This is the accepted version of the article: Gueye M.N., Vercouter A., Jouclas R., Guérin D., Lemaur V., Schweicher G., Lenfant S., Antidormi A., Geerts Y., Melis C., Cornil J., Vuillaume D.. Thermal conductivity of benzothieno-benzothiophene derivatives at the nanoscale. Nanoscale, (2021). 13. : 3800 - . 10.1039/d0nr08619c. Available at: https://dx.doi.org/10.1039/d0nr08619c Thermal conduc.vity of benzothieno-benzothiophene deriva.ves at the nanoscale Maga$e N. Gueye,a,# Alexandre Vercouter,b,# Rémy Jouclas,c David Guérin,a Vincent Lemaur,b Guillaume Schweicherc,*,c Stéphane Lenfant,a Aleandro AnHdormi,d Yves Geerts,c,e Claudio Melis,f Jérôme Cornilb,* and Dominique Vuillaumea,* a) Ins'tute for Electronics Microelectronics and Nanotechnology (IEMN), CNRS, Av. Poincaré, Villeneuve d'Ascq, France. b) Laboratory for Chemistry of Novel Materials, University of Mons, Place du Parc 20, Mons, Belgium. c) Laboratoire de Chimie des Polymères, Faculté des Sciences, Université Libre de Bruxelles (ULB), Boulevard du Triomphe, 1050, Brussels, Belgium. d) Catalan Ins'tute of Nanoscience and Nanotechnology (ICN2), CSIC and BIST, Campus UAB, Bellaterra, 08193, Barcelona, Spain. e) Interna'onal Solvay Ins'tutes for Physics and Chemistry, Brussels, Belgium. f) Dipar'mento di Fisica, Universita di Cagliari, CiWadella Universitaria, 09042 Monserrato (Ca), Italy. We study by scanning thermal microscopy the nanoscale thermal conductance of films (40-400 nm thick) of [1]benzothieno[3,2-b][1]benzothiophene (BTBT) and 2,7-dioctyl[1]benzothieno[3,2-b][1]benzothiophene (C8-BTBT-C8). We demonstrate that the out-of-plane thermal conduc'vity is significant along the interlayer direc'on, larger for BTBT (0.63 ± 0.12 W m-1 K-1) compared to C8-BTBT- C8 (0.25 ± 0.13 W m-1 K-1). These results are supported by molecular dynamics calcula'ons (Approach to Equilibrium Molecular Dynamics method) performed on the corresponding molecular crystals. The calcula'ons point to significant thermal conduc'vity (3D-like) values along the 3 crystalline direc'ons, with anisotropy factors between the crystalline direc'ons below 1.8 for BTBT and below 2.8 for C8-BTBT-C8, in deep contrast with the charge transport proper'es featuring a two-dimensional character for these materials. In agreement with the experiments, the calcula'ons yield larger values in BTBT compared to C8-BTBT-C8 (0.6-1.3 W m-1 K-1 versus 0.3-0.7 W m-1 K-1, respec'vely). The weak thickness dependence of the nanoscale thermal resistance is in agreement with a simple analy'cal model. Keywords: thermal conducHvity, scanning thermal microscopy (SThM), approach to equilibrium molecular dynamics (AEMD), organic semiconductor, organic thermoelectricity. 2 INTRODUCTION. Organic materials have recently aWracted a great interest for poten'al thermoelectric applica'on because commercial modules are built with materials such as bismuth telluride (Bi2Te3)-based alloys1 which are toxic, expensive, and energy-consuming for processing. The candidates range from polymers (like PEDOT:PSS, PEDOT:OTf, and other deriva'ves) with high electrical conduc'vity and Seebeck coefficient (up to ∼5,000-6,000 S cm-1 and ∼1 mV K-1),2-5 down to self-assembled monolayers and single molecule junc'ons (based mainly on alkyl chains, oligo(phenylene ethynylene)s, benzene, C60,…).6-13 In the laWer case, quantum interference effects can be exploited to tailor and op'mize the thermoelectric proper'es of the molecules.14-17 In contrast to the charge transport proper'es, the thermal conduc'vity (κ) of thin films of small π- conjugated organic semiconductors (OSCs) appears less studied at both the experimental and theore'cal levels. The in-plane κ of various molecular thin films (pentacene, TPD, Alq3, C60, PCBM, rubrene, DNTT, ….)4, 18-26 has been measured in the range 0.1 - 0.8 W m-1 K-1 at the macroscale (ac-calorimetry, 3ω Joule hea'ng, 'me domain thermo-reflectance….). Only a very few reports have been published on the out-of-plane κ of these organic materials at the nanoscale, e.g. using the scanning thermal microscope (SThM).27, 28 On the theore'cal side, the predic'on of κ for OSCs and the defini'on of structure- property rela'onships is also scarcely addressed. The κ values can be es'mated by models29 based on: (i) collec've excita'ons of phonons via the resolu'on of the Boltzmann Transport Equa'on (BTE); (ii) atomic displacements via Molecular Dynamics (MD) approaches such as the Green-Kubo formalism or the Non Equilibrium Molecular Dynamics method (see below). The BTE approach is known to be far much prohibi've in terms of computa'onal cost than MD-based techniques. Thus, κ of a few molecules (pentacene, C60, PCBM, H2Pc, TPD,..)30-38 3 have been essen'ally calculated by various methods belonging to the “MD-class” in order to es'mate the anisotropy of κ along the long axis of the molecules versus in the perpendicular intralayer direc'ons. In this work, we measure by SThM the out-of-plane κ at the nanoscale of thin films (40-400 nm thick) of [1]benzothieno[3,2-b][1]benzothiophene (BTBT) and 2,7-dioctyl[1]benzothieno[3,2-b][1]benzothiophene (C8-BTBT-C8).39, 40 These molecules have promising performances for organic electronics with reproducible transistor hole mobili'es in excess of 10 cm2 V-1 s-1 and up to ∼ 200 cm2 V-1 s-1 at the local scale.40, 41 Since their thermoelectric proper'es were not inves'gated experimentally up to now, this has s'mulated the present experimental and theore'cal works to determine the structure-property rela'onships of BTBT deriva'ves, especially the role of alkyl chains on the thermal transport. We demonstrate (SThM) that κBTBT = 0.63 ± 0.12 W m-1 K-1 is larger than κC8-BTBT-C8 = 0.25 ± 0.13 W m-1 K-1. The nanoscale thermal resistance is weakly dependent on the film thickness, as predicted by a simple analy'cal model of the constric'on thermal resistance for a surface coated by a thin film. The experimental results are supported by the theore'cal es'mates of κ obtained by the AEMD (Approach to the Equilibrium Molecular Dynamics) method.42 We find κ along the long c- axis of the molecules larger for BTBT (1.04 W m-1 K-1) than for C8-BTBT-C8 (0.72 W m-1 K-1). The results also point to a decrease of κ in the a-b plane upon alkyla'on (from 0.6-1.3 W m-1 K-1 for BTBT to 0.26-0.33 W m-1 K-1 for C8-BTBT-C8). RESULTS. Scanning thermal microscopy results. Figure 1 shows the typical topographic and thermal voltage images of a C8-BTBT- C8 film prepared by spin-coa'ng (see Methods). The organic thin film has a staircase topography (Fig. 1-a, height profile Fig. 1-c, red curve) with an 4 incomplete surface coverage leaving apparent several zones of the underlying Si/ SiO2 substrate. The same type of "staircase" topography is observed for all samples (Figs. S1 and S2 in the ESI), in agreement with previous results.43 In contrast, the SThM thermal voltage (Fig. 1-b) shows a featureless structure (Fig. 1-c, blue curve) for all samples (Figs. S1 and S2 in the ESI). The thermal voltage VSThM is related to the thermal resistance of the sample by: ( ) VSThM ∝ Tsample −Tamb ∝ RthQ! (1) where Tsample is the temperature of the surface sample, Tamb the ambient temperature, Rth the thermal resistance of the sample and Q! the thermal flux. We analyzed the SThM image to determine the thermal resistance of the films at various loca'ons with various thicknesses t, taking the thermal resistance of the Si/SiO2 substrate as a reference (Fig. 1-d). Figure 1. (a) Topographic (10 μm x 10 μm) and (b) SThM thermal voltage (at the output of the Wheatstone bridge, 3ω-SThM method) images of a C8-BTBT-C8 film 5 spin-coated on Si/SiO2 substrate, (c) height (red) and thermal voltage profiles (blue). (d) SchemaHc of the nanoscale SThM measurement. On the substrate, the thermal resistance is Rth-ox=Rt-ox+Rsub, with Rt-ox the thermal resistance of the Hp- SiO2 interface and Rsub the thermal resistance of the substrate. Over the organic domains, we have Rth-org=Rt-org+Rorg(t)+Rorg-ox+Rsub, with Rt-org the thermal resistance of the Hp-organic interface, Rorg(t) the thermal resistance of the organic film of thickness t and Rorg-ox the thermal resistance of the organic-oxide interface (Fig. 1-d). We determined κ using the null-point method, NP-SThM.44 This differen'al method is suitable to remove the parasi'c contribu'ons (air conduc'on, etc…). When the 'p contacts the sample surface (C), both the sample and parasi'c thermal contribu'ons are involved, whereas, just before physical 'p contact (non contact, NC), only the parasi'c