A form-stable phase change material made with a cellulose acetate nanofibrous mat from bicomponent electrospinning and incorporated capric–myristic–stearic acid ternary eutectic mixture for thermal energy storage/retrieval

Abstract

An innovative type of form-stable phase change material (PCM) was prepared by incorporating a capric–myristic–stearic acid (CMS) ternary eutectic mixture with a cellulose acetate (CA) nanofibrous mat that was derived from electrospinning a binary mixture of CA/polyvinylpyrrolidone (PVP) and subsequent selective dissolution of PVP component from the obtained bicomponent nanofibrous mat. PVP removal from the CA/PVP bicomponent nanofibers created nanoporous features on the resultant CA nanofiber surface and increased CMS incorporation capability of the nanofibrous mat. Morphology, thermal behavior and durability, and thermal energy storage/retrieval capacity of the prepared CMS/CA nanofibrous form-stable PCM were investigated. This form-stable PCM could maintain well the PCM characteristics even after multiple thermal cycle uses and demonstrated great thermal storage/retrieval capability and temperature regulation ability.

---

1. Introduction

Thermal energy storage and retrieval have drawn continuous attention for decades. Phase change materials (PCMs), the material that stores and releases thermal energy in the process of a state/phase change, play an important role in the applications of various fields such as temperature-regulating textiles, solar energy utilization, building energy efficiency (e.g., wallboards/concretes, roofs and/or floor boards), waste heat recovery, active and passive solar energy storage systems, and air-conditioning system. Based on their transition phases, PCMs can be classified into four categories: solid to solid, solid to liquid, solid to gas, and liquid to gas. However, solid to solid transition goes through change of crystalline phase and has small latent heat storage density while solid to gas or liquid to gas transition have large volume changes associated with phase transition, which limit the uses of solid to solid or gas as well as liquid to gas PCMs in latent heat thermal storage systems (LHTES). Therefore, solid to liquid PCMs are one of the most promising LHTES materials.^1–3

Among all investigated solid to liquid PCMs, fatty acids (CH₃(CH₂)_2nCOOH) and their eutectics have attracted growing attention recently as they possess desirable thermodynamic and kinetic characteristics for low temperature LHTES such as appropriate phase transition temperature, high energy storage capacity, small volume change, little or no super-cooling, low corrosiveness and non-toxicity, good chemical and thermal stability, and cost-effectiveness. With increase of carbon atoms in fatty acid molecules, melting temperature, melting heat, and degree of crystallization of corresponding fatty acid gradually increase.^2–9 Furthermore eutectic mixture of fatty acids can be used to adjust phase transition temperature and melting heat of LHTES into preferable range for certain applications.^5–12 Nonetheless there is a serious problem for solid–liquid fatty acids and/or their eutectics in practice, i.e. large leakage may occur during solid–liquid phase transition. Therefore fabrication of form-stable PCMs have been extensively investigated in recent years.^12–19

Direct incorporation of molten PCMs into porous matrix is a quite straightforward method to obtain form-stable PCMs. It is well known that overlaid nanofibrous mat from electrospinning possesses high porosity and large specific surface area, which makes it an excellent candidate to incorporate substantial amount of fatty acid and/or their eutectics for form-stable PCMs. In our previous research, fatty acids and/or their eutectics were successfully absorbed in and then supported by electrospun nanofibrous mats including polyamide 6, polyacrylonitrile and its derived carbon, and SiO₂.^17–19 The prepared PCM materials could well-retain their overall shape and weight when fatty acids and/or their eutectics were in molten state. In other words, electrospun nanofibrous mat is a good porous matrix to prevent solid–liquid PCMs from flowing/leaking in practice.

