One pot green synthesis of graphene–iron oxide nanocomposite (GINC): an efficient material for enhancement of thermoelectric performance

Abstract

We report for the first time, a green method for graphene–iron oxide nanocomposite (GINC) synthesis by dispersing graphene and nano iron oxide (Fe₂O₃) in ethanol via ultrasonication followed by micro-wave irradiation. This is a simple method of making a broader range of graphene–metal oxide nanocomposites with excellent dispersion of 3D nanoparticles over 2D graphene. In addition, we have also demonstrated the synthesis of highly conductive PVAc–GINC and PVAc–graphene composites by ultrasonication followed by hot compaction for thermoelectric application. Graphene and GINC concentration were judiciously varied and optimized for the sake of high electrical conductivity and Seebeck coefficient. The fabricated PVAc–GINC film exhibited a conductivity of 2.18 × 10⁴ S m⁻¹ with a Seebeck coefficient of 38.8 μV K⁻¹. Hence, the power factor (PF) reaches 32.90 μW m⁻¹ K⁻², which is 27 fold higher than the thermoelectric material based on PVAc–graphene composite. This PF value is found to be the maximal ever reported without using conducting polymer.

---

1. Introduction

Graphene was discovered by Geim and Novoselov¹ in 2004, who were awarded the noble prize in physics in the year 2010. Since then, a number of researchers have been trying to extract effective properties out of it. It has the potential to attract widespread attention due to its unique properties like high carrier mobility,² room temperature quantum effect and ambipolar electric field effect. These unique properties of graphene are effectively reduced by increasing the number of graphene layers until it reaches a 3D form^3,4 i.e. graphite. It is a bulk form of graphene with multilayer structure. Graphene with 10 or less graphitic layer exhibit distinct properties compared to the bulk form.⁵ These distinct properties make graphene one interesting material for composite, electronic and advance mechanical resonator.^6,7

Thermoelectricity is an upcoming field, which deals with smart materials to convert heat into electricity. Recently, thermoelectric materials have gained tremendous research interests as a clean and green energy source that helps in energy harvesting from waste heat. The efficiency of thermoelectric materials is expressed as ZT = S²σT/κ, where ‘S’ denoted by thermopower i.e. Seebeck coefficient, σ is denoted by electrical conductivity, κ is thermal conductivity and T is absolute temperature.⁸ Nowadays, some inorganic materials, such as Ag-doped Cu₂Se and Cu₂Te, SrTiO₃ by Pr doping, copper telluride, Bi₂Te₃ nanowires, magnesium silicide, n-type SiGe bulk alloys, Cu_xS and β-Zn₄Sb₃ (ref. 9–17) showed improved thermoelectric performance. Among these materials, Bi₂Te₃ is found to be the best room temperature thermoelectric material.^18–21 However, organic materials like conjugated polymers are becoming potential candidates due to their enhanced thermoelectric power factor and figure of merit.^22–25 In addition, their low thermal conductivity (k = 0.1–0.5 W m⁻¹ K⁻¹),^26,27 easy processibility, flexibility, non toxicity, stability and cheaper cost in comparison to chalcogenides make them more beneficial. In addition, they present several other advantages like mechanical robustness for better durability, application as a thermoelectric paint over armor vehicle for stealth application and cloths made of flexible thermoelectric materials based fabric for energy harvesting from body heat. These applications are very difficult to achieve in case of chalcogenide based material.

In polymer composite, higher power factor, i.e. S²σ, can be achieved by two mechanisms: doping the polymer^28–30 or blending with different conducting nano fillers^31–33 like CNT^34,35 and graphenes.^36–39 Studies on the thermoelectric properties of these composites have found them to be competitive with chalcogenides, however their efficiency is still lower.^28,31,32 Among several polymers, (3,4-ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS),^23,40 poly(3-hexylthiophene) (P3HT),⁴¹ and polyaniline (PANI)^38,42–44 are regularly used due to their intrinsic high electrical conductivity. The electrical properties of the polymer can easily be enhanced without affecting the thermal conductivity and mechanical flexibility.⁴⁵ The reason of this enhancement in electrical conductivity is usually explained by percolation law. This law predicts a drastic increase in electrical conductivity after reaching to a percolation threshold.^46,47

Here, we demonstrated the simple preparation method of GINC. Fe₃O₄ based GINC has been reported to be prepared following the solvothermal reaction.⁴⁸ However, novelty of our methodology is its simplicity, economic nature and eco-friendliness. The required raw chemical grades and instrumentation are described in detail in the ESI (see S-1 in the ESI†). The structure, composition and morphology of the samples have been characterized by Raman spectroscopy, Raman mapping, Fourier transform infrared spectroscopy (FTIR), environmental scanning electron microscope (ESEM), high resolution transmission electron microscopy (HRTEM), powdered X-ray diffraction (XRD) and BET experiment.