thermal contribu'ons are involved. We measured the thermal voltage VSThM-z traces at several places on the organic films and on the substrate zones. The 'p temperature is determined from VSThM (see the ESI). Figure 2 show 25 typical 'p temperature versus distance (z-trace) curves measured on the C8-BTBT-C8 domain and on the nearby apparent Si/SiO2 substrate (circled bullets on the SThM images in the insets). When approaching the heated 'p to the surface, the 'p temperature decreases gradually because the heat transfer through the air gap is increased. At contact, we observe a sudden decrease from TNC to TC, due to the addi'onal heat flux through the 'p- sample contact. Figure 2-c shows the TC versus TNC-TC curves for the C8-BTBT-C8 and BTBT samples and on the apparent substrate zone, where TC and TNC are the measured temperature at contact and non-contact condi'ons. 6 Figure 2. 25 Hp temperature versus Hp verHcal displacement curves (z-trace, approach, 0 corresponds to the Hp retracted) measured at VDC = 0.6 V on (a) a C8- BTBT-C8 domain (sample #5), (b) on the Si/SiO2 substrate (as indicated by the circled bullets). (c) Temperature jump TNC-TC versus temperature at Hp contact TC for C8-BTBT-C8 (◆), BTBT (●) and the Si/SiO2 substrate (◼). The increasing TC corresponds to a supply voltage VDC from 0.6 - 0.9 V (by step of 0.1 V) of the Wheatstone bridge (0.6 - 1 V, step of 0.1 V on the substrate). Solid lines are the linear fits. Each data point (C8-BTBT-C8 and BTBT) is the average of 3 measurements at 3 different locaHons (25 VSThM-z traces at each locaHon). The data points for the substrate are averaged from the data acquired on the 2 samples. 7 The thermal conduc'vity is determined using the following rela'on:44 ⎡ 1 ⎤ ( TC −Tamb = ⎢α + β ⎥ TNC −TC ) (2) ⎣ κ ⎦ where α and β are calibrated parameters related to the SThM equipment and 'p (Fig. S3 in the ESI) and Tamb the room temperature (α = 25.6 W/m.K, β = 21.6 K/K and Tamb = 22.5 °C). From a linear fit on these data, we get: κC8-BTBT-C8 = 0.25 ± 0.13 W m-1 K-1, κBTBT = 0.63 ± 0.12 W m-1 K-1 and κsub = 1.57 ± 0.43 W m-1 K-1. A key finding is that the thermal conduc'vity of BTBT is larger than that of C8-BTBT- C8, in full consistency with our AEMD simula'ons (see below). For the Si/SiO2 substrate, we found a value close to that of bulk SiO2, in agreement with previous measurements by NP-SThM44 showing that the effec've κ is that of bulk SiO2 (see Fig. 6-b in Ref. 44 and Fig. S4) if the SiO2 thickness is larger than ∼ 100 nm (here 200-500 nm). The ra'o of the thermal voltage measured on the organic domain over that on the substrate (Fig. 1), VSThM-org/VSThM-sub is related to the ra'o of the corresponding thermal resistance of each zone. For the Si/SiO2 substrate, the constric'on thermal resistance45 is Rsub=1/4rκox = 9x106 K W-1 with κox the "bulk" SiO2 value (1.4 W m-1 K-1) and r is the radius of the SThM 'p thermal contact (es'mated to be ≈ 20 nm, see the ESI). In order to determine the Rorg(t) from this SThM voltage ra'o, we need to es'mate the various interfacial resistances (Kapitza resistance)46 which cannot be directly measured here. Reported values for a large variety of interfaces27, 47-54 range typically between 10-9 and 10-6 m2 K W-1. For simplicity, we consider the lower limit value for all interfaces. In that case, the interface thermal resistances are negligible (8x105 K W-1) compared to Rsub and Rorg (Rsub=1/4rκox = 9x106 K W-1, Rorg=1/4rκorg = 8 2-5x107 K W-1, considering the values of κorg determined above). In this oversimplified case, the ra'o of the thermal voltage measured on the organic * domain versus over the substrate is given by VSThM−org = Rorg (t ) , assuming VSThM−sub Rsub the same thermal flux Q! on the oxide and the organic domain (see the ESI), * where Rorg (t ) is the effec've constric'on thermal resistance measured by the sharp SThM 'p at the surface of the organic thin film. We consider a simple analy'cal model derived by Dryden55 for film with t/r>2 (here t > 40 nm): 1 1 ⎛r⎞ ⎛ 2 ⎞ * Rorg (t ) = − ln ⎜ ⎟ (3) 4rκ org 2π rκ org ⎜⎝ t ⎟⎠ ⎝ 1+ κ org κ sub ⎠ The first term on the right-hand side stands for the constric'on thermal resistance of the bulk organic film (t → ∞) and the second term represents the effect of the underlying substrate covered by the film of thickness t. Figure 3 * shows the measured Rorg (t ) obtained from VSThM-org/VSThM-sub ra'os picked up at various loca'ons on the organic films with various thicknesses (Fig. 1, Figs. S1 and S2, in the ESI) and on the underlying substrate, taking Rsub=1/4rκox = 9x106 K W-1. 