Article www.acsnano.org Defect Engineering in Solution-Processed Polycrystalline SnSe Leads to High Thermoelectric Performance Yu Liu, Mariano Calcabrini, Yuan Yu, Seungho Lee, Cheng Chang, Jérémy David, Tanmoy Ghosh, Maria Chiara Spadaro, Chenyang Xie, Oana Cojocaru-Mirédin, Jordi Arbiol, and Maria Ibáñez* Cite This: ACS Nano 2022, 16, 78−88 Read Online ACCESS Metrics & More Article Recommendations * sı Supporting Information ABSTRACT: SnSe has emerged as one of the most promising materials for thermoelectric energy conversion due to its extraordinary performance in its single-crystal form and its low- cost constituent elements. However, to achieve an economic impact, the polycrystalline counterpart needs to replicate the performance of the single crystal. Herein, we optimize the thermoelectric performance of polycrystalline SnSe produced by consolidating solution-processed and surface-engineered SnSe particles. In particular, the SnSe particles are coated with CdSe molecular complexes that crystallize during the sintering process, forming CdSe nanoparticles. The presence of CdSe nanoparticles inhibits SnSe grain growth during the consolidation step due to Zener pinning, yielding a material with a high density of grain boundaries. Moreover, the resulting SnSe−CdSe nanocomposites present a large number of defects at different length scales, which significantly reduce the thermal conductivity. The produced SnSe−CdSe nanocomposites exhibit thermoelectric figures of merit up to 2.2 at 786 K, which is among the highest reported for solution-processed SnSe. KEYWORDS: tin selenide, nanocomposite, grain growth, Zener pinning, thermoelectricity, annealing, solution processing INTRODUCTION Different approaches to overcome these limitations include Materials able to reversibly convert heat into electricity, i.e., chemical reduction of oxide species,16,17 liquid-phase compac- thermoelectric materials, require high electrical conductivity tion18 and hot deformation processes to promote texture,19 (σ), high Seebeck coefficient (S), and low thermal conductivity and doping control with different alkali (K, Na, and Li)20−22 (κ). The optimization of these three strongly interrelated and transition (Ag, Cu, Zn, and Cd) metals.23−29 Additionally, properties involves tuning the electronic structure of the approaches have been scrutinized to further enhance the material and the charge and phonon scattering mechanisms.1−4 polycrystalline materials’ performance, such as alloying with Zhao et al. discovered in 2014 that SnSe has an outstanding SnS,30 Pb,31 and Ge32 and the introduction of different thermoelectric performance, originating an outburst of nanofeatures such as nanoporous or nanoprecipitates (i.e., research on the material.5−12 The highest figure of merit (zT InSey,33 AgSnSe2,34 PbSe,35 and Ag8SnSe636). = σS2Tκ−1) obtained to date in p-type SnSe is ∼2.6 at 923 K Herein we report a simple and scalable synthesis route to along the b-axis in pristine SnSe single crystals13 and ∼2.8 at produce SnSe−CdSe nanocomposites based on the aqueous 773 K along the a-axis in Br-doped n-type SnSe single synthesis of SnSe particles and their surface treatment with crystals.14 However, the high cost and stagnant production of CdSe molecular complexes. Such surface treatment allows single crystals, together with their poor mechanical properties, engineering of the material microstructure by promoting defect limit the large-scale use of SnSe in thermoelectric devices.15 A potential solution is shifting to polycrystalline SnSe-based Received: August 5, 2021 materials. The problem is that polycrystalline SnSe suffers from Accepted: September 17, 2021 lower thermoelectric performance due to oxidation leading to Published: September 22, 2021 higher thermal conductivities, partial loss of anisotropy diminishing electrical conductivity, and imprecise control of the doping level.16,17 © 2021 The Authors. Published by American Chemical Society https://doi.org/10.1021/acsnano.1c06720 78 ACS Nano 2022, 16, 78−88 ACS Nano www.acsnano.org Article Figure 1. Scheme of the fabrication process of SnSe−CdSe nanocomposites. (1−3) Aqueous synthesis and purification of SnSe particles; (4) preparation of CdSe molecular solution (CdSe ink); (5) blending the CdSe ink with SnSe particles; (6) purification of the CdSe surface- treated SnSe particles and SnSe−CdSe powder vacuum drying; (7) annealing; (8) spark plasma sintering (SPS) for producing cylinders (⌀ = 8.6 mm × h = 12 mm); and (9) slicing for transport measurements. formation at all length scales. In particular, during the of chalcogenidocadmates such as (Cd 2 Se 3 ) n 2n − or processing, CdSe complexes crystallize, forming CdSe nano- CdSe22−.39−42 The adsorption of CdSe species on the SnSe particles (NPs) at the surface of SnSe particles. CdSe NPs surface was verified by tracking the color change of the hinders grain growth during consolidation yielding a material solution. The vivid orange color of the CdSe solution changed with a high density of multiscale defects (point defects, to a slightly yellow color after SnSe particles were introduced dislocations, planar defects, and nanostructures). The presence in the solution (Figure S1). Finally, CdSe surface-treated SnSe of CdSe NPs together with the significantly higher defect particles were precipitated from solution, washed twice with content results in a reduction of the thermal conductivity by 2- acetone, and dried under vacuum for further processing into fold with respect to bare SnSe produced by untreated SnSe cylindrical pellets. The complete material fabrication process is particles. Overall, the strategy presented here produces illustrated in Figure 1. inexpensive and highly stable polycrystalline SnSe with a zT Figure 2a shows scanning electron microscopy (SEM) of ca. 2.2 at 786 K. images of the pellets obtained from SnSe-x%CdSe particles. In the presence of CdSe, the sintered materials present smaller RESULTS AND DISCUSSION crystal domains than bare SnSe despite having all similar SnSe particles were prepared in water, using Se powder and tin densities (Table S1). Lower CdSe content than the estimated chloride hydrate as precursors.37 The obtained SnSe particles to coat the whole SnSe particles (1 mol %) resulted in larger were purified to remove unbound ionic impurities by washing grain sizes, but still smaller than that without any surface them with water and ethanol. To produce dense SnSe treatment. Above 2 mol %, the final grain size of all the pellets polycrystalline materials, the purified particles were then analyzed is practically the same. dried under vacuum, annealed in forming gas (95% N2 + 5% X-ray diffraction (XRD) data showed no evidence of the H2)16,17 at 500 °C, and consolidated into cylindrical pellets presence of secondary phases in any nanocomposite (Figures through spark