Foldable Thermoelectric Materials: Improvement of the Thermoelectric

Subscriber access provided by Northern Illinois University

Article

Foldable Thermoelectric Materials: Improvement of the Thermoelectric Performance of Directly Spun CNT Webs by Individual Control of Electrical and Thermal Conductivity Cheng Jin An, Young Hun Kang, A-Young Lee, Kwang-Suk Jang, Youngjin Jeong, and Song Yun Cho ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.6b04485 • Publication Date (Web): 09 Aug 2016 Downloaded from http://pubs.acs.org on August 10, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Foldable Thermoelectric Materials: Improvement of the Thermoelectric Performance of Directly Spun CNT Webs by Individual Control of Electrical and Thermal Conductivity Cheng Jin An1, Young Hun Kang1, A-Young Lee2, Kwang-Suk Jang1, Youngjin Jeong2, and Song Yun Cho1*

1

Division of Advanced Materials, Korea Research Institute of Chemical Technology, 141

Gajeong-ro, Yuseong-gu, Daejeon 34114, Republic of Korea 2

Department of Organic Materials and Fiber Engineering, Soongsil University, 369 Sangdo-ro,

Dongjak-gu, Seoul 06978, South Korea KEYWORDS: CNT web; gold dopant; conducting polymers; thermoelectrics; flexible thermoelectric generator

*

E-mail: [email protected]

ACS Paragon Plus Environment

1


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 31

Abstract We suggest the fabrication of foldable thermoelectric (TE) materials by embedding conducting polymers into Au-doped CNT webs. The CNT bundles which are interconnected by a direct spinning method to form 3D networks without interfacial contact resistance, provide both high electrical conductivity and high carrier mobility. The ZT value of spun CNT web is significantly enhanced through two simple processes. Decorating the porous CNT webs with Au nanoparticles increases the electrical conductivity, resulting in an optimal ZT of 0.163, which represents a more than twofold improvement compared to the ZT of pristine CNT webs (0.079). After decoration, polyaniline (PANI) is integrated into the Au-doped CNT webs both to improve the Seebeck coefficient by an energy-filtering effect, and to decrease the thermal conductivity by the phonon scattering effect. This leads to a ZT of 0.203, which is one of the highest ZT values reported for organic TE materials. Moreover, Au-doped CNT/PANI web is ultralightweight, free-standing, thermally stable, and mechanically robust, which makes it a viable candidate for a hybrid TE conversion device for wearable electronics. When a 20 K temperature gradient is applied to the TE module consisting of seven p-n couples, 1.74 µW of power is generated.


2

Page 3 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


INTRODUCTION The demand for sustainable energy has triggered extensive research on different types of energy conversion systems in the past decades. Thermoelectric (TE) materials, which can directly convert waste heat into usable electricity, have received considerable attention, with promising applications to energy harvesting.1-2 However, current commercial TE devices typically consist of expensive, toxic, rigid, and heavy inorganic materials, hindering the full utilization of their unique benefits. Recently, efforts to alleviate such problems have focused on polymer materials, such as poly(3,4-ethylenedioxythiophene) (PEDOT), polyaniline (PANI), and poly(3hexylthiophene) (P3HT), because they have intrinsically low thermal conductivity, good mechanical flexibility, low weight, easy processability, and low cost.3-5 Organic TE materials have been developed particularly to harvest low-grade waste heat from the human body to supply electricity to small devices such as bio-sensors, wireless communication units, and wearable electronics.6 However, the performance of organic TE materials remains insufficient for commercialization. The performance of TE materials is evaluated by the dimensionless figure of merit given by ZT = S2σT/κ, where S, σ, T, and κ are the Seebeck coefficient, electrical conductivity, absolute temperature, and thermal conductivity of the material, respectively.7-8 A larger value of ZT indicates a higher TE conversion efficiency, which requires a high power factor (S2σ) and a low thermal conductivity (κ). Unfortunately, the figure of merit of conventional bulk materials cannot be infinitely increased, because the parameters are interrelated: a large Seebeck coefficient requires a low carrier concentration, which decreases the electrical conductivity; a large electrical conductivity is always paired with a high thermal conductivity.9 Therefore, improving TE performance was achieved slowly until the proposal of two promising approaches, including


3


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 31

the reduction of lattice thermal conductivity by phonon scattering10-11 and the enhancement of the TE power factor (S2σ) by quantum confinement,12-15 energy filtering,16-17 or tuning the electronic band structure (i.e. the density of states) of the material.18 The incorporation of lowerdimensional structures could create sharp features in the electronic density of states, which could result in increased asymmetry of the differential conductivity with respect to the Fermi energy.19 In this respect, low-dimensional carbon materials have been exploited to improve the TE performance of conducting polymers in composite form. Nanocomposites have been demonstrated as especially effective in improving material performances by synergistically combining the advantages of each component.20 In 2009, Meng et al. reported that multi-walled carbon nanotube (MWCNT) sheet/PANI nanocomposites prepared by a two-step method showed simultaneously

enhanced

electrical

conductivities

and

Seebeck

coefficients.21

These

enhancements were attributed to the size-dependent energy-filtering effect at the nano-interfaces between the PANI coated layer and the carbon nanotubes (CNTs). In 2010, Yao et al. prepared a single-walled CNT/PANI hybrid composite by in-situ polymerization, demonstrating the formation of an ordered polymer chain structure via strong π–π conjugation interactions between the π bonds on the surfaces of the CNTs and PANI molecular chains.22 The ordered molecular structure increased the carrier mobility, thereby improving both the electrical conductivity and the Seebeck coefficient. In 2011, Yu et al.23 reported a notable increase in the electrical conductivity of CNT/PEDOT composites without a decrease in the Seebeck coefficient by the addition of CNTs. Despite these approaches of using carbon-based nanomaterials such as CNTs, graphene, and graphene nanoplates as fillers for conducting polymers to increase the TE properties of materials, the TE properties of such carbon/polymer composites are still low compared to those of inorganic-based materials. This is because it is extremely difficult to


4

Page 5 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


homogenously disperse carbon-containing materials into the polymer matrix, but good dispersions are essential to obtain optimal TE performances. Additionally, the electrical properties of the highly conductive CNTs were drastically weakened by the less conductive polymer

matrix.

