betaFeedback

WorksPhysics and Astronomy

Effects of W alloying on the electronic structure, phase stability and thermoelectric power factor in epitaxial CrN thin films

Abstract

CrN-based alloy thin films are of interest as thermoelectric materials for energy harvesting. Ab initio calculations show that dilute alloying of CrN with 3 at.% W substituting Cr, induce flat electronic bands and push the Fermi level EF into the conduction band, while retaining dispersive Cr 3d bands. These features are conducive for both high electrical conductivity σand high Seebeck coefficient α, and hence the thermoelectric power factor α^2σ. To investigate this possibility, epitaxial CrWxNz films were grown on c-plane sapphire by dc-magnetron sputtering. However, even films with the lowest W concentration (x = 0.03) in our study contained metallic h-Cr2N, which is not conducive for a high α. Nevertheless, the films exhibit a sizeable power factor of α^2σ~ 4.7 x 10-4 Wm-1K-2 due to high σ~ 700 Scm-1, and a moderate α~ -25 ~{^^^^00b5}V/K. Increasing h-Cr2N fractions in the 0.03 < x \le 0.19 range monotonically increases σ, but severely diminishes αleading to two orders of magnitude decrease in α^2σ. This trend continues with x > 0.19 due to W precipitation. These findings indicate that dilute W additions below its solubility limit in CrN is important for realizing high thermoelectric power factor in CrWxNz alloy films.

below its solubility limit in CrN is important for realizing high thermoelectric power factor in CrWxNz\mathrm{CrW}_{x}\mathrm{N}_{z} alloy films.

Introduction

Transition metal nitrides (TMNs) constitute an important class of materials that exhibit a combination of high hardness, mechanical strength, thermal stability and corrosion resistance [1]. TMNs such as TiN, HfN and ZrN have been used as protective and decorative coatings, and nanolayers of metallic TMNs are used as diffusion barriers in nanodevice wiring architectures in integrated circuits [2, 3, 4, 5]. Narrow-bandgap thin films of TMNs, such as, CrN- and ScN-based alloys, have been considered for thermoelectric and piezoelectric energy conversion [6, 7, 8, 9]. ScAlN films exhibit large piezoelectric responses at 500 °C making them suitable for high-temperature device applications [10, 11, 12].

Thermoelectric conversion efficiency is related to a dimensionless figure of merit ZT=α2σTκZT = \frac{\alpha^{2}\sigma T}{\kappa}, where TT is the absolute temperature, α\alpha is the Seebeck coefficient, σ\sigma the electrical conductivity, κ\kappa the thermal conductivity [13, 14] and α2σ\alpha^{2}\sigma is the thermoelectric power factor.

Realizing high ZTZT is a fundamental challenge that requires the circumvention of unfavorable couplings between α\alpha, σ\sigma and κ\kappa. ZTZT enhancement strategies include doping (less than a few at.%) [15] and/or alloying (several at.% or more) [16] to increase α2σ\alpha^{2}\sigma by manipulating the electronic structure, and decreasing κ\kappa by impurity- and nanostructuring-induced phonon scattering [17, 18].

CrN-based alloys have inherently high thermoelectric power factors [19, 20, 21] due to the narrow bandgap of CrN (Eg∼0.2 eVE_g \sim0.2\,\mathrm{eV}) which simultaneously induces high σ\sigma and high α\alpha underpinned by high charge carrier entropy near the Fermi level EFE_{\mathrm{F}}. These attributes, together with an intrinsically low κ∼2 Wm−1K−1\kappa\sim2\,\mathrm{Wm}^{-1}\mathrm{K}^{-1} when compared to other TMNs [22], due to low phonon lifetimes arising from spin-lattice coupling [23, 24], make CrN an attractive candidate for enhancing ZTZT through doping and alloying. Substituting Cr (3d53d^5) with W (5d45d^4) in CrN offers the possibility to not only manipulate σ\sigma and α\alpha, but also decrease κ\kappa due to the atomic mass contrast between W and Cr. Indeed, adding W or V to CrN increase α2σ\alpha^2\sigma, but ZTZT increases only for W additions which decrease κ\kappa, [25] unlike V additions which increase κ\kappa [26]. Understanding the effects of W alloying on the electronic structure, phase stability and film morphology are crucial for understanding and tuning the thermoelectric properties to realize high ZTZT in CrN-based films.

Here, we show from ab initio calculations that dilute substitutions of Cr with W in CrN result in flat electronic bands near the Fermi level EFE_{\mathrm{F}}, which is pushed into the conduction band. These features are conducive for high σ\sigma as well as high α\alpha, and hence, a high power-factor α2σ\alpha^2\sigma.

To examine this experimentally, we attempted to grow epitaxial CrWxNz\mathrm{CrW}_x\mathrm{N}_z films, and measured σ\sigma and α\alpha as a function of W content xx. We find that W additions monotonically increase σ\sigma, but severely diminish α\alpha due to increasing fractions of metallic hh-Cr2N\mathrm{Cr}_2\mathrm{N} and W precipitation. These findings indicate that restricting W content to below its solubility limit in CrN is crucial for realizing high α2σ\alpha^2\sigma in CrWxNz\mathrm{CrW}_x\mathrm{N}_z alloys.