plasma sintering (experimental details can be 2b and S2). The grain size of the nanocomposites is stable and found in the “Methods” section). In the case of SnSe−CdSe hardly changes after the consolidation step, as observed in the nanocomposites, SnSe particles were mixed with CdSe structural analysis of the pellets subjected to further thermal molecular complexes (x mol %; x = 1, 2, 3, and 4) in N- treatment such as heating and cooling cycles from room methyl formamide for 48 h before annealing. The CdSe temperature to 823 K carried out during the transport molecular solution was prepared by dissolving stoichiometric measurements (Figure S3). These results evidence the high amounts of CdO and Se powder in a thiol−amine mixture stability of the grain size. (1,2-ethaneditiol, en; ethylenediamine, EDT) at room temper- To understand the mechanism that inhibits grain growth in ature in an inert atmosphere (Figure 1).38 On the basis of the presence of CdSe, we analyzed the bare SnSe and SnSe-3% previous studies of hydrazinium-based CdSe solutions, we CdSe samples after each processing step (surface treatment, hypothesize that the molecular solute is composed of a variety annealing, and consolidation, Figure 3). SEM images indicate 79 https://doi.org/10.1021/acsnano.1c06720 ACS Nano 2022, 16, 78−88 ACS Nano www.acsnano.org Article that the particle morphology is merely affected by the CdSe surface treatment. Differences appear between the CdSe treated and the untreated samples upon annealing. Despite the fact that both materials show grain growth, the average grain size of bare SnSe increases from 150 ± 50 nm to 680 ± 400 nm, which is 2.5 times larger than that in the presence of CdSe (Figures S4 and S5). The difference in grain growth is more evident after the pressure-assisted sintering step through spark plasma sintering (SPS, 45 MPa, 500 °C). In the presence of CdSe, grain growth is hindered, resulting in relatively smaller crystal domains (Figures 3a and S6) and thus in a higher volume fraction of grain boundaries for the SnSe−CdSe nanocomposites compared to bare SnSe. SnSe samples have an average grain size of ca. 3.5 ± 3.0 μm while SnSe-3%CdSe nanocomposites display grain sizes of ca. 0.3 ± 0.1 μm as shown by electron backscattering diffraction (EBSD) inverse pole figure maps (Figures 3b, S5, and S7). The sintering process of surface-treated SnSe particles involves the crystallization of CdSe complexes into CdSe NPs.38 Control experiments revealed that the complete Figure 2. (a) Representative SEM images of the pellets obtained from SnSe-x%CdSe (x = 0, 1, 2, 3, and 4) particles; (b) the crystallization of CdSe complexes occurs at 350 °C, yielding corresponding XRD patterns in the direction perpendicular to the ca. 4 nm CdSe NPs (determined by XRD data, Figure S8). At pressing axis including the SnSe reference pattern (Powder this same temperature, untreated SnSe particles hardly show Diffraction File (PDF) no. 00−048−1224, Joint Committee on any difference in grain size compared to the as-synthesized Powder Diffraction Standards (JCPDS)). SnSe particles (Figure S9). Considering that the CdSe crystallization temperature is unaffected by the presence of SnSe particles, we conclude that at 350 °C, there is a homogeneous distribution of CdSe NPs at the surface of SnSe particles (Figure S10).43 On the basis of the CdSe and SnSe average particle size, CdSe NPs cover approximately 68% of the total surface of SnSe for the SnSe-3%CdSe. Therefore, the corresponding powder densification and grain growth are strongly affected by the changes in atomic diffusion due to the presence of secondary phase CdSe NPs (Figure 3). In the classical theory of grain growth, the average grain growth rate is proportional to the average rate of grain boundary movement.44 However, when second-phase particles are present at the grain boundaries, the driving force for their migrations is reduced, hindering grain growth.45 This is known as the Zener pinning effect.46 Zener pinning occurs when a grain boundary encounters a second-phase particle, as the particle exerts a drag force on the grain boundary.47 In such a situation, the growing grain is subject to two opposing pressures: the driving pressure for growth and the Zener pinning pressure arising from the particles. For a grain to grow further, the net driving pressure should be positive. By adjusting the volume fraction of second-phase particles and their mean radius is possible to kinetically control the final grain size.48 The effect of CdSe NPs in the microstructure of SnSe−CdSe nanocomposites is in accordance with grain growth stagnation predicted by the Zener pinning effect. In Figure 2, one can see that as we increase the content of CdSe molecular solute, the average grain size is reduced. Moreover, considering that bare SnSe and all SnSe−CdSe pellets have the same density despite their difference in grain size, we believe that the presence of Figure 3. (a) SEM images of SnSe particles, the corresponding annealed powder, and the consolidated pellet without (top, black) CdSe NPs favors atomic diffusion along the grain boundaries, and with (bottom, blue) the CdSe-surface treatment. (b) EBSD i.e., the densification rate. In contrast, atomic diffusion across inverse pole figure maps for bare SnSe (left) and SnSe-3%CdSe the grain boundaries is strongly hampered, limiting grain (right) pellets; inset images correspond to the particle size growth. The result is a material with the same density but histograms. much smaller grain domains than bare SnSe.44 One key point to achieving grain growth inhibition through such a surface treatment is the proper selection of the metal 80 https://doi.org/10.1021/acsnano.1c06720 ACS Nano 2022, 16, 78−88 ACS Nano www.acsnano.org Article Figure 4. Representative SEM images of (a) PbS-surface-treated SnSe particles, (b) the corresponding annealed powder, and (c) consolidated pellet. (d) XRD diffraction patterns of recrystallized PbS molecular complexes (blue), the surface-modified particles (black), and consolidated pellet in the perpendicular direction to the pressing axis (green). The inset picture corresponds to the PbS molecular solution. The pattern in gray corresponds to the SnSe reference pattern PDF no. 00−048−1224 (JCDPS) and in dark blue to the PbS reference pattern PDF no. 03−065−0241 (JCDPS). (e) Magnification of the XRD pattern for the consolidated pellets with (green) and without (black) PbS surface treatment showing the peak shift due to partial alloying of PbS. chalcogenide complex. During the thermal processing, the Electronic Transport Properties. The electrical con- material selected should form NPs at the SnSe particle surface ductivity (σ), Seebeck coefficient (S), thermal conductivity instead of diffusing into its crystal structure creating a solid (κ), and calculated figure of merit (zT) of bare SnSe and solution. To satisfy this condition is necessary to choose a SnSe−CdSe nanocomposites with different content of CdSe material that possesses a positive enthalpy of mixing with a were measured in the direction parallel (Figure S17) and miscibility gap over the processing temperature range. In this perpendicular (Figure S18) to the pressure axis. Because case, the enthalpy is a driving force for segregation that materials’ zTs in the parallel direction are larger than those in prevents the formation of a solid solution.44 As observed in the normal one, we discuss here transport properties in the their phase diagram,49 CdSe and SnSe are immiscible in the parallel direction. Also, for clarity, the discussion focuses on whole range of processing temperatures (Figure S13). XRD comparing bare SnSe and the best performing SnSe−CdSe data corroborates this fact, as there are no changes in the nanocomposite, which corresponds to 3 mol %, referred to as lattice parameters between bare SnSe and SnSe−CdSe SnSe-3%CdSe (Figure 5). The data corresponding to other nanocomposites (Figures S14 and S15). SnSe-x%CdSe samples (x = 1, 2, and 4) can be found in To prove this idea, we chose a different molecular complex Figures S17 and S18. with no miscibility gap (PbS) to treat SnSe particles. PbS is Bare SnSe shows a higher electrical conductivity and a lower known to form a stable solid solution with SnSe up to positive Seebeck coefficient than those of SnSe-3%CdSe in the concentrations of 20%.50 When comparing the two composite whole temperature range (Figures 5a and b). The Hall carrier materials at different processing stages, we observe that in the concentration (pH) of SnSe-3%CdSe is lower (ca. pH = 9 × presence of PbS the grain coarsening is enhanced significantly 1018 cm−3) than the one obtained for bare SnSe (ca. pH = 2 × already in the annealing step and yields pellets with even larger 1019 cm−3). CdSe is generally an n-type semiconductor,51 and grains than bare SnSe (Figure 4). We associate this with the it has a much lower electron affinity than that of p-type SnSe. fact that atomic diffusion across the grain boundaries is Thus, within the SnSe matrix, CdSe injects free electrons to promoted during the solid-solution formation, enhancing grain SnSe, inducing band bending in the regions close to CdSe growth. XRD analysis of the SnSe-3%PbS pellet corroborates domains. Owing to the small size of the CdSe crystal domains, the solid-solution formation as the lattice parameter changes, its Fermi level becomes pinned near the valence band edge, from a = 11.494 Å to a = 11.515 Å (Figures 4e and S16). thus trapping holes from the SnSe matrix. Thus, the hole 81 https://doi.org/10.1021/acsnano.1c06720 ACS Nano 2022, 16, 78−88 ACS Nano www.acsnano.org Article carrier concentration is reduced in the presence of CdSe, which directly translates into an increase of the Seebeck coefficient. Moreover, the presence of larger energy barriers at the interface between CdSe and SnSe allows for even higher Seebeck coefficients due to the filtering of high energy carriers (Figure S19).52 The temperature dependence of the electrical conductivity and the Seebeck coefficient of both materials are similar to those previously reported polycrystalline SnSe.20,31,32,53 To understand the underlying transport mechanism, we compare the hall mobilities with the calculated weighted mobilities (μW, Figure 5c) according to the equation54 ÄÅ ÄÅ ÉÑ ÉÑ ÅÅ ÑÑ ÅÅ expÅÅÅÅ e |S| − 2ÑÑÑÑ ÑÑ Å ÅÅ Å ÇÅ B ÑÑÖ ÑÑ 3/2 Å Å ÄÅ ÉÑ + ÄÅ ÉÑ ÑÑÑÑ 3 e |S| Å Å 8πe(2mekBT ) ÅÅ 1 + expÅÅ Å Ñ Å 5e |S| ÑÑ ÑÑ 5ÑÑÑÑ 1 + expÅÅÅÅ k 3 − 5ÑÑÑ ÑÑ 3h σ k π 2 kB ÅÅÅÇ ÅÅ kB + μW = ÅÇ ÑÖ ÅÇ B ÑÖ ÑÑÖ −5 e |S | where h, kB, e, and me, are the Planck’s constant, Boltzmann’s constant, electron charge, and electrons mass. Both mobilities indicate the same behavior. Bare SnSe and SnSe-3%CdSe nanocomposite show ther- mally activated conductivity from room temperature up to ca. 500−600 K. In this temperature range, the increase of mobility with temperature reflects the presence of potential barriers due to charge accumulation at the grain boundaries.55,56 The energy barrier height (Eb; μ ∝ T−1/2 exp(−Eb/kBT)) for SnSe− CdSe nanocomposites is 190 meV, while for SnSe, it is 113 meV (Figure 5d). The larger energy barrier of the SnSe-3% CdSe material results in lower mobility at room temper- ature.56,57 At high temperatures, the difference in mobility between both materials is practically negligible. With Figure 5. (a) Electrical conductivity, σ. (b) Seebeck coefficient, S. (c) Hall and weighted mobilities, μH and μw, respectively. (d) increasing temperature, the thermally excited carriers reduce Fitting for the energy barrier, Eb, (μ ∝ T−1/2 exp(−Eb/kBT)). (e) the effect of the potential barriers on mobility, and the Thermal conductivity, κ. (f) Figure of merit, zT, for bare SnSe dominant scattering mechanism in both materials is acoustic nanomaterial (black) and SnSe-3%CdSe nanocomposite (blue). phonon scattering. The Seebeck coefficient in both materials peaks at ca. 650 K, indicating the onset for bipolar conduction. As the material transitions from the Pnma to the Cmcm phase, the changes in Figure 6. (a) APT characterization showing the 3D distribution of Na and (b) TEM images of different grain boundaries in bare SnSe. (c) APT characterization showing the 3D distribution of Na and (d) TEM images of SnSe-3%CdSe nanocomposites. The correspondence between defects in APT and TEM is highlighted with red, blue, and orange dashed boxes. 82 https://doi.org/10.1021/acsnano.1c06720 ACS Nano 2022, 16, 78−88 ACS Nano www.acsnano.org Article Figure 7. (a) SEM image of the tip (top) used for the APT elemental analysis (bottom) showing the enhanced presence of Na and Cd at the grain boundary (GB). (b) CdSe NP with Fm3m cubic phase with the corresponding indexed power spectrum from the region marked in yellow and the phase filtered images highlighting the CdSe NP (red) and the SnSe matrix (green). (c) 3D distribution of elements from APT analysis in a region containing a grain boundary and a CdSe NP; (d) composition profile across a CdSe NP. the local bonding translate into differences in the electronic surface of 2.0 at% Na in green, allowing identifying the 3D structure, including a reduction of the bandgap from 0.61 to distribution of Na in the pellets. In both materials, bare SnSe 0.39 eV that favors the thermal excitation of minority and SnSe-3%CdSe, Na is found within the grains, at carriers.13,31 Above 800 K, the material is fully converted to dislocations and grain boundaries, and in Na-rich precipitates Cmcm (Figures S14 and S15) and both the Seebeck coefficient (Figures 6a,c and 7). Elemental analysis by inductively coupled and electrical conductivity stabilize (Figure S20). plasma optical emission spectroscopy (ICP-OES) and