In addition, an in depth study of thermoelectric properties (electrical conductivity, Seebeck coefficient and power factor) of PVAc polymer with different fillers like GINC or graphene have been evaluated and presented. In spite of high thermal conductivity, these materials could be engineered in a simple way to enhance thermoelectric properties of synthesized polymer nanocomposite. PVAc was selected because of its good adhesive nature and binding capability with lower thermal conductivity. These properties help to increase the filler loading and efficiency of thermoelectric material. Graphene were used as a substrate, which helps to increase the electrical conductivity drastically. A few oxides, such as iron oxide,⁴⁹ nickel oxide, cadmium oxide and doped zirconium oxide exhibit impressive TE properties for various applications. There are many known stable iron oxide stoichiometries available. Amongst them, Fe₂O₃ is the compound that has been studied for its TE property. An analysis of a limited number of reports suggests that Fe₂O₃ can be a promising TMO for TE application at high temperatures as it exhibits high TPFs at room as well as elevated temperatures. Fe₂O₃ thin films have been shown to exhibit peak S i.e. Seebeck coefficient of 1650 μV K⁻¹ in the temperature range of 270–290 K. A peak σ (i.e. electrical conductivity) of 5.5 × 10³ S m⁻¹ has been reported in the same temperature range, resulting in a large TPF of 1.5 × 10⁴ μW m⁻¹ K⁻². In GINC, nano Fe₂O₃ was decorated over the graphene sheet during exfoliation. After nano Fe₂O₃ decoration, the staking nature of graphene sheet reduced drastically. Hence it reduced the transformation process of graphene to graphite.

2. Experimental

During nanocomposite preparation, nano iron oxide and nano graphene were prepared separately. Nano iron oxide was prepared from iron nitrate i.e. Fe(NO₃)₃ in a two step process. Firstly, preparation of citrate polymeric gel followed by calcination of gel (see also S-2 in the ESI†). The graphene was prepared in three steps as reported in the literature.⁴⁹ In the first step, graphite oxide was prepared from graphitic flakes by Hummers method. In the second step, thermally expanded graphene oxide (TEGO) was prepared by thermal expansion/exfoliation at 1050 °C (Ar, 30 s). Finally, graphene nanosheet (GNS) was obtained by hydrogen reduction of TEGO at 400 °C for 2 hours.

For making GINC, 50 mg graphene was first dispersed in absolute ethanol by ultrasonication for 40 min. After ultrasonication, nano iron oxide was added into the graphene dispersion with ultrasonication continued for 120 min. After ultrasonication, the dispersed composite solution was kept as ambient condition for drying. After evaporation of ethanol, nanocomposite was deposited on the Petri dish. The deposited nanocomposite was placed into the microwave reactor for 2 min for better exfoliation. After cooling, the GINC was collected in the sample vial (Scheme 1).

Scheme 1 Schematic representation of synthesis of graphene–iron oxide nanocomposite (GINC).

3. Results and discussion

3.1 Characterization of GINC

Fig. S-3 (see ESI†) represents the color variation between graphene (black), nano iron oxide (red) and GINC (brown). Environmental scanning electron microscopy (ESEM) images and FESEM images (see Fig. 1) depict continuous dispersion of iron oxide nano particle over graphene nano sheet. Uniform decoration of nano iron oxide was observed over the graphene sheet.

Fig. 1 (I and II), ESEM images and (III), FESEM images of graphene and graphene iron oxide nanocomposite (GINC) (see also S-4 in the ESI†).