Methods

Computational details

The electronic structures of cubic Cr1−xWxN\mathrm{Cr}_{1-x}\mathrm{W}_x\mathrm{N} with x=0x = 0 and x=0.03x = 0.03 in double layer antiferromagnetic (AFM [110]2[110]_2) configuration were computed by density functional theory calculations (DFT) using projector-augmented-wave (PAW) potentials as implemented in the Vienna ab initio simulation package (VASP) code [27, 28]. The exchange-correlation energy was treated within the LDA + U scheme with U=3U = 3 eV [29] to consider Coulombic interactions in the narrow Cr 3d3d bands, and counteract bandgap underestimation or absence by local approaches. The cutoff energy of plane waves for the expansion of wavefunctions was set to 520 eV. Brillouin zone sampling was carried out on a 5×5×55 \times5 \times5 Monkhorst-Pack kk mesh, and the electronic optimization criterion was set to 10−510^{-5} eV. A 2×2×22 \times2 \times2 supercell of 64 atoms was used to analyze the equation of state (EOS) to estimate the equilibrium lattice constant a0a_0. The crystal structure was visualized using Vesta [30] and the electronic structures were plotted using Sumo package [31].

Experimental details

CrWxNz\mathrm{CrW}_{x}\mathrm{N}_{z} thin films with 0≤x≤0.250 \leq x \leq0.25 were grown in an ultra-high vacuum chamber with a base pressure of ∼10−6\sim10^{-6} Pa [29] by DC magnetron sputtering onto 10 x 10 mm2^2 cc-plane sapphire substrates (Alineason Materials Technologies). Prior to film growth, the substrates were ultrasonicated successively in acetone and ethanol for 10 minutes each, followed by blow drying with a N2\mathrm{N}_2 gas jet. The films were sputter-deposited from 50-mm-diameter disc-shaped targets of Cr and W (both 99.95% purity) using a 0.33 Pa (∼2.5\sim2.5 mTorr) plasma struck with a 40/60 Ar/N2\mathrm{N}_2 mixture. The W level was varied by fixing the Cr target power PCrP_{\mathrm{Cr}} constant and varying the W target power PWP_{\mathrm{W}} in the 0.1≤PW/PCr≤0.650.1 \leq P_{\mathrm{W}}/P_{\mathrm{Cr}} \leq0.65 range. During film deposition, the substrates were at 600 °C and rotated at 15 rpm.

Phase purity and crystal structure were characterized by X-ray diffractometry (XRD) using a PANanalytical X’Pert Pro instrument with a Cu-Kα\alpha radiation source operated at 45 kV and 40 mA. Pole figure measurements were carried out by scanning along in-plane rotation ϕ\phi and out-of-plane tilt ψ\psi angles at 2.5° steps per second using cross-slit primary optics and parallel plate collimator secondary optics. Film thickness, roughness, and density were estimated by analyses of X-ray reflectivity (XRR) data acquired using a hybrid-mirror module with a 0.5° incidence slit and a 0.125° post-diffraction slit. Film morphology was examined by scanning electron microscopy (SEM) in a Sigma 300 Zeiss instrument operated at 3 kV using Inlens detector. Measurement of composition of the films was carried out by combination of Rutherford backscattering spectrometry (RBS) and time-of-flight elastic recoil detection analysis (ToF-ERDA). Experiments were performed using 5 MV 15-SDH2 Pelletron accelerator at Tandem Laboratory, Uppsala University [32]. 2 MeV He+\mathrm{He}^{+} primary ion beam was incident at 5∘5^\circ to the sample normal and the backscattered particles were detected at an angle of 170∘170^\circ. The acquired spectra was processed and analysed using the SIMNRA [33] software, version 7.03. For ToF-ERDA measurements, 44 MeV iodine (136I8+{}^{136}\mathrm{I}^{8+}) beam was incident at an angle of 67.5∘67.5^\circ with the sample surface normal. The incident ions created recoils which passed through the time-of-flight detector and gas ionization chamber, placed at an angle of 45∘45^\circ relative to the incident beam. The elemental depth profile was extracted from the time and energy coincidence spectra using Potku 2.0 code [34].

Room-temperature electrical conductivity σ\sigma was measured by using a Jandel Model RM3000 four-point-probe station. Seebeck coefficient, α\alpha was measured using a home-made setup equipped with Peltier devices that maintain desired temperature gradient across the sample. In this setup, temperatures are measured with K-type thermocouples, and the Seebeck voltage is measured with a Keithley 2001 multimeter as described elsewhere [35]. Charge carrier density nn and mobility μ\mu were measured by room-temperature Hall measurements in a home-made setup with a 0.485 T permanent magnet. The Hall voltage was measured in the van der Pauw configuration with a Keithley 2100 multimeter, with a 40-mA current supplied by a Keithley 2400 source.

Results and discussion

Ab-initio calculations

CrN undergoes a magnetostructural transition at the Néel temperature TN∼280T_\mathrm{N} \sim280 K, wherein the antiferromagnetic low-temperature orthorhombic phase is converted to the hightemperature paramagnetic rocksalt phase [36]. However, cubic CrN is known to energetically favor the antiferromagnetic spin configuration with alternating double layers of identical spin perpendicular to the crystallographic [110] direction (AFM-1102_2) as shown in Figure 1a [33]. The equilibrium lattice parameter a0a_0 of cubic CrN was estimated by fitting energy (E) vs volume (V) to obtain a0∼4.13a_0 \sim4.13 Å for CrN (Figure 1b), which agrees well with prior theoretical and experimental reports [38, 39]. For Cr1−x_{1-x}Wx_xN with x=0.03x = 0.03, we obtain a0∼4.14a_0 \sim4.14 Å for W occupancy of either Cr↑\mathrm{Cr}^{\uparrow} or Cr↓\mathrm{Cr}^{\downarrow} sites, suggesting no observable site-occupancy effect. Thus, we calculated the electronic properties only for W substitution at Cr↑\mathrm{Cr}^{\uparrow} sites.

