Exceptional thermoelectric properties of flexible organic−inorganic hybrids with monodispersed and periodic nanophase

Flexible organic−inorganic hybrids are promising thermoelectric materials to recycle waste heat in versatile formats. However, current organic/inorganic hybrids suffer from inferior thermoelectric properties due to aggregate nanostructures. Here we demonstrate flexible organic−inorganic hybrids where size-tunable Bi2Te3 nanoparticles are discontinuously monodispersed in the continuous conductive polymer phase, completely distinct from traditional bi-continuous hybrids. Periodic nanofillers significantly scatter phonons while continuous conducting polymer phase provides favored electronic transport, resulting in ultrahigh power factor of ~1350 μW m−1 K−2 and ultralow in-plane thermal conductivity of ~0.7 W m−1 K−1. Consequently, figure-of-merit (ZT) of 0.58 is obtained at room temperature, outperforming all reported organic materials and organic−inorganic hybrids. Thermoelectric properties of as-fabricated hybrids show negligible change for bending 100 cycles, indicating superior mechanical flexibility. These findings provide significant scientific foundation for shaping flexible thermoelectric functionality via synergistic integration of organic and inorganic components.

1. Thermoelectric properties of PEDOT film sometimes change from sample to sample, depending on processing. What values of PEDOT films are used in the mixture model for the calculation of Seebeck and electrical conductivity of hybrid films? 2. The uniform thick films were used for in-plane thermal conductivity measurement by a 3w method since it is a challenging to perform the measurement of films with a thickness less than 100nm.In this case, please clarify if the electrical conductivity and Seebeck coeffcient were also measured on the same thick films?
3. The air stability test was lasted up to only 24h. From Figure S12, the electrical conductivity decays slightly. How about the TE performance if extending the text up to one week?
Reviewer #2 (Remarks to the Author): In this manuscript, Wang et al reported flexible organic/inorganic hybrids with monodispersed and periodic naophase. In fact, an organic/inorganic hybrid thermoelectric material is an attractive research area and the development of effective strategy to improve the TE performance is of vital importance. As for the hybrid materials, the conventional introduction of nanoparticles usually suffers from aggregation problem, leading to a challenge in fine-tuning the interface properties. In this work, the authors fabricated the hybrid films with PEDOT and fine-tuned Bi2Te3 nanoparticles. An exciting result claimed is the achievement of a high power factor over 1000 µW m-1 k-2 and low thermal conductivity of 0.7 W m-1 K-1. The concept and the experimental results are interesting, but I have the following concerns about the main claim of the authors. 1: The high performance is the main claim of this manuscript. Notably, the VPP deposited PEDOT (without Bi2Te3) show a high conductivity of 1350 S cm-1 and high Seebeck coefficient of 40 µV/K, which yield a high power factor over 200 µW m-1 k-2 (Fig. 3). This performance is very high compared with previous reports without fine-tuned doping level (Nat. Mater. 2011, 10, 429). How could this happen? A discussion should be added. 2: As mentioned above, one feature in this study is that the Seebeck coefficient is large. So, the authors should make extra effort to motivate that the measurement of Seebeck coefficient is free of major errors. There are errors from the temperature measurement and from the voltage reading due to geometrical parameters (no due to instrumentation). The authors should show estimate the error on the Seebeck measurement due to the geometrical feature of the electrodes chosen according to the article [S. van Reenen, M. Kemerink, Organic Electronics, 15 (2014) 2250]. The device geometry for the measurement of Seebeck coefficient should be provided. Moreover, I suggest the deposition of thermal resistor on the hybrid film to measure the temperature difference. The utilization of thermocouple can easily lead to contact problem and result in measurement errors in temperature difference. 3: As for the measurement of thermal conductivity, it seems that it is not so easy to fabricated thick film (over 500 nm) with method illustrated by Fig. 1. The authors should provide detailed fabricated method, whereas the characterization details are also needed to confirm that the thick film have a same or similar morphology or structure. What is Seebeck coefficient and electrical conductivity of the thick sample? By the way, what is thickness of PEDOT film in a typical thin film and so-called thick film. 4：In the section of "Estimation of the Bi2Te3 nanoparticle fraction in hybrid films" in supplementary information, the thickness of hybrid films and Bi2Te3 nanoparticle are presented by d0 and d1, respectively. What is the relationship among d0, d1 and the 'typical average thickess of film' ('Electrical conductivity measurement' section in supplementary information). How to determine the exact thickness of the sample if these three thickness are different? 5: The SEM image of the device for thermal conductivity measurement should be provided in Fig  s14. 6: As for the mechanical flexibility, I recommend the authors to perform the measurement of Seebeck coefficient upon different bending to make the result more convincing. 7: The fourth step in Figure 1 schematically showed that the SiO2 is removed and Bi2Te3 is directly contact with Si substrate. I'm wondering whether the TE material is directly attached to Si substrate or not. Would the conductive Si substrate influence the Seebeck coefficient or electrical conductivity characterization? It is necessary to fabricate material on insulating substrate like glass to measure the TE performance.
Reviewer #3 (Remarks to the Author): The presented hybrid film for themoelectricity is interesting, and the fabrication method is also interesting. However, I have the following comments. * Flexible materials with high ZT The authors mentioned that flexible materials with high ZT doesn't exist in the abstract. However, the flexible Bi2Te3 films in both p and n type had been presented as follows; K. Kato et al., Journal of electronic materials, Vol.43, No.6 (2014) pp.1733-1739 * For N-type The present study is only for p-type. N-type material is faborable to make a thermoelectric module.