PANI/graphene/PANI/double-walled

carbon

nanotube

(DWNT)

nanocomposites with ordered molecular structures formed using layer-by-layer deposition were reported by Cho et al. in 2015.24 The PANI-covered DWNT bridged gaps between the graphene sheets and increased the electrical conductivity and Seebeck coefficient simultaneously by increasing the carrier mobility, resulting in the maximum power factor of 1825 µW m−1 K−2. Toshima et al. also demonstrated the fabrication of three-component hybrid organic materials consisting of n-poly(nickel 1,1,2,2-ethenetetrathiolate) (n-PETT)/CNT/poly(vinyl chloride) (PVC), where n-PETT promoted smooth contact of and good charge transport between the CNT bundles in the films.25 Such bridging materials can effectively connect each CNT bundle, but large contact resistance remains between the bridging conductive polymers and the carbon materials because of the heterogeneous nature of the interface. Here, we suggest the fabrication of high-performance organic TE materials by embedding conducting polymers into Au-doped CNT webs. CNT bundles of unmeasurable length were innately interconnected by a direct spinning method to form 3D networks without interfacial contact resistance, providing both high electrical conductivity and high carrier mobility. The selfsustainable homogeneous network structure of the porous CNT web is promising as a template for polymer compositing, because it avoids the conventional CNT dispersion problem as well as electron transport interference by the polymer. To further enhance the conductivity of the CNT web without significantly affecting thermal power, the directly spun CNT web was decorated with Au nanoparticles (NPs). Consequently, the power factor reached a value 2.3 times higher


5


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 31

than that of the pristine CNT web by increasing the electrical conductivity from 1090 to 5134 S cm-1. This power factor is larger than that of PbTe and close to that of bulk Bi2Te3.14,26 Conducting polymers with low thermal conductivities were embedded in the Au-doped CNT webs by the vacuum filtration of the polymer solutions to induce a phonon scattering effect, reducing the thermal conductivity by forming a nanoscale heterostructure. Accordingly, the thermal conductivity of the Au-doped CNT was reduced from 7.09 to 3.61 W m-1 k-1 by PANI infiltration. Finally, the dimensionless figure of merit of the Au-doped CNT/polymer composite was dramatically increased from 0.079 to 0.203.

RESULTS AND DISCUSSION Figure 1 shows the overall fabrication scheme and corresponding scanning electron microscopy (SEM) images for ultra-lightweight, free-standing, foldable TE materials consisting of Au-doped porous CNT webs with conducting polymers. Large-area CNT webbing was assembled by direct spinning from a chemical vapor deposition zone with continuous processing (Figure 1a), as described recently (see experimental section). The CNT webs on the wheel were then tailored and cut with scissors into pieces measuring 5 × 5 cm (Figure 1b). Au decoration was then performed by immersing the CNT webs in 2 mM AuCl3 solution for 10 min. After Au treatment, the webs were rinsed thoroughly with deionized water and dried in an oven at 313 K (Figure 1c). The CNT web doped with the 2 mM AuCl3 solution exhibits the maximum power factor with both significantly enhanced conductivity and some loss of thermal power (Supporting Information, Figures 1 and 2). Following Au decoration, the Au-doped CNT web was infiltrated with conducting polymers through vacuum filtration (Figure 1d). Although vacuum filtration is commonly used to separate particles from suspensions, it was adapted to embed various


6

Page 7 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


polymers efficiently within the tightly percolated CNT bundles. Polymer solutions of 1 wt% P3HT, PANI, or PEDOT:polystyrene sulfonate (PSS) were infiltrated into the voids of the CNT web (Figure 1e). For PEDOT:PSS, the 30-100 nm sizes of the particles hindered their incorporation into the CNT webs. Finally, Au electrodes of 100 nm in thickness were deposited on top of the free-standing Au-doped CNT/polymer webs in order to measure the Seebeck coefficient. One side of each sample was heated gradually from room temperature to form temperature gradients of 10 K, while the opposite side was maintained at room temperature using a cooling system. The temperature difference and Seebeck voltage were monitored simultaneously. The morphologies of the CNT web-based composites were investigated by SEM. The surface SEM image clearly shows that the pristine CNT web consists of interconnected 3D networks (Figure 1g). A number of spherical Au NPs with diameters of 150-200 nm are observed on the surface of the CNT web after treatment with Au ions (Figure 1h). While the pure and Audoped CNT bundles have similar diameters of approximately 10-30 nm (Figures 1g and h), the Au-doped CNT/polymer bundles show slightly larger diameters than the composite polymers without Au doping, probably because the π-bonded surfaces of the CNTs interact strongly with the conjugated structure of the polymer. This structural property may facilitate the formation of a tubular coating layer at the surface of the nanotubes (Figures 1i and j).27 Although some aggregated polymer particles are observed on the surface of the Au-doped CNT/polymer webs, as indicated with red arrows, the electrically conductive pathway structure is retained, facilitating carrier transport through the highly conductive CNT network. The cross-sectional SEM image shows long straight strands of CNT bundles in all directions with a film thickness of 1 µm (inset of Figure 1j). The chemical composition of the surfaces were investigated by X-ray


7


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 31

photoelectron spectroscopy (XPS) to identify the different components in the composites (Supporting Information, Figure 3). Figure 2 shows the in-plane TE transport properties of the Au-doped CNT/PANI web compared to those of the host materials, such as the pristine CNT web, Au-doped CNT web, and Au-doped CNT/P3HT web. While the Seebeck coefficient of the CNT web is slightly decreased after Au-doping (Figure 2a), the electrical conductivity is significantly enhanced from 1090 to 5134 S cm-1 (Figure 2b). As a result of the significant enhancement of electrical conductivity, the power factor is doubled in the Au-doped CNT webs, reaching 3548 µW m-1 K-2 (inset of Figure 2b), which is the highest value reported for organic TE materials and approaches that of the most promising inorganic material Bi2Te3.14,26 Notably, the rate of increase in the thermal conductivity induced by Au-doping is very low compared to that of the electrical conductivity, probably because of the porous structure of the CNT webs. The porosity may provide a scattering effect for phonons at the interfaces between the CNTs and air. Such scattering can impede drastic increases in the thermal conductivity caused by Au doping. To further control the thermal conductivity, various low-thermal conducting polymers were uniformly embedded in the Audoped CNT webs, which suffered from slight losses in electrical conductivity by polymeric insulation and interfacial contact resistance (Supporting Information, Figure 4). Despite the slight decrease in electrical conductivity, however, the Seebeck coefficient of the resultant webs was significantly enhanced in comparison with that of Au-doped CNT webs. This increase in the Seebeck coefficient may be ascribed to the preferential transfer of high-energy carriers to pass, while the polymer blocked that of low-energy carriers.28 This energy-filtering effect increases the mean carrier energy in carrier transport with an enhanced Seebeck coefficient.27 In detail, the Au-CNT/PANI webs show higher electrical conductivities and Seebeck coefficients than the Au-