Although incorporation of fatty acids and/or their eutectics into electrospun nanofibrous mats has been investigated, incorporation capability of the nanofibrous mat still needs to be improved. In addition, the investigation of ternary fatty acid eutectic mixture as PCM has rarely been reported; the resulting composite PCMs prepared by physical absorption had not adjustable dimensions/sizes. Hence, in this study, cellulose acetate/polyvinylpyrrolidone (CA/PVP) were electrospun to bicomponent nanofibers and PVP was then selectively dissolved by water treatment.^20–23 The motivation of PVP removal was to create nanoporous structure on nanofiber surface and increase incorporation capability of fatty acids and/or their eutectics of the nanofibrous mat. Capric–myristic–stearic acid (CMS) ternary eutectic mixture was prepared as a model PCM followed by being absorbed in and supported by water-treated CA/PVP nanofibrous mat to fabricate form-stable PCMs. Morphological structures, PCM incorporation capability, thermal energy storage/retrieval capacity of the electrospun nanofibrous mats as well as resultant form-stable PCMs were investigated. The prepared CA/CMS nanofibrous form-stable PCMs exhibited desired morphological and physical properties for storing/retrieving thermal energy.

2. Experimental

2.1. Materials

Cellulose acetate (CA, M_n = 29 000), polyvinylpyrrolidone (PVP, M_w = 1 300 000), acetone, N,N-dimethylacetamide (DMAc), powders of capric acid (CA), myristic acid (MA) and stearic acid (SA) were supplied by Shanghai Chemical Regents Co. (Shanghai, China). The chemicals/materials were used as received without further purification.

2.2. Preparation of fatty acid ternary eutectic mixture

Capric–myristic–stearic acid (CMS) ternary eutectic mixture was prepared using theoretical calculation and experimental procedure in literature.^5,6,8–10 Mass ratio of CA, MA and SA in the eutectic mixture was determined as 76.97 : 18.55 : 4.48. The CMS mixture was heated to 80 °C, held at this temperature for 2 h, and then magnetically stirred for 2 min to ensure homogeneity. Subsequently the CMS mixture was cooled down to room temperature in air to get it ready for PCM use.

2.3. Preparation of CA based nanofibrous mat

Spin dope for electrospinning is prepared by dissolving CA in 2 : 1 (w : w) acetone/DMAc mixture solvents at 6 wt% followed by dissolving PVP powder in the above-prepared CA solution at CA/PVP mass ratio of 75/25, 60/40, 50/50, 43/57 and 37/63, respectively. These CA/PVP spin dopes were magnetically stirred at room temperature for 12 h before electrospinning. Electrospinning was then performed at 12 kV with feeding rate of spin dope at 1 ml h⁻¹. An aluminum foil covered roller with diameter of ∼25 cm was used to collect nanofibers at rotating speed of 100 rpm.

The obtained CA/PVP bicomponent nanofibrous mats were dried under vacuum at 60 °C to remove residual solvents and named as CA/PVP-1, CA/PVP-2, CA/PVP-3, CA/PVP-4, and CA/PVP-5, based on their mass ratios of 75/25, 60/40, 50/50, 43/57 and 37/63, respectively. The CA/PVP bicomponent nanofibrous mats were further immersed in deionized water for 24 h under ultrasonication to dissolve PVP. The water treated CA/PVP nanofibrous mats were consequently named as CA-1, CA-2, CA-3, CA-4, and CA-5, respectively, and dried at 60 °C for 10 h to get ready for form-stable PCM fabrication.

2.4. Fabrication of form-stable PCMs

The prepared CMS ternary eutectic mixture was placed first in a beaker at 60 °C to get completely melted. Electrospun nanofibrous mats were respectively immersed into the molten CMS ternary eutectic mixture for 60 min to ensure saturate absorption. The nanofibrous mats with absorbed CMS ternary eutectic mixture were finally hung in an oven at 60 °C for 10 h to remove excess CMS on surface of the nanofibrous mats.

2.5. Characterization

Fourier transform infrared (FTIR) spectra of nanofibrous mats were acquired from a Nicolet MAGNA-IR 750 spectrometer in the wavenumber range of 400 to 4000 cm⁻¹ with a resolution of 4 cm⁻¹. A SU1510 scanning electron microscope (SEM) was employed to examine morphologies of the prepared samples. Prior to SEM examination, all specimens were sputter-coated with gold to avoid charge accumulation.