Cubic CrN crystal structure, Murnaghan fits, and density of states plots

Figure 1. (a) A sketch of the cubic CrN crystal structure showing Cr↑\mathrm{Cr}^{\uparrow} (red) and Cr↓\mathrm{Cr}^{\downarrow} (green) sites, with W substitution at the Cr↑\mathrm{Cr}^{\uparrow} sites. (b) Murnaghan fits for CrN and Cr0.97_{0.97}W0.03_{0.03}N. Density of states plots for (c) CrN and (d) Cr0.97_{0.97}W0.03_{0.03}N.

For pristine defect-free CrN, the calculated bandgap Eg∼0.2E_g \sim0.2 eV agrees with prior work [37], and the Cr 3d states are the primary contributors to the states near EFE_F (see Figure 1c).

For Cr0.97W0.03N\mathrm{Cr}_{0.97}\mathrm{W}_{0.03}\mathrm{N}, the EFE_{\mathrm{F}} is inside the conduction band (Figure 1d) and is associated with a significantly higher density of states near EFE_{\mathrm{F}} than in CrN. This W-induced EFE_{\mathrm{F}} shift into the conduction band is likely to produce metal-like electrical transport, commonly observed in degenerate semiconductors.

Electronic structure calculations for CrN along the high symmetry X-Γ\Gamma-K-W-X path reveal dispersive Cr 3d3d bands with conduction band minima at Γ\Gamma, and a bandgap of Eg∼0.2E_{\mathrm{g}} \sim0.2 eV (Figure 2a). Substituting Cr with 3 at.% W, i.e., x=0.03x = 0.03, largely preserves the CrN band structure, but results in additional flat bands with a strong W-dd character near EFE_{\mathrm{F}}. The simultaneous presence of dispersive and flat bands, and high density of states near EFE_{\mathrm{F}} are conducive for high α2σ\alpha^{2}\sigma in Cr1−xWxN\mathrm{Cr}_{1-x}\mathrm{W}_{x}\mathrm{N} with x=0.03x = 0.03. Dispersive bands favor high carrier mobility μ\mu, and hence, high σ\sigma. Flat narrow bands favor high α\alpha due to the high effective mass m∗m^{*} as seen from α=8π2kB23eh2m∗T(π3n)2/3\alpha= \frac{8\pi^{2}k_{\mathrm{B}}^{2}}{3eh^{2}}m^{*}T\left(\frac{\pi}{3n}\right)^{2/3}, where nn is the charge carrier density [13]. Thus, higher m∗m^{*} tends to decrease σ\sigma leading to the inverse coupling between α\alpha and σ\sigma. Thus, the presence of both narrow as well as highly dispersive bands are necessary for realizing high α2σ\alpha^{2}\sigma [36].

Electronic band structures of CrN and Cr0.97W0.03N

Figure 2. Electronic band structures computed by ab initio calculations along X-Γ\Gamma-K-W-X for (a) CrN and (b) Cr0.97W0.03N\mathrm{Cr}_{0.97}\mathrm{W}_{0.03}\mathrm{N}.

Epitaxial CrWxN\mathrm{CrW}_{x}\mathrm{N} growth and structural analysis

To experimentally investigate our theoretical predictions, we attempted to grow epitaxial Cr1−xWxNδ\mathrm{Cr}_{1-x}\mathrm{W}_{x}\mathrm{N}_{\delta} films. We chose a growth temperature of 600 °C because epitaxial rocksalt CrN is favored in the 550 °C ≤T≤\le T \le 700 °C range [41], and metallic hh-Cr2_2N formation at higher temperatures is not conducive for high α\alpha. Cr target power PCrP_{\mathrm{Cr}} is also a crucial parameter that influences CrN and hh-Cr2_2N phase selection and fractions, and Cr vacancy concentration that can impact σ\sigma [42]. Results of our studies of CrN films grown with 42 W ≤PCr≤\le P_{\mathrm{Cr}} \le 50 W showed that the highest α2σ\alpha^{2}\sigma was obtained for PCr=44P_{\mathrm{Cr}} = 44 (see supplementary information Figure S1). Accordingly, CrWxNz\mathrm{CrW}_{x}\mathrm{N}_{z} films were deposited for different PWP_{\mathrm{W}} at a fixed PCr=44P_{\mathrm{Cr}} = 44 W.

X-ray diffractograms and thermoelectric power factor of CrN films

Figure S1. (a) X-ray diffractograms, and (b) thermoelectric power factor, α2σ\alpha^{2}\sigma of CrN films as a function of Cr target power PCrP_{\mathrm{Cr}}.