9 * Figure 3. EffecHve constricHon thermal resistance Rorg (t ) measured at the surface of the organic film for the C8-BTBT-C8 samples (closed symbols) and for the BTBT samples (semi-open symbols). Data and fit (lines): red = sample #1, blue = sample #2, green = sample #3, and orange = sample #4 as defined in the ESI. * For t → ∞, Rorg (t ) tends to saturate at a larger value for C8-BTBT-C8 than for BTBT although the difference is weak. This trend is in agreement with the NP-SThM determina'on that κBTBT > κC8-BTBT-C8. From the fits of Eq. 3 (solid lines in Fig. 4), we obtain a mean value (see the ESI) κBTBT = 1.37 ± 0.01 W m-1 K-1 and κC8- BTBT-C8 = 1.35 ± 0.01 W m-1 K-1 which are not significantly different. This implies that, even though the same trends are observed as with the NP-SThM method, this image analysis approach is perturbed by the interfacial thermal resistance, which more or less masks the actual values of the organic film thermal resistance. Previous works 27, 28, 49, 50, 54 also reported SThM 'p-organic materials and organic- SiO2 interfacial thermal resistances larger than 107 K W-1 (or > ∼ 10-8 m2 K W-1), i.e., larger than Rorg of our C8-BTBT-C8 and BTBT films. 10 Theory. We use the Approach to the Equilibrium Molecular Dynamics (AEMD) method42 to compute the la~ce κ. We ignored the contribu'on of the electronic κ at this stage because the thermal transport is barely controlled by electrons in most neutral and weakly doped organic semiconductors (OSCs).56 This argument is supported by the newly developed molecular Wiedemann-Franz model57 which predicts an electronic contribu'on to the heat transport smaller by several orders of magnitude than the corresponding la~ce contribu'on. In brief, the key steps of the AEMD methodology are the following: (i) applying a perfectly monitored thermal pulse on a simula'on box; (ii) recovering the ini'al thermodynamic equilibrium during a fast transient regime; (iii) fi~ng the 'me-decaying temperature difference between the right and le• parts of the system from a reliable solu'on of the one-dimensional heat equa'on ∂T = D ∂ T2 in order to 2 ∂t ∂x evaluate the thermal diffusivity, D.58 This alterna've scheme has the benefit to be far less 'me-consuming than the previously cited methods due to the rapid dissipa'on of the thermal gradient. Then κ=DρCP is finally obtained, provided the density, ρ, and the specific heat, CP, of the system are known (see details in the ESI). It is also of prime importance to account for the size-dependence of the κ values deduced from this approach since phonons having a mean free path larger than the cell dimension do not effec'vely contribute to κ. An extrapola'on procedure is thus needed to get rid of these size effects59 and to extract a quasi- length-free la~ce κ from the linear regression of 1/κ versus 1/L (L the length of the box size along the direc'on of heat propaga'on). 11 a b ! Figure 4. Inverse of the lance thermal conducHvity as a funcHon of the inverse of the simulaHon box length along direcHons a, b and c for (a) BTBT and (b) C8-BTBT- C8. Each inset represents the molecular packing of BTBT and C8-BTBT-C8 in the ab plane. For sake of clarity, we did not represent the alkyl chains in (b). The two compounds crystallize into a monoclinic structure (a = 5.854 Å, b = 7.960 Å, c = 11.809 Å, α = 105.990° for BTBT and a = 5.927 Å, b = 7.880 Å, c = 29.180 Å, β = 92.443° for C8-BTBT-C8; respecHvely); both of them exhibit a very similar layered herringbone packing,60 as shown in the insets. 