Design and morphology control of a thiophene derivative through electrospraying using various solvents

Abstract

In the present work, electrospraying of an organic molecule is carried out using various solvents, obtaining fibril structures along with a range of distinct morphologies. Solvent characteristics play a major role in determining the morphology of the organic material. A thiophene derivative (7,9-di(thiophen-2-yl)-8H-cyclopenta[a]acenaphthylen-8-one) (DTCPA) of donor–acceptor–donor (DAD) architecture is used to study this solvent effect. Seven solvents with decreasing vapour pressure are selected for experiments. Electrospraying is conducted at a solution concentration of 1.5 wt% and a constant applied voltage of 15 kV. Gradual transformation in morphology of the electrospun product from spiked-spheres to only spikes is observed. A mechanism describing this transformation is proposed based on electron micrograph analysis and XRD analysis. These data indicate that the morphological change is due to the synergistic effect of both vapour pressure and dielectric constant of the solvents. Through a reasonable control of the crystallite size and morphology along with the proposal of the transformation mechanism, this study elucidates electrospraying as a prospective method for designing architectures in organic electronics.

---

Introduction

Conjugated molecules such as acetylenes and thiophenes have always amazed and excited chemists due to their tailorable charge transportation and electrochemical properties, and therefore find numerous applications ranging from organic field effect transistors (OFETs), organic light emitting diodes (OLEDs), sensors, organic photovoltaics (OPVs),¹ to electromagnetic interference (EMI) shielding materials and radio frequency identification (RFID) tags.^2,3 A conjugated molecule generally has enhanced charge transportation properties along the chain backbone when polymerized. However, due to issues such as insolubility, conjugated polymers face a drawback for progressing further in these applications. On the other hand, fabrication of small molecules are in the limelight, as these molecules are easily soluble, have a conjugated structure, offer better control over morphology and have better crystallinity than their respective polymers and therefore, are considered as good candidates for organic devices. Furthermore, in terms of the purity of materials, small molecules have fewer by-products after the purification process. Therefore research attention on small molecules has been expanding and in some instances they have even replaced conjugated polymers in OPVs⁴ OFETs, and sensors.⁵ The potential focus in the near future would therefore be to increase the size of crystallites for better charge transport properties.

One of the simplest and effective method to increase crystallites size is through solvent management.⁶ Solvent plays a major role in dissolving the solute by creating a uniform phase and thus governs the evolved morphology. However simple it may seem, but crystallinity studies with respect to solvents of these conducting molecules are yet to be explored completely. Solvent processing methods used to fabricate small molecules are simple techniques such as solution casting, dip coating, spin coating, or more advanced methods such as drawing, template synthesis and electrospraying.^7,8

Electrospraying is a technique to control and modify morphology with various controlling parameters such as solvents, solution concentration, applied voltage, tip to collector distance and flow rate. It is a modified form of electrospinning and occurs with insufficient solution viscosity. With this technique, micro and nanostructures^9,10 are obtained by applying high voltages and it has been employed quite often in drug delivery,¹¹ where drug is loaded in these electrosprayed hollow spheres to release or trigger later into the target.¹² Micro or nano-structures have increased surface area and are therefore beneficially utilized in super absorbers and air filters.¹³ Moreover, apart from generating geometrical structures that result in increase in surface area, material specific applications were also reported, for conducting¹⁴ and hydrophobic^6,15 materials as well as in biological^16,17 and chemical sensors.⁵ Therefore to sum up, materials such as commercial polymers,¹⁸ biopolymers,¹⁹ nanoparticles blends,²⁰ dispersed metal ions in solution and conducting polymers or molecules can be electrosprayed.²¹

Studies for selecting suitable solvents for electrospraying^7,22 or electrospinning^23,24 has been addressed to find the regime of solubility using 28 solvents and some of its binary mixtures on poly methylsilesquioxane.²⁵ Other studies showed the effect of solvent evaporation on morphology,^15,26 pore formation,²⁷ microspheres^28,29 and molecular stacking.^30,31