8

Page 9 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


CNT/P3HT webs, probably because of the arrangement of the PANI chains by the π–π interactions between PANI and the CNTs. This results in increased carrier mobility, thus enhancing the electrical conductivity and Seebeck coefficient. The strong interactions between the CNTs and PANI can also facilitate mutual charge transfer. Meanwhile, the lower electrical conductivity of the Au-doped CNT/P3HT webs can be understood by assuming that the less electrically conductive P3HT with weaker π–π interaction with the CNT webs interferes significantly with electron transfer. Even though the power factor of the Au-doped CNT web after PANI treatment decreases to 2454 µW m-1 K-2, this value is still the highest for organic TE materials. The in-plane thermal conductivity of the Au-doped CNT web is reduced to 4.53 and 3.61 W m-1 K-1 by P3HT and PANI treatment, respectively (Figure 2c). Such decreases in thermal conductivity results from the increased number of phonon scattering sites on the CNTs wrapped with low-thermal-conductivity polymers,29 which form nanoscale heterostructures by enlarging the interfaces between CNT and polymer, polymer and air, and CNT and air. As a result of the significant decrease in the thermal conductivity and the increase in Seebeck coefficient, the in-plane ZT value is increased from 0.15 for the Au-doped CNT web to 0.203 for the Au-doped CNT/PANI web at 300 K (Figure 2d). Consequently, the ZT value of the directly spun CNT web was finally increased from 0.079 to 0.203 through the two post-treatment processes of Au doping and polymer embedding. Cross-sectional transmission electron microscope (TEM) images provide more details on the structures of the Au-doped CNT/PANI webs (Figure 3). Polymer-CNT entanglements and several voids are observed in the cross-sectional TEM image (Figure 3a), confirming the porous 3D network structure of the Au-doped CNT/PANI webs. This porous structure would greatly benefit electron transport through the network, while effectively suppressing thermal heat


9


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 31

transport.30 Elemental nitrogen is detected in the energy dispersive X-ray spectroscopy (EDXS) mapping results (Figure 3b), illustrating that the PANI polymer is homogenously dispersed within the webs. The N peak on the EDXS spectrum from the cross-section further clarifies the PANI distribution on the Au-doped CNT/PANI web (inset of Figure 3b). Furthermore, distinct boundaries of individual CNTs are observed in the high-resolution image (Figure 3c), in which the crystallized CNT bundle is homogeneously wrapped with conductive PANI and PANI backbone chains are well-oriented along the CNT axes because of the strong chemical interactions between the PANI and CNTs (Figure 3d).31 To provide further insight on the electronic and lattice structure of the carbon materials, Raman spectroscopy was performed using an excitation wavelength of 514 nm.32 The Raman analysis obviously confirms that the enhanced electrical conductivity of the CNT web results from the chemical doping effect of the Au NPs (Figure 3d). The peak positions of the G bands (corresponding to the in-plane stretching E2g mode) in the Raman spectra of CNTs depend on the doping effects of the CNTs.33 The upward shift in the G band positions by the phonon stiffening effect indicates p-type doping from charge extraction. Compared to the Au-doped CNT/P3HT composites, the G band of the Au-doped CNT/PANI composites is slightly shifted towards longer wave numbers, which indicates strong interactions between the constituent materials.34-35 This difference in G-band shifts can be ascribed to the stronger π-π stacking interactions between the conjugated structures of PANI and CNTs than that occurring between P3HT and CNTs. The simpler chemical structure of PANI provides less steric hindrance than P3HT, and PANI also assumes an expanded conformation in the m-cresol solvent.36 Moreover, the Raman spectrum of the Au-doped CNT/PANI web shows the relative suppression of the G-band intensity compared to that of the Au-doped CNT/P3HT. The reduction in G-band intensity can be attributed to the interference of carbon oscillation in the


10

Page 11 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


plane direction of the CNTs by the PANI attached to the CNT surfaces.33 The highly ordered arrangement of the PANI backbone chains enhances the carrier mobility by both decreasing the π–π conjugated defects and increasing the effective degree of electron delocalization, thus enhancing the TE properties, with an especially significant improvement of the Seebeck coefficient in the hybrid nanocomposite.36 The thermal stability of the composites is investigated by measuring the resistance change depending on annealing temperature (Figure 4a). The Au-doped CNT/PANI web was annealed without encapsulation at various temperatures for 20 min in air. The web demonstrates high stability below 423 K with no change in the resistance; however, the resistance becomes slightly increased at 473 K because of the weak degradation of the network. The degradation is then activated and accelerated at 523 K, showing the maximum resistance. With further annealing at higher temperatures, the resistance is continuously decreased to values beyond the initial value. This thermal behavior demonstrates that the Au-doped CNT/PANI web is thermally stable, maintaining a high Seebeck coefficient even at temperatures approaching 423 K. We also investigate the internal resistance (R0) stability of the composites under folding stresses, as illustrated in the inset of Figure 4b. After measuring the resistance of the original state of the Audoped CNT/PANI sheet, it was completely folded and unfolded repeatedly. No significant change occurs in the internal resistance under 180° folding (less than 0.35%) and the following 180° folding (less than 0.47%). Furthermore, the sheet is repeatedly folded up to 50 cycles (25 180° folds and 25 -180° folds). Despite the repeated folding cycles, the sheet exhibits a very small change in internal resistance of less than 1.6%. After the folding tests, we crumpled the sample by hand. Although the increase of the relative resistance with crumpling stress is slightly higher than that with folding stress, the TE sheets maintain stable resistance changes of 4.4%