12 Figure 4 display the inverse of the calculated la~ce κ as a func'on of the inverse of the super cell length. The linear extrapola'on procedure provides κ values of 1.31, 0.59 and 1.04 W m-1 K-1 along direc'ons a, b and c for BTBT and 0.33, 0.26 and 0.72 W m-1 K-1 along the same direc'ons for C8-BTBT-C8 (Table 1). There is a fairly good quan'ta've agreement between experimental measurements and theore'cal es'mates. Interes'ngly, the calcula'ons indicate that the overall calculated conduc'vity (along the 3 crystal axes) of BTBT is higher than for C8-BTBT-C8 (Table 1), in full consistency with the experimental measurements probing essen'ally heat transport along the c axis. Theory BTBT C8-BTBT-C8 a-axis 1.31 0.33 b-axis 0.59 0.26 c-axis 1.04 0.72 isotropic approxima.on 0.95 0.46 Experiment null-point SThM 0.63 ± 0.12 0.25 ± 0.13 Table 1. Calculated and measured thermal conducHvity values (W m-1 K-1) DISCUSSION. Since no theore'cal and experimental data of the out-of-plane κ was previously reported for these BTBT deriva'ves, we compare our values with other results obtained for various OSCs. The measured nanoscale out-of-plane κC8-BTBT-C8 = 0.25 ± 0.13 W m-1 K-1, is on par with values reported for other small molecule thin films by SThM: 0.15-0.20 W m-1 K-1 for methylstyryl-benzene,28 0.15-0.4 W m-1 K-1 for CuPc (Cu-phthalocyanine) and 0.15-0.25 W m-1 K-1 for PbCu.27 They are also consistent with the corresponding values obtained by characteriza'on methods 13 at the macroscale on rubrene (0.07 W m-1 K-1)23 and DNTT (0.45 W m-1 K-1).24 The experimental out-of-plane κBTBT = 0.63 ± 0.12 W m-1 K-1 lies at the highest limit of reported values: 0.45 W m-1 K-1 for DNTT24 (to the best of our knowledge and excluding the highly unusual value of 21 W m-1 K-1 reported for crystal of TIPS- pentacene61, 62). The calcula'ons of the la~ce κ were performed on single crystals while the SThM measurement was carried out on polycrystalline samples containing a certain amount of disorder. It is also known (X-ray diffrac'on) that a disordered layer of less than ca. 10 nm exists at the SiO2/BTBT interfaces.39 To cope with this "experimental" disorder we compare the measured values with an effec've isotropic κeff defined as63 κ eff = κ cκ ab where the in-plane κ is κ ab = κ aκ b . The calculated values are given in Table 1. A ra'o of 2-2.5 between the thermal conduc'vi'es of BTBT and C8-BTBT-C8 is observed at both the experimental and theore'cal level. For both materials, the agreement between the evaluated κeff and the measurements (Table 1) is excellent, suppor'ng the existence of a thin disordered layer at the SiO2/BTBT interface with a lower thermal conduc'vity (i.e. ab-plane). We do find a rather isotropic (3D) behavior of the heat propaga'on in the two deriva'ves in deep contrast with the charge transport proper'es featuring a two-dimensional character in presence of a herringbone arrangement;60, 64, 65 note that a significant thermal conduc'vity along the three crystallines axes has also been reported theore'cally for the DNTT single crystal36. The κ ra'os for a/ b, c/a and c/b are respec'vely 2.22, 0.79 and 1.76 for BTBT and 1.27, 2.18 and 2.77 for C8-BTBT-C8. Morever, we evidence a no'ceable drop in the ab-plane κ when octyl chains are added on each side of the aroma'c cores, with anisotropic factors of ∼ 3.97 and ∼ 2.27. This complements the theore'cal 14 study of Shi et al. 32 showing that κ is marginally affected by the presence of terminal saturated chains in Cn-BTBT-Cn (with n = 8;10;12). It is worth stressing that the drop in thermal conduc'vity upon addi'on of saturated chains has been also clearly observed in recent κ measurements performed on DNTT and C8- DNTT-C8 thin films in the a-b plane, and fully supported again by our AEMD calcula'ons.66 To gain a deeper insight into the underlying physical mechanisms, we explore the spa'al character of the vibra'onal modes, by es'ma'ng their par'cipa'on ra'o (PR).67 This parameter is a quan'ta've measure of the spa'al extension of vibra'ons, allowing to classify them into extended (large PR) and localized modes (PR ～ 0). It has been shown68 that extended modes are generally more effec've in transpor'ng heat across the material than localized modes. The details of the calcula'ons are given in the ESI. 0.5 BTBT 0.45 C8-BTBT-C8 0.4 0.35 0.3 PR 0.25 0.2 0.15 0.1 0.05 0 0 500 1000 1500 2000 2500 3000 -1 ν (cm ) Figure 5. EsHmated ParHcipaHon RaHo (PR) as a funcHon of the vibraHonal frequency for BTBT (blue) and C8-BTBT-C8 (red). 