In the current work, effect of various solvents, their solubilities and rate of evaporation at ambient temperature, on the morphology of conducting organic molecule when electrosprayed is evaluated. The organic small molecule studied here is a thiophene derivative (7,9-di(thiophen-2-yl)-8H-cyclopenta[a]acenaphthylen-8-one) (DTCPA) having donor–acceptor–donor (DAD) moieties. It has been studied for feasibility in OPVs via spincoating technique in poly DAD form with [6,6]-phenyl-C61-butyric acid methyl ester (PCBM) using chlorobenzene as solvent³² and as poly(3-hexyl thiophene) : DTCPA composites using chloroform as solvent.³³ Furthermore, properties of a single crystal DTCPA was evaluated, where two orders of increase in conductivity along with charge density analysis was observed.³⁴ It was also blended with poly(ethylene oxide) obtaining electrospun nanofibers of barb wires architecture.³⁵ Therefore, further solvent based studies would provide useful insights for applications such as transistors and sensors. The primary objective of this study is to analyze the morphological change of the electrosprayed particles with respect to solvents. Almost common process parameters are employed and their morphological and structural data are critically analyzed. The change in morphology with respect to change in solvents could be used based on application that is on one hand as a drug carrier through the hollow microspheres and on other hand as control on crystallites size. As these crystallites can be used as nucleating site to enhance the crystallinity or can be combine with other donor acceptor materials as a composite for heterojunction devices.

Experimental

Materials

DTCPA (Fig. 1) has two thiophene groups as donors and a ketone group as acceptor moieties with an absorption range from 300–900 nm, has an optical band gap from 1.49 eV to 1.45 eV and was synthesized as reported.⁴ It is dissolved in a range of seven solvents with varying vapour pressures at 25 °C and dielectric constants and showed partial to high solubility (Table 1). All solvents used are of analytical grade and are purchased from S D Fine Chem. Ltd. Solutions with 1.5 wt% of DTCPA are prepared by approximately 8 hours of stirring.

Fig. 1 Molecular structure of 7,9-di(thiophen-2-yl)-8H-cyclopenta[a]acenaphthylen-8-one (DTCPA).

Table 1 Design of experiments and used solvents, properties³⁶

	Solvents	Vapour pressure (kPa)	Dielectric constant	Voltage (kV)	Flow rate (ml h^−1)	Distance (cm)	Morphology
1	Acetone (Ace)	30.6	20.7	15	2	12	Very small spikes in splattered spheres
2	Chloroform (CF)	26.2	4.8	15	1.5	12	Small spikes on spheres
3	Tetrahydrofuran (THF)	21.6	7.6	15	1.5	12	Spikes on well-developed spheres
4	1,2-Dichloroethane (12DCE)	11.6	10.4	15	1.5	12	Spikes on distorted spheres
5	Chlorobenzene (CB)	1.6	2.7	15	1.5	12	Spikes on a disappeared spheres
6	1,2-Dichlorobenzene (12DCB)	0.2	9.8	15	0.8	16	Large spikes
7	N-Methyl-2-pyrrolidinone (NMP)	0.05	32.0	15	0.8	16	Very large spikes

---

Methods

Electrospraying. Electrospraying is carried out using high voltage, applied using Gamma high voltage RR 50 with voltage control of ±1%. Flow rate is maintained through Holmarc-SPLF-2D infusion pump and aluminium plate of 15 cm × 15 cm wrapped with aluminium foil is used as grounded collector. A median applied voltage of 15 kV is selected from a range of 8–20 kV and process duration of 10 minutes giving a visible uniform deposition on collector. With respect to applied voltage other electrospraying parameters like flow rate and tip to collector distance or flight time are varied as given in Table 1. Flow rate and tip to collector distance is optimized to allow more number of solvents to be accommodated in a common conditions window. All the experiments are carried out at ambient temperature.