RBS and ERDA analyses showed that all the CrWxNz\mathrm{CrW}_{x}\mathrm{N}_{z} films were nitrogen-deficient (see Table I). Increasing PWP_{\mathrm{W}} in the 0.1≤PW/PCr≤0.650.1 \le P_{\mathrm{W}}/P_{\mathrm{Cr}} \le0.65 range increased the W content from x=0.03x = 0.03 to 0.25, decreased the nitrogen content from z=0.91z = 0.91 to 0.21, and increased oxygen contamination from 1.7 at. % to 5.5 at.%. The large decreases in nitrogen content correlating with increasing W content suggests the formation of a Cr-rich Cr2_2N, corroborated by XRD analyses described below.

CrWNON/(Cr+W)N/(\mathrm{Cr}+\mathrm{W})
51.6-46.91.70.91
50.53.443.42.80.80
58.46.830.64.20.47
55.818.921.04.30.28
53.025.116.45.50.21

Table I. Elemental compositions of CrWN films determined by RBS and ToF-ERDA.

X-ray diffractograms and lattice parameters

Figure 3. X-ray diffractograms from (a) CrWxNz\mathrm{CrW}_x\mathrm{N}_z films with 0.00≤x≤0.250.00 \le x \le0.25. Here, * represents the substrate 006 peak at 2θ∼41.67∘2\theta\sim41.67^\circ (b) Lattice parameters of CrN and hh-Cr2N\mathrm{Cr}_2\mathrm{N} for CrWxNz\mathrm{CrW}_x\mathrm{N}_z films with 0.00≤x≤0.250.00 \le x \le0.25.

X-ray diffractograms show that W additions shift the 111 and 222 CrN Bragg peaks to higher angles, i.e., lower d-spacings, indicative of W-induced unit cell compression. Even for x∼0.03x \sim0.03, the lowest W content explored in this study, we observe the hh-Cr2N\mathrm{Cr}_2\mathrm{N} (0002) reflection at 2θ∼40.3∘2\theta\sim40.3^\circ. The results indicate that the solubility limit of W in CrN is less than x∼0.03x \sim0.03 even under the far-from equilibrium conditions involved during sputter deposition. Our attempts to obtain CrN films with lower W contents were unsuccessful due to the instability of the plasma at low PWP_{\mathrm{W}}. Thus, alternative approaches need to be implemented for achieving desired dilute W concentrations in CrN films.

At x=0.07x = 0.07, the CrN peaks continue to shift further but considerably decrease in intensity, while the hh-Cr2N\mathrm{Cr}_2\mathrm{N} 0002 peak intensity increases and the hh-Cr2N\mathrm{Cr}_2\mathrm{N} 0004 peak appears at 2θ∼86.5∘2\theta\sim86.5^\circ.

No rocksalt CrN peaks were detectable in diffractograms from films with x=0.19x = 0.19, whereas the hh-Cr2_2N peaks were stronger. Films with x=0.25x = 0.25 exhibit 110 and 220 Bragg peaks at 2θ∼40.55∘2\theta\sim40.55^\circ and ∼88.1∘\sim88.1^\circ from metallic W, showing the coexistence of metallic hh-Cr2_2N and W.

X-ray pole figures of the 111 and 002 reflections from the CrNz_z and CrW0.25_{0.25}Nz_z films, respectively, show distinct spot patterns at specific angles (Figure 4) confirming the epitaxial growth of CrWx_xNz_z films. Scans of the {111}\{111\} poles with Bragg angle 2θ=37.4∘2\theta= 37.4^\circ show a central peak and 6 peaks at ψ∼71∘\psi\sim71^\circ (Figure 4a). Here, the presence of 6 peaks rather than 3 expected is indicative of twin domains seen in epitaxial films of rocksalt structure materials grown on cc-plane sapphire. Pole figure performed at 2θ=73.1∘2\theta= 73.1^\circ on the CrW0.25_{0.25}Nz_z film reveals the presence of 6 poles at ψ∼65∘\psi\sim65^\circ with a symmetry of 3 (Figure 4(b)). These poles correspond to the {033ˉ1}\{03\bar{3}1\} reflection from hh-Cr2_2N and demonstrating its epitaxial character.

Pole figure scans from CrN$_z$ and CrW$_{0.25}$N$_z$ films

Figure 4. Pole figure scans from CrNz_z and CrW0.25_{0.25}Nz_z film grown on cc-plane sapphire for (a) {111}\{111\} reflections at 2θ=37.4∘2\theta= 37.4^\circ and (b) {033ˉ1}\{03\bar{3}1\} reflections at 2θ=73.1∘2\theta= 73.1^\circ, respectively.

Analyses of XRR data (See supplementary, Figure S1) show that all the CrWx_xNz_z films were ∼60\sim60 nm thick with a roughness of ≤3\le3 nm, irrespective of the W content. The film density increased from ∼6.7\sim6.7 to 11.1 g cm−311.1\ \mathrm{g\ cm}^{-3} as the W content was increased from x=0.03x = 0.03 to x=0.25x = 0.25.

The baseline density for CrN films was ∼5.8 g cm−3\sim5.8\ \mathrm{g\ cm}^{-3} which is ∼98%\sim98\% of the theoretical density of CrN.

Morphological studies

W alloying had a significant effect on the CrWx_xNz_z film morphology. Pristine CrN film surface exhibit large triangular grains (Fig. 5(a)) indicative of 111-oriented rocksalt films [43].