Thermal behaviors of individual fatty acids (CA, MA, and SA), CMS ternary eutectic mixture, and form-stable PCMs were characterized using a Q200 DSC thermal analyzer in N₂ atmosphere at N₂ flow rate of 25 ml min⁻¹. DSC curves were recorded from −10 to 100 °C with a scanning rate of 5 °C min⁻¹. The DSC data were reproducible and corresponding standard deviations were typically less than ±2.0%. The extrapolated peak onset temperatures (T_e, defined as the intersection between the tangent to the maximum rising slope of the peak and extrapolated sample baseline), melting peak temperature (T_m) and crystallization peak temperature (T_c) are used to characterize the melting and freezing characteristics. Melting/crystallization heats (ΔH_m/ΔH_c) of fatty acids and their ternary eutectic mixture were calculated based upon area under the melting/crystallization DSC peaks through the thermal analysis software that came with the DSC instrument.

Thermal durability of the CA nanofiber based form-stable PCM was characterized by thermal cycling test. In a full thermal cycle, the prepared form-stable PCM was first placed in a 2 ml centrifugal tube, next heated to melt in a 50 °C water bath, and finally cooled down in a 4 °C ice/water bath. Thermal behavior of the form-stable PCM before and after 20 thermal cycles was compared.

Thermal energy storage/retrieval capability of the fabricated form-stable PCMs was evaluated and compared with non-PCM containing water-treated CA/PVP nanofibrous mats by using a laboratory-made set-up. In the process of evaluation, certain amount of the form-stable PCM was placed in a glass bottle and a thermocouple was then inserted and tightly wrapped with the PCM. The glass bottle was subsequently placed in a 50 °C water bath and kept there until the PCM's temperature stabilized. The glass bottle was further transferred to a 4 °C ice/water bath and consequent temperature changes of the form-stable PCM and control sample were recorded using a computer program. The temperature change profile of each sample was obtained by averaging three measurements.

3. Results and discussion

3.1. Thermal properties of individual fatty acids and CA–MA–SA ternary eutectic mixture

DSC curves of CA, MA, SA as well as CA–MA–SA (CMS) ternary eutectic mixture are shown in Fig. 1. Corresponding thermal data including melting peak temperature (T_m), crystallization peak temperature (T_c), extrapolated peak onset temperatures in the process of melting or crystallization (T_e), melting heat (ΔH_m), and crystallization heat (ΔH_c) of these PCMs are summarized in Table 1. Phase transition temperatures as well as melting/crystallization heats of individual fatty acids increased with the size of their molecules from CA (C₁₀H₂₀O₂) to MA (C₁₄H₂₈O₂) to SA (C₁₈H₃₆O₂). The phase transition temperatures of these individual fatty acids, however, are relatively high for low temperature LHTES in practice. Fortunately, the prepared CMS ternary mixture (CA : MA : SA = 76.97 : 18.55 : 4.48) exhibited desired phase transition temperature and presented single endothermic/exothermic peak, indicating eutectic state of the mixture. The melting and crystallization temperatures of the CMS ternary eutectic mixture were 23.6 and 13.1 °C, respectively, and corresponding phase change heat was 122.8 and 119.7 kJ kg⁻¹, respectively. The phase transition temperature of the CMS eutectic mixture was much lower than that of individual fatty acids while the latent heat was still good for PCM uses. The fatty acids and/or their eutectics as PCMs have been applied in several energy storage applications such as wallboard with the PCM energy storage system,²⁴ the solar heat storage and photovoltaic system,^25,26 condensing heat recovery of the air conditioning system,²⁷ the temperature adaptable greenhouses, thermo-regulating fibers and smart textile materials.¹

Fig. 1 DSC curves of CA, MA, SA and CMS ternary eutectic mixture.

Table 1 Thermal behaviors of CA, MA, SA, CMS ternary eutectic mixture and CMS/CA nanofiber based form-stable PCM^b

Samples	Heating process			Cooling process
	Te (°C)	Tm (°C)	ΔHm (kJ kg^−1)	Te (°C)	Tc (°C)	ΔHc (kJ kg^−1)
a This specimen underwent 20 thermal cycles before DSC test.b Tm – melting peak temperature; Tc – crystallization peak temperature; Te – extrapolated peak onset temperatures in the process of melting or crystallization; ΔHm – melting heat; ΔHc – crystallization heat.
CA	31.1	32.7	166.7	29.3	29.3	163.1
MA	53.7	56.1	187.3	52.1	51.9	184.9
SA	69.0	70.5	222.8	67.1	66.5	226.7
CMS	17.4	23.6	122.8	15.3	13.1	119.7
Form-stable PCM	17.7	21.9	101.8	15.5	14.7	99.0
Form-stable PCM^a	17.4	21.9	100.5	14.3	13.0	97.4

