressure drop at the shell side often can not meet the target of the heat transfer coefficient. So, this work discusses the performances of circumferential overlap trisection helical baffle heat exchanger with incline angles of 12°, 16°, 24° 20° and 28°, and gives the comparison between it and the segmental baffle heat exchanger in order to research the application of circumferential overlap trisection helical baffles scheme in the smaller incline angle range.

2 Performance test rig

The heat exchanger performance test rig was built by Shanghai Institute of Technology and Southeast University, China. It contains two separate systems of oil-water and water-water heat transfer. There are three kinds of heat exchangers in the test rig, shell-and-tube heat exchanger, plate heat exchanger and double pipe heat exchanger. This work chooses the oil-water system and shell-and-tube heat exchanger. The tube-side fluid is municipal water, and the shell side fluid is heat transfer oil. The schematic diagram of test rig is shown in Fig. 1.

Figure 2 shows the photos of five tube bundles in the test helical baffle heat exchanger with incline angles of 12°d) ckness , 16°, 20°, 24° and 28° and one tube bundle in the segmental baffled heat exchanger. The cylinder shell in the test rig can be shared by two kinds of heat exchanger and the tube bundles are replaceable. One-way counter- flow layouts are used in both shell and tube sides. The inner diameter of the shell is F81 mm, and the diameter of the baffles is F79 mm; the effective geometry (outer diameter×thickness×length) of the heat transfer tubes is F10 mm×l mm×832 mm. The number of heat transfer tubes is 16 and the tubes are equilateral triangularly arranged. There are also 3 sets of rods and spanning tubes to fix the baffles. The other geometric parameters are listed in Table 1. Figure 3 shows the projection view of baffles and the tube layout of helical heat exchangers (left) and segmental heat exchangers (right). In the circumferential overlap trisection helical baffle heat exchanger, the two straight edges of each baffle are widened and the baffle occupies more than one third cross sections, which makes the circumferential overlap area of the adjacent baffles be accommodated by a row of tubes for dampening reverse leakage. Both mass flow rate and volume flow rate of heating oil are measured with Emerson F025 mass flow meter (uncertainty ± 0.15%), while the volume flow rate of cooling water is measured with a turbo-flow-meter (uncertainty ± 0.5%). The temperature is measured with platinum resistance thermometers (uncertainty ± 0.15°C). And the shell side pressure drop is measured with a Rosemount 3051S differential pressure transmitter (uncertainty ± 0.15%). The temperature and pressure measuring points are arranged at the tubular bent conjunctions to the inlets and outlets of the testing heat exchanger. The experimental data were collected by an Agilent 34970A data acquisition instrument and processed by operation software programmed on the platform of LabVIEW.

Fig. 1 Schematic diagram of heat transfer test rig:

Fig. 2 Photo of tube bundles of testing heat exchangers:

Table 1 Geometric parameters of testing helical and segmental baffled heat exchangers

3 Method and analysis

3.1 Experiment data process

During the oil-to-water performance tests, the flow rate of heat transfer oil was adjusted by changing the oil pump motor frequency from 25 Hz to 50 Hz by 5 Hz step with an inverter; and its inlet temperature was kept at 85 (±0.5) °C by a electric heater controlled with a solid state relay. The inlet temperature and the flow rate of cooling water in tube-side were kept at 11(±0.4) °C and 0.20 (±2.5%) kg/s (Reynolds number about 2000).

The heat exchanger overall heat transfer coefficient K can be obtained from the Eq. (1). Because of the heat exchanger is new, scale formation can be ignored. The flow of tube side heat transfer coefficient h_i can be estimated by Eq. (2) [24], and the shell side heat transfer coefficient h_o can be calculated by Eq. (3).

Fig. 3 Projection view of baffles and tube layout of helical heat exchangers (a) and segmental heat exchangers (b)

                                (1)

               (2)

                    (3)

where Q_i and Q_o are are the transferring heat at tube-side and shell-side, respectively; A is the heat transfer area; △t_m is the logarithmic mean temperature difference; l is the length of heat exchange tube; d_o and d_i are the tube outer and inner diameters respectively; λ is the thermal conductivity of material of the tubes; μ is fluid dynamic viscosity.