In contrast, films with x=0.03x = 0.03 show highly elongated grains with an average width of ∼70\sim70 nm, however, the average lengths of these grains are difficult to calculate due to continuous and curved domains (Fig. 5(b)). Films with higher W contents show a different morphology indicative of fine hh-Cr2_2N grains. However, for x=0.07x = 0.07, we observe a dense morphology (Fig. 5(c)). Films with x=0.19x = 0.19 and 0.250.25 (Fig. 5(d-e)) both exhibit similar features having average grain size ∼40\sim40 nm, but with the presence of voids. Increasing the W content leads to grain refinement, hh-Cr2_2N formation and W precipitation. The grain size distribution for the corresponding films is plotted in Fig 5(f-h). Figure S3 show elemental mapping of films with x=0.25x = 0.25, showing that the W precipitates are homogeneously distributed throughout the film.

Elemental EDX maps of Cr and W from CrW$_{0.25}$N$_z$ films

Figure S3. (a-c) Elemental EDX maps of Cr and W from CrW0.25_{0.25}Nz_z films.

Morphological images and grain size distributions of CrWₓN_z films for x = 0, 0.03, 0.07, 0.19, and 0.25, with corresponding grain size distributions

Figure 5.

Image not extracted

Figure 5 (PDF p. 12). (a-e) SEM micrographs from CrWx_xNz_z films with 0.00≤x≤0.250.00 \le x \le0.25, and (f–h) corresponding grain size distribution plot with Gaussian fit.

Thermoelectric transport properties

The room-temperature values of σ\sigma and α\alpha measured from the films are plotted in Figure 6 as a function W alloying content. Pristine CrN shows σ∼197\sigma\sim197 Scm−1^{-1} which agrees with prior reports for CrN films deposited under similar conditions [5]. Adding W increases σ\sigma to ∼700\sim700 Scm−1^{-1} for x=0.03x = 0.03, which is in accordance with the W-induced changes in the electronic band structure and increase in the density of states at the Fermi level indicated by our ab initio calculations.

Seebeck coefficient, electrical conductivity, corresponding power factor, carrier density, and mobility as functions of W content

Figure 6. (a) Seebeck coefficient, electrical conductivity, and corresponding power factor, and (b) carrier density nn and mobility μ\mu extracted from Hall measurements in CrWx_xNz_z films.

For x>0.03x > 0.03, however, our theoretical calculations do not directly apply because of hh-Cr2_2N phase formation that was not considered in our computations. Films with higher W contents in the 0.07≤x≤0.190.07 \le x \le0.19 range exhibit no considerable changes in σ\sigma suggesting that the increase in σ\sigma expected from increased metallic hh-Cr2_2N fraction is countered by the decreased rocksalt CrN fraction. At x=0.25x = 0.25, σ\sigma increases to ∼1800\sim1800 Scm−1^{-1} due to W precipitation [44].

The thermoelectric properties and the Hall measurement analyses are shown in Fig. 5. All the films show negative α\alpha, indicating n-type behavior, i.e., electrons are the majority charge carriers. Pristine CrN shows high α∼−155\alpha\sim-155 μ\muV/K which reduces greatly to α∼−25\alpha\sim-25 μ\muV/K for x=0.03x = 0.03. Experimental observation of lower α\alpha for x=0.03x = 0.03 as contrary to the theoretical calculations infers one needs to explore much lower W content CrN films to avoid formation of secondary phases. With even higher W content (x>0.03x > 0.03), α\alpha reduces further and saturates around α∼5\alpha\sim5 μ\muV/K due to the increased metallic hh-Cr2_2N phase.

The room-temperature power factor α2σ∼4.7×10−4\alpha^2\sigma\sim4.7 \times10^{-4} Wm−1^{-1}K−2^{-2} for CrN is in close agreement as reported by Kerdsongpanya et al., (∼5×10−4\sim5 \times10^{-4} Wm−1^{-1}K−2^{-2}) [21], but lower than those of Gharavi et al., (∼9\sim9-11 ×10−3\times10^{-3} Wm−1^{-1}K−2^{-2}) [45]. However, due to the reduced α\alpha, all the films with W shows lowered α2σ\alpha^2\sigma values.

Analyses of Hall measurements of the CrWx_xNz_z films (Fig. 5 (b)) indicate W additions result in a up to four-orders-of-magnitude increase in charge carrier density nn, accompanied by a two-orders of magnitude decrease in carrier mobility μ\mu. The large increase in carrier density are due to metallic hh-Cr2_2N [46] and W precipitation. The measured n∼5.5×1019n \sim5.5 \times10^{19} cm−3^{-3} for CrN agrees well with prior reports [21]. The low μ\mu is typical of metals and is attributable to grain-boundary and W impurity scattering mechanisms. Since α\alpha and σ\sigma are strongly and adversely coupled through nn, the monotonic rise in nn increases σ\sigma but decreases α\alpha, leading to a net decrease in α2σ\alpha^2\sigma.

Thus, it is crucial to restrict W additions to below its solubility limit in CrN to preclude the formation of hh-Cr2_2N and W precipitation to realize high α2σ\alpha^2\sigma.