15 The par'cipa'on ra'o of vibra'ons in BTBT and C8-BTBT-C8 are shown in Fig. 5 as a func'on of their frequency. We observe that PR in BTBT takes larger values than C8-BTBT-C8 over the whole frequency spectrum. This shows that BTBT hosts vibra'onal modes are spa'ally more extended than those in C8-BTBT-C8. Hence, BTBT alkyla'on leads to a strong localiza'on of the vibra'onal modes and consequently to a strong reduc'on of the overall thermal conduc'vity. CONCLUSION. In summary, we demonstrate that the thermal conduc'vity of BTBT, κBTBT = 0.63 ± 0.12 W m-1 K-1 is larger than for C8-BTBT-C8, κC8-BTBT-C8 = 0.25 ± 0.13 W m-1 K-1. The nanoscale thermal resistance is weakly dependent on the film thickness, as predicted by a simple analy'cal model of the constric'on thermal resistance for a surface coated by a thin film. The experimental results are further supported by the theore'cal es'mates of the thermal conduc'vity obtained by the AEMD (Approach to the Equilibrium Molecular Dynamics) method. Moreover, we have not only demonstrated the drop of the thermal conduc'vity upon introduc'on of saturated chains but we have also provided a clear ra'onale based on the degree of delocaliza'on of the intermolecular vibra'onal modes. Finally, our calcula'ons point to significant thermal conduc'vity (3D-like) values along the 3 crystalline direc'ons, with anisotropy factors between the crystalline direc'ons below 1.8 for BTBT and below 2.8 for C8-BTBT-C8, in deep contrast with the charge transport proper'es featuring a two-dimensional character for these materials. METHODS. Synthesis and sample fabricaAon. BTBT and C8-BTBT-C8 were synthesized according to the reported procedures.69, 70 The thin films were deposited by spin- 16 coa'ng on thermal SiO2/Si-n+ according to previously reported methods.39, 43 Nine samples (5 C8-BTBT-C8, 4 BTBT) were prepared using various spin-coa'ng parameters to vary the film thickness (see the ESI). Scanning Thermal Microscope (SThM). SThM71, 72 were carried out with a Bruker ICON machine equipped with the Anasys SThM module. All measurements were done at room temperature in an air-condi'oned laboratory (Tamb = 22.5 °C, rela've humidity of 35-40 %). In the scanning mode, the topography and thermal voltage were recorded simultaneously. We used both the DC method or the 3ω- SThM.73 The thermal conduc'vity is determined by using the null-point SThM method (see the ESI).44 CalculaAons of the laDce thermal conducAvity. All MD simula'ons have been performed in the LAMMPS so•ware package74 by means of the Op'mized Poten'als for Liquid Simula'ons All-Atoms (OPLS-AA) force-field (see the ESI). In addi'on, we also es'mated the par'cipa'on ra'o to characterize the vibra'onal proper'es of both BTBT and C8-BTBT-C8. This parameter is a quan'ta've measure of the spa'al extension of vibra'ons (see the ESI). ASSOCIATED CONTENT Electronic Supplementary Informa'on (ESI) is available: Sample fabrica'on, SThM methods, theore'cal methods, addi'onal SThM images, NP-SThM calibra'on, es'ma'on of the constric'on thermal resistance of the Si/SiO2 substrate, es'ma'on of the radius of the thermal contact between 'p and surface, fiWed parameters of Eq. 3. AUTHOR CONTRIBUTIONS M.N.G. prepared the samples, carried out and analyzed the SThM measurements with the help of D.G. and S.L., respec'vely. A.V. carried out the calcula'ons of the thermal conduc'vity under the supervision of V.L. and J.C.. R.J., G.S. and Y.G. synthesized and assessed purity of the materials. A.A. and C.M. carried out the 17 calcula'ons of the vibra'onal proper'es. D.V. conducted the project, analyzed the results and wrote the paper with the help of all the authors. All authors have given approval to the final version of the manuscript. # These authors (M.N.G. and A.V.) contributed equally to this work. CONFLICTS OF INTEREST. The authors declare no compe'ng financial interest. ACKNOWLEDGEMENTS. M.G., D.G., S.L. and D.V. thank the ANR for financial support (ANR-16-CE05-0029). Y.G. is thankful to the Belgian Na'onal Fund for Scien'fic Research (FNRS) for financial support through research projects BTBT (# 2.4565.11), Phasetrans (# T. 0058.14), Pi-Fast (# T.0072.18), 2D to 3D (# 30489208), and DIFFRA (# U.G001.19). Financial supports from the French Community of Belgian (ARC # 20061) is also acknowledged. G.S. acknowledges postdoctoral fellowship support from the FNRS. The work in the Laboratory for Chemistry of Novel Materials was supported by the Consor'um des Équipements de Calcul Intensif (CÉCI), funded by the Fonds de la Recherche Scien'fique de Belgique (F.R.S.