energy- Thermal Transport Properties. The thermal conductiv- dispersive X-ray spectrometry (EDS) indicate that both pellets ities of both nanomaterials follow the same trend in the whole contain the same amount of Na (Table S2), yet the defect temperature range. From room temperature up to ca. 790 K, concentration of defects associate with Na differ in their the values decrease monotonically. At higher temperatures, due concentration. The comparison between low-magnification to the SnSe phase transition to the higher symmetry Cmcm TEM images (Figure 6b,d) and APT data clearly reveals the phase, the thermal conductivity increases (Figure S20). The different microstructure of both materials. SnSe-3%CdSe temperature at which the Pnma fully converts into the Cmcm nanocomposites presented a larger density of grain boundaries, phase was analyzed by differential scanning calorimeter (Figure dislocations, planar defects, and Na-rich precipitates (Figures S21b) and temperature-dependent XRD measurements 6d and S24). (Figure S15). Both analyses indicate a complete phase Figure 7 shows the analysis carried out with APT and transition at 800 K. HRTEM for the sample SnSe-3%CdSe. Both techniques In the whole temperature range, the κ values of SnSe-3% revealed the presence of CdSe NPs in the range between 1 CdSe nanocomposites are almost 50% lower than those of bare and 20 nm (Figures 7a−c and S24). In particular, thanks to SnSe (Figure S22). At the temperature where the thermal power spectrum analysis, we identify that CdSe NP possesses a diffusivity is minimum (786 K), the Cp value measured in the Fm3m cubic structure (Figure 7b). Finally, APT uncovered the SnSe-3%CdSe nanocomposite is 0.263 J g−1 K−1 (Figure S21b) presence of Cd at the grain boundaries (Figures 7a,d) together leading to a thermal conductivity of 0.20 W m−1 K−1 (κlattice = with the segregation of Na and depletion of Sn. Due to the 0.14 W m−1 K−1). This value is similar to the lowest reported variation of chemical composition and likely the structure at values for polycrystalline SnSe (Figure S23b).16,25,32 the grain boundary, the most accurate way to define the To comprehend the origin of such low values, bare SnSe and interfacial material at the grain boundary would be grain SnSe−CdSe nanocomposites were further investigated by boundary complexions,60 which can generally scatter phonons transmission electron microscopy (TEM) and atom probe more strongly than a pure grain boundary.61 Such grain tomography (APT, Figures 6).58,59 boundary complexions are also associated with the enhanced Na is an unavoidable impurity, yet beneficial for doping, due energy barrier for holes that reduces carrier mobility and to the nature of the aqueous-based synthesis and the use of Na enhances the Seebeck coefficient through energy filtering salts in the reaction. Figures 6a,c show the isocomposition effects seen in the charge transport analysis. 83 https://doi.org/10.1021/acsnano.1c06720 ACS Nano 2022, 16, 78−88 ACS Nano www.acsnano.org Article The dominant phonon modes in SnSe have mean free paths parent supernatant solution was carefully discarded. The remaining of a similar length to the defects found in SnSe-3%CdSe crude solution (ca. 120 mL) was purified by 3 precipitation/ nanocomposite, such as atomic impurities Na and Cd, CdSe redispersion cycles with deionized water and ethanol alternatively. In NPs and Na-rich precipitates ranging from 1 to 20 nm, and the first cycle, 120 mL of deionized water were added into the crude solution, and the particles were separated by centrifugation (6000 grain boundary complexions.62 This explains the significant rpm, 5000 g, 1 min). Then, the particles were redispersed in 105 mL reduction in the overall thermal conductivity with respect to of ethanol and centrifuged (8000 rpm, 5 min). In the second cycle, bare SnSe, where the density of such defects is significantly 120 mL of deionized water was added to solubilize the remaining lower. Moreover, the surface treatment with CdSe molecular impurities and disperse the particles, which were precipitated (9000 solution may reduce the overall content of oxide species in the rpm, 5 min). Afterward, 105 mL of ethanol were employed to final composite, therefore, having also a positive effect on the redisperse and precipitate the particles (8000 rpm, 5 min). These thermal conductivity.16,17 Such reduction of the thermal same steps were repeated for a third purification cycle. Washed conductivity leads to a state-of-the-art figure of merit in particles were dried under vacuum overnight at room temperature and solution-processed SnSe of 2.2 at 786 K, comparable to the kept in the glovebox for further use. CdSe Molecular Complexes Preparation. The CdSe molecular best performing polycrystalline SnSe (Figure S23a). Addition- complexes (87 mg/mL) were obtained by mixing stoichiometric ally, the results obtained from SnSe-3%CdSe pellets exhibit amounts of CdO (4 mmol) and Se powder (4 mmol) with en (8 mL) excellent stability (Figure S26) and repeatability from sample and EDT (0.8 mL) in a N2-filled vial following a modified approach to to sample (Figure S27). the developed by Brutchey et al.38 The mixture was agitated for ∼5 min at room temperature until complete dissolution. All the CdSe CONCLUSIONS molecular complexes solutions were prepared fresh before blending with SnSe particles in MFA, due to their limited stability. Similarly, In summary, we presented a scalable, simple, and economical PbS molecular complexes (109 mg/mL) were obtained by mixing 1 method to produce high-performance polycrystalline SnSe mmol of PbO with 2 mL of en and 0.2 mL of EDT in a N2-filled vial. thermoelectric materials. Specifically, we optimized a water- Particle Surface Treatment. All surface treatments were carried based synthesis protocol to obtain large quantities of SnSe out in an inert atmosphere (N2). Different amounts (1%: 0.44 mL, particles and developed a surface treatment to (i) inhibit grain 2%: 0.88 mL, 3%: 1.32 mL, and 4%: 1.76 mL) of CdSe molecular growth during consolidation and operation, (ii) introduce complexes solution were mixed with different amounts of MFA (ca. CdSe NPs within the range 1−20 nm, (iii) create grain 50 μL/mL). We denoted these mixtures as CdSe-MFA solutions. Then, these CdSe-MFA solutions were combined with 4.0 g of dried boundary complexions, and (iv) enhance the number of SnSe particles and vigorously stirred at room temperature for 48 h. defects at multiple length scales, such as atomic impurities, After that, acetone was added to the mixture and the particles were planar defects, dislocations, and Na-rich precipitates. The precipitated by centrifugation. Subsequently, the CdSe molecular presence of scattering sources at all relevant length scales complex capped SnSe particles were washed one more time with improved the figure of merit from zT = 1.3 in bare SnSe to zT acetone, centrifuged, and dried under vacuum yielding a fine powder. = 2.2 in SnSe−CdSe nanocomposites. We believe that the SnSe particles were treated with 3% PbS molecular complexes strategy presented here for inhibiting grain growth is of great following the exact same process as for the SnSe−CdSe system. significance beyond the thermoelectric field because it tackles Bulk Nanomaterial Consolidation. Dried SnSe-x%CdSe (x = 0, grain growth in semiconductor nanocomposites. 