High resolution transmission electron microscope (HRTEM) and SAED pattern were captured to inspect the quality of nano iron oxide decoration in the graphene layers. Fig. 2 shows HRTEM micrographs with SAED pattern of graphene, nano iron oxide and GINC. The results confirm that the size of iron oxide and graphene are in the nano scale. Graphene sheets are clearly visible. Both nano iron oxide and graphene were found to be crystalline in nature as confirmed by SAED pattern. According to Fig. 2c, iron oxides nano particles were placed over the graphene sheet forming the nano composite. It shows nicely arranged lattice fringes as demonstrated in the HRTEM images. GINC was also found to be crystalline in nature.

Fig. 2 HR-TEM image (I and II) and SAED pattern (III) of (a) nano iron oxide, (b) graphene, (c) graphene–iron oxide nanocomposite (GINC) (see also S-4 in the ESI†).

Fig. 3 shows FTIR traces of graphene, iron oxide and GINC. GINC spectra reveal the presence Fe₂O₃ signature with low defect content in graphene. Raman spectroscopy was used to examine the quality of graphene sheet before and after nanocomposite formation by above mentioned techniques. The most pronounced Raman traces are D band at 1310 cm⁻¹ corresponding to defect and G band at 1575 cm⁻¹ corresponding to in plane vibration of sp² carbon. 2D band at 2627 cm⁻¹, which is generated by a two phonon double resonance process, has also been observed. The lower intensity D band indicates the presence of a small amount of defect on graphene flakes. Coleman and coworkers suggested⁴² that the defects were mainly present at the edges of the flakes while the basal plane was found to be free of defects. The I(D)/I(G) ratio of graphene–iron oxide composite increased 2 times (0.993) with respect to pure graphene (0.497). Several defects with sp² domain were formed during nanocomposite preparation. Fig. 3b shows their characteristic Raman signatures. The decoration of nano iron oxide over graphene substrate is confirmed by Raman mapping (see Fig. 3c).

Fig. 3 (a) FTIR and (b) Raman spectra of graphene, nano iron oxide, GINC (c) Raman mapping of GINC (see also S-5 in the ESI†).

Fig. 4, depicts the typical XRD scans of GINC, nano iron oxide and graphene. All the peaks are assigned to their respective crystallographic phases. X-ray diffractogram of GINC (Fig. 4) reveals characteristics peaks corresponding to graphene as well as those owing to nano Fe₂O₃. It is revealed that during processing, the crystallographic planes of graphene and iron oxide remain intact. According to the Scherrer equation, the calculated crystallite size of GNS and nano Fe₂O₃ were 28 nm and 38 nm, respectively. From BET analysis (see S-6 in the ESI†) the size of GINC was found to be 43 nm.

Fig. 4 XRD profiles of nano iron oxide, graphene and GINC.

3.2 Thermoelectric application of GINC

Thermoelectric properties mainly consist of two parameters, i.e. Seebeck coefficient or thermopower, electrical conductivity. These properties were measured and the power factor (PF) was calculated. For the measurement of above mentioned properties, synthesis of thermoelectric polymer nanocomposite and its property measurement procedures are as follows:

Synthesis of thermoelectric polymer nanocomposite. Polymer nanocomposites were prepared through very simple method. Both fillers (i.e. graphene, nano iron oxide and GINC) and PVAc water emulsion were dispersed ultrasonically for 30 min at 35 kHz. After ultrasonication, mechanical stirring continued for 4 hours at 250 rpm. After getting proper dispersion, casting was carried out in an aluminum tray. Complete drying was performed by placing the sample in a vacuum oven at 105 °C for 4 h. Now, the dried samples were subjected to hot pressing at 120 °C with 5 min preheating time and 3 min compression (pressure 10 T approx.). By following the above mentioned methodology, polymer nanocomposite sheets were prepared and subjected to the measurement of thermoelectric properties.