Conclusions

Ab initio calculations indicate that dilute W additions to CrN are attractive to increase the thermoelectric power factor due to the emergence of flat bands near the Fermi level pushed into the conduction band. Epitaxial growth of dc-magnetron sputtered W alloyed CrN films showed metallic hh-Cr2_2N formation even for the smallest W contents explored in this study. Despite this, the power factor is considerable at x=0.08x = 0.08 due to high σ∼700\sigma\sim700 S cm−1^{-1} and sizeable $\alpha = -25 μV/K\mu\mathrm{V}/\mathrm{K}. Increasing W content severely diminishes α\alpha despite the monotonic increase in σ\sigma due to increasing hh-Cr2_2N fraction and W precipitation. These findings indicate that limiting W alloying to below its solubility limit in CrN is crucial for realizing a high power factor.

Conflict of interest

The authors declare no conflict of interest.

Data availability

Data is available from the corresponding author on reasonable request.

Acknowledgements

The authors acknowledge funding from the Swedish Government Strategic Research Area in Materials Science on Functional Materials at Linköping University (Faculty Grant SFO-Mat-LiU No. 2009 00971), the Knut and Alice Wallenberg foundation through the Wallenberg Academy Fellows program (KAW-2020.0196), the Swedish Research Council (VR) under Project No. 2021-03826, and the Swedish Energy Agency under project number 52740-1. NKS and PE acknowledge support from the Foundation Olle Engkvist Byggmästare, grant no. 218-0076. NKS acknowledges Dr Rui Shu for his help during the initial experimental phase. GR gratefully acknowledges funding from the US National Science Foundation grant CMMI 2135725 through the BRITE program. DG acknowledges support from the Swedish Research Council (VR) International Postdoc Grant 2023-00208. BA acknowledges financial support from the Swedish Research Council (VR) through Grant No. 2019-05403 and 2023-05194. The computations were enabled by resources provided by the National Academic Infrastructure for Supercomputing in Sweden (NAISS), at the National Supercomputer Center (NSC), partially funded by the Swedish Research Council through grant agreements no. 2022-06725.

Prior to depositing films with different W contents, CrN films were deposited at different Cr target power PCrP_{\mathrm{Cr}} to preclude the formation of Cr2N\mathrm{Cr}_{2}\mathrm{N} formation and Cr-deficient films. X-ray diffractograms from films deposited with 42 W ≤PCr≤\le P_{\mathrm{Cr}} \le 46 W (Figure S1) show 111 and 222 Bragg peaks from rocksalt CrN indicating a lattice parameter of a0∼a_{0} \sim 4.17 Å [9]. The 0006 peak is from the sapphire substrate. At PCr=50P_{\mathrm{Cr}} = 50 W, we observe large 0002 peaks from hh-Cr2N\mathrm{Cr}_{2}\mathrm{N} corresponding to c∼c \sim 4.48 Å, and a weak CrN 111 peak. Among these, CrN film grown at PCr=P_{\mathrm{Cr}} = 44 W showed the highest thermoelectric power factor, α2σ\alpha^{2}\sigma (Figure S1).

The XRR data was fitted to extract the film thickness, roughness and density (see Figure S2).

Measured XRR data and fits for CrWₓN𝓏 films and extracted film thickness and density

Figure S2. (a) Measured XRR data and fits for CrWx_xNz_z (0≤x≤0.250 \le x \le0.25) films and (b) the extracted values of film thickness and density.

CrWx_xNz_z films with the highest W content of x=0.25x=0.25 showed homogenous contrast from elemental W indicating homogeneous distribution of W and Cr (Figure S3).