---

3.2. CMS incorporation capacity of CA/PVP nanofibrous mats

Incorporation capability of CMS eutectic mixture in nanofibrous mat is extremely important for the form-stand PCMs and it directly determines the amount of CMS in the fabricated form-stable PCMs. More CMS eutectic mixture in the form-stand PCMs, higher thermal storage/retrieval capability and temperature regulation competence. CMS incorporation capacity of nanofibrous mat was calculated using the following equation:
where m₀ and m are the mass of nanofibrous mats before and after absorption of CMS ternary eutectic mixture, respectively. Typical standard deviation of the measured CMS incorporation capacity of nanofibrous mats was less than ±2.0% and the average was adopted to characterize and compare all the nanofibrous mats.

CMS incorporation capacities of CA/PVP bicomponent nanofibrous mats before and after water treatment were plotted in Fig. 2. The good CMS incorporation capacity of as-electrospun CA/PVP bicomponent nanofibrous mat at ∼58.2% suggested that nanofibrous mat was in favor of accommodating large amount of PCMs due to their inter-fiber porous structure and high surface-to-volume ratio. CMS incorporation capacities of water-treated CA/PVP nanofibrous mats were 70.1% for CA-1, 77.4% for CA-2, 80.1% for CA-3, 83.3% for CA-4, and 72.4% for CA-5, respectively. Since the CA-4 nanofibrous mat, i.e. water treated CA/PVP nanofibrous mat with original mass ratio CA/PVP = 43/57, showed the highest CMS incorporation capacity, it was chosen to be used to make CMS based form-stable PCM in the following research.

Fig. 2 CMS absorption capacities of CA/PVP bicomponent nanofibrous mats before and after water treatment.

3.3. Characterization of CA/PVP nanofibrous mats and prepared form-stable PCMs

Removal of PVP from CA/PVP bicomponent nanofibrous mats was verified by comparing FTIR spectra of CA/PVP bicomponent nanofibrous mats before and after water treatment (Fig. 3). The FTIR spectrum of CA/PVP bicomponent nanofibers showed not only stretching vibration of ester carbonyl group at ∼1734 cm⁻¹ and stretching vibration of C–O group at ∼1228 cm⁻¹ and ∼1032 cm⁻¹, which are characteristic peaks of CA,^22,23,28,29 but also stretching vibration of amide carbonyl group at ∼1656 cm⁻¹, bending vibration of –CH₂– group at ∼1498 cm⁻¹ and 1422 cm⁻¹, and stretching vibration of tertiary C–N group at ∼1295 cm⁻¹, which are characteristic peaks of PVP.^30,31 After water treatment, the characteristic peaks of CA were still present while the characteristic peaks of PVP significantly weakened or disappeared, indicating most of PVP, if not all, was removed by water treatment.

Fig. 3 FTIR spectra of CA/PVP bicomponent nanofibers before and after water treatment: (a) before water treatment; (b) after water treatment.

Visual appearance of CA/PVP bicomponent nanofibrous mats showed little difference before and after water treatment (Fig. 4(a) and (b)). The prepared form-stable PCM also retained original mat shape and appearance (Fig. 4(c)), indicating that electrospun nanofibrous mats are mechanically strong enough for form-stable PCM preparation. A closer look at as-electrospun CA/PVP bicomponent nanofibers by SEM indicated that these nanofibers are perfectly cylindrical and quite smooth with uniform diameters of ∼1 μm (Fig. 5(a) and (b)). As a comparison, water-treated CA/PVP nanofibers presented large size variation and rough surface with long and narrow grooves, dents, and pores on surface (Fig. 5(c) and (d)). Slight fiber mergence was also observed. The rough surface of water-treated CA/PVP nanofibers would further facilitate CMS incorporation in the nanofibrous mat in addition to the pre-existing inter-fiber pores. SEM of the form-stable PCM that consists of water-treated CA/PVP nanofibrous mat and absorbed CMS ternary eutectic mixture demonstrated that the fibrous mat morphology was retained but CMS covered nanofiber surface and most of inter-fiber pores (Fig. 6(a)). After 20 thermal cycles, no structural degradation was observed for the form-stable PCM, i.e. shape and morphology of the form-stable PCM was generally maintained except for some CMS reorganization due to phase transition (Fig. 6(b)), suggesting little leakage during the PCM phase transition and excellent thermal durability of the prepared form-stable PCM.