1, 2, 3, and 4) nanocomposites were first annealed at 500 °C for 60 min under a forming gas (95% N2 + 5% H2) flow inside a tube furnace with the heating rate of around 10 °C/min. Afterward, the annealed METHODS nanopowder was ground with an agate mortar and loaded into a Chemicals. Tin(II) chloride dihydrate (SnCl2·2H2O, 98%), graphite die lined with graphite paper inside the glovebox. The sodium hydroxide (NaOH, pellets 98%), sodium borohydride nanopowder was then consolidated into cylinders under vacuum (Ø (NaBH4, 98%), cadmium oxide (CdO, 99.99%), lead(II) oxide 8.6 mm × h = 12 mm) in an AGUS PECS SPS System-Model SPS (PbO, 99.99%), and N-methylformamide (MFA, 99%) were 210Sx by applying an axial pressure of 45 MPa at 500 °C for 5 min. purchased from Fisher Scientific. Selenium powder (Se, 100 mesh, All consolidated cylinders presented relative densities above 92% of ≥99.5%), ethylenediamine (en, 99%), 1, 2-ethanedithiol (EDT, the theoretical value. Finally, all the cylindrical pellets were annealed ≥95.0%), anhydrous acetone (extra dry), and ethanol (95%) were in forming gas (95% N2 + 5% H2) static atmosphere for 1 h at 550 °C purchased from Sigma-Aldrich. All chemicals were used as received (ca. 4 °C/min). These cylinders were cut in two normal directions, without further purification. Syntheses were carried out using a i.e., parallel to the pressing direction and within the cylinder plane, vacuum/dry argon Schlenk line. into discs and rectangular bars (Figure S28). SnSe Particle Synthesis. SnSe particles were prepared following Structural and Chemical Characterization. XRD measure- previously reported by Gregory et al.,37 with slight modifications. In a ments were carried out on a D8 ADVANCE diffractometer (Bruker, typical synthesis, NaBH4 (6.053 g, 160 mmol) was first dissolved in Germany) with Cu Kα radiation from 20 to 60° with a resolution of 400 mL of deionized water, and then Se powder (6.314 g, 80 mmol) 0.01° and time step of 0.1 s. Temperature-dependent XRD patterns was slowly added into the solution. Stirring should be avoided during were collected in a reactor chamber with sample spinning and height this step due to the strong evolution of hydrogen gas. Once the correction (XRK 900, Anton Paar). The temperature was ramped bubbling finished, stirring was resumed under Ar flow until the from 50 to 550 °C at a rate of 50 °C/min and held for 10 min before solution became transparent indicating the complete reduction of Se. each measurement. The size and morphology of initial particles, In parallel, NaOH (30 g, 750 mmol) and SnCl2·2H2O (16.247 g, 72 annealed nanopowders, and sintered pellets were examined by field- mmol) were mixed with 360 mL of deionized water. The mixture was emission scanning electron microscopy operated at 5.0 kV (FE-SEM stirred at room temperature under Ar flow until complete dissolution. Merlin VP Compact, Zeiss). The overall material composition was At this point, the solution was heated under reflux to its boiling point investigated by using EDS (Octane Elite EDS, Oxford) attached to (ca. 101.3 °C). The freshly prepared Se solution was rapidly injected the SEM operated at 15.0 kV. EBSD studies were used to estimate the into the boiling Sn (II) solution, and the temperature dropped to ∼70 average grain size of the consolidated materials. The samples for °C. Upon injection, the reaction mixture turned black indicating the EBSD were polished with a Neutral Alumina Suspension (OP-AN particle formation. The temperature was allowed to recover to ca. Struers No. 40700054). The EBSD measurements were carried out on 101.3 °C, and this temperature was maintained for 2 h. To purify the Zeiss FEG SEM Merlin microscope with OXFORD EBSD Symmetry as-synthesized particles, the mixture was decanted, and the trans- and OXFORD EDS Ultim Max 170 using the software Aztec 4.3HF1 84 https://doi.org/10.1021/acsnano.1c06720 ACS Nano 2022, 16, 78−88 ACS Nano www.acsnano.org Article and a microscope at an accelerating voltage of 200 kV with a JEOL GmbH (Germany). Temperature dependent Hall charge carrier silicon drift detector (SDD). Scanning transmission electron micro- concentrations (pH) and mobilities (μH) were measured from 300 to scope (STEM) characterization of the CdSe surface-coated SnSe 750 K with the Van der Pauw method using a magnetic field of 0.6 T samples has been carried out using a JEOL JEM2800 microscope (ezHEMS, NanoMagnetics) (Figure S29a). Values provided corre- operated at 200 kV with a point to point resolution of 0.14 nm. spond to the average of 10 measurements, and the estimated error is HRTEM was also done with (TECNAI F20, FEI) microscope ca. 10%. operated at 200 kV with a point to point resolution of 0.14 nm. TEM specimens were prepared by Focused Ion Beam (FIB) (Helios Dual Beam Nanolab, FEI). In order to evaluate the crystal structure, 3D ASSOCIATED CONTENT atomic models of the matrix and the precipitates were created using the Rhodius software platform.63,64 These models were used for the * sı Supporting Information HRTEM image simulation with the STEM_CELL software.65,66 The The Supporting Information is available free of charge at models used for the simulation of the image were those created by https://pubs.acs.org/doi/10.1021/acsnano.1c06720. Rhodius. Needle-shaped APT specimens were prepared by a standard “lift-out” method in an SEM/FIB dual-beam focused ion beam Additional characterization data including TEM, microscope (Helios NanoLab 650, FEI). APT measurements were HRTEM, SEM, EBSD, XRD, thermoelectric properties, carried out on a local electrode atom probe (LEAP 4000X Si, and literature comparison (PDF) Cameca) by applying 10 ps, 5 pJ ultraviolet laser pulses (355 nm). The pulse repetition rate was 200 kHz, and the detection rate was set as 1 ion per 100 pulses (1%) on average. The measurement base AUTHOR INFORMATION temperature of specimen was 30 K to minimize surface migration. The detection efficiency was 50% owing to the open area of the Corresponding Author microchannel plates, and the ion flight path was 160 mm. APT data Maria Ibáñez − IST Austria, 3400 Klosterneuburg, Austria; was processed with the software package IVAS 3.8.0. The overall orcid.org/0000-0001-5013-2843; Email: mibanez@ material composition was investigated by using an Oxford EDS ist.ac.at apparatus attached to the Zeiss