11


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 31

with crumpling. As shown in Figure 4c, the sheet surface is significantly crinkled after the folding and crumpling, but the Au-doped CNT/PANI web in the mechanically deformed portion retains micro- and nanostructural integrity without notable disconnection or collapse (Figure 4d). Additionally, the Au-doped CNT/PANI web is less than one-fifth of the weight of a feather of similar size (Supporting Information, Figure 5). We believe that these TE webs can properly work on curved thermal energy sources, such as the human body and automobile engine parts, as energy harvesting devices. We then fabricated TE modules containing Au-doped CNT/PANI webs and polyethylenimine (PEI)-doped CNT webs as p- and n-type materials, respectively. Each p- and n-type leg was electrically connected by Ag paste by a dispensing process (Figure 5a). The free-standing webs were attached to a flexible polyethylene terephthalate (PET) substrate to create an electrical connection without significant loss of contact resistance. The n-type TE materials as counterparts to the p-type webs were prepared by dipping a CNT web into a 5 wt% PEI solution in dimethyl sulfoxide (DMSO) for 10 min, followed by washing with DI water. The PEI-doped CNT web possesses a negative Seebeck coefficient of -129 µV K-1 and electrical conductivity of 1026 S cm-1 (Supporting Information, Figure 6). The n-type TE module consisting of 14 legs of 2 mm in width and 15 mm in length shows a maximum power output of 191 nW; the p-type TE module consisting of the Au-doped CNT/PANI web exhibits a maximum power output of 307 nW at ∆T = 10 K. The flexible p-n junction-type TE module generates a high open-circuit voltage (Voc) of 10.87 mV, with a maximum power output of 376 nW at ∆T = 10 K (Figure 5b), although the thermal properties of the n-type legs are not optimized and matched with those of the p-type materials. The maximum power output of the p-n junction-type TE module is observed at the load resistance of 70 Ω, between the matched external loads for both the p- and n-type TE


12

Page 13 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


modules (Figure 5c). These modules have relatively low internal resistances compared to the previously reported organic TE modules, permitting the connection of more legs serially for various applications. The maximum power output of the p-n junction TE module as a function of the temperature gradient is presented in Figure 5d; the inset figure represents the typical powercurrent curves of the module for different temperature gradients. The maximum power output of the p-n junction-type TE module is increased with the increase of the temperature gradient, approaching 1.7 µW at ∆T = 20 K, corresponding to 0.4 µW/cm2 and 550 µW/g. We believe that this is the first report of microwatt-scale generation by an organic TE device, operating at temperature gradients as low as ∆T = 20 K, which can be obtained from the human body. Therefore, the first target application of the hybrid TE generator demonstrated here can be as a wearable power source to directly supply sufficient electrical power to low-power wireless and medical sensors on packages or textiles, following the “internet-of-things” concept.37-38

CONCLUSION We prepared Au-doped CNT/PANI webs with high electrical conductivities (1106 S cm-1) and high Seebeck coefficients (150.86 µV K-1) at room temperature. The ZT value of the TE device was significantly enhanced through two simple processes. Decorating the porous CNT webs with Au NPs increased the electrical conductivity with a small increase in the thermal conductivity, resulting in an optimal ZT of 0.163, which represented a more than twofold improvement compared to the ZT of pristine CNT webs (0.079). After decoration, conjugated PANI was integrated into the Au-doped CNT webs both to improve the Seebeck coefficient by inducing an energy-filtering effect, and to decrease the thermal conductivity by inducing the phonon scattering effect. This led to a ZT of 0.203, which is one of the highest ZT values reported for


13


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 31

organic TE materials. Moreover, the ultra-lightweight, free-standing, thermally stable, and mechanically robust Au-doped CNT/PANI web is a viable candidate for a hybrid TE conversion device for wearable electronics. Upon the application of a 20 K temperature gradient on the TE module, 11.2 mV of TE voltage was created, generating 1.74 µW of power. The TE properties could be further optimized by properly choosing a conjugated polymer and a highly conductive 3D porous structure. Furthermore, TE generators with highly improved power output can be developed by perfectly matching n-type materials or optimizing device geometries. The novel strategy of integrating a polymer into highly conductive porous carbon materials provides a new fabrication route for high-performance organic TE materials.

EXPERIMENTAL SECTION Preparation of CNT Web. Continuous CNT webs were produced in a heat furnace from a liquid feedstock consisting of acetone as a carbon source, ferrocene (0.2 wt%) as a catalyst precursor, and thiophene (0.8 wt%) as a promoter. The furnace temperature was maintained at 1473 K and the carrier H2 gas was injected at a rate of 1000 sccm. The rate of liquid injection (acetone, ferrocene, and thiophene) was 10 mL/h. The synthesis was conducted by injecting the mixture into a heated reactor under flowing H2, as described elsewhere.39-40 The CNTs were organized into continuous tubular forms, referred to as a CNT sock. The CNT sock was wound onto a winder at the end of the reactor. The CNT web was densified by immersion in dimethyl sulfoxide (DMSO) and drying in an oven at 373 K.41 Au Treatment on CNT Web. AuCl3 (purity 99%, Sigma-Aldrich) powder was dissolved in a mixed solvent of ethanol (Sigma-Aldrich) and DI water (1:1 in volume ratio) at concentrations of 1, 2, 5, and 10 mM. The amount of Au was controlled by adjusting the concentration of Au


14

Page 15 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


dopant for a constant reaction time. Au doping was achieved by placing the pristine CNT webs in AuCl3 solution for 10 min, followed by rising with DI water until neutrality was reached and drying in an oven at 40 °C. When the AuCl3 solution is applied to the CNT web, Au3+ is reduced to Au0 by withdrawing electrons from the CNT because it has a lower reduction potential, resulting in the formation of well-scattered Au particles on the CNT webs (Figure S1, Supporting information). The reduction potential of CNT is +0.22 V against a standard hydrogen electrode (SHE), which is much lower than that of AuCl4 (+1.002 V vs. SHE).42 Embedding

Conducting

Polymers

in

Au-doped

CNT

Web.

Poly(3,4-

ethylenedioxythiophene) (P3HT, Mw=50-70 k, regioregular electronic grade) was purchased from RIEKE METALS and dissolved in 1,2-dichlorobenzene (Sigma-Aldrich) at a concentration of 1 mg/mL. The P3HT was heavily doped with anhydrous FeCl3 in a nitromethane solution with a molar ratio of 1:1. Polyaniline (PANI, emeraldine base, average Mw = ~5000, Sigma-Aldrich) solution was prepared by dissolving PANI in m-cresol (Sigma-Aldrich) at 1 mg/mL under continuous stirring overnight, followed by bath sonication for 2 h. Here, m-cresol was selected as the solvent for PANI because the molecular conformation of PANI changes from a compacted coil to an expanded coil by dissolution in m-cresol.43 A camphor sulfonic acid (CSA)-doped PANI solution was prepared by adding CSA powder into the PANI solution at the molar ratio of 1:1. By adding CSA, the dark-blue PANI solution was changed to a dark green solution. The different conducting polymer solutions infiltrated the Au-doped CNT webs by vacuum filtration. In this process, the positively charged conjugated polymer is physically adsorbed by the negatively charged CNT web.24 The solution-passing speed was controlled to be very low, permitting sufficient adsorption of the polymer solution onto the surface of the CNT web.