* Interfacial resistance
The electrical conductivity of the composite follow the highest predicted value in the present study, however, those values follow the lowest value mostly mentioned below (sometime, lower than the lowest value due to interfacial electrical resistance).
On the other hand, interfacial thermal resistance is very high. The order is 10^-5 K/W/m2. Thermal resistance is well reviewed in the following article. The order of the resistance is 10^-10 to 10^-7 K/W/m2. T. Zhan et al., Scientific report, Vol.7 (2017) 7109.
The electrical resistance is really low although the thermal resistance is 100 times higher than other reported value at the present study. The mechanism of the present study should be explained to have the readers' understanding.  Fig. 3a and 3b. The VPP process of preparation of PEDOT for all PEDOT films as well as PEDOT/Bi 2 Te 3 hybrid films were conducted under the same conditions. We added the sentence "The electrical conductivity and Seebeck coefficient values of PEDOT and Bi 2 Te 3 used in the models were based on the average values of more than three specimens" on Page 7 in the revised manuscript.

Comment 2:
The uniform thick films were used for in-plane thermal conductivity measurement by a 3w method since it is a challenging to perform the measurement of films with a thickness less than 100 nm. In this case, please clarify if the electrical conductivity and Seebeck coefficient were also measured on the same thick films?
Response: The in-plane electrical conductivity and Seebeck coefficient were measured on the thin films with thickness less than 100 nm. Since it is very difficult to perform the in-plane thermal conductivity measurement of films with a thickness less than 100 nm, thus we prepared thick hybrid films (thickness of ~0.75 µm) for thermal conductivity measurement by a 3ω method, in order to estimate the ZT values of as-prepared hybrid films. All of these have been clarified in the section of "thermal conductivity measurement" in the revised Supplementary Information.

Comment 3:
The air stability test was lasted up to only 24h. From Figure S12, the electrical conductivity decays slightly. How about the TE performance if extending the text up to one week?

Response:
The air stability has been tested by extending the duration up to one week according to your suggestion. As shown in the following figure, all the films were relatively stable in air at room temperature, displaying slightly changes in electrical conductivity after one week. These results were added in Supplementary Fig. 15 in the revised Supplementary Information.  (Fig. 3). This performance is very high compared with previous reports without fine-tuned doping level (Nat. Mater. 2011, 10, 429). How could this happen? A discussion should be added.