Characterization. Morphological analysis of electrosprayed samples are carried out by 5 nm gold sputtering and subsequent imaging of the samples at 5 kV with SE2 detector using Carl Zeiss Ultra55, field emission scanning electron microscope (FESEM). Crystallite size measurements are carried out on FESEM images employing image J® software. Structural analysis is performed using X-Pert PRO, PANalytical XRD with 2θ value ranging from 5–60°. XRD samples are prepared by electrospraying on glass slides, fixed on aluminium foil collector.

Results and discussion

The change in morphology obtained by electrospraying at applied voltage of 15 kV, and modifying the flow rate and tip to collector distance is shown in FESEM images (Fig. 2 to 4), there is a sweeping change in morphology from Fig. 2d to 4g. Transformation in morphology from very small spikes around splattered spheres to exclusively very large crystallites is observed with variation in crystallites size from 3–4 μm to 500 μm. Higher magnification images along with spikes size distribution for each solvent are shown. Fig. 2a–c corresponds to as synthesized DTCPA powder with irregular structures lacking any particular morphology.

Fig. 2 FESEM images and crystallite size distribution of (a–c) as synthesized DTCPA and (d–f) electrosprayed DTCPA using acetone.

Effect of higher vapour pressure of solvents

DTCPA in acetone. Crystallites with more or less same size are randomly distributed with approximate crystallites size of around 3–4 μm, Fig. 2d–f. Due to its high volatility, issues like drying up and blocking of needle are observed, and hence these are overcome by slightly increasing the flow rate from 1.5 ml h⁻¹ to 2 ml h⁻¹.

DTCPA in chloroform. Spikes on spheres are observed, the spheres are of 4–6 μm, and the spikes emerging from sphere surface are of 1–2 μm, Fig. 3a–c. Since structures like spikes on spheres are difficult to design and may require two system or two step process, this idea can be employed in drug delivery systems and sensors as well.^5,12 As these spikes have enhanced the surface area drastically compared to just spheres, with an inherent advantage of conducting nature of the material, this leads to enhancement in sensitivity.

Fig. 3 FESEM images of electrosprayed DTCPA using (a–c) chloroform; (d–f) tetrahydrofuran and (g–i) 1,2-dichloroethane.

DTCPA in tetrahydrofuran. The morphology is similar to chloroform system, but spikes are on well-developed spheres with slight increase in the crystallites size to 4–5 μm are observed, Fig. 3d–f. The spheres are also observed to be hollow (see Fig. S1–S3 in the ESI†). This is due to the evaporation of the solvent during the flight time, leading to thermodynamic instability.²⁶ Phase separation occurs forming (i) material rich outer surface; as the solvent evaporates first and then (ii) solvent rich inner surface; giving rise to hollow spheres as solvent eventually evaporates leaving behind the hollow spheres. Since vapour pressures of CF and THF are closely spaced (26.2 and 21.6), slight change in spheres morphology is observed.

DTCPA in 1,2-dichloroethane. Here, all spheres are distorted and as in the previous case, spheres are found to be hollow (Fig. 3g–i) with spikes of 5–7 μm, but spikes probability on sphere has reduced. Also, there are only few spheres along with independent spikes formation are observed, therefore this solvent could be considered as transition solvent from spike-sphere to only spike morphology.

Effect of lower vapour pressure of solvent

DTCPA in chlorobenzene. Due to the big leap in the vapour pressure from 11.6 kPa of 12DCE to that of 1.6 kPa of CB, spheres have completely disappeared giving way to spike crystallites with slight increase in crystallites size of 6–8 μm, Fig. 4a–c. Additionally, there is a considerable increase in the number of crystallites on the collector.

Fig. 4 FESEM images of electrosprayed DTCPA using (a–c) chlorobenzene (d–f) 1,2-dichlorobenzene and (g–i) N-methyl-2-pyrrolidinone.

DTCPA in 1,2-dichlorobenzene. As the vapour pressure of solvent has further decreased, there is formation of larger spikes of 25–30 μm, Fig. 4d–f. Here the solvent evaporation during flight is minimal and solvent evaporation primarily takes place on collector, hence crystallites had sufficient time to grow in length.