15


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 31

Fabrication of the TE Module. The TE module composed of Au-doped CNT/PANI webs and polyethyleneimine-doped CNT webs as p- and n-type materials, respectively, was fabricated. The both type webs were alternatively attached on flexible plastic substrates via wet transfer method. Then, the TE webs were electrically connected by Ag paste through dispensing equipment (Musashi Shotmaster 200DS-S, Musashi Engineering, Inc.), followed by sintering at 383 K on a hot plate for 30 min. The legs of the CNT web, measuring 15 mm in length and 2 mm in width, were arranged at intervals of 2 mm on a PET substrate. (Figure 7, Supporting Information) Experimental Setup for Seebeck Coefficient Measurement. The Seebeck coefficients of the free-standing CNT webs were measured by a custom system. Dot-shaped Au electrodes were deposited by thermal evaporation on the tops of the samples through a shadow mask in order to minimize the contact resistance with the probe (Figure 8a, Supporting Information). The distance between the two electrodes was 3 cm and the diameters of the circular electrodes were 5 mm. The samples were then placed on top of two Peltier plates and a temperature difference was applied across the sample by applying an electrical current through the two Peltier plates (Figure 8b, Supporting Information). A Keithely 2460 sourcemeter was used to control the temperature of the hot and Peltier plates (Figure 8c, Supporting Information). A pair of thermocouples was placed onto the surface of the sample to detect differences in local temperature. The thermocouples provided fast and precise responses to the local temperature changes of the sample. The temperature gradient between the electrodes was varied gradually from 1 to 10 K. Meanwhile, a pair of probes was directed toward the previously deposited Au electrodes to measure the TE voltage, which ensured that the measured voltage corresponded exactly to the recorded temperature difference. By varying the temperature difference through the control of


16

Page 17 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


the current to the Peltier devices, the TE voltage (V) versus temperature difference (∆T) was automatically recorded by a Keithley 2182A nano-voltmeter (Figure 8c, Supporting information). The Seebeck coefficient was then obtained by analyzing the linear regression slope of the ∆V-∆T curve. The typical ∆V-∆T curve obtained in measuring the Au-doped CNT/PANI webs is shown in Figure 9d, where a Seebeck coefficient of 150 µV/K is obtained. Notably, the experimental setup was calibrated with a well-calibrated commercial Bi2Te3 sample to confirm the accuracy of the measurement of the Seebeck coefficient. Measurements of the Electrical Conductivity. The electrical conductivity was measured by the standard van der Pauw direct-current four-probe method.44 All measurements were conducted at room temperature using a Keithley 195A digital multimeter and a Keithley 220 programmable current source. The thicknesses of the films were determined by an alpha-step surface profiler (αstep DC50, KLA Tencor). Measurements of the Thermal Conductivity. The thermal conductivity κ was calculated by the equation κ = αρCp, where α is the thermal diffusivity, ρ is the density calculated by measuring the weight and volume of the webs, and Cp is the specific heat capacity at constant pressure. The in-plane thermal diffusivity α of the free-standing webs was measured using a scanning laser heating analyzer (Laser PIT, Ulvac-Riko, Inc.)45 at 300 K. Cp was measured using a modulated differential scanning calorimeter (MDSC Q200, TA Instruments). The experimental results of the thermal conductivity measurements of the webs are summarized in Table S1. The thermal conductivity was measured in the same direction as the Seebeck coefficient measurements. The in-plane thermal diffusivity of Au-doped/PANI web was reduced to 8.06 W m-1 K-1 from that of the Au-doped CNT web, indicating that the PANI coating efficiently suppressed phonon transfer along the direction of the CNT web direction. The thermal


17


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 31

conductivity of the Au-doped CNT/PANI web is similar to those reported for other CNT web/polymer composites.46 Measurements of Power-generation Characteristics. The power generation characteristics of the TE module were measured by a custom system, as reported previously.47 One side of the TE module was heated by applying an electrical current through a hot Peltier plate; simultaneously, the other side was cooled by a cooling fan to maintain a constant temperature of 300 K. We set the temperature difference of ∆T = 10 K in the in-plane direction between the two sides of the device. The temperature difference of the surfaces of the device was measured by thermocouples. This system measured the voltage–current and power output–current curves by varying the load resistance from 0 to 600 Ω. Characterization of Structures. The surface morphology was characterized by SEM (SigmaHD, Carl Zeiss). The cross-section TEM sample was prepared using the ex situ lift-out technique using focused ion beam machining (FIB, FB-2100, Hitachi) and imaged with a fieldemission TEM (Tecnai F30 S-Twin, FEI). Raman spectra were obtained by a high-resolution dispersive Raman spectroscope (XploRATMPLUS, Horiba Jobin Yvon).

ASSOCIATED CONTENT Supporting Information SEM images and thermoelectric properties of CNT webs depending on Au dopant concentration. XPS spectra for various CNT webs. Thermoelectric properties of Au-doped CNT webs depending on treated conducting polymers. Weight of Au-doped CNT/PANI web. Thermoelectric properties of n-type CNT webs. Photographs of thermoelectric generators and measurement set-up.


18

Page 19 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


AUTHOR INFORMATION Corresponding Author *E-mail: [email protected] Notes The authors declare no competing financial interest.

ACKNOWLEDGEMENTS This work was supported by a grant from the KRICT Core Project (KK 1507-C06) and the R&D Convergence Program of National Research Council of Science and Technology of the Republic of Korea.