Response:
The VPP process used in this work is referred to another paper (Nat. Mater. 2014, 13, 190), as we mentioned on Page 5 in the revised manuscript that "PEDOT was polymerized with Bi 2 Te 3 nanoparticles via a modified vapor-phase polymerization (VPP) process 30,31 ." In the reference, the electrical conductivity of VPP deposited PEDOT films can be 1200~1500 S cm -1 , and the Seebeck coefficient is in the range of 35~50 µV K -1 . Our results are comparable to previously reported values in the references. Response: Thank you for providing us with this important article (Organic Electronics, 2014, 15, 2250. We strongly agree with you that the temperature measurement and the voltage reading due to geometrical parameters will greatly affect the accuracy of the Seebeck coefficient measurement. In fact, we also carefully read this paper before and thereby chose the geometrical feature of the electrodes according to this article. Similar to this article as well as a previous paper reported by our group (Nanoscale, 2016, 8, 8033. Supporting Information), a set of parallel and narrow line-shaped (1 mm × 7 mm) gold electrodes with thickness of 150 nm and spacing of 10 mm were used, in order to get an accurate determination of the actual Seebeck coefficient. In addition, both the thermocouples and the sample surface in the region of the thermocouples were erased by a swab with ethanol each time in order to avoid the contact problem and get accurate measurement in temperature difference. The details for the measurement of Seebeck coefficient have been added in the section "Seebeck coefficient measurement" in the revised Supplementary Information. More importantly, two different reference samples, n-type Bi 2 Te 3 sample with Seebeck coefficient of -180 µV K -1 and constantan with Seebeck coefficient of -36 µV K -1 which were obtained by a well-calibrated commercial instrument ZEM-3, were tested in order to confirm the accuracy of this method. The Seebeck coefficient measurement curves of this two samples are illustrated as follows. It reveals the measurement errors for Seebeck coefficient were less than 6%.  Response: Although the electrical conductivity and Seebeck coefficient were measured from thin films with thickness less than 100 nm, it is still very difficult to perform the in-plane thermal conductivity measurement for these thin films. Thus we prepared thick hybrid films (thickness of ~0.75 µm) for thermal conductivity measurement by a 3ω method. Compared with the method for preparing thin hybrid films, several modifications were made to prepare thick hybrid films.
Firstly, 1 µm thick SiO 2 layer was deposited on the Si wafer via PEVCD in the step of "pretreatment of sustrates". Then, the etching time for removing the exposed SiO 2 by RIE was increased to 30 min in the step of "fabrication of Bi 2 Te 3 nanoparticle arrays". Also, the thickness of deposited Bi 2 Te 3 film was fixed to 0.7 µm. Finally, the parameters of spin coating in the step of VPP was changed to 1500 rpm for 25 s. The VPP process was repeated for three times to obtain desired thick hybird films. Taken as an example, the SEM image of as-prepared PEDOT/31 vol% Bi 2 Te 3 (100) hybrid thick film is given in Supplementary Fig. 10 in the revised Supplementary Information, which displays a very similar morphology as compared to the corresponding thin film. All of the above discussions have been provided in the section of "thermal conductivity measurement" in the revised Supplementary Information according to your comment. Both the electrical conductivity and Seebeck coefficient of the as-prepared thick films are just a little lower than the corresponding thin film. For instance, the electrical conductivity and Seebeck coefficient of PEDOT/31 vol% Bi 2 Te 3 (100) hybrid thick film are 391 S cm -1 and 152 µV K -1 . The thickness of as-prepared PEDOT thin film is ~56 nm. We modified the Method section with proper revision in the revised manuscript (Page 15). For the PEDOT thick films for thermal conductivity measurement, the detailed preparation method was also provided in the section of "thermal conductivity measurement" in the revised Supplementary Information. The thickness of as-prepared PEDOT thick film was measured to be ~0.62 µm. In the section of "Electrical conductivity measurement" in Supplementary Information, the typical average thickness of film means the thickness of hybrid films (d 0 ). Indeed, these thicknesses show a little difference. Thus, in our work, we estimated the Bi 2 Te 3 nanoparticle fraction in hybrid films and calculated the electrical conductivity of all films with the tested film thickness.
Thank you for your valuable comment. We have made revisions in the section of "Estimation of the Bi 2 Te 3 nanoparticle fraction in hybrid films" and deleted the misleading sentence "typical average thickness of film" in the section of "Electrical conductivity measurement" in the revised Supplementary Information.

Comment 5:
The SEM image of the device for thermal conductivity measurement should be provided in Fig s14. Response: Thank you for your suggestion. The characterizations of the device for thermal conductivity measurement (see below) have been provided in the revised Supplementary   Information (Supplementary Fig. 12). Response: According to your suggestion, the Seebeck coefficient of PEDOT/31 vol% Bi 2 Te 3 (100) hybrid film upon different bending radii were measured. The results were listed as following. The temperature difference was generated by heating one side of the hybrid film with a flexible heater which was connected with a Keithley 2400 SourceMeter. The Seebeck coefficient kept stable upon different bending radii, only ~10% changes even at a very low curve radius of 3.5 mm. These results were added in the revised Supplementary Information   (Supplementary Fig. 19).