Auriga SEM at 15.0 kV and by optical emission spectroscopy by means of inductively coupled plasma (ICP) Authors on the ICPE-9820 system. Thermoelectric Property Measurement. Seebeck coefficients Yu Liu − IST Austria, 3400 Klosterneuburg, Austria; were measured by using a static direct current (DC) method. orcid.org/0000-0001-7313-6740 Electrical resistivities were obtained by a standard four-probe method. Mariano Calcabrini − IST Austria, 3400 Klosterneuburg, Both Seebeck coefficient and electrical resistivity were simultaneously Austria measured in a Linseiss LSR-3 system from room temperature to ca. Yuan Yu − RWTH Aachen, I. Physikalisches Institut (IA), 790 K, under helium atmosphere. Samples were held between two 52074 Aachen, Germany; orcid.org/0000-0002-3148- alumel electrodes and two probe thermocouples with spring-loaded 6600 pressure contacts. A resistive heater on the lower electrode created Seungho Lee − IST Austria, 3400 Klosterneuburg, Austria; temperature differentials in the sample to determine the Seebeck coefficient. We estimate an error of ca. 4% in the measurement of orcid.org/0000-0002-6962-8598 both electrical conductivity and Seebeck coefficient. Combining the Cheng Chang − IST Austria, 3400 Klosterneuburg, Austria; uncertainties of electrical conductivity and Seebeck coefficient, the orcid.org/0000-0002-9515-4277 uncertainty of the power factor is ca. 12%. The results presented here Jérémy David − Catalan Institute of Nanoscience and are an average of the results obtained after measuring 3 pellets Nanotechnology (ICN2), CSIC and BIST, 08193 Barcelona, produced under identical conditions; measurements between different Catalonia, Spain samples have standard deviations below 10%. Additionally, each pellet Tanmoy Ghosh − IST Austria, 3400 Klosterneuburg, Austria was measured 3 times; a difference in the temperature dependent Maria Chiara Spadaro − Catalan Institute of Nanoscience and thermoelectric properties is only found between the first up and down Nanotechnology (ICN2), CSIC and BIST, 08193 Barcelona, measurements (Figure S26). In subsequent measurements, the values remain stable. To avoid this hysteresis, before transport measurements Catalonia, Spain; orcid.org/0000-0002-6540-0377 the samples undergo a temperature treatment detailed in “Bulk Chenyang Xie − Department of Physics, INTE & Barcelona Nanomaterial Consolidation” section. A Xenon Flash Apparatus (XFA Multiscale Res. Center, Universitat Politècnica de Catalunya, 500, Linseis) and a Laser Flash Analyzer (LFA 1000, Linseis) were 08930 Barcelona, Catalunya, Spain used to determine thermal diffusivity of the samples with an estimated Oana Cojocaru-Mirédin − RWTH Aachen, I. Physikalisches error of ca. 5%. The total thermal conductivity was calculated by κ = Institut (IA), 52074 Aachen, Germany; orcid.org/0000- λCpρ, where λ is the thermal diffusivity, Cp is the heat capacity, and ρ 0001-6543-203X is the mass density. Temperature-dependent Cp values were also Jordi Arbiol − Catalan Institute of Nanoscience and evaluated by DSC (DSC 404 F3, Netzsch) and calculated by the Cp Nanotechnology (ICN2), CSIC and BIST, 08193 Barcelona, ratio method with a sapphire standard using the Proteus software Catalonia, Spain; ICREA, 08010 Barcelona, Catalonia, (Netzsch). Measurements were carried out under high purity N2 flow in a Pt−Rh/Al2O3 crucible. The sample was first preheated to 50 °C Spain; orcid.org/0000-0002-0695-1726 and keep for 10 min to avoid a heat hook and then heated to 550 °C Complete contact information is available at: with a rate of 10 °C/min and keep for 10 min, before cooling inside https://pubs.acs.org/10.1021/acsnano.1c06720 the instrument. Cp was also estimated from the Dulong−Petit limit (3R law). The densities (ρ) were measured using the Archimedes’ Author Contributions method with a ca. 2% error. Figure S21d displays a comparison of the zT values obtained when considering the experimental Cp or the The manuscript was written through the contributions of all calculated Cp. As a consequence, the combined uncertainty for all authors. All authors have approved the final version of the measurements involved in zT determination shown in the plot is manuscript. estimated to be ca. 17%. To avoid cluttering the plots, error bars were not included in the figures. The results reported in this work were Notes measured at our lab in IST Austria as well as by Linseis Messgeräte The authors declare no competing financial interest. 85 https://doi.org/10.1021/acsnano.1c06720 ACS Nano 2022, 16, 78−88 ACS Nano www.acsnano.org Article ACKNOWLEDGMENTS (10) Shi, W.; Gao, M.; Wei, J.; Gao, J.; Fan, C.; Ashalley, E.; Li, H.; Wang, Z. Tin Selenide (SnSe): Growth, Properties, and Applications. This work was financially supported by IST Austria and the Adv. Sci. 2018, 5, 1700602. Werner Siemens Foundation. Y.L. acknowledges funding from (11) Qin, B.; Wang, D.; Liu, X.; Qin, Y.; Dong, J.-F.; Luo, J.; Li, J.- the European Union’s Horizon 2020 research and innovation W.; Liu, W.; Tan, G.; Tang, X.; et al. Momentum and Energy program under the Marie Sklodowska-Curie grant agreement Multiband Alignment Enable Power Generation and Thermoelectric No. 754411. S.L. and M.C. received funding from the Cooling. Science 2021, 373, 556−561. European Union’s Horizon 2020 research and innovation (12) Xiao, Y.; Zhao, L.-D. Seeking New, Highly Effective program under the Marie Skłodowska-Curie Grant Agreement Thermoelectrics. Science 2020, 367, 1196−1197. No. 665385. J.D. acknowledges funding from the European (13) Zhao, L.-D.; Lo, S.-H.; Zhang, Y.; Sun, H.; Tan, G.; Uher, C.; Union’s Horizon 2020 research and innovation program under Wolverton, C.; Dravid, V. P.; Kanatzidis, M. G. Ultralow Thermal the Marie Sklodowska-Curie grant agreement no. 665919 (P- Conductivity and High Thermoelectric Figure of Merit in SnSe SPHERE) cofunded by Severo Ochoa Programme. C.C. Crystals. Nature 2014, 508, 373−377. (14) Chang, C.; Wu, M.; He, D.; Pei, Y.; Wu, C.-F. F.; Wu, X.; Yu, acknowledges funding from the FWF “Lise Meitner Fellow- H.; Zhu, F.; Wang, K.; Chen, Y.; Huang, L.; Li, J. F.; He, J.; Zhao, L.- ship” grant agreement M 2889-N. Y.Y. and O.C.-M. acknowl- D. 3D Charge and 2D Phonon Transports Leading to High Out-of- edge the financial support from DFG within the project SFB Plane ZT in n-Type SnSe Crystals. Science 2018, 360, 778−783. 917: Nanoswitches. M.C.S. received funding from the (15) Zhao, L.-D.; Chang, C.; Tan, G.; Kanatzidis, M. G. SnSe: A European Union’s Horizon 2020 research and innovation Remarkable New Thermoelectric Material. Energy Environ. Sci. 2016, programme under the Marie Skłodowska-Curie grant agree- 9, 3044−3060. ment No. 754510 (PROBIST) and the Severo Ochoa (16) Lee, Y. K.; Luo, Z.; Cho, S. P.; Kanatzidis, M. G.; Chung, I. programme. J.D. received funding from the European Union’s Surface Oxide Removal for Polycrystalline SnSe Reveals Near-Single- Horizon 2020 research and innovation programme under the Crystal Thermoelectric Performance. Joule 2019, 3, 719−731. Marie Sklodowska-Curie grant agreement No. 665919 (P- (17) Zhou, C.; Lee, Y. K.; Yu, Y.; Byun, S.; Luo, Z.