Using dimensionless numbers to process the experiment data is significant for analyzing data, fitting formula of heat transfer and resistance, and the heat exchanger design. Because the shell-side Reynolds number Re₀ demonstrates no additional merit but difficult in calculating the cross section of the helical channel, the shell-side axial Reynolds number Re_z0 defined in Eq. (4) is used reasonably as the independent variable for comparing different schemes, which actually corresponds to the flow rate in the shell side and more fair to compare the performance of different schemes. Equation (5) reflects the relationship between Re_z0 and the axial velocity w₀. The shell-side Nusselt number Nu₀ and the shell-side hydraulic diameter d_h0 are respectively defined in Eqs. (6) and (7). For the similar reason that the friction factor is quite difficult to determine because of the complexity of flow in the helical channel, so the shell-side axial Euler number Eu_z0 is adopted and defined in Eq. (8) based on the Fanning formula to reflect the shell side flow friction factor of different baffle schemes. The shell side pressure drop can be calculated by the Eq. (8) as long as the shell-side axial Euler number Eu_z0 is obtained from the correlation equation. Therefore, this paper uses comprehensive index Nu₀·Re_z0^-1 to analyze and compare the heat transfer capability of all heat exchanger schemes.

                                (4)

                     (5)

                                (6)

          (7)

                              (8)

where G₀ is the mass flow rate of the shell side; D_s, d_r and a are the shell inner diameter, the spanning tube diameter and the tube pitch, respectively; N and n are the tube number and rod number in the radius of the shell, respectively.

3.2 Experimental error analysis

The experimental error of direct measurement parameters can transfer to the indirect ones. The direct measurement parameters include length and diameter of the heat exchanger, the fluid temperature and pressure of inlet and outlet, the flow rate of tube and shell side, and so on. And the direct measurement parameters include the quantity of heat transferred in tube-side and shell- side, the overall heat transfer coefficient and the heat transfer coefficient of shell side, and others. dengl as tal

The indirect experimental error of overall heat transfer coefficient K is estimated with Eq. (9).

                         (9)

The indirect experimental error of the quantity of heat transferred Q is estimated with Eqs. (10) and (11).

                              (10)

                  (11)

Through analyzing the main measuring instrument accuracy of the test rig, the relative experimental error of direct measurement parameters of six heat exchanger schemes are shown in Table 2.

The relative experimental error of indirect measurement parameters of six heat exchanger schemes is obtained by the experimental error transfer theory are listed in Table 3.

Table 3 shows that the maximum relative experimental error of indirect measurement parameters is ±6.45%, thus the experiment data are effective and reliable.

4 Experimental results and discussions

As all the experiments were completed within the same shell and both the size and the number of the replaceable tube bundles are identical, the shell side flow rate G₀ can be reasonably considered independent variable for comparing characteristics of different schemes.

Figures 4 to 7 show the curves of the overall h.t.c. K, the shell side h.t.c. h_o, the shell side pressure drop △p_o and the comprehensive index of h_o/△p_o varied with the shell side flow rate G_o of oil-to-water heat transfer. From the figures it can be seen that the overall h.t.c. K, the shell side h.t.c. h_o and the shell side pressure drop △p_o increases with the increase of the shell side flow rate G_o for all schemes. These parameters increase with the decrease of the baffle incline angle at a constant flow rate for helical schemes. The segmental baffled scheme has the lowest average values of overall h.t.c. and shell side h.t.c. and the second highest average value of shell side pressure drop. Within the testing scope，unlike other helical baffled schemes，12° curve of the overall h.t.c. K and the shell side h.t.c. h_o have shape increase gradient at smaller flow rate G_o and then rise slowly with the increase of flow rate G_o. All of the values of 20° are very close to those of 24° correspondingly. As the flow rate is only changed with the frequency of the pump motor, it can be seen from the Fig. 6 that the flow rate increases but the pressure drop reduces somewhat with the increase of the incline angles at the same pump motor frequency. Figure 7 shows that the curves of the comprehensive index h_o/△p_o of the 12° is highest, followed with 20°, 24° and 28° sequentially, and the segmental baffled scheme is lowest. The ratios of average values of overall h.t.c. K, shell side h.t.c. h_o, pressure drops △p_o and comprehensive index h_o·△p_o^-1 of the 12° helical baffled scheme over those of the segmental baffled scheme are 1.416, 1.664, 1.039 and 1.596, respectively. It shows that h_o/△p_o decreases with the increase of the shell side flow rate for all schemes.