Thermoelectric power/Seebeck coefficient (S) measurements. To measure the thermopower as a function of temperature, a 30 mm × 6 mm × 1 mm pieces⁵⁰ of the polymer nanocomposite film was cut and placed on a thermal insulating fiberglass. A Peltier heater was positioned at one end of the sample with a thermally conductive epoxy (electrically insulating 2763 Stycast), while the other end, a piece of copper (drainage of heat) made contact with the Peltier cooling module. The temperature gradient and voltage drop along the film was measured with thermocouples arranged in series (electrically insulated from the sample with 2763 Stycast) with two copper wires. To be sure that the thermal gradient and the voltage drop were being measured at the same place, two small Cu films were attached to the PVAc–GINC film with thermally/electrically conducting silver epoxy (Dupont 4929N). The thermocouple and the voltage wires were attached to these Cu films. The thermoelectric voltages were monitored with respect to temperature difference by Keithley 2182A nanovoltmeter. The base temperature was changed with Peltier cooling module. The thermoelectric power was determined by two independent means: after reaching a steady state through an applied current to the heater and by fitting the linear V vs. ΔT response to a heating pulse. The deviation between both methods and between different experimentation was always lower than 5%.

Electrical resistivity measurements. Due to the high electrical conductivity of the composite, electrical resistivity was measured using delta mode four probe method. The smallest possible current (100 mA) was obtained using Keithley 6220 generator and voltage was monitored with a Keithley 2182A nanovoltmeter. The smallest possible current was used in order to avoid heating of the sample at low temperature. To measure electrical conductivity, polymer nanocomposite sample with a dimension of 8 mm × 3 mm × 1 mm have been prepared and subjected to the test. The detailed procedure for electrical resistivity measurement is given in S-7 in the ESI.†

From Fig. 5a, it can be realized that the electrical conductivity of PVAc–GINC composite increased up to 4–5 order compared to PVAc graphene composite. Raw PVAc is having an electrical conductivity of 10⁻¹³ S m⁻¹. The electrical conductivity was measured at ambient conditions. After two months, they showed identical results. This indicates the good stability of the nanocomposite over a period of time. The Seebeck coefficient (see Fig. 5b) also exhibits an increasing trend and reaches a maximum with 80 wt% filler concentration, then decreases. Fig. 5c exhibits the variation of power factor (PF) as a function of filler concentration. According to Fig. 5c, PF increases and reaches a very high value, 32 μW m⁻¹ K⁻² at 80 wt% filler concentration. This value is found to be the highest ever reported in the literature for PVAc based system without containing conducting polymer. The improvement of electrical conductivity follows the percolation law of the composite, which predicts an enhancement of electrical conductivity up to a critical concentration level of filler. These phenomena come into play, when two dissimilar materials with a large difference in electrical conductivity are mixed.

Fig. 5 (a) Electrical conductivity (b) Seebeck coefficient (c) power factor as a function of graphene concentration at room temperature (300 K). The arrow indicates the enhancement of all the three properties, i.e. electrical conductivity, Seebeck coefficient and power factor at a concentration level of 80–90 wt%.

During the study of thermoelectric properties, PVAc–GINC composite showed a 10 fold increase in electrical conductivity and a two fold increase in Seebeck coefficient was observed compared to the PVAc–graphene composite with equal filler loading (80 wt%). Hence, the calculated power factor for PVAc–GINC composite (density: 1.47 g cm⁻³) increases up to 27 times compared to PVAc–graphene composite (density: 1.32 g cm⁻³). For PVAc–GINC composite, thermal conductivity is found to be 3.21 W mK⁻¹. Hence, ZT approaches 0.0031. This is one of the noble findings. In GINC, nano iron oxides were decorated over 2D graphene sheet. Presence of nano iron oxide particle helps to destroy thermally conductive network but electrical network remains intact.^51,52 When GINC is employed as conducting filler, it not only decouples σ and S, but also enhances both parameters simultaneously. However, the enhancement of Seebeck coefficient is marginal with respect to electrical conductivity in the case of PVAc–GINC composite. We are reporting this feature for the first time for GINC. The thermal insulating nature of the PVAC helps to reduce the thermal conductivity of the matrix, which is very important in thermoelectric field (Fig. 6).

Fig. 6 Schematic diagram to explain enhancement of electrical conductivity and Seebeck coefficient simultaneously. () Represent electrical conduction and () thermal conduction.