-FNRS) under Grant # 2.5020.11. J.C. and A.V. are FNRS research fellows. REFERENCES. 1. H. J. Goldsmid, in Semiconductors and Semimetals, ed. T. M. TriW, Elsevier, 2001, vol. 69, pp. 1-24. 2. O. Bubnova, Z. U. Khan, A. Mal', S. Braun, M. Fahlman, M. Berggren and X. Crispin, Nat. Mater., 2011, 10, 429-433. 3. M. N. Gueye, A. Carella, N. Massonnet, E. Yvenou, S. Brenet, J. Faure- Vincent, S. Pouget, F. Rieutord, H. Okuno, A. Benayad, R. Demadrille and J.-P. Simonato, Chem. Mater, 2016, 28, 3462-3468. 18 4. B. Russ, A. Glaudell, J. J. Urban, M. L. Chabinyc and R. A. Segalman, Nature Reviews Materials, 2016, 1, 16050. 5. M. N. Gueye, A. Carella, J. Faure-Vincent, R. Demadrille and J.-P. Simonato, Progress in Materials Science, 2020, 108, 100616. 6. M. Paulsson and S. DaWa, Phys. Rev. B, 2003, 67, 241403(R). 7. P. Reddy, S.-Y. Jang, R. A. Segalman and A. Majumdar, Science, 2007, 315, 1568-1571. 8. J. A. Malen, P. Doak, K. Bahe', T. D. Tilley, R. A. Segalman and A. Majumdar, Nano Le$, 2009, 9, 1164-1169. 9. C. Evangeli, K. Gillemot, E. Leary, M. T. González, G. Rubio-Bollinger, C. J. Lambert and N. Agraït, Nano Le$, 2013, 13, 2141-2145. 10. T. Meier, F. Menges, P. Nirmalraj, H. Hölscher, H. Riel and B. Gotsmann, Phys. Rev. Le$., 2014, 113, 060801. 11. L. Rincon-Garcia, C. Evangeli, G. Rubio-Bollinger and N. Agraït, Chem. Soc. Rev., 2016, 45, 4285-4306. 12. L. Cui, S. Hur, Z. A. Akbar, J. C. Klöckner, W. Jeong, F. Pauly, S.-Y. Jang, P. Reddy and E. Meyhofer, Nature, 2019, 572, 628-633. 13. N. Mosso, H. Sadeghi, A. Gemma, S. Sangtarash, U. Drechsler, C. Lambert and B. Gotsmann, Nano Le$, 2019, 19, 7614-7622. 14. C. J. Lambert, Chem. Soc. Rev., 2015, 44, 875-888. 15. J. C. Klöckner, J. C. Cuevas and F. Pauly, Phys. Rev. B, 2017, 96, 245419. 16. R. Miao, H. Xu, M. Skripnik, L. Cui, K. Wang, K. G. L. Pedersen, M. Leijnse, F. Pauly, K. Wärnmark, E. Meyhofer, P. Reddy and H. Linke, Nano Le$, 2018, 18, 5666-5672. 17. P. Gehring, J. M. Thijssen and H. S. J. Van Der Zant, Nature Reviews Physics, 2019, 1, 381-396. 18. R. C. Yu, N. Tea, M. B. Salamon, D. Lorents and R. Malhotra, Phys. Rev. Le$., 1992, 68, 2050-2053. 19 19. N. Kim, B. Domercq, S. Yoo, A. Christensen, B. Kippelen and S. Graham, Appl. Phys. Le$., 2005, 87, 241908. 20. Y. Okada, M. Uno, Y. Nakazawa, K. Sasai, K. Matsukawa, M. Yoshimura, Y. Kitaoka, Y. Mori and J. Takeya, Phys. Rev. B, 2011, 83, 113305. 21. J. C. Duda, P. E. Hopkins, Y. Shen and M. C. Gupta, Phys. Rev. Le$., 2013, 110, 015902. 22. X. Wang, C. D. Liman, N. D. Treat, M. L. Chabinyc and D. G. Cahill, Phys. Rev. B, 2013, 88, 075310. 23. H. Zhang and J. W. Brill, J. Appl. Phys., 2013, 114, 043508. 24. X. Wang, K. D. Parrish, J. A. Malen and P. K. L. Chan, ScienHfic Reports, 2015, 5, 16095. 25. Y. Yao, M. Shahi, M. M. Payne, J. E. Anthony and J. W. Brill, J. Mater. Chem. C, 2016, 4, 8817-8821. 26. T. Nomoto, S. Imajo, S. Yamashita, H. Akutsu, Y. Nakazawa and A. I. Krivchikov, Journal of Thermal Analysis and Calorimetry, 2018, 135, 2831-2836. 27. D. Trefon-Radziejewska, J. Juszczyk, A. Fleming, N. Horny, J. S. Antoniow, M. Chirtoc, A. Kaźmierczak-Bałata and J. Bodzenta, SyntheHc Metals, 2017, 232, 72-78. 28. Y. Zhang, C. Zhang, D. Wei, X. Bai and X. Xu, CrystEngComm, 2019, 21, 5402-5409. 29. G. Fugallo and L. Colombo, Physica Scripta, 2018, 93, 043002. 30. D. Wang, L. Tang, M. Long and Z. Shuai, J. Phys. Chem. C, 2011, 115, 5940-5946. 31. J. Chen, D. Wang and Z. Shuai, Journal of Chemical Theory and Computa'on, 2012, 8, 3338-3347. 32. W. Shi, J. Chen, J. Xi, D. Wang and Z. Shuai, Chem. Mater, 2014, 26, 2669-2677. 33. L. Chen, X. Wang and S. Kumar, ScienHfic Reports, 2015, 5, 12763. 34. J. Y. Kim and J. C. Grossman, Nano Le$, 2016, 16, 4203-4209. 20 35. A. Giri and P. E. Hopkins, J. Phys. Chem. Le$., 2017, 8, 2153-2157. 36. X. Wang, J. Zhang, Y. Chen and P. K. L. Chan, Nanoscale, 2017, 9, 2262-2271. 37. H. Kojima, M. Nakagawa, R. Abe, F. Fujiwara, Y. Yakiyama, H. Sakurai and M. Nakamura, Chemistry Le$ers, 2018, 47, 524-527. 38. R. Sasaki, Y. Takahashi, Y. Hayashi and S. Kawauchi, J. Phys. Chem. B, 2020, 124, 881-889. 