The results indicate that the smaller the baffle incline angle is, the higher the overall h.t.c., shell side h.t.c. and pressure drop of the testing helical baffle schemes are. The shell side h.t.c. h_o and comprehensive index h_o·△p_o^-1 of the 12° helical scheme are above 50% higher than those of the segmental baffled scheme correspondingly, while the pressure drop is very close, which are attractively indicating the superior features of the cothSTHXs. The results also indicated that it is unnecessary for cothSTHXs to deliberately pursue large baffle incline angle with greater difficulty in manufacturing helical baffles for small volumetric liquid fluids.

Table 2 Direct measurement parameters’ relative experimental error of six kinds of heat exchangers

Table 3 Indirect measurement parameters’ relative experimental error of six kinds of heat exchangers

Fig. 4 Overall heat transfer coefficient versus shell side flow rate

Fig. 5 Shell side heat transfer coefficient versus shell side flow rate

Fig. 6 Shell side pressure drop versus shell side flow rate

Fig. 7 Comprehensive index of h_o·△p_o^-1 versus shell side flow rate

Figure 8 shows the shell side heat transfer coefficient h_o versus shell side pressure drop △p_o of six schemes. The h_o-△p_o curve is often used to compare comprehensive performance of heat exchangers. It shows that the change regulation of shell side h.t.c. h_o versus the shell side pressure drop △p_o in all schemes is the same with those in Fig. 7.

Fig. 8 Shell side heat transfer coefficient versus shell side pressure drop of six schemes

Figures 9 and 10 show the regulation that the shell side axial Euler number Eu_z0 and the shell side Nusselt number Nu₀ change with the shell side axial Reynolds number Re_z0. Unlike the linear-like increasing curves of Nusselt number, each curve of Euler number has shaped decrease gradient at smaller axial Reynolds number and then falls slower with the increase of axial Reynolds number. Figure 11 shows the comprehensive index of Nu₀·Re_z0^-1 varies with the shell side axial Reynolds number Re_z0 of these schemes. The positive proportional trend of the curves of comprehensive index Nu₀·Re_z0^-1 versus Reynolds number is quite different from that of the decreased trend of the comprehensive index h_o·△p_o^-1 versus shell side flow rate as shown in Fig. 7.

Fig. 9 Shell side Nusselt numbers versus shell side axial Reynolds number

Fig. 10 Shell side axial Euler numbers versus shell side axial Reynolds number

Fig. 11 Comprehensive indexes Nu₀·Eu_z0^-1 versus shell side axial Reynolds number

The above test results show that the best incline angle for the circumferential overlap trisection helical baffle heat exchangers with small shell side flow rate (such as oil cooler) is around 12°. Small incline angle scheme not only has the highest heat transfer coefficient and the highest comprehensive evaluation index, but also easy to manufacture. However, whether this conclusion is suitable for larger volume flow rate or larger size heat exchangers has yet to be confirmed by further studies.

5 Conclusions

1) The testing schemes on oil-water test rig include five circumferential overlap trisection helical baffle heat exchangers (cothSTHXs) with incline angles of 12°, 16°, 20°, 24°, 28° and a segmental baffled one. Incline angle of baffle has important influence on the heat transfer performance of cothSTHXs.