DTCPA in N-methyl-2-pyrrolidinone. This solvent had the lowest vapour pressure among the selected solvents and hence has affected the crystallites size to greater extent with its crystallites length distribution mainly around 300–400 μm, Fig. 4g–i. Furthermore, these crystallites are observed to have branch formation. With 10 μm scale bar image, it can be depicted as fibres with a width of around 1 μm.

For the solvents 12DCB and NMP, because of low vapour pressure, the flight time given for solvent to evaporate being inadequate, wet deposition on the collector surface is observed. To overcome such an undesirable deposition, (i) distances are increased from 12 cm to 16 cm and (ii) flow rates are gradually decreased from 1.5 ml h⁻¹ to 0.6 ml h⁻¹ and optimized at 0.8 ml h⁻¹ to balance the conditions with other solvents, Table 1.

Possible mechanism for progressive evolution in morphology

DTCPA solutions, unlike polymer solutions possess insufficient viscosity and hence behave like liquids. When such a solution is taken for fabrication using high voltage, it apparently causes breaking of jet from the needle tip and results in formation of droplets which eventually form spheres. This electrospraying phenomenon is same for all the solvents regardless of their volatility. However, due to varying vapour pressure of solvents, there is change in size of the structure in both spheres and spikes. That is, as time taken by the solvents for evaporation is increased, spikes length increases, Fig. 5. The same hypothesis also holds good for spheres with an additional factor of droplet size and its weight as seen in Fig. 6, there is upper limit, above which collapsed spheres are observed.

Fig. 5 Crystallite size increments with respect to solvent vapour pressure.

Fig. 6 Spheres size increments with respect to solvent vapour pressure.

For the first four solvents (acetone, CF, THF, 12DCE), due to small size and instant solidification of droplets, the spheres remain intact on the collector, as schematically represented in the first case of “rapid solvent evaporation” (Fig. 7). However, for CB, 12DCB and NMP, because of low evaporation rates the radius of the droplet that emerged at the needle tip remained almost of the same size throughout its flight time. Hence, such large spheres collapsed on collector due to their own mass, and this could also be due to other factors such as low molecular weight of the material and less viscosity. Consequently, circumferential spikes pattern around the disappeared spheres are seen, as shown in the second case of “solvent evaporation at collector” (Fig. 7). This proposed mechanism is supported with FESEM images, Fig. 8.

Fig. 7 Possible mechanism for the evolved morphologies involving various solvents.

Fig. 8 FESEM images to support mechanism of collapsing spheres with (a) acetone (b) CB (c) 12DCB and (d) NMP.

For solvents such as acetone, CB, 12DCB and NMP, spheres appeared to be absent (Fig. 2d and 4a, d and g). However the FESEM images, in (Fig. 8a–d) suggest that sphere formation occurs, but they collapse leaving behind the characteristic splattered circular droplet mark around which these crystallite spikes get arranged. As shown in Fig. 8, four collapsed spheres are highlighted for acetone (Fig. 8a), whereas for CB (Fig. 8b) and 12DCB (Fig. 8c), a single collapsed sphere could be imaged at 10 μm scale. While for NMP (Fig. 8d), single collapsed sphere could be captured only at 100 μm scale, directly indicating the impact of vapour pressures.

Good solubility may not always lead to good electrospinnability.^25,37 Rapid solvent evaporation decreases the stretching factor of the droplets.³⁸ As observed with acetone possessing highest volatility, intact spheres were not formed instead it resulted in burst spheres. This can be attributed to its high dielectric constant compared to other solvents in the vapour pressure range. Higher dielectric constant increases the jet path due to higher net charge density in the solution and by introducing bending instabilities. This facilitates quick solvent evaporation yielding smaller burst spheres.³⁹ In the case of acetone, both vapour pressure and dielectric constant have significant effect in reducing the electrosprayed structures to just spikes or crystallites. Whereas in the case of NMP, inspite of having the highest dielectric constant and increased jet path, it could not aid much in the evaporation of solvent due to very low vapour pressure of 0.05 kPa, consequently leaving a very large collapsed sphere. Since both acetone and NMP are strong polar solvents having dielectric constant more than 15, the molecular dipoles (solvent) tend to cancel out the electric charges present in the given solution. Hence, for these solvents dielectric effects are prominent along with vapour pressures. Furthermore, these results demonstrate that (i) vapour pressure and then (ii) dielectric constant both affect the evolving morphology, Table 1 (morphology column).