REFERENCES (1) Bell, L. E. Cooling, Heating, Generating Power, and Recovering Waste Heat with Thermoelectric Systems. Science 2008, 321, 1457-1461. (2) Dresselhaus, M. S.; Chen, G.; Tang, M. Y.; Yang, R. G.; Lee, H.; Wang, D. Z.; Ren, Z. F.; Fleurial, J. P.; Gogna, P. New Directions for Low-dimensional Thermoelectric Materials. Adv. Mater. 2007, 19, 1043-1053. (3) Poehler, T. O.; Katz, H. E. Prospects for Polymer-based Thermoelectrics: State of the Art and Theoretical Analysis. Energy Environ. Sci. 2012, 5, 8110–8115. (4) Zhang, Q.; Sun, Y.; Xu, W.; Zhu, D. Organic Thermoelectric Materials: Emerging Green Energy Materials Converting Heat to Electricity Directly and Efficiently. Adv. Mater. 2014, 26, 6829–6851.


19


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 31

(5) Bae, E. J.; Kang, Y. H.; Jang, K.-S.; Lee, C.; Cho, S. Y. Solution Synthesis of Telluride based Nano-barbell Structures Coated with PEDOT:PSS for Spray-printed Thermoelectric Generators. Nanoscale 2016, 8, 10885-10890. (6) Fukumaru, T.; Fujigaya, T.; Nakashima, N. Development of N-type Cobaltoceneencapsulated Carbon Nanotubes with Remarkable Thermoelectric Property. Sci. Rep. 2015, 5, 7951. (7) Goldsmid, H. J. Thermoelectric Refrigeration, 1st ed; Plenum: New York, 1964. (8) Snyder, G. J.; Toberer, E. S. Complex Thermoelectric Materials. Nat. Mater. 2008, 7, 105– 114. (9) Bejan, A.; Kraus, A. D. Heat Transfer Handbook, 1st ed; Wiley: New York, 2003. (10) Zebarjadi, M.; Esfarjani, K.; Dresselhaus, M. S.; Ren, Z. F.; Chen, G. Perspectives on Thermoelectrics: From Fundamentals to Device Applications. Energy Environ. Sci. 2012, 5, 5147–5162. (11) Hsu, K. F.; Loo, S.; Guo, F.; Chen, W.; Dyck, J. S.; Uher, C.; Hogan, T.; Polychroniadis, E. K.; Kanatzidis1, M. G. Cubic AgPb(m)SbTe(2+m): Bulk Thermoelectric Materials with High Figure of Merit. Science 2004, 303, 818–821. (12) Hicks, L. D.; Dresselhaus, M. S. Thermoelectric Figure of Merit of One Dimensional Conductor. Phys. Rev. B 1993, 47, 16631–16634. (13) Venkatasubramanian, R.; Siivola, E.; Colpitts, T.; O’Quinn, B. Thin-film Thermoelectric Devices with High Room-temperature Figures of Merit. Nature 2001, 413, 597–602. (14) Poudel, B.; Hao, Q.; Ma, Y.; Lan, Y.; Minnich, A.; Yu, B.; Yan, X.; Wang, D.; Muto, A.; Vashaee, D.; Chen, X.; Liu, J.; Dresselhaus, M. S.; Chen, G.; Ren, Z. High-thermoelectric


20

Page 21 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Performance of Nanostructured Bismuth Antimony Telluride Bulk Alloys. Science 2008, 320, 634–638. (15) Hochbaum, A. I.; Yang, P. Semiconductor Nanowires for Energy Conversion. Chem. Rev. 2010, 110, 527–546. (16) Moizhes, B. Y.; Nemchinsky, V. A. in Proc. for the 11th Int. Conf. on Thermoelectrics, University of Texas, Arlington, TX 1992. (17) Rowe, D. M. CRC Handbook of Thermoelectrics, 1st ed; CRC Press: New York, 1995. (18) Lee, J.; Hao, Q.; Ma, Y.; Lan, Y.; Minnich, A.; Yu, B.; Yan, X.; Wang, D.; Muto, A.; Vashaee, D.; Chen, X.; Liu, J.; Dresselhaus, M. S.; Chen, G.; Ren, Z. A Novel Approach for Fabrication of Bismuth-silicon Dioxide Core-shell Structures by Atomic Layer Deposition. J. Mater. Chem. 2009, 19, 7050–7054. (19) Vineis, C. J.; Shakouri, A.; Majumdar, A.; Kanatzidis, M. G. Nanostructured Thermoelectrics: Big Efficiency Gains from Small Features. Adv. Mater. 2010, 22, 3970–3980. (20) Zhao, W.; Tan, H. T.; Tan, L. P.; Fan, S.; Hng, H. H.; Boey, Y. C. F.; Beloborodov, I.; Yan, Q. n-Type Carbon Nanotubes/Silver Telluride Nanohybrid Buckypaper with a HighThermoelectric Figure of Merit. ACS Appl. Mater. Interfaces 2014, 6, 4940−4946. (21) Meng, C. Z.; Liu, C. H.; Fan, S. S. A Promising Approach to Enhanced Thermoelectric Properties using Carbon Nanotube Networks. Adv. Mater. 2010, 22, 535-539. (22) Yao, Q.; Chen, L. D.; Zhang, W. Q.; Liufu, S. C.; Chen, X. H. Enhanced Thermoelectric Performance of Single-walled Carbon Nanotubes/Polyaniline Hybrid Nanocomposites. ACS Nano 2010, 4, 2445-2451. (23) Yu, C.; Choi, K.; Yin, L.; Grunlan, J. C. Light-weight Flexible Carbon Nanotube based Organic Composites with Large Thermoelectric Power Factors. ACS Nano 2011, 5, 7885-7892.