Figure |
The Seebeck coefficient S of prepared PEDOT/Bi 2 Te 3 (100) hybrid film with ~31 vol% Bi 2 Te 3 nanoparticle fraction as a function of curve radius, where S 0 is the Seebeck coefficient before bending. Figure 1 schematically showed that the SiO 2 is removed and After completely removing exposed SiO 2 , Bi 2 Te 3 was directly deposited on exposed Si substrate.

Comment 7: The fourth step in
Then, SiO 2 (under Cr protective layer) can be easily and fastly dissolved by HF solution in the fourth step, leaving Bi 2 Te 3 nanoparticle arrays on the Si substrates which was confirmed by SEM images (Fig. 2f-j).
In order to exclude the effect of substrate, we used nonconductive Si wafer as substrate, as mentioned in the Method. We are sorry for the unclear expression. The detailed information of used Si wafer is provided on Page 13 in the revised manuscript: Silicon wafers (undoped, resistivity ρ > 10000 ohm cm) were purchased from University Wafer. Before using the silicon wafers, we also tested and confirmed that these silicon wafers were nonconductive.

Reviewer #3: Comment 0: The presented hybrid film for themoelectricity is interesting, and the fabrication method is also interesting.
Response: Thank you very much for your positive comments.
Comment 1: Flexible materials with high ZT. The authors mentioned that flexible materials with high ZT doesn't exist in the abstract. However, the flexible Bi 2 Te 3 films in both p and n type had been presented as follows; K. Kato et al., Journal of electronic materials, Vol.43, No.6 (2014) pp.1733-1739.
Response: Thank you for your comment. To make accurate statement, we modified the sentence from "such a ZT value is much higher than any existing flexible materials" to "outperforming all reported organic materials and organic/inorganic hybrids" in the Abstract. In our work, we report flexible organic/inorganic hybrids with high thermoelectric properties, which hold lower density, better flexibility and less harmful element contents in comparison to neat Bi 2 Te 3 films.
Additionally, K. Kato et al. reported both p-type and n-type flexible Bi 2 Te 3 films with high ZT values of ~1, which were estimated with out-of-plane thermal conductivity.

Comment 2:
For N-type. The present study is only for p-type. N-type material is favorable to make a thermoelectric module.

Response:
A thermoelectric generator is made of both p-type and n-type materials. Thus it is necessary to explore both p-type and n-type materials with comparable thermoelectric properties.
Although we only report p-type high-performance thermoelectric materials, we present a facile but robust method for preparing organic/inorganic hybrids. We are confident that n-type thermoelectric materials with high thermoelectric performance can be achieved by adopting the preparation method with n-type materials in the future work. Thanks again for your suggestions.  ACS Applied Materials & Interfaces, Vol.2, No.11 (2010) pp.3170-3178. On the other hand, interfacial thermal resistance is very high. The order is 10^-5 K/W/m2. Thermal resistance is well reviewed in the following article. The order of the resistance is 10^-10 to 10^-7 K/W/m2. T. Zhan et al., Scientific Report, Vol.7 (2017) 7109. The electrical resistance is really low although the thermal resistance is 100 times higher than other reported value at the present study. The mechanism of the present study should be explained to have the readers' understanding.
Response: Thank you very much for your comment.
(1) Regarding to the electrical conductivity. We also noticed the results reported by B. Zhang et al. (ACS Applied Materials & Interfaces, 2010, 2, 3170), and mentioned this reference (Ref. 13) in the Introduction. In Ref. 13, the inorganic particles severely aggregated and sank to the bottom of hybrids, forming a two-layer structure. This made the electrical conductivity of hybrids far below the calculated values based on mixture models. On the other hand, the electrical conductivity (1000 S cm -1 ) of Bi 2 Te 3 used in the mixture models was taken from conventional sintered Bi 2 Te 3 bulk, which should be very different with their samples that were prepared by simple solution process. As to our samples, the electrical conductivity of hybrid films gradually approached to the lowest predicted values with increased Bi 2 Te 3 nanoparticle fraction (Fig. 3a). It was mainly caused by the increase of the interfacial surface-to-volume ratio which scattered more carriers, and thereby decreasing the electrical conductivity.
(2) Regarding to the interfacial thermal and electrical resistance. Thank you very much for providing us with this useful article (Scientific Report, 2017, 7, 7109). This paper well reviewed the thermal resistance for lots of selected hybrid systems, and the order of the resistance is from 10 -10 to 10 -7 m 2 K W -1 . However, there are also lots of reported thermal resistances which are in the similar order (10 -6 to 10 -5 m 2 K W -1 ) with our sample. .) The size of nanoparticles in the hybrids may be similar to the mean free path of phonons but much longer than that of charge carriers. Therefore, thermal transport can be significantly suppressed due to the boundary phonon scattering effect, while electrical transport is slightly affected. (Ref: Advanced Materials, 2007, 19, 1043 Consequently, the electrical resistance is low although the thermal resistance is high. All the discussions are added and highlighted on Page 11 in the revised manuscript.
The authors have addressed all of my concerns. I think that now the manuscript can be accepted for publication in Nat Comm. One thing I want to stress again that the all the thermoelectric properties for zT calculation should be measured on the same sample along the same direction, otherwise the zT value may have a large deviation from the true one. In this work, the thermal conductivity was measured on a thick film but the electrical properties were measured on a thin film. The authors should bare mind that such measurements may result in a relatively large error for zT estimation.
Reviewer #2 (Remarks to the Author): I suggest acceptance of the manuscript since appropriate revision has been made according to the reveiwer's comments.