2) The heat transfer performance test results are presented including both overall h.t.c. K and shell-side h.t.c. h_o, pressure drop △p_o and the comprehensive index h_o·△p_o^-1 vary with the shell side flow rate. In the testing scope, the small angled helical scheme demonstrates better performance that the shell side heat transfer coefficient h_o and comprehensive index h_o·△p_o^-1 of the 12° helical scheme are about 66.4% and 59.6% higher than those of the segmental baffled one with approximate pressure drop, which also verifies the superior heat transfer features of the cothSTHXs.

3) The regulation that the shell side axial Euler number Eu_z0, the shell side Nusselt number Nu₀ and the comprehensive index Nu₀·Eu_z0^-1 change with the shell side axial Reynolds number Re_zo are studied. In the testing scope, the shell-side Nusselt number Nu₀ and the comprehensive index Nu₀·Eu_z0^-1 increase with the increase of the shell side axial Reynolds number for all cothSTHXs and demonstrate that the small inclined angle helical scheme has better performance.

References

[1] YOU Yong-hua, FAN Ai-wu, HUANG Su-yi, LIU Wei. Numerical modeling and experimental validation of heat transfer and flow resistance on the shell side of a shell-and-tube heat exchanger with flower baffles [J]. International Journal of Heat and Mass Transfer, 2012, 55: 7561-7569.

[2] YOU Yong-hua, FAN Ai-wu, LAI Xue-jiang, HUANG Su-yi, LIU Wei. Experimental and numerical investigations of shell-side thermo- hydraulic performances for shell-and-tube heat exchanger with trefoil-hole baffle [J]. Applied Thermal Engineering, 2013, 50: 950-956.

[3] LUTCHA J, NEMCANSKY J. Performance improvement of tubular heat exchangers by helical baffles [J]. Chemical Engineering Research & Design, 1990, 68(3): 263-270.

[4] JAFARI N M R, SHAFEGHAT A. Fluid flow analysis and extension of rapid design algorithm for helical baffle heat exchangers [J]. Applied Thermal Engineering, 2008, 28(11/12): 1324-1332.

[5] STEHLIK P, NEMCANSKY J, KRAL D. Comparison of correction factors for shell-and-tube heat exchangers with segmental or helical baffles [J]. Heat Transfer Engineering, 1994, 15(1): 55-65.

[6] STEHLIK P, WADEKAR V. Different strategies to improve industrial heat exchange [J]. Heat Transfer Engineering, 2002, 23(6): 36-48.

[7] SUN Qi, CHEN Jia-jia, ZHU Ying, WANG Hai-xiu, WANG Shu-li. Hydrodynamic studies on the shells of the heat exchangers with overlap helical baffles [J]. Chemical Engineering & Machinery, 2008, 35(1): 10-13. (in Chinese)

[8] WANG Liang, LUO Lai-qin, WANG Qiu-wang, ZENG Min, TAO Wen-quan. Effect of inserting block plates on pressure drop and heat transfer in shell-and-tube heat exchangers with helical baffles [J]. Journal of Engineering Thermo Physics, 2001, 22(Suppl): 173-176. (in Chinese)

[9] ZENG Min, PENG Bo-tao, YU Peng-qing，CHEN Qiu-yang, WANG Qiu-wang, HUANG Yan-ping, XIAO Zhe-jun. Experimental study of heat transfer and flow resistance characteristics for shell-and-tube heat exchangers with continuous helical baffles [J]. Nuclear Power Engineering, 2006, 27(Suppl): 102-106. (in Chinese)

[10] WANG Qiu-wang, CHEN Qiu-yang, CHEN Gui-dong, ZENG Min. Numerical investigation on combined multiple shell-pass shell-and- tube heat exchanger with continuous helical baffles [J]. International Journal of Heat and Mass Transfer, 2009, 52(5/6): 1214-1222.