21


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 31

(24) Cho, C.; Stevens, B.; Hsu, J.-H.; Bureau, R.; Hagen, D. A.; Regev, O.; Yu, C.; Grunlan, J. C. Completely Organic Multilayer Thin Film with Thermoelectric Power Factor Rivaling Inorganic Tellurides. Adv. Mater. 2015, 27, 2996-3001. (25) Toshima, N.; Oshima, K.; Anno, H.; Nishinaka, T.; Ichikawa, S.; Iwata, A.; Shiraishi, Y. Novel Hybrid Organic Thermoelectric Materials: Three-component Hybrid Films Consisting of a Nanoparticle Polymer Complex, Carbon Nanotubes, and Vinyl Polymer. Adv. Mater. 2015, 27, 2246–2251. (26) Martin, J.; Nolas, G. S.; Zhang, W.; Chen, L. PbTe Nanocomposites Synthesized from PbTe Nanocrystals. Appl. Phys. Lett. 2007, 90, 222112. (27) Yu, A.; Ramesh, P.; Sun, X.; Bekyarova, E.; Itkis, M. E.; Haddon, R. C. Enhanced Thermal Conductivity in a Hybrid Graphite Nanoplatelet-Carbon Nanotube Filler for Epoxy Composites. Adv. Mater. 2008, 20, 4740–4744. (28) Ko, D.-K.; Kang, Y.; Murray, C. B. Enhanced Thermopower via Carrier Energy Filtering in Solution-Processable Pt-Sb2Te3 Nanocomposites. Nano Lett. 2011, 11, 2841–2844. (29) Zhou, C.; Dun, C.; Wang, Q.; Wang, K.; Shi, Z.; Carroll, D. L.; Liu, G.; Qiao, G. Nanowires as Building Blocks to Fabricate Flexible Thermoelectric Fabric: The Case of Copper Telluride Nanowires. ACS Appl. Mater. Interfaces 2015, 7, 21015−21020. (30) Zhang, K.; Zhang, Y.; Wang, S. R. Enhancing Thermoelectric Properties of Organic Composites through Hierarchical Nanostructures. Sci. Rep. 2013, 3, 3448. (31) Wang, Q.; Yao, Q.; Chang, J.; Chen, L. Enhanced Thermoelectric Properties of CNT/PANI Composite Nanofibers by Highly Orienting the Arrangement of Polymer Chains. J. Mater. Chem. 2012, 22, 17612–17618.


22

Page 23 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(32) Ferrari, A. C.; Basko, D. M. Raman Spectroscopy as a Versatile Tool for Studying the Properties of Graphene. Nat. Nanotechnol. 2013, 8, 235–246. (33) Pisana, S.; Lazzeri, M.; Casiraghi, C.; Novoselov, K. S.; Geim, A. K.; Ferrari, A. C.; Mauri, F. Breakdown of the Adiabatic Born–oppenheimer Approximation in Graphene. Nat. Mater. 2007, 6, 198-201. (34) Xu, K.; Chen, G.; Qiu, D. Convenient Construction of Poly(3,4-ethylenedioxythiophene)– Graphene Pie-like Structure with Enhanced Thermoelectric Performance. J. Mater. Chem. A 2013, 1, 12395–12399. (35) Hellstrom, S.; L. Vosgueritchian, M.; Stoltenberg, R. M.; Irfan, I.; Hammock, M.; Wang, Y. B.; Jia, C.; Guo, X.; Gao, Y.; Bao, Z. Strong and Stable Doping of Carbon Nanotubes and Graphene by MoOx for Transparent Electrodes. Nano Lett. 2012, 12, 3574−3580 (36) Yu, C.; Murali, A.; Choia, K.; Ryua, Y. Air-stable Fabric Thermoelectric Modules Made of N- and P-type Carbon Nanotubes. Energy Environ. Sci. 2012, 5, 9481–9486. (37) Wusten, J.; Potje-Kamloth, K. Organic Thermogenerators for Energy Autarkic Systems on Flexible Substrates. J. Phys. D: Appl. Phys. 2008, 41, 135113. (38) Bae, E. J.; Kang, Y. H.; Jang, K.-S.; Cho, S. Y. Enhancement of Thermoelectric Properties of PEDOT:PSS and Tellurium-PEDOT:PSS Hybrid Composites by Simple Chemical Treatment. Sci. Rep. 2016, 6, 18805. (39) Li, Y.-L.; Kinloch, I. A.; Windle, A. H. Direct Spinning of Carbon Nanotube Fibers from Chemical Vapor Deposition Synthesis. Science 2004, 304, 276–278. (40) Zhong, X. H.; Li, Y. L.; Liu, Y. K.; Qiao, X. H.; Feng, Y.; Liang, J.; Jin, J.; Zhu, L.; Hou, F.; Li, J.-Y. Continuous Multilayered Carbon Nanotube Yarns. Adv. Mater. 2009, 21, 1–5.


23


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 31

(41) Liu, K.; Sun, Y.; Zhou, R.; Zhu, H.; Wang, J.; Liu, L.; Jiang, K. Carbon Nanotube Yarns with High Tensile Strength Made by a Twisting and Shrinking Method. Nanotechnology 2010, 21, 045708. (42) Kim, K. K.; Reina, A.; Shi, Y.; Park, H.; Li, L.-J.; Lee, Y. H.; Kong, J. Enhancing the Conductivity of Transparent Graphene Films via Doping. Nanotechnology 2010, 21, 285205. (43) Yao, Q.; Wang, Q.; Wang, L. M.; Wang, Y.; Sun, J.; Zeng, H. R.; Jin, Z. Y.; Huang, X. L.; Chen, L. D. The Synergic Regulation of Conductivity and Seebeck Coefficient in Pure Polyaniline by Chemically Changing the Ordered Degree of Molecular Chains. J. Mater. Chem. A 2014, 2, 2634-2640. (44) Van der Pauw, L. T. A Method of Measuring Specific Resistivity and Hall Effect of Discs of Arbitrary Shape. Philips Res. Rep. 1958, 13, 1-9. (45) Veca, L. M.; Meziani, M. J.; Wang, W.; Wang, X.; Lu, F. S.; Zhang, P. Y.; Lin, Y.; Fee, R.; Connell, J. W.; Sun, Y. P. Carbon Nanosheets for Polymeric Nanocomposites with High Thermal Conductivity. Adv. Mater. 2009, 20, 2088–2092. (46) Wang, X.; Yong, Z. Z.; Li, Q. W.; Bradford, P. D.; Liu, W.; Tucker, D. S.; Cai, W.; Wang, H.; Yuan, F. G.; Zhu, Y. T. Ultrastrong, Stiff and Multifunctional Carbon Nanotube Composites. Mater. Res. Lett. 2013, 1, 19-25. (47) Hong, C. T.; Kang, Y. H.; Ryu, J.; Cho, S. Y.; Jang, K.-S. Spray-printed CNT/P3HT Organic Thermoelectric Films and Power Generators. J. Mater. Chem. A 2015, 3, 21428–21433.