In the case of acetone, due to high volatility the electrosprayed material had inadequate time to form spheres and hence only spikes are deposited on the collector, implying before full development of spheres, splattering on the collector occurs. For the next three solvents used; CF, THF and 12DCE, formation of spikes on spheres with variation in spike size and shape of spheres are observed. Whereas, for the last three solvents CB, 12DCB and NMP, due to low vapour pressure, formation of larger spikes with collapsed spheres are observed. Solvents from acetone to 12DCE have poor to partial solubility, leaving the crystallites as such by just forming spheres due to rapid solvent evaporation. Solvents from CB to NMP support high solubility and hence help in enhancing the crystallites size. Table 2 summarizes the solvent evaporation rate with respect to electrospraying flight time.

Table 2 Deduce relation between solvent evaporation time and flight time

Solvents	Observation
Acetone	Solvent evaporation time < flight time
CF, THF and 12DCE	Solvent evaporation time ≅ flight time
CB, 12DCB and NMP	Solvent evaporation time > flight time

---

Structural analysis

X-ray diffraction (XRD) analysis is performed on samples electrosprayed over glass substrates to study the orientation of crystallites. The XRD analysis peak intensities show variation with respect to solvents, Fig. 9. The peaks are normalized and two peak regions are observed between 2θ ranges of (i) 8.5–9° and (ii) 17–17.4°. A similar observation was reported in the study of DTCPA single crystal.³⁴ There are two peaks in between 8.5–9°, at 8.6° and 8.75°. It is observed that 8.6° peak is prominent in CF, CB and 12DCB and 8.75° peak is comparatively higher for acetone, THF and NMP. A broad hump with no particular peak either at 8.6° or 8.75° is observed for 12DCE. In the 17–17.4° region, 17.2° peak is observed to be associated with prominent 8.6° peak, hence is seen only in CF, CB and 12DCB samples. Therefore, it can be speculated that there are two types of crystallite orientations. The crystallite would first like to orient itself with the 8.75° peak. NMP, due to its lowest evaporation rate, has allowed the crystallites to orient the most in only one particular direction, giving highest crystallinity, whereas in acetone and THF, there is a slight reduction in the 8.75° peak intensity with the appearance of 8.6° peak (stronger in THF). As the 8.6° peak becomes more prominent, it diminishes 8.75° peak and contributes the growth in 17.2° peak.

Fig. 9 XRD spectra for various solvent dependent electrosprayed DTCPA samples.

In other words, NMP imparts a particular orientation, acetone; the solvent with highest evaporation rate takes the second position as rapid solvent evaporation hardly allows material to take other orientations. These two solvents having high dielectric constants have increased jet path, which leads to increase in directional orientation. Apart from vapour pressure, dielectric constant also plays an important role in determining the structural orientation. THF, barely prefers one particular orientation. It is followed by 12DCE, a transition solvent, allowing both preferred orientations and having no particular order, which behaves like an amorphous material (Fig. 3g). The other solvents, CF, CB and 12DCB affect structural alignment the least, and hence multiple peaks are observed.

In case of NMP, and acetone, both vapour pressure and dielectric constant play a major role in peak intensity. THF and 12DCE, with intermediate vapour pressures and dielectric constants, behave like amorphous materials. Whereas, in the case of CF, CB and 12DCB, due to low dielectric constants, only vapour pressure effect is significant.