To justify the novelty of the present work, a comparative summary of the latest results based on PVAc matrix (see Table 1) and other composites of inorganic and organic materials have been highlighted in S-8 and S-9 in ESI.†

Table 1 Summary of thermoelectric properties of various PVAc based carbon material composites

Sample	σ, S m^−1	S, μV k^−1	κ, W mK^−1	Calculated PF (S^2σ) μW m^−1 K^−2
PVAc + CNT (20%) [ref. 52]	4800 (300 K)	40–50 (300 K)	0.18–0.34 at 300 K	PF = 7.8–12
PVAc + SWCNT (40%) [ref. 53]	900	40	0.25	PF = 1.44
PVAc + SWCNT (3 wt%) + GA [ref. 54]	22–49	39–42	0.22–0.25	PF = 0.033
PVAc + Au + CNT [ref. 55]	10^5	—	Unaffected	Unaffected
PVAc + DOC + MWCNT (7–12%)	32–63	5–10	0.13–0.17	PF = 0.34–0.50
PVAc + TCPP + MWCNT (7–12%)	10–100	22–26	0.14	PF = 0.079–0.34
PVAc + DOC + DWCNT (7–12%)	—	50–70	0.15	PF = 0.045–0.096
PVAc + TCPP + DWCNT (7–12%) [ref. 56]	—	70–82	0.155–0.16	PF = 0–0.204
PVAc + polyethyleneimine (10 wt%) + CNT with 99% purity (20 wt%) + SDBS (20–60 wt%)	420–1250	−66–75	—	PF = 1.89–7.03
PVAc + CNT with 99% purity (20 wt%) + SDBS (20 wt%) + PEI (0–40 wt%)	320–430	−65–80	—	PF = 1.35–2.752
PVAc + CNT with 99% purity (20 wt%) + SDBS (40 wt%) + PEI (0–40 wt%), composition IX [ref. 57]	440–920	−110–110	—	PF = 5.32–11.13
PVAc + Au deposited CNT (0–20 wt%) + PEDOT:PSS (15% vol replacement by Au) [ref. 55]	6 × 10^5	2.5	—	PF = 3.75
PEDOT:PSS + PVAc + CNT (35–75%) [ref. 50]	5 × 10^4 to 1.35 × 10^5	19–34	0.2–0.4	PF = 30–110
Present work
PVAc + GINC (80 wt%)	2.18 × 10^4	38.8	—	PF = 32.90
PVAc + graphene (95%)	2.89 × 10^3	20.7	—	PF = 1.24

---

4. Conclusions

In summary, a novel, economical and green route has been developed by using ultrasonication and microwave irradiation. Other metal oxide–graphene composites can also be processed easily by following the above mentioned method. Large scale production can be achieved by using this simple method. In addition, we have also demonstrated the synthesis of highly conductive PVAc–GINC and PVAc–graphene composites by ultrasonication followed by hot compaction for thermoelectric application. Graphene and GINC concentration were judiciously varied and optimized for the sake of high electrical conductivity and Seebeck coefficient. The fabricated PVAc–GINC film exhibited a conductivity of 2.18 × 10⁴ S m⁻¹ with a Seebeck coefficient of 38.8 μV K⁻¹. Hence, the power factor (PF) reaches 32.90 μW m⁻¹ K⁻², which is 27 fold higher than the thermoelectric material based on PVAc–graphene composite. This PF value is found to be the highest ever reported without using conducting polymer. We also expect superior thermoelectric properties by incorporating conducting polymer and ionic liquid.

Acknowledgements

The authors are thankful to Dr Tejeshree Bhave, Dr Prasanth Alegaonkar, DIAT, Pune and Dr P. Wadgaonkar and his group.