-Z.; Lee, H.; Ge, SPHERE) cofunded by Severo Ochoa Programme. The ICN2 B.; Lee, Y.-L.; Chen, X.; Lee, J. Y.; Cojocaru-Mirédin, O.; Chang, H.; is funded by the CERCA Program/Generalitat de Catalunya Im, J.; Cho, S.-P.; Wuttig, M.; Dravid, V. P.; Kanatzidis, M. G.; and by the Severo Ochoa program of the Spanish Ministry of Chung, I. Polycrystalline SnSe with a Thermoelectric Figure of Merit Economy, Industry, and Competitiveness (MINECO, grant Greater than the Single Crystal. Nat. Mater. 2021, DOI: 10.1038/ s41563-021-01064-6. no. SEV-2017-0706). ICN2 acknowledges funding from (18) Li, M.; Liu, Y.; Zhang, Y.; Zuo, Y.; Li, J.; Lim, K. H.; Cadavid, Generalitat de Catalunya 2017 SGR 327 and the Spanish D.; Ng, K. M.; Cabot, A. Crystallographically Textured SnSe MINECO project NANOGEN (PID2020-116093RB-C43). Nanomaterials Produced from the Liquid Phase Sintering of This project received funding from the European Union’s Nanocrystals. Dalt. Trans. 2019, 48, 3641−3647. Horizon 2020 research and innovation program under grant (19) Liang, S.; Xu, J.; Noudem, J. G.; Wang, H.; Tan, X.; Liu, G.-Q.; agreement No. 823717-ESTEEM3. The FIB sample prepara- Shao, H.; Yu, B.; Yue, S.; Jiang, J. Thermoelectric Properties of tion was conducted in the LMA-INA-Universidad de Zaragoza. Textured Polycrystalline Na0.03Sn0.97Se Enhanced by Hot Deforma- tion. J. Mater. Chem. A 2018, 6, 23730−23735. REFERENCES (20) Wei, T. R.; Tan, G.; Zhang, X.; Wu, C. F.; Li, J. F.; Dravid, V. P.; Snyder, G. J.; Kanatzidis, M. G. Distinct Impact of Alkali-Ion (1) Ortega, S.; Ibáñez, M.; Liu, Y.; Zhang, Y.; Kovalenko, M. V.; Doping on Electrical Transport Properties of Thermoelectric p-Type Cadavid, D.; Cabot, A. Bottom-Up Engineering of Thermoelectric Polycrystalline SnSe. J. Am. Chem. Soc. 2016, 138, 8875−8882. Nanomaterials and Devices from Solution-Processed Nanoparticle (21) Ge, Z.-H.; Song, D.; Chong, X.; Zheng, F.; Jin, L.; Qian, X.; Building Blocks. Chem. Soc. Rev. 2017, 46, 3510−3528. Zheng, L.; Dunin-Borkowski, R. E.; Qin, P.; Feng, J.; Zhao, L.-D. (2) Tan, G.; Zhao, L.-D.; Kanatzidis, M. G. Rationally Designing Boosting the Thermoelectric Performance of (Na,K)-Codoped High-Performance Bulk Thermoelectric Materials. Chem. Rev. 2016, Polycrystalline SnSe by Synergistic Tailoring of the Band Structure 116, 12123−12149. and Atomic-Scale Defect Phonon Scattering. J. Am. Chem. Soc. 2017, (3) Ibáñez, M.; Luo, Z.; Genc, A.; Piveteau, L.; Ortega, S.; Cadavid, 139, 9714−9720. D.; Dobrozhan, O.; Liu, Y.; Nachtegaal, M.; Zebarjadi, M.; et al. High- (22) Chen, Y. X.; Ge, Z.-H.; Yin, M.; Feng, D.; Huang, X. Q.; Zhao, Performance Thermoelectric Nanocomposites from Nanocrystal W.; He, J. Understanding of the Extremely Low Thermal Building Blocks. Nat. Commun. 2016, 7, 10766. (4) Liu, Y.; Ibáñez, M. Tidying up the Mess. Science 2021, 371, 678− Conductivity in High-Performance Polycrystalline SnSe through 679. Potassium Doping. Adv. Funct. Mater. 2016, 26, 6836−6845. (5) Zhao, L.-D.; Chang, C.; Tan, G.; Kanatzidis, M. G. SnSe: A (23) Chen, C. L.; Wang, H.; Chen, Y. Y.; Day, T.; Snyder, G. J. Remarkable New Thermoelectric Material. Energy Environ. Sci. 2016, Thermoelectric Properties of p-Type Polycrystalline SnSe Doped with 9, 3044−3060. Ag. J. Mater. Chem. A 2014, 2, 11171−11176. (6) Chen, Z.-G.; Shi, X.; Zhao, L.-D.; Zou, J. High-Performance (24) Shi, X.; Zheng, K.; Hong, M.; Liu, W.; Moshwan, R.; Wang, Y.; SnSe Thermoelectric Materials: Progress and Future Challenge. Prog. Qu, X.; Chen, Z. G.; Zou, J. Boosting the Thermoelectric Mater. Sci. 2018, 97, 283−346. Performance of p-Type Heavily Cu-Doped Polycrystalline SnSe: Via (7) Banik, A.; Roychowdhury, S.; Biswas, K. The Journey of Tin Inducing Intensive Crystal Imperfections and Defect Phonon Chalcogenides towards High-Performance Thermoelectrics and Scattering. Chem. Sci. 2018, 9, 7376−7389. Topological Materials. Chem. Commun. 2018, 54, 6573−6590. (25) Liu, J.; Wang, P.; Wang, M.; Xu, R.; Zhang, J.; Liu, J.; Li, D.; (8) Shi, X.-L.; Tao, X.; Zou, J.; Chen, Z.-G. High-Performance Liang, N.; Du, Y.; Chen, G.; Tang, G. Achieving High Thermoelectric Thermoelectric SnSe: Aqueous Synthesis, Innovations, and Chal- Performance with Pb and Zn Codoped Polycrystalline SnSe via Phase lenges. Adv. Sci. 2020, 7, 1902923. Separation and Nanostructuring Strategies. Nano Energy 2018, 53, (9) Qin, B.; Zhang, Y.; Wang, D.; Zhao, Q.; Gu, B.; Zhang, H.; Wu, 683−689. H.; Ye, B.; Pennycook, S. J.; Zhao, L.-D. Ultrahigh Average ZT (26) Wang, X.; Xu, J.; Liu, G.-Q.; Tan, X.; Li, D.; Shao, H.; Tan, T.; Realized in p-Type SnSe Crystalline Thermoelectrics through Jiang, J. Texturing Degree Boosts Thermoelectric Performance of Producing Extrinsic Vacancies. J. Am. Chem. Soc. 2020, 142, 5901− Silver-Doped Polycrystalline SnSe. NPG Asia Mater. 2017, 9, 5909. No. e426. 86 https://doi.org/10.1021/acsnano.1c06720 ACS Nano 2022, 16, 78−88 ACS Nano www.acsnano.org Article (27) Shi, X.; Wu, A.; Feng, T.; Zheng, K.; Liu, W.; Sun, Q.; Hong, (43) Webber, D. H.; Brutchey, R. L. Alkahest for V 2 VI 3 M.; Pantelides, S. T.; Chen, Z.-G.; Zou, J. High Thermoelectric Chalcogenides: Dissolution of Nine Bulk Semiconductors in a Performance in p-Type Polycrystalline Cd-Doped SnSe Achieved by a Diamine-Dithiol Solvent Mixture. J. Am. Chem. Soc. 2013, 135, Combination of Cation Vacancies and Localized Lattice Engineering. 15722−15725. Adv. Energy Mater. 2019, 9, 1803242. (44) Kang, S.-J. L. Sintering: Densification, Grain Growth and (28) Li, S.; Lou, X.; Li, X.; Zhang, J.; Li, D.; Deng, H.; Liu, J.; Tang, Microstructure; Elsevier: Oxford, 2005. G. Realization of High Thermoelectric Performance in Polycrystalline (45) Chen, D.; Ghoneim, T.; Kulkarni, Y. Effect of Pinning Particles Tin Selenide through Schottky Vacancies and Endotaxial Nano- on Grain Boundary Motion from Interface Random Walk. Appl. Phys. structuring. Chem. Mater. 2020, 32, 9761−9770. Lett. 2017, 111, 161606. (29) Zhang, L.; Wang, J.; Sun, Q.; Qin, P.; Cheng, Z.; Ge, Z.; Li, Z.; (46) Nes, E.; Ryum, N.; Hunderi, O. On the Zener Drag. Acta Dou, S. Three-Stage Inter-Orthorhombic Evolution and High Metall. 1985, 33, 11−22. Thermoelectric Performance in Ag-Doped Nanolaminar SnSe (47) Manohar, P. A.; Ferry, M.; Chandra, T. Five Decades of the Polycrystals. Adv. Energy Mater. 2017, 7, 1700573. Zener Equation. ISIJ Int. 1998, 38, 913−924. (30) Lin, C.-C.; Lydia, R.; Yun, J. H.; Lee, H. S.; Rhyee, J. S. (48) Humphreys, F. J.; Hatherly, M. Recrystallization and Related Extremely Low Lattice Thermal Conductivity and Point Defect Annealing Phenomena; Elsevier: Oxford, 2012. Scattering of Phonons in Ag-Doped (SnSe)1−x(SnS)x Compounds. (49) Leute, V.; Menge, D. Thermodynamic Investigations of the Chem. Mater. 