Finally, experiments are also performed on varying voltages of 8–20 kV at same flow rates and distances, and similar morphology is observed with changes in deposition density and patterning. Since the focus is on transformation in morphology due to solvents, this study is confined to solvent dependent morphology.

Conclusions

Influence of solvent on crystallite size of DTCPA was analysed using electrospraying. Solutions of DTCPA with acetone, chloroform, tetrahydrofuran, 1,2-dichloroethane, chlorobenzene, 1,2-dichlorobenzene and N-methyl-2-pyrrolidinone were electrosprayed by optimizing and maintaining almost same parametric conditions. Solvents were selected in the decreasing order of vapour pressures from 30.6 for acetone to 0.05 kPa for that of NMP. Morphological and structural properties were analysed with respect to solvents. It was observed that transformation from spiked spheres to spheres with tuning in the spiked length was feasible and possible mechanism was demonstrated through supporting micrographs. XRD analysis was carried out to understand the change in crystallite size and crystallinity and it was found that there was a definite trend in the peaks with respect to vapour pressure and dielectric constant of solvents. This transformation from spiked spheres to spheres with tuning in the spiked length can also be utilized for other small molecules. This data also suggests that change in solvent can be employed for sphere-spike morphology as drug carrier for hollow microspheres and for spike morphology as one has control on crystallites size, it can be employed to blend it with photoactive materials or as composite in heterojunction of organic photovoltaics.

Acknowledgements

Authors would like to acknowledge Department of Science and Technology, India, DST: SR/S3/ME/0051/2012 for financial support. Also, authors would like to acknowledge Dr Ranjith K. for synthesizing DTCPA material for a case study, which was used in this work.