Notes and references

1. K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y. S. Zhang, V. Dubonos, I. V. Grigorieva and A. A. Firsov, Science, 2004, 306, 666.
2. K. S. Novoselov and A. K. Geim, Nat. Mater., 2007, 6, 183.
3. W. Zhang, W. He and X. Jing, J. Phys. Chem. B, 2010, 114, 10368.
4. K. S. Novoselov, D. Jiang, F. Schedin, T. J. Booth, V. V. Khotkevich, S. V. Morozov and A. K. Geim, Proc. Natl. Acad. Sci., 2005, 102, 10451.
5. B. Partoens and F. M. Peeters, Phys. Rev. B: Condens. Matter Mater. Phys., 2006, 74, 075404.
6. A. Schlierf, P. Samorì and V. Palermo, J. Mater. Chem. C, 2014, 2, 3129.
7. A. K. Geim, Science, 2009, 324, 1530.
8. L. D. Zhao, S. H. Lo, Y. Zhang, H. Sun, G. Tan, C. Uher, C. Wolverton, P. V. Dravid and M. G. Kanatzidis, Nature, 2014, 508, 373.
9. S. Ballikaya, H. Chi, R. James Salvador and C. Uher, J. Mater. Chem. A, 2013, 1, 12478.
10. M. U. Kahaly and U. Schwingenschlögl, J. Mater. Chem. A, 2014, 2, 10379.
11. C. Nethravathi, C. R. Rajamathi, M. Rajamathi, R. Maki, T. Mori, D. Golberg and Y. Bando, J. Mater. Chem. A, 2014, 2, 985.
12. H. Y. Lv, H. J. Liu, J. Shi, X. F. Tang and C. Uher, J. Mater. Chem. A, 2013, 1, 6831.
13. S. K. Bux, M. T. Yeung, E. S. Toberer, G. J. Snyder, R. B. Kaner and J. P. Fleurial, J. Mater. Chem., 2011, 21, 12259.
14. R. Basu, S. Bhattacharya, R. Bhatt, M. Roy, S. Ahmad, A. Singh, M. Navaneethan, Y. Hayakawa, D. K. Aswal and S. K. Gupta, J. Mater. Chem. A, 2014, 2, 6922.
15. Q. Jiang, H. Yan, J. Khaliq, Y. Shen, K. Simpson and M. J. Reece, J. Mater. Chem. A, 2014, 2, 9486.
16. L. Hu, H. Gao, X. Liu, H. Xie, J. Shen, T. Zhu and X. Zhao, J. Mater. Chem., 2012, 22, 16484.
17. E. S. Toberer, P. Rauwel, S. Gariel, J. Taft and G. J. Snyder, J. Mater. Chem., 2010, 20, 9877.
18. X. Yan, B. Poudel, Y. Ma, W. S. Liu, G. Joshi, H. Wang, Y. Lan, D. Wang, Z. G. Chenand and F. Ren, Nano Lett., 2010, 10, 3373.
19. H. Li, X. Tang, Q. Zhang and C. Uher, Appl. Phys. Lett., 2008, 93, 252109.
20. M. Martín-González, O. Caballero-Calero and P. Díaz-Chao, Renewable Sustainable Energy Rev., 2013, 24, 288.
21. C. V. Manzano, A. A. Rojas, M. Decepida, B. Abad, Y. Feliz, O. Caballero-Calero, D. A. Borca-Tasciuc and M. Martin-Gonzalez, J. Solid State Electrochem., 2013, 17, 2071.
22. R. B. Aïch, N. Blouin, A. Bouchard and M. Leclerc, Chem. Mater., 2009, 21, 751.
23. O. Bubnova, Z. U. Khan, A. Malti, S. Braun, M. Fahlman, M. Berggren and X. Crispin, Nat. Mater., 2011, 10, 429.
24. T. Park, C. Park, B. Kim, H. Shin and E. Kim, Energy Environ. Sci., 2013, 6, 788.
25. G. H. Kim, L. Shao, K. Zhang and K. P. Pipe, Nat. Mater., 2013, 12, 719.
26. C. L. Choy, Polymer, 1977, 18, 984.
27. M. Hu, D. Yu and J. Wei, Polym. Test., 2007, 26, 333.
28. H. Kaneko, T. Ishiguro, A. Takahashi and J. Tsukamoto, Synth. Met., 1993, 57, 4900.
29. E. R. Holland, S. J. Pomfret, P. N. Adams, L. Abell and P. Monkman, Synth. Met., 1997, 84, 777.
30. J. Li, X. Tang, H. Li, Y. Yan and Q. Zhang, Synth. Met., 2010, 160, 1153.
31. Y. Qi, Z. Wang, M. Zhang, F. Yang and X. Wang, J. Mater. Chem. A, 2013, 1, 6110.