2017, 29, 5344−5352. Quasibinary Systems (CdKSn1‑K)Te, (Cd KSn 1‑K)Se, And Sn- (31) Lee, Y. K.; Ahn, K.; Cha, J.; Zhou, C.; Kim, H. S.; Choi, G.; (SeLTe1‑L). Z. Phys. Chem. 1992, 176, 47−64. Chae, S. I.; Park, J.-H.; Cho, S.-P.; Park, S. H.; et al. Enhancing p- (50) Abrikosov, N. Kh.; Bankina, V. F.; Poretskaya, L. V.; Shelimova, Type Thermoelectric Performances of Polycrystalline SnSe via L. E.; Skudnova, E. V. Semiconducting II−VI, IV−VI and V−VI Tuning Phase Transition Temperature. J. Am. Chem. Soc. 2017, Compounds; Plenum Press: New York, 1969; pp 143−145. 139, 10887−10896. (51) Shen, W. P.; Kwok, H. S. Highly Doped p-Type and n-Type (32) Chandra, S.; Biswas, K. Realization of High Thermoelectric ZnS, ZnSe, CdS and CdSe Thin Films Growth by Pulsed Laser Figure of Merit in Solution Synthesized 2D SnSe Nanoplates via Ge Deposition. Mater. Res. Soc. Symp. Proc. 1994, 340, 457−463. Alloying. J. Am. Chem. Soc. 2019, 141, 6141−6145. (52) Gayner, C.; Amouyal, Y. Energy Filtering of Charge Carriers: (33) Shi, X.; Wu, A.; Liu, W.; Moshwan, R.; Wang, Y.; Chen, Z.-G.; Current Trends, Challenges, and Prospects for Thermoelectric Zou, J. Polycrystalline SnSe with Extraordinary Thermoelectric Materials. Adv. Funct. Mater. 2020, 30, 1901789. Property via Nanoporous Design. ACS Nano 2018, 12, 11417−11425. (53) Wei, W.; Chang, C.; Yang, T.; Liu, J.; Tang, H.; Zhang, J.; Li, (34) Wang, H. X.; Mao, L. S.; Tan, X. J.; Liu, G. Q.; Xu, J. T.; Shao, Y.; Xu, F.; Zhang, Z.; Li, J.-F.; Tang, G. Achieving High H. Z.; Hu, H. Y.; Jiang, J. Nontrivial Thermoelectric Behavior in Thermoelectric Figure of Merit in Polycrystalline SnSe via Cubic SnSe Driven by Spin-Orbit Coupling. Nano Energy 2018, 51, Introducing Sn Vacancies. J. Am. Chem. Soc. 2018, 140, 499−505. 649−655. (54) Snyder, G. J.; Snyder, A. H.; Wood, M.; Gurunathan, R.; (35) Tang, G.; Wei, W.; Zhang, J.; Li, Y.; Wang, X.; Xu, G.; Chang, Snyder, B. H.; Niu, C. Weighted Mobility. Adv. Mater. 2020, 32, C.; Wang, Z.; Du, Y.; Zhao, L.-D. Realizing High Figure of Merit in 2001537. Phase-Separated Polycrystalline. J. Am. Chem. Soc. 2016, 138, 13647− (55) Kuo, J. J.; Kang, S. D.; Imasato, K.; Tamaki, H.; Ohno, S.; 13654. Kanno, T.; Snyder, G. J. Grain Boundary Dominated Charge (36) Luo, Y.; Cai, S.; Hua, X.; Chen, H.; Liang, Q.; Du, C.; Zheng, Transport in Mg3Sb2-Based Compounds. Energy Environ. Sci. 2018, Y.; Shen, J.; Xu, J.; Wolverton, C.; Dravid, V. P.; Yan, Q.; Kanatzidis, 11, 429−434. M. G. High Thermoelectric Performance in Polycrystalline SnSe via (56) Seto, J. Y. W. The Electrical Properties of Polycrystalline Silicon Dual-Doping with Ag/Na and Nanostructuring with Ag8SnSe6. Adv. Films. J. Appl. Phys. 1975, 46, 5247−5254. Energy Mater. 2019, 9, 1803072. (57) Wei, T.-R.; Wu, C.-F.; Zhang, X.; Tan, Q.; Sun, L.; Pan, Y.; Li, (37) Han, G.; Popuri, S. R.; Greer, H. F.; Bos, J.-W. G.; Zhou, W.; J.-F. Thermoelectric Transport Properties of Pristine and Na-Doped Knox, A. R.; Montecucco, A.; Siviter, J.; Man, E. A.; Macauley, M.; SnSe1−xTex Polycrystals. Phys. Chem. Chem. Phys. 2015, 17, 30102− Paul, D. J.; Li, W.-g.; Paul, M. C.; Gao, M.; Sweet, T.; Freer, R.; 30109. Azough, F.; Baig, H.; Sellami, N.; Mallick, T. K.; Gregory, D. H. Facile (58) Yu, Y.; Zhou, C.; Zhang, S.; Zhu, M.; Wuttig, M.; Scheu, C.; Surfactant-Free Synthesis of p-Type SnSe Nanoplates with Excep- Raabe, D.; Snyder, G. J.; Gault, B.; Cojocaru-Mirédin, O. Revealing tional Thermoelectric Power Factors. Angew. Chem., Int. Ed. 2016, 55, Nano-Chemistry at Lattice Defects in Thermoelectric Materials Using 6433−6437. Atom Probe Tomography. Mater. Today 2020, 32, 260−274. (38) McCarthy, C. L.; Webber, D. H.; Schueller, E. C.; Brutchey, R. (59) Cojocaru-Mirédin, O.; Abdellaoui, L.; Nagli, M.; Zhang, S.; Yu, L. Solution-Phase Conversion of Bulk Metal Oxides to Metal Y.; Scheu, C.; Raabe, D.; Wuttig, M.; Amouyal, Y. Role of Chalcogenides Using a Simple Thiol−Amine Solvent Mixture. Nanostructuring and Microstructuring in Silver Antimony Telluride Angew. Chem., Int. Ed. 2015, 54, 8378−8381. Compounds for Thermoelectric Applications. ACS Appl. Mater. (39) Dolzhnikov, D. S.; Zhang, H.; Jang, J.; Son, J. S.; Panthani, M. Interfaces 2017, 9, 14779−14790. G.; Shibata, T.; Chattopadhyay, S.; Talapin, D. V. Composition- (60) Cantwell, P. R.; Tang, M.; Dillon, S. J.; Luo, J.; Rohrer, G. S.; Matched Molecular ″Solders″ for Semiconductors. Science 2015, 347, Harmer, M. P. Grain Boundary Complexions. Acta Mater. 2014, 62, 425−428. 1−48. (40) Hudson, M. H.; Dolzhnikov, D. S.; Filatov, A. S.; Janke, E. M.; (61) Zong, P.; Hanus, R.; Dylla, M.; Tang, Y.; Liao, J.; Zhang, Q.; Jang, J.; Lee, B.; Sun, C.; Talapin, D. V. New Forms of CdSe: Snyder, G. J.; Chen, L. Skutterudite with Graphene-Modified Grain- Molecular Wires, Gels, and Ordered Mesoporous Assemblies. J. Am. Boundary Complexion Enhances ZT Enabling High-Efficiency Chem. Soc. 2017, 139, 3368−3377. Thermoelectric Device. Energy Environ. Sci. 2017, 10, 183−191. (41) McCarthy, C. L.; Brutchey, R. L. Solution Processing of (62) Guo, R.; Wang, X.; Kuang, Y.; Huang, B. First-Principles Study Chalcogenide Materials Using Thiol-Amine ″Alkahest″ Solvent of Anisotropic Thermoelectric Transport Properties of IV-VI Systems. Chem. Commun. 2017, 53, 4888−4902. Semiconductor Compounds SnSe and SnS. Phys. Rev. B: Condens. (42) Ibáñez, M.; Hasler, R.; Genç, A.; Liu, Y.; Kuster, B.; Schuster, Matter Mater. Phys. 2015, 92, 115202. M.; Dobrozhan, O.; Cadavid, D.; Arbiol, J.; Cabot, A.; Kovalenko, M. (63) Bernal, S.; Botana, F. J.; Calvino, J. J.; Lopez-Cartes, C.; Perez- V. Ligand-Mediated Band Engineering in Bottom-Up Assembled Omil, J. A.; Rodrıguez-Izquierdo, J. M. The Interpretation of HREM SnTe Nanocomposites for Thermoelectric Energy Conversion. J. Am. Images of Supported Metal Catalysts Using Image Simulation: Profile Chem. Soc. 2019, 141, 8025−8029. View Images. Ultramicroscopy 1998, 72, 135−164. 87 https://doi.org/10.1021/acsnano.1c06720 ACS Nano 2022, 16, 78−88 ACS Nano www.acsnano.org Article (64) Arbiol, J.; Cirera, A.; Peiró, F.; Cornet, A.; Morante, J. R.; Delgado, J. J.; Calvino, J. J. Optimization of Tin Dioxide Nanosticks Faceting for the Improvement of Palladium Nanocluster Epitaxy. Appl. Phys. Lett. 2002, 80, 329−331. (65) Grillo, V.; Rotunno, E. STEM_CELL: A Software Tool for Electron Microscopy: Part ISimulations. Ultramicroscopy 2013, 125, 97−111. (66) Spadaro, M. C.; D’Addato, S.; Luches, P.; Valeri, S.; Grillo, V.; Rotunno, E.; Roldan, M. A.; Pennycook, S. J.; Ferretti, A. M.; Capetti, E.; et al. Tunability of Exchange Bias in Ni@NiO Core-Shell Nanoparticles Obtained by Sequential Layer Deposition. Nano- technology 2015, 26, 405704. 88 https://doi.org/10.1021/acsnano.1c06720 ACS Nano 2022, 16, 78−88