Reviewer #3 (Remarks to the Author):
Thank you for the response, but I have the following questions.
(1) Dr. Kato made a Bi2Te3-PEDOT:PSS thick film. Out-of-plane thermal conductivity was used to calculated the ZT, but their film was enough thick for thermal conductivity measurement of the composite. Sub-micron structure smaller than the thickness was made in the film. Their ZT estimation was not bad.
(2) I understand the response for comment 2.
(3) I have read the papers pointed out in the response. I found that 10^-5 m^2 K/W order thermal resistance had been reported. In one of the papers (Journal of Power Technologies, 2014, 94, 270), the authors explained that the high thermal resistance is caused by the imperfections at the interfaces in Line 2 on Page 279. In the present study, authors should explain the low electrical interfacial resistance mechanism at the imperfection interface.
In addition, low thermal conductivity is explained by long phonon mean free path in the response. However, the reported lattice thermal conductivity of Bismuth Telluride nano-wire with 52nm diameter is about 0.5 W/m/K (Journal of Applied Physics, 2009, 105, 104318). They explained that the thermal conductivity reduction was only 20% due to the short phonon mean free path in Bi2Te3.
We roughly calculated lattice thermal conductivity by using Lorentz number in the present study. The calculated lattice thermal conductivity was about 0.4 W/m/K for PEDOT/Bi2Te3(100). The measured thermal conductivity in the present study was lower than that of nano-wire. More explanation is necessary for the publication.

Response:
Thank you for your recommendation. Our preparation method is suitable for preparing organic/inorganic hybrid films with thickness less than 100 nm, so the electrical conductivity and Seebeck coefficient were measured from thin films. We agree that the ZT evaluation should be performed on the same sample. However, to the best of our knowledge, it is still lack of accurate methods for performing the in-plane thermal conductivity measurement of nano-thick films with substrates. Thus, we prepared thick films (thickness of ~0.75 µm) for the 3ω thermal conductivity measurement. This also is a common solution in previous works (Nat. Mater., 2011, 10, 429. Nat. Mater., 2013, 12, 719. J. Am. Chem. Soc., 2017. The total thermal conductivity contains the lattice thermal conductivity and electronic thermal conductivity. In our hybrid films, the lattice thermal conductivity is mainly dominated by the film morphology. The thick films for the 3ω thermal conductivity measurement display a very similar morphology as compared to the corresponding thin films. For example, the SEM image of as-prepared PEDOT/31 vol% Bi 2 Te 3 (100) hybrid thick film is given in Supplementary Fig. 10. While the electronic thermal conductivity is related to the electrical conductivity. Our thick films for thermal conductivity measurement only possesses 10~20% lower electrical conductivity as compared to the corresponding thin films. For example, the electrical conductivity of PEDOT/31 vol% Bi 2 Te 3 (100) hybrid thick film is ~391 S cm -1 , which is ~483 S cm -1 for the corresponding thin film. It may not result in a relatively large error for the electronic thermal conductivity. Therefore, the total thermal conductivity of thick film may have limited deviation from the thin one. We have added the above discussion in the revised supplementary information (Page 10).
Thanks again for your very useful suggestions.

Reviewer #2:
Comment: I suggest acceptance of the manuscript since appropriate revision has been made according to the reviewer's comments.

Response:
Thank you very much for your recommendation.