32. J. R. Szczech, J. M. Higgins and S. Jin, J. Mater. Chem., 2011, 21, 4037.
33. R. Benoit, V. Hornebecq, F. Weill, L. Lecren, X. Bourrat and M. T. Delapierre, J. Mater. Chem. A, 2013, 1, 14221.
34. C. Yu, Y. S. Kim, D. Kim and J. C. Grunlan, Nano Lett., 2008, 8, 4428.
35. C. Bounioux, P. Diaz-Chao, M. Campoy-Quiles, M. S. Martin-Gonzalez, A. R. Goni, R. Yerushalmi-Rozen and C. Muller, Energy Environ. Sci., 2013, 6, 918.
36. Y. Lu, Y. Song and F. Wang, Mater. Chem. Phys., 2013, 138, 238.
37. Y. Du, S. Z. Shen, W. Yang, R. Donelson, K. Cai and P. S. Casey, Synth. Met., 2012, 161, 2688.
38. J. Xiang and L. T. Drzal, Polymer, 2012, 53, 4202.
39. M. Scholdt, H. Do, J. Lang, A. Gall, A. Colsmann, U. Lemmer, J. Koenig, M. Winkler and H. Boettner, J. Electron. Mater., 2010, 39, 1589.
40. H. Park, S. H. Lee, F. S. Kim, H. H. Choi, I. W. Cheong and J. H. Kim, J. Mater. Chem. A, 2014, 2, 6532.
41. A. W. Musumeci, G. G. Silva, J. W. Liu, W. N. Martens and E. R. Waclawik, Polymer, 2007, 48, 1667.
42. M. Lotya, Y. Hernandez, P. J. King, R. J. Smith, V. Nicolosi, L. S. Karlsson, F. M. Blighe, S. De, Z. M. Wang, I. T. McGovern, G. S. Duesberg and J. N. Coleman, Liquid phase production of graphene by exfoliation of graphite in surfactant/water solutions, J. Am. Chem. Soc., 2009, 131, 3611.
43. Y. W. Park, Y. S. Lee, C. Park, L. W. Shacklette and R. H. Baughman, Solid State Commun., 1987, 63, 1063.
44. K. Lee, S. Cho, H. P. Sung, A. J. Heeger, C. W. Lee and S. H. Lee, Nature, 2006, 441, 65.
45. K. I. Winey and R. A. Vaia, MRS Bull., 2007, 32, 314.
46. R. D. Sherman, L. M. Middleman and S. M. Jacobs, Polym. Eng. Sci., 1983, 23, 36.
47. S. Kirkpatrick, Rev. Mod. Phys., 1973, 45, 574.
48. C. Fu, G. Zhao, H. Zhang and S. Li, Int. J. Electrochem. Sci., 2014, 9, 46.
49. S. Walia, S. Balendhran, H. Nili, S. Zhuiykov, G. Rosengarten, Q. H. Wang, M. Bhaskaran, S. Sriram, M. S. Strano and K. Kalantar-zadeh, Prog. Mater. Sci., 2013, 58(8), 1443–1489.
50. G. Zhou, D. W. Wang, F. Li, L. Zhang, N. Li, Z. S. Wu, L. Wen, G. Q. Lu and H. M. Cheng, Chem. Mater., 2010, 22, 5306.
51. D. Kim, M. Sc. thesis, Texas A&M University, 2009.
52. T. Mori, T. Nishimura, K. Yamaura and E. T. Muromachi, J. Appl. Phys., 2007, 101, 093714.
53. C. Yu, K. Choi, L. Yin and J. C. Grunlan, ACS Nano, 2011, 5, 7885.
54. M. He, F. Qiu and Z. Lin, Energy Environ. Sci., 2013, 6, 1352.
55. Y. S. Kim, D. Kim, K. J. Martin, C. Yu and J. C. Grunlan, Macromol. Mater. Eng., 2010, 295, 431.
56. K. Choi and C. Yu, PLoS One, 2012, 7(9), e44977.
57. G. P. Moriarty, J. N. Wheeler, C. Yu and J. C. Grunlan, Carbon, 2012, 50, 885.

---

Footnote

† Electronic supplementary information (ESI) available: Details of raw materials, instrumentation, nano Fe₂O₃ synthesis, photograph of nano iron oxide, graphene and GINC, additional HRTEM images, BET data, Raman mapping, detail procedure for thermoelectric properties measurement, summary of thermoelectric properties of the best composite of inorganic and organic materials. See DOI: 10.1039/c4ra14655g

---