References

  1. [1]R. S. Ningthoujam and N. S. Gajbhiye, Synthesis, Electron Transport Properties of Transition Metal Nitrides and Applications, Prog. Mater. Sci. 70, 50 (2015).DOI
  2. [2]D. D. Kumar, N. Kumar, S. Kalaiselvam, S. Dash, and R. Jayavel, Wear Resistant Super-Hard Multilayer Transition Metal-Nitride Coatings, Surfaces and Interfaces 7, 74 (2017).DOI
  3. [3]F. Lévy, P. Hones, P. E. Schmid, R. Sanjinés, M. Diserens, and C. Wiemer, Electronic States and Mechanical Properties in Transition Metal Nitrides, Surf. Coatings Technol. 120–121, 284 (1999).DOI
  4. [4]M. A. Vaslyyev, B. N. Mordyuk, S. I. Sidorenko, S. M. Voloshko, A. P. Burmak, I. O. Krulhov, and V. I. Zakiev, Characterization of ZrN Coating Low-Temperature Deposited on the Preliminary Ar + Ions Treated 2024 Al-Alloy, Surf. Coatings Technol. 361, 413 (2019).
  5. [5]G. Wang, Y. Si, M. Wen, J. Qiu, S. Zhang, Q. Song, W. Wang, X. Yang, and P. Ren, Microstructure, Mechanical and Tribological Behaviors of Hard-yet-Tough Hf-Ag-N Coating, J. Mater. Res. Technol. 22, 2030 (2023).DOI
  6. [6]S. Kerdsongpanya, N. Van Nong, N. Pryds, A. Žukauskaitė, J. Jensen, J. Birch, J. Lu, L. Hultman, G. Wingqvist, and P. Eklund, Anomalously High Thermoelectric Power Factor in Epitaxial ScN Thin Films, Appl. Phys. Lett. 99, (2011).DOI
  7. [7]P. V. Burmistrova, J. Maassen, T. Favaloro, B. Saha, S. Salamat, Y. Rui Koh, M. S. Lundstrom, A. Shakouri, and T. D. Sands, Thermoelectric Properties of Epitaxial ScN Films Deposited by Reactive Magnetron Sputtering onto MgO(001) Substrates, J. Appl. Phys. 113, (2013).DOI
  8. [8]C. X. Quintela, B. Rodríguez-González, and F. Rivadulla, Thermoelectric Properties of Heavy-Element Doped CrN, Appl. Phys. Lett. 104, 1 (2014).DOI
  9. [9]C. X. Quintela et al., Epitaxial CrN Thin Films with High Thermoelectric Figure of Merit, Adv. Mater. 27, 3032 (2015).DOI
  10. [10]M. Akiyama, T. Kamohara, K. Kano, A. Teshigahara, Y. Takeuchi, and N. Kawahara, Enhancement of Piezoelectric Response in Scandium Aluminum Nitride Alloy Thin Films Prepared by Dual Reactive Cosputtering, Adv. Mater. 21, 593 (2009).DOI
  11. [11]F. Tasnádi, B. Alling, C. Höglund, G. Wingqvist, J. Birch, L. Hultman, and I. A. Abrikosov, Origin of the Anomalous Piezoelectric Response in Wurtzite ScAl1-XN Alloys, Phys. Rev. Lett. 104, 1 (2010).
  12. [12]A. Zukauskaite, G. Wingqvist, J. Palisaitis, J. Jensen, P. O. A. Persson, R. Matloub, P. Muralt, Y. Kim, J. Birch, and L. Hultman, Microstructure and Dielectric Properties of Piezoelectric Magnetron Sputtered W-Sc XAl 1-XN Thin Films, J. Appl. Phys. 111, 1 (2012).
  13. [13]G. J. Snyder and E. S. Toberer, Complex TE Materials, Nat. Mater. 7, 105 (2008).
  14. [14]S. Bathula, M. Jayasimhadri, B. Gahtori, N. K. Singh, K. Tyagi, A. K. Srivastava, and A. Dhar, The Role of Nanoscale Defect Features in Enhancing the Thermoelectric Performance of P-Type Nanostructured SiGe Alloys, Nanoscale 7, 12474 (2015).DOI
  15. [15]R. J. Mehta, Y. Zhang, C. Karthik, B. Singh, R. W. Siegel, T. Borca-Tasciuc, and G. Ramanath, A New Class of Doped Nanobulk High-Figure-of-Merit Thermoelectrics by Scalable Bottom-up Assembly, Nat. Mater. 11, 233 (2012).DOI
  16. [16]D. Wines, F. Ersan, and C. Ataca, Engineering the Electronic, Thermoelectric, and Excitonic Properties of Two-Dimensional Group-III Nitrides through Alloying for Optoelectronic Devices (B1-XAlXN, Al1-XGaXN, and Ga1-XInXN), ACS Appl. Mater. Interfaces 12, 46416 (2020).DOI
  17. [17]J. R. Sootsman, D. Y. Chung, and M. G. Kanatzidis, New and Old Concepts in Thermoelectric Materials, Angew. Chemie - Int. Ed. 48, 8616 (2009).DOI
  18. [18]C. J. Vineis, A. Shakouri, A. Majumdar, and M. G. Kanatzidis, Nanostructured Thermoelectrics: Big Efficiency Gains from Small Features, Adv. Mater. 22, 3970 (2010).DOI
  19. [19]P. Eklund, S. Kerdsongpanya, and B. Alling, Transition-Metal-Nitride-Based Thin Films as Novel Energy Harvesting Materials, J. Mater. Chem. C 4, 3905 (2016).pubmed.ncbi.nlm.nih.gov/27358737
  20. [20]S. Kerdsongpanya, B. Alling, and P. Eklund, Effect of Point Defects on the Electronic Density of States of ScN Studied by First-Principles Calculations and Implications for Thermoelectric Properties, Phys. Rev. B - Condens. Matter Mater. Phys. 86, 1 (2012).DOI
  21. [21]S. Kerdsongpanya, B. Sun, F. Eriksson, J. Jensen, J. Lu, Y. K. Koh, N. Van Nong, B. Balke, B. Alling, and P. Eklund, Experimental and Theoretical Investigation of Cr1-XScX/N Solid Solutions for Thermoelectrics, J. Appl. Phys. 120, (2016).