Reviewer #3:
Comment 1: Dr. Kato made a Bi 2 Te 3 -PEDOT:PSS thick film. Out-of-plane thermal conductivity was used to calculated the ZT, but their film was enough thick for thermal conductivity measurement of the composite. Sub-micron structure smaller than the thickness was made in the film. Their ZT estimation was not bad.

Response:
Thank you for your instructive comment. We agree with you. In Dr. Kato's paper (Journal of Electronic Materials, 2014, 43, 1733, they prepared both p-type and n-type Bi 2 Te 3 films on insulating porous polyimide (PI) substrates to obtain flexible films. It is an interesting work, which provides an effective way for preparing high-performance and flexible Bi 2 Te 3 films.
Differently, we offer another way for making flexible organic/inorganic hybrids with high thermoelectric properties in our work. The matrix (highly conductive PEDOT) is organic component, which holds intrinsically low density and excellent flexibility.
Comment 2: I understand the response for comment 2.

Response:
Thank you.
Comment 3: I have read the papers pointed out in the response. I found that 10^-5 m^2 K/W order thermal resistance had been reported. In one of the papers (Journal of Power Technologies, 2014, 94, 270), the authors explained that the high thermal resistance is caused by the imperfections at the interfaces in Line 2 on Page 279. In the present study, authors should explain the low electrical interfacial resistance mechanism at the imperfection interface.
In addition, low thermal conductivity is explained by long phonon mean free path in the response. However, the reported lattice thermal conductivity of Bismuth Telluride nano-wire with 52nm diameter is about 0.5 W/m/K (Journal of Applied Physics, 2009, 105, 104318). They explained that the thermal conductivity reduction was only 20% due to the short phonon mean free path in Bi 2 Te 3 . We roughly calculated lattice thermal conductivity by using Lorentz number in the present study. The calculated lattice thermal conductivity was about 0.4 W/m/K for PEDOT/Bi 2 Te 3 (100). The measured thermal conductivity in the present study was lower than that of nano-wire. More explanation is necessary for the publication.

Response:
Thank you for your instructive comments. We have made further explanation for the high interfacial thermal resistance and low interfacial electrical resistance at the organic/inorganic hybrid interface in the revised manuscript.
The low interfacial thermal conductance is possibly caused by the strong acoustic mismatch due to different phonon densities and velocities between polymer matrix (PEDOT) and inorganic filler (Bi 2 Te 3 ) (Ref. 3,43 in the revised manuscript). Furthermore, the phonon mean free path in PEDOT might be in the order of 10 1~1 0 2 nm as simulated in Ref 44, which is comparable to the size of Bi 2 Te 3 nanoparticles in the hybrid films. This will enhance interfacial phonon scattering and thereby the thermal transport can be considerably suppressed (Ref. 3,12).
Regarding to the electrical transport, the highly conductive PEDOT matrix in hybrids is continuous, providing hole transport paths. More importantly, PEDOT is in situ polymerized with the presence of Bi 2 Te 3 nanoparticles. This is helpful for intimate contact between PEDOT and Bi 2 Te 3 and thereby enhancing charge transport across the PEDOT-Bi 2 Te 3 interfaces (Ref. 33). So, we do not think the PEDOT-Bi 2 Te 3 interfaces are imperfection. Many works have reported in situ synthesized polymer composites with enhanced electrical transport properties (Ref. 3,45). Besides, the mean free path of holes in PEDOT was calculated to be 10 0~1 0 1 nm (Ref. 44), which is much smaller than the size of Bi 2 Te 3 nanoparticles (~40-500 nm), suggesting a negligible degradation to carrier transport. Thus, the hole transport can be minimally affected as compared to the phonon transport.
Experimental measurement of phonon mean free path, such as using frequency domain thermoreflectance method, is helpful to illustrate the interfacial thermal/electrical conductance in the future work. (Nature Communications volume 4, 1640 (2013)) The paper (Journal of Applied Physics, 2009, 105, 104318) reported the thermal conductivity of individual bismuth telluride nanowires with a 52nm diameter, where thermal transport is along the continuous single nanowire. That is significantly different from our research here. Our Bi 2 Te 3 /PEDOT hybrid film consists of numerous discontinuous and isolated Bi 2 Te 3 particles embedded in the PEDOT polymer matrix. The thermal transport mechanism should be very different. The factors on the thermal conductivity of our hybrid film are complex, including the volume fraction, morphology, the thermal conductivity of conducting polymer and the inorganic fillers, and especially the interfacial thermal conductance. The interfacial effect may contribute to the lower lattice thermal conductivity of our hybrid sample as explained above.