Fig. 4 Photographs of as-electrospun CA/PVP bicomponent nanofibrous mat (a), water treated CA/PVP nanofibrous mat (b), and form-stable PCM (c).

Fig. 5 SEM images of CA/PVP bicomponent nanofibers before water treatment: (a) 10 000×; (b) 30 000×; and after water treatment: (c) 10 000×; (d) 30 000×.

Fig. 6 Representative SEM images of CA based form-stable PCM prepared from water-treated CA/PVP nanofibrous mat with absorbed CMS ternary eutectic mixture: (a) before thermal cycles; (b) after thermal cycles.

3.4. Thermal behavior and durability of the CMS/CA nanofiber based form-stable PCM

Thermal behaviors of the CMS/CA nanofiber based form-stable PCM in the process of heating/cooling were characterized using DSC (Fig. 7 and Table 1). Compare to pure CMS ternary eutectic mixture (Table 1), melting peak temperature (T_m) of the prepared form-stable PCM was slightly lower while the opposite was observed for corresponding crystallization peak temperature (T_c). This might be attributed to three aspects: (1) weak attractive interaction between fatty acid molecules and CA substrates;^5,10 (2) depression of phase transition temperatures of CMS within porous nanofibrous substances;^5,17,32 (3) degree of super-cooling was smaller in the form-stable PCM than in neat CMS powders in the process of phase transition.³³ Theoretically, melting/crystallization heat of the form-stable PCM can be calculated through multiplying corresponding phase transition heat of pure CMS by mass percentage of CMS in the form-stable PCM. The experimental melting/crystallization heat of the prepared form-stable PCM (Table 1) was close to theoretical values from pure CMS (Table 1) and corresponding CMS incorporation capacity of CA-4 nanofibrous mat (83.3%).

Fig. 7 DSC curves of CA based form-stable PCM consisting of water treated CA/PVP nanofibrous mats and CMS ternary eutectic mixture before and after 20 thermal cycles.

The comparisons on thermal energy storage/retrieval properties of the obtained CMS/CA nanofiber based form-stable PCM with those of the different composite PCMs in literature were listed in Table 2. It was observed that CMS/CA nanofiber based form-stable PCM had a comparably higher melting/crystallization heat for thermal storage/retrieval. The fabricated form-stable PCM demonstrated high thermal storage capacity and temperature regulation ability.

Table 2 Comparisons on thermal energy storage/retrieval properties of the prepared form-stable PCM with those of some composite PCMs in literatures

Composite PCMs	Tm (°C)	ΔHm (kJ kg^−1)	Tc (°C)	ΔHc (kJ kg^−1)	Ref.
MA–PA–SA(92.9 wt%)/EG	41.64	153.5	42.99	151.4	5
LA–MA–PA(94.7 wt%)/EG	30.94	135.9	—	—	6
CA–MA–PA(92.9 wt%)/EG	18.61	128.2	16.58	124.5	8
LA–MA–SA(92.3 wt%)/EG	29.05	137.1	29.38	131.3	9
LA–PA–SA(55.0 wt%)/EP	31.8	81.5	30.3	81.3	10
CA–LA–PA(80 wt%)/CNTs	17.2	101.6	8.1	100.4	12
CA–MA–SA(83.3 wt%)/CA nanofibers	21.9	101.8	14.7	99.0	Present study

---

Thermal durability of the CMS/CA nanofiber based form-stable PCM was characterized by thermal cycling test. Thermal properties of the form-stable PCM after 20 thermal cycles showed little change in phase transition temperature and heat, indicating that the form-stable PCM can well maintain their PCM characteristics even after multiple thermal cycle uses (Fig. 7 and Table 1). This is consistent with the morphology characterization and indicates that water treated CA/PVP nanofibrous mat with incorporated CMS is an excellent form-stable PCM for thermal energy storage/retrieval uses.