24

Page 25 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


a

c

b

P3HT PANI

CNT web on a wheel

Free-standing CNT web

f

Au-doped CNT web

e

Au-doped CNT/ polymer web

Seebeck coefficient measurements

g

d

Au-doped CNT web

h

j

i P3HT

Au NP

PANI

Figure 1. Overall scheme for fabricating ultra-lightweight, free-standing, and foldable Audoped CNT/PANI webs and corresponding surface morphologies. (a) The CNT web was prepared using direct spinning method. (b) The CNT web on the wheel was tailored and cut with scissors into pieces of 5 × 5 cm. (c) Au decoration was performed by immersing the CNT web in 2 mM AuCl3 solution for 10 min. (d, e) Au-doped CNT web was infiltrated with a conducting polymer through vacuum filtration. (f) The Seebeck coefficient was measured in air. SEM micrographs show surface morphologies of (g) pristine CNT web, (h) Au-doped CNT web with a number of spherical Au NPs, (i) Au-doped CNT/P3HT web, and (j) Au-doped CNT/PANI web. The scale bar indicates 1 µm. Some segregated polymer particles were observed on the surface of Au-doped CNT/polymer webs.


25

c

150

100

50

0

web web web web CNT ed CNT T/P3HT NT/PANI N p C o C d Au- -doped u-doped A Au

5000 4000 3000

-2

4000

-1

3500 3000 2500 2000 1500 1000 500 0

web web web web CNT ed CNT T/P3HT NT/PANI C op CN Au-d -doped u-doped u A A

2000 1000 0

web web web web CNT ed CNT T/P3HT NT/PANI N p C o C d Au- -doped u-doped A Au

d 0.25

10

-1

-1

6000

Page 26 of 31

Power Factor (µW m K )

-1

b 200

-1

Seebeck Coefficient (µV K )

a

Thermal Conductivity (W m K )

8

Figure of Merit ZT

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Electrical Conductivity (S cm )


6 4 2 0

web web web web CNT ed CNT T/P3HT NT/PANI N p C o C Au-d -doped u-doped A Au

0.20

0.15

0.10

0.05

0.00

web web web web CNT ed CNT T/P3HT NT/PANI N p C o C Au-d -doped u-doped A Au

Figure 2. Significantly enhanced in-plane TE figure of merit ZT. The figure of merit was increased by approximately three times in the Au-doped CNT/PANI web compared to that of the pristine CNT web. (a) In-plane Seebeck coefficient. (b) In-plane electrical conductivity (Inset: in-plane power factor). (c) In-plane thermal conductivity. (d) In-plane TE figure of merit ZT. All TE properties for the CNT webs are averaged from ten different circuits.


26

Page 27 of 31

a

b Counts

C

N 0.0

Void

0.5

1.0

1.5

Energy (KeV)

e d

c PANI

CNT

CNT Ordered PANI on the CNT surface

CNT Au-doped CNT Au-doped CNT/P3HT Au-doped CNT/PANI

Intensity (a. u.)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1200

1400

1600

1800

-1

Wavenumber (cm )

Figure 3. Characterizations of Au-doped CNT/polymer webs. (a) High-resolution TEM image. (b) EDXS mapping of Au-doped CNT/PANI web (N, green; C, red) (Inset: EDS spectrum). C and N are uniformly dispersed over the web, which implies that PANI is embedded homogeneously (scale bar: 40 nm). (c) Magnified TEM image. (d) Schematic representation of the highly ordered arrangement of the PANI backbone on the surface of the CNT induced by the synergistic effects of the m-cresol solvent process and the π–π conjugation between PANI and CNTs. (e) Raman spectra for different CNT webs.


27


b 1.4

Normalized Resistance (R/R0)

a Normalized Resistance (R/R0)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 31

Annealing for 20 min

1.2

1.0

0.8

0.6

300

400

500

600

Temperature (K)

700

c 3.0

Folding

2.5

300 µm

2.0

Unfolding 1.5

d

Unfoldi ng

1.0

0.5

0

10

20

30

40

50 Crumpling 60

µm 22 µm

Folding Cycles

Figure 4. Thermal and mechanical stability of Au-doped CNT/PANI webs. (a) The resistance R as a function of temperature and (b) number of folding cycles for Au-doped CNT/PANI webs, where R0 is the corresponding resistance of the original state before annealing and folding. No significant change was observed in the normalized resistance under the folding and subsequent crumpling tests. (c) Low-resolution SEM image and (d) magnified SEM image of crumpled Au-doped CNT/PANI web. While the whole area was wrinkled after folding and crumpling, the Au-doped CNT/PANI web maintained its structure in the distorted part.


28

Page 29 of 31

b

V V

Power Output (nW)

400

12 p type n type p-n type

300

10

Voltage Output (mV)

a

8 200

6 4

100 2 0 0

30

60

90

120

150

1.5

Power Output (µ W)

300

1.2

200

0.9

0.5

Power Output (µ W)

p type n type p-n type

at △T = 20 K

2.0

at △T = 15 K 1.5

at △T = 10 K

0.3

0.5 0.0

0

100

200

300

0.2

Current Output (µA)

0.6

100

0.4

at △T = 5 K 1.0

0.1

0.3

0

0.0

0.0

0

100

200

300

400

500

2

4

6

Load Resistance (Ω)

8

10

12

∆T (K)

14

16

18

20

-1

d 1.8

400

Power Output per Unit Weight (µW g )

-2

c

Power Output per Unit Area (µ W cm )

Current Output (µA)

Power Output (nW)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1000

800

600

400

200

0

Figure 5. Power generation of the organic TE module. (a) Schematic drawing of TE module. (b) Power output-current and voltage-current curves and (c) power output-load resistance curves of a flexible 14-leg TE device composed of Au-doped CNT/PANI webs (ptype), PEI-doped CNT webs (n-type), and Ag electrodes on a PET substrate. All dashed semicircle lines were drawn by calculation and estimation based on the measured Voc and short circuit current of the TEGs, while the particular solid circles are based on the measured power output values. The Voc of 10.87 mV and the R0 of 70 Ω are generated at the temperature gradient of ∆T = 10 K. The maximum power (Pmax) of 376 nW was obtained at a matching condition from the power-current curves. (d) Maximum power output (black), maximum power output per area (red), and maximum power output per


29


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 31

weight at ∆T = 20 K. The power output per unit area and unit weight was defined by dividing the power output by the area occupied by the legs and the weight of the legs, respectively. The inset figure represents a typical current-power curve of the p-n TE module at different temperature gradients.


30

Page 31 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


TOC


31

Foldable Thermoelectric Materials: Improvement of the Thermoelectric

Recommend Documents