[11] WANG Qiu-wang, ZENG Min, MA Ting, DU Xue-ping, YANG Jian-feng. Recent development and application of several high- efficiency surface heat exchangers for energy conversion and utilization [J]. Applied Energy, 2014, 135: 748-777.

[12] JI Shui, DU Wen-jing, CHENG Lin. Numerical studies on double shell-pass heat exchanger with continuous helical baffles [J]. Journal of Engineering Thermo Physics, 2010, 31(4): 651-654. (in Chinese)

[13] JI Shui, DU Wen-jing, WANG Peng, CHENG Lin. Field synergy analysis on shell-side flow and heat transfer of heat exchanger with staggered overlap helical baffles [J]. Proceedings of the CSEE, 2011, 31(20): 75-79. (in Chinese)

[14] SONG Xiao-ping, PEI Zhi-zhong. Shell and tube heat exchanger with anti-short circuit spiral baffle plate [J]. Petro-Chemical Equipment Technology, 2007, 28(3): 13-17. (in Chinese)

[15] LIN Yu-juan, LIU Dan, YANG Xiao-bo, LI Lei-peng. Study on the best spiral angle of helical baffle heat exchanger based on htri software [J]. Science Technology and Engineering, 2012, 12(5): 1181-1184. (in Chinese)

[16] CHEN Ya-ping, SHENG Yan-jun, DONG Cong, WU Jia-feng. Numerical simulation on flow field in circumferential overlap trisection helical baffle heat exchanger [J]. Applied Thermal Engineering, 2013, 50(1): 1035-1043.

[17] DONG Cong, CHEN Ya-ping. Impact of block plates on the flow and heat transfer performance of middle-axial-overlap helical baffle heat exchangers [J]. Journal of Mechanical Engineering, 2014, 50(6): 135-140. (in Chinese)

[18] WANG Wei-han, CHEN Ya-ping, CAO Rui-bing, SHI Ming-heng. Analysis of secondary flow in shell-side channel of trisection helix heat exchangers [J]. Journal of Southeast University: English Edition, 2010, 26(3): 426-430.

[19] DONG Cong, CHEN Ya-ping, WU Jia-feng. Comparison of heat transfer performances of helix baffled heat exchangers with different baffle configurations [J]. Chinese Journal of Chemical Engineering, 2015, 23: 255-261.

[20] WANG Si-min, WEN Jian. Experiment on heat transfer performance of helical baffled heat exchanger without short circuit flow [J]. Journal of Xi’an Jiaotong University, 2012, 46(9): 12-15, 42. (in Chinese)

[21] WEN Jian, YANG Hui-zhu, WANG Si-min, XU Shi-feng, XUE Yu-lan, TUO Han-fei. Numerical investigation on baffle configuration improvement of the heat exchanger with helical baffles [J]. Energy Conversion and Management, 2015, 89: 438-448.

[22] DU Wen-jing, WANG Hong-fu, CAO Xing, CHENG Lin. Heat transfer and fluid flow on shell-side of heat exchangers with novel sextant sector helical baffles [J]. CIESC Journal, 2013, 64(9): 3123-3129. (in Chinese)

[23] GAO Xing, DU Wen-jing, CHENG Lin. Performance of heat exchanger with novel overlapped helical baffles [J]. Journal of Engineering Thermo Physics, 2013, 34(6): 1130-1132. (in Chinese)

[24] YANG Shi-ming, TAO Wen-quan. Heat transfer [M]. Fourth Edition. Beijing: Higher Education Press, 2006: 251-252. (in Chinese)

(Edited by DENG Lü-xiang)

Foundation item: Project(50976035) supported by the National Natural Science Foundation of China; Project(4521ZK120064004) supported by the Science and Technology Commission Green Energy and Power Engineering of Special Fund Project of Shanghai, China

Received date: 2015-06-23; Accepted date: 2015-11-19

Corresponding author: CHENG Dao-lai, Professor, PhD; Tel: +86-13311998959; E-mail: darkliu@sohu.com