3.5. Thermal storage/retrieval capability of the CMS/CA nanofiber based form-stable PCM

Thermal storage/retrieval capability of the CMS/CA nanofiber based form-stable PCM was evaluated by monitoring temperature change of a 50 °C sample in a 4 °C environment (Fig. 8). For a control sample, i.e. water treated CA/PVP bicomponent nanofibrous mat, temperature dropped quickly. It took only 7 min for the control sample to cool from 50 °C down to 12 °C. In contrast, temperature of a form-stable PCM sample, i.e. water treated CA/PVP nanofibrous mat with CMS ternary eutectic mixture, decreased much more slowly. Compared to the control sample, the same temperature change from 50 °C to 12 °C took almost 5 times longer for the form-stable PCM. There was a constant temperature plateau at ∼17 °C in the temperature change profile of the form-stable PCM sample and it lasted approximately 12 min. It is obvious that the CMS/CA nanofiber based form-stable PCM has great thermal storage/retrieval capability and temperature regulation ability.

Fig. 8 Temperature change profiles of initial 50 °C samples in a 4 °C environment: blue – water treated CA/PVP bicomponent nanofibrous mat (control sample); red – CMS/CA nanofiber based form-stable PCM.

4. Conclusions

In this research, capric–myristic–stearic acid (CMS) ternary eutectic mixture (CA : MA : SA = 76.97 : 18.55 : 4.48) was employed and incorporated into water-treated cellulose acetate/polyvinylpyrrolidone (CA/PVP) electrospun nanofibrous mat to fabricate a novel form-stable PCMs. Selective removal of PVP upon water dissolution created nanoporous feature on the resultant CA nanofiber surface, which facilitated CMS incorporation in the nanofibrous mat. The water-treated CA/PVP nanofibrous mat with original mass ratio CA/PVP = 43/57 showed the highest CMS incorporation capacity at ∼83.3 wt%. The CMS/CA nanofibrous form-stable PCM well maintained their structural integrity as well as PCM characteristics after 20 thermal cycle uses. Compared to the control sample (water treated CA/PVP nanofibrous mat), it took almost 5 times longer for the CMS/CA nanofibrous form-stable PCM to change its temperature from 50 °C to 12 °C in a 4 °C environment. CMS/CA nanofiber based form-stable PCM presented great thermal storage/retrieval capability and demonstrated great potential for temperature regulation applications.

Acknowledgements

This research was financially supported by the China Postdoctoral Science Foundation (No. 2015T80496 and No. 2014M560391), the Six Talent Peaks Project in Jiangsu Province (No. 2014-XCL001) and the Fundamental Research Funds for the Central Universities (No. JUSRP51505).