References

1. Y.-Z. Long, M. Yu, B. Sun, C.-Z. Gu and Z. Fan, Chem. Soc. Rev., 2012, 41, 4560–4580.
2. Y.-Z. Long, M.-M. Li, C. Gu, M. Wan, J.-L. Duvail, Z. Liu and Z. Fan, Prog. Polym. Sci., 2011, 36, 1415–1442.
3. R. Rotzoll, S. Mohapatra, V. Olariu, R. Wenz, M. Grigas, K. Dimmler, O. Shchekin and A. Dodabalapur, Appl. Phys. Lett., 2006, 88, 123502.
4. K. Ranjith, S. K. Swathi, P. Kumar and P. C. Ramamurthy, J. Mater. Sci., 2011, 46, 2259–2266.
5. A. H. Khoshaman and B. Bahreyni, Sens. Actuators, B, 2012, 162, 114–119.
6. E. Simsek, K. Acatay and Y. Z. Menceloglu, Langmuir, 2012, 28, 14192–14201.
7. C. H. Park and J. Lee, J. Appl. Polym. Sci., 2009, 114, 430–437.
8. O. V. Kim and P. F. Dunn, Langmuir, 2010, 26, 15807–15813.
9. D. Fantini, M. Zanetti and L. Costa, Macromol. Rapid Commun., 2006, 27, 2038–2042.
10. B. Liu, X. Zhao, K. Nakata and A. Fujishima, J. Mater. Chem. A, 2013, 1, 4993–5000.
11. V. R. Sinha, K. Bansal, R. Kaushik, R. Kumria and A. Trehan, Int. J. Pharm., 2004, 278, 1–23.
12. A. Bohr, J. Kristensen, E. Stride, M. Dyas and M. Edirisinghe, Int. J. Pharm., 2011, 412, 59–67.
13. J. H. Jung, J. E. Lee and G.-N. Bae, J. Aerosol Sci., 2013, 57, 185–193.
14. D.-R. Chen, D. Y. H. Pui and S. L. Kaufman, J. Aerosol Sci., 1995, 26, 963–977.
15. J. Y. Park, K. O. Oh, J. C. Won, H. Han, H. M. Jung and Y. S. Kim, J. Mater. Chem., 2012, 22, 16005–16010.
16. L. Sun, X. Yu, M. Sun, H. Wang, S. Xu, J. D. Dixon, Y. A. Wang, Y. Li, Q. Yang and X. Xu, J. Colloid Interface Sci., 2011, 358, 73–80.
17. P. Ke, X.-N. Jiao, X.-H. Ge, W.-M. Xiao and B. Yu, RSC Adv., 2014, 4, 39704–39724.
18. J. H. Moon, G.-R. Yi, S.-M. Yang, D. J. Pine and S. B. Park, Adv. Mater., 2004, 16, 605–609.
19. K. Na, J. Jung, J. Lee and J. Hyun, Langmuir, 2010, 26, 11165–11169.
20. H. Chen, J. D. Snyder and Y. A. Elabd, Macromolecules, 2008, 41, 128–135.
21. E. Scholten, H. Dhamankar, L. Bromberg, G. C. Rutledge and T. A. Hatton, Langmuir, 2011, 27, 6683–6688.
22. A. Bohr, M. Yang, S. Baldursdóttir, J. Kristensen, M. Dyas, E. Stride and M. Edirisinghe, Polymer, 2012, 53, 3220–3229.
23. J. Zheng, H. Zhang, Z. Zhao and C. C. Han, Polymer, 2012, 53, 546–554.
24. L. Li, Z. Jiang, M. Li, R. Li and T. Fang, RSC Adv., 2014, 4, 52973–52985.
25. C. J. Luo, M. Nangrejo and M. Edirisinghe, Polymer, 2010, 51, 1654–1662.
26. J. Yao, L. Kuang Lim, J. Xie, J. Hua and C.-H. Wang, J. Aerosol Sci., 2008, 39, 987–1002.
27. L. Natarajan, J. New, A. Dasari, S. Yu and M. A. Manan, RSC Adv., 2014, 4, 44082–44088.
28. Y. Wu and R. L. Clark, J. Colloid Interface Sci., 2007, 310, 529–535.
29. Q. Zhang, J. Liu, X. Wang, M. Li and J. Yang, Colloid Polym. Sci., 2010, 288, 1385–1391.
30. X.-Y. Zhao, W. Yang, C. Li, X. Wang, S. L. Lim, D. Qi, R. Wang, Z.-Q. Gao, B.-X. Mi, Z.-K. Chen, W. Huang and W. Deng, Sol. Energy Mater. Sol. Cells, 2015, 134, 140–147.
31. Y. Liao, T. Fukuda and N. Kamata, Phys. Status Solidi RRL, 2014, 8, 154–157.
32. K. Ranjith, S. K. Swathi, P. Kumar and P. C. Ramamurthy, Sol. Energy Mater. Sol. Cells, 2012, 98, 448–454.
33. S. K. Swathi, K. Ranjith, P. Kumar and P. C. Ramamurthy, Sol. Energy Mater. Sol. Cells, 2012, 96, 101–107.
34. M. Bai, S. P. Thomas, R. Kottokkaran, S. K. Nayak, P. C. Ramamurthy and T. N. Guru Row, Cryst. Growth Des., 2014, 14, 459–466.
35. S. R. Mishra, K. Ranjith, S. K. Swathi and P. C. Ramamurthy, Appl. Phys. Lett., 2012, 100, 013302.
36. CRC Handbook of Chemistry and Physics, ed. D. R. Lide, CRC Press, Boca Raton, Fla, 87th edn, 2006.
37. J. Ju, Y. Yamagata and T. Higuchi, Adv. Mater., 2009, 21, 4343–4347.
38. T. Fukuda, H. Asaki, T. Asano, K. Takagi, Z. Honda, N. Kamata, J. Ju and Y. Yamagata, Thin Solid Films, 2011, 520, 600–605.
39. W. K. Son, J. H. Youk, T. S. Lee and W. H. Park, Polymer, 2004, 45, 2959–2966.

---

Footnote

† Electronic supplementary information (ESI) available. FESEM images of CF, THF and 12DCE to demonstrate the hollow part inside the microspheres. See DOI: 10.1039/c5ra06468f

---