39. G. Gbabode, M. Dohr, C. Niebel, J.-Y. Balandier, C. Ruzié, P. Négrier, D. Mondieig, Y. H. Geerts, R. Resel and M. Sferrazza, ACS Appl. Mater. Interfaces, 2014, 6, 13413-13421. 40. Y. Tsutsui, G. Schweicher, B. ChaWopadhyay, T. Sakurai, J.-B. Arlin, C. Ruzié, A. Aliev, A. Ciesielski, S. Colella, A. R. Kennedy, V. Lemaur, Y. Olivier, R. Hadji, L. Sanguinet, F. Castet, S. Osella, D. Dudenko, D. Beljonne, J. Cornil, P. Samorì, S. Seki and Y. H. Geerts, Adv. Mater., 2016, 28, 7106-7114. 41. G. Schweicher, V. Lemaur, C. Niebel, C. Ruzié, Y. Diao, O. Goto, W.-Y. Lee, Y. Kim, J.-B. Arlin, J. Karpinska, A. R. Kennedy, S. R. Parkin, Y. Olivier, S. C. B. Mannsfeld, J. Cornil, Y. H. Geerts and Z. Bao, Adv. Mater., 2015, 27, 3066-3072. 42. C. Melis, R. DeWori, S. Vandermeulen and L. Colombo, The European Physical Journal B, 2014, 87, 96. 43. M. Dohr, H. M. A. Ehmann, A. O. F. Jones, I. Salzmann, Q. Shen, C. Teichert, C. R. xe, G. Schweicher, Y. H. Geerts, R. Resel, M. Sferrazza and O. Werzer, Sov Ma$er, 2017, 13, 2322-2329. 44. K. Kim, J. Chung, G. Hwang, O. Kwon and J. S. Lee, ACS Nano, 2011, 5, 8700-8709. 45. H. S. Carslaw and J. C. Jaeger, ConducHon of Heat in Solids, Oxford University Press, 1959. 46. P. L. Kapitza, J. Phys. (Moscow), 1941, 4, 181. 47. S. M. Lee and D. G. Cahill, J. Appl. Phys., 1997, 81, 2590-2595. 48. H.-K. Lyeo and D. G. Cahill, Physical Review B, 2006, 73, 234. 21 49. Z. Chen, W. Jang, W. Bao, C. N. Lau and C. Dames, Appl. Phys. Le$., 2009, 95, 161910. 50. M. D. Losego, M. E. Grady, N. R. SoWos, D. G. Cahill and P. V. Braun, Nature Materials, 2012, 11, 502-506. 51. F. Menges, H. Riel, A. Stemmer, C. Dimitrakopoulos and B. Gotsmann, Phys. Rev. Le$., 2013, 111, 205901. 52. R. Cheaito, J. T. Gaskins, M. E. Caplan, B. F. Donovan, B. M. Foley, A. Giri, J. C. Duda, C. J. Szwejkowski, C. Constan'n, H. J. Brown-Shaklee, J. F. Ihlefeld and P. E. Hopkins, Physical Review B, 2015, 91, 035432. 53. Z. Ding, J.-W. Jiang, Q.-X. Pei and Y.-W. Zhang, Nanotechnology, 2015, 26, 065703. 54. C. Evangeli, J. Spiece, S. Sangtarash, A. J. Molina Mendoza, M. Mucientes, T. Mueller, C. Lambert, H. Sadeghi and O. Kolosov, Adv. Electron. Mater., 2019, 7, 1900331. 55. J. R. Dryden, J. Heat Transfer, 1983, 105, 408-410. 56. D. Wang, W. Shi, J. Chen, J. Xi and Z. Shuai, Physical Chemistry Chemical Physics, 2012, 14, 16505-16520. 57. G. T. Craven and A. Nitzan, Nano Le$, 2019, 20, 989-993. 58. Y. He, I. Savić, D. Donadio and G. Galli, Phys. Chem. Chem. Phys., 2012, 14, 16209-16222. 59. D. P. Sellan, E. S. Landry, J. E. Turney, A. J. H. McGaughey and C. H. Amon, Phys. Rev. B, 2010, 81, 214305. 60. G. Schweicher, G. D'Avino, M. T. Ruggiero, D. J. Harkin, K. Broch, D. Venkateshvaran, G. Liu, A. Richard, C. Ruzié, J. Armstrong, A. R. Kennedy, K. Shankland, K. Takimiya, Y. H. Geerts, J. A. Zeitler, S. Fra'ni and H. Sirringhaus, Adv. Mater., 2019, 31, 1902407. 61. H. Zhang, Y. Yao, M. M. Payne, J. E. Anthony and J. W. Brill, Appl. Phys. Le$., 2014, 105, 073302. 22 62. J. W. Brill, M. Shahi, M. M. Payne, J. Edberg, Y. Yao, X. Crispin and J. E. Anthony, J. Appl. Phys., 2015, 118, 235501. 63. Y. S. Muzychka, M. M. Yovanovich and J. R. Culham, Journal of Thermophysics and Heat Transfer, 2004, 18, 45-51. 64. S. Haas, Y. Takahashi, K. Takimiya and T. Hasegawa, Appl. Phys. Le$., 2009, 95, 022111. 65. W. Xie, K. Willa, Y. Wu, R. Häusermann, K. Takimiya, B. Batlogg and C. D. Frisbie, Adv. Mater., 2013, 25, 3478-3484. 66. E. Selezneva, A. Vercouter and G. Schweicher, et al., to be submi$ed. 67. R. J. Bell, in Methods in Computa'onal Physics: Advances in Research and Applica'ons, ed. G. Gilat, Elsevier, 1976, vol. 15, pp. 215-276. 68. A. Cappai, A. An'dormi, A. Bosin, D. Narducci, L. Colombo and C. Melis, Physical Review Materials, 2020, 4, 035401. 69. M. Saito, I. Osaka, E. Miyazaki, K. Takimiya, H. Kuwabara and M. Ikeda, Tetrahedron Le$ers, 2011, 52, 285-288. 70. C. Grigoriadis, C. Niebel, C. Ruzié, Y. H. Geerts and G. Floudas, J. Phys. Chem. B, 2014, 118, 1443-1451. 71. A. Majumdar, Annu. Rev. Mater. Sci., 1999, 29, 505-585. 72. S. Gomès, A. Assy and P.-O. Chapuis, phys. stat. sol. (a), 2015, 212, 477-494. 73. S. Lefèvre and S. Volz, Rev. Sci. Instrum., 2005, 76, 033701. 74. S. Plimpton, Journal of Computa'onal Physics, 1995, 117, 1-19. 23