Notes and references

1. Y. P. Yuan, N. Zhang, W. Q. Tao, X. L. Cao and Y. L. He, Renewable Sustainable Energy Rev., 2014, 29, 482–498.
2. Q. Q. Tang, J. Sun, S. M. Yu and G. C. Wang, RSC Adv., 2014, 4, 36584–36590.
3. K. Pielichowska and K. Pielichowski, Prog. Mater. Sci., 2014, 65, 67–123.
4. W. D. Liang, G. D. Zhang, H. X. Sun, Z. Q. Zhu and A. Li, RSC Adv., 2013, 3, 18022–18027.
5. X. J. Yang, Y. P. Yuan, N. Zhang, X. L. Cao and C. Liu, Sol. Energy, 2014, 99, 259–266.
6. N. Zhang, Y. P. Yuan, X. Wang, X. L. Cao, X. J. Yang and S. C. Hu, Chem. Eng. J., 2013, 231, 214–219.
7. F. Tang, D. Su, Y. J. Tang and G. Y. Fang, Sol. Energy Mater. Sol. Cells, 2015, 141, 218–224.
8. Y. G. Yuan, Y. P. Yuan, N. Zhang, Y. X. Du and X. L. Cao, Mater. Lett., 2014, 125, 154–157.
9. C. Liu, Y. P. Yuan, N. Zhang, X. L. Cao and X. J. Yang, Mater. Lett., 2014, 120, 43–46.
10. N. Zhang, Y. P. Yuan, Y. G. Yuan, T. Y. Li and X. L. Cao, Energy Build., 2014, 82, 505–511.
11. M. Li, Z. S. Wu and H. T. Kao, Sol. Energy Mater. Sol. Cells, 2011, 95, 2412–2416.
12. X. Meng, H. Z. Zhang, L. X. Sun, F. Xu, Q. Z. Jiao, Z. M. Zhao, J. Zhang, H. Y. Zhou, Y. Sawada and Y. L. Liu, J. Therm. Anal. Calorim., 2013, 111, 377–384.
13. G. R. Dheep and A. Sreekumar, Energy Convers. Manage., 2014, 83, 133–148.
14. W. D. Liang, P. S. Chen, H. X. Sun, Z. Q. Zhu and A. Li, RSC Adv., 2014, 4, 38535–38541.
15. M. M. Kenisarin and K. M. Kenisarina, Renewable Sustainable Energy Rev., 2012, 16, 1999–2040.
16. W. Hu and X. Yu, RSC Adv., 2012, 2, 5580–5584.
17. Y. B. Cai, C. T. Gao, T. Zhang, Z. Zhang, Q. F. Wei, J. M. Du, Y. Hu and L. Song, Renewable Energy, 2013, 57, 163–170.
18. Y. B. Cai, X. Zong, J. J. Zhang, Y. Y. Hu, Q. F. Wei, G. F. He, X. X. Wang, Y. Zhao and H. Fong, Sol. Energy Mater. Sol. Cells, 2013, 109, 160–168.
19. X. Zong, Y. B. Cai, G. Y. Sun, Y. Zhao, F. L. Huang, L. Song, Y. Hu, H. Fong and Q. F. Wei, Sol. Energy Mater. Sol. Cells, 2015, 132, 183–190.
20. L. Zhang and Y. L. Hsieh, J. Nanosci. Nanotechnol., 2008, 8, 4461–4469.
21. E. Saljoughi and T. Mohammadi, Desalination, 2009, 249, 850–854.
22. M. M. Castillo-Ortega, A. Najera-Luna, D. E. Rodriguez-Felix, J. C. Encinas, F. Rodriguez-Felix, J. Romero and P. J. Herrera-Franco, Mater. Sci. Eng., C, 2011, 31, 1772–1778.
23. T. Xiang, Z. L. Zhang, H. Q. Liu, Z. Z. Yin, L. Li and X. M. Liu, Sci. China: Chem., 2013, 56, 567–575.
24. D. A. Neeper, Sol. Energy, 2000, 68, 393–403.
25. S. Tarhan, A. Sair and M. H. Yardim, Energy Convers. Manage., 2006, 47, 2143–2154.
26. A. Hasan, S. J. McCormack, M. J. Huang and B. Norton, Sol. Energy, 2010, 84, 1601–1612.
27. L. Lv, X. Zhang, J. Ye, P. Yang and Q. Sheng, Energy Technol., 2007, 28, 353–355.
28. D. Zavastin, I. Cretescu, M. Bezdadea, M. Bourceanu, M. Dragan, G. Lisa, I. Mangalagiu, V. Vasic and J. Savic, Colloids Surf., A, 2010, 370, 120–128.
29. S. Ramesh, R. Shanti and E. Morris, Carbohydr. Polym., 2013, 91, 14–21.
30. J. Bai, Y. X. Li, C. Q. Zhang, X. F. Liang and Q. B. Yang, Colloids Surf., A, 2008, 329, 165–168.
31. M. J. Gao, L. Sun, Z. Q. Wang and Y. B. Zhao, Mater. Sci. Eng., C, 2013, 33, 397–404.
32. G. Y. Fang, H. Li and X. Liu, Mater. Chem. Phys., 2010, 122, 533–536.
33. M. Mehrali, S. T. Latibari, M. Mehrali, H. S. C. Metselaar and M. Silakhori, Energy Convers. Manage., 2013, 67, 275–282.

---