  22. [22]W. Lengauer, Nitrides: Transition Metal Solid-State Chemistry, in Encyclopedia of Inorganic and Bioinorganic Chemistry (John Wiley & Sons, Ltd, 2015), pp. 1–24.DOI
  23. [23]I. Stockem, A. Bergman, A. Glensk, T. Hickel, F. Körmann, B. Grabowski, J. Neugebauer, and B. Alling, Anomalous Phonon Lifetime Shortening in Paramagnetic CrN Caused by Spin-Lattice Coupling: A Combined Spin and Ab Initio Molecular Dynamics Study, Phys. Rev. Lett. 121, 125902 (2018).
  24. [24]O. Jankovský, D. Sedmidubský, Š. Huber, P. Šimek, and Z. Sofer, Synthesis, Magnetic and Transport Properties of Oxygen-Free CrN Ceramics, J. Eur. Ceram. Soc. 34, 4131 (2014).DOI
  25. [25]C. X. Quintela, F. Rivadulla, and J. M. Rivas, Thermoelectric Properties of Stoichiometric and Hole-Doped CrN, Appl. Phys. Lett. 94, (2009).DOI
  26. [26]F. Rivadulla et al., Reduction of the Bulk Modulus at High PressureinCrN, Nat. Mater. 8, 947 (2009).DOI
  27. [27]G. Kresse and J. Hafner, Ab Initio Molecular Dynamics for Liquid Metals, Phys. Rev. B 47, 558 (1993).
  28. [28]G. Kresse and J. Furthmüller, Efficiency of Ab-Initio Total Energy Calculations for Metals and Semiconductors Using a Plane-Wave Basis Set, Comput. Mater. Sci. 6, 15 (1996).DOI
  29. [29]S. Dudarev and G. Botton, Electron-Energy-Loss Spectra and the Structural Stability of Nickel Oxide: An LSDA+U Study, Phys. Rev. B - Condens. Matter Mater. Phys. 57, 1505 (1998).DOI
  30. [30]K. Momma and F. Izumi, VESTA 3 for Three-Dimensional Visualization of Crystal, Volumetric and Morphology Data, J. Appl. Crystallogr. 44, 1272 (2011).DOI
  31. [31]A. M Ganose, A. J Jackson, and D. O Scanlon, Sumo: Command-Line Tools for Plotting and Analysis of Periodic Ab Initio Calculations, J. Open Source Softw. 3, 717 (2018).DOI
  32. [32]P. Ström and D. Primetzhofer, Ion Beam Tools for Nondestructive In-Situ and in-Operando Composition Analysis and Modification of Materials at the Tandem Laboratory in Uppsala, J. Instrum. 17, (2022).DOI
  33. [33]M. Mayer, SIMNRA User’s Guide, Max-Planck-Institut Für Plasmaphys. Garching, Ger. Report IPP, (1997).
  34. [34]K. Arstila et al., Potku – New Analysis Software for Heavy Ion Elastic Recoil Detection Analysis, Nucl. Instruments Methods Phys. Res. Sect. B Beam Interact. with Mater. Atoms 331, 34 (2014).DOI
  35. [35]B. Xin, A. Le Febvrier, L. Wang, N. Solin, B. Paul, and P. Eklund, Growth and Optical Properties of CaxCoO2 Thin Films, Mater. Des. 210, 110033 (2021).DOI
  36. [36]A. S. Botana, F. Tran, V. Pardo, D. Baldomir, and P. Blaha, Electronic Structure of CrN: A Comparison between Different Exchange Correlation Potentials, Phys. Rev. B - Condens. Matter Mater. Phys. 85, 1 (2012).DOI
  37. [37]L. Zhou, K. Fritz, D. Holec, M. Bartosik, and B. Grabowski, Structural Stability and Thermodynamics of CrN Magnetic Phases from Ab Initio Calculations and Experiment, Phys. Rev. B 184102, 1 (2014).
  38. [38]B. Alling, T. Marten, and I. A. Abrikosov, Effect of Magnetic Disorder and Strong Electron Correlations on the Thermodynamics of CrN, Phys. Rev. B - Condens. Matter Mater. Phys. 82, 1 (2010).DOI
  39. [39]L. M. Corliss, N. Elliott, and J. M. Hastings, Antiferromagnetic Structure of CrN, Phys. Rev. 117, 929 (1960).DOI
  40. [41]J. E. G. and I. P. D. Gall, C.-S. Shin, T. Spila, M. Ode´n, M. J. H. Senna, Growth of Single-Crystal CrN on MgO(001): Effects of Low- Energy Ion-Irradiation on Surface Morphological Evolution and Physical Properties, J. Appl. Phys. 91, 3589 (2002).DOI
  41. [42]A. Le Febvrier, D. Gambino, F. Giovannelli, B. Bakhit, S. Hurand, G. Abadias, B. Alling, and P. Eklund, P -Type Behavior of CrN Thin Films via Control of Point Defects, Phys. Rev. B 105, 1 (2022).
  42. [43]P. A. Stampe, M. Bullock, W. P. Tucker, and R. J. Kennedy, Growth of MgO Thin Films on M-, A-, C- and R-Plane Sapphire by Laser Ablation, J. Phys. D. Appl. Phys. 32, 1778 (1999).DOI
  43. [44]M. A. Gharavi, G. Greczynski, F. Eriksson, J. Lu, B. Balke, D. Fournier, A. le Febvrier, C. Pallier, and P. Eklund, Synthesis and Characterization of Single-Phase Epitaxial Cr2N Thin Films by Reactive Magnetron Sputtering, J. Mater. Sci. 54, 1434 (2019).DOI
  44. [45]M. A. Gharavi, D. Gambino, A. le Febvrier, F. Eriksson, R. Armiento, B. Alling, and P. Eklund, High Thermoelectric Power Factor of Pure and Vanadium-Alloyed Chromium Nitride Thin Films, Mater. Today Commun. 28, 102493 (2021).DOI
  45. [46]B. Biswas, S. Chakraborty, O. Chowdhury, D. Rao, A. I. K. Pillai, V. Bhatia, M. Garbrecht, J. P. Feser, and B. Saha, In-Plane Cr2 N-CrN Metal-Semiconductor Heterostructure with Improved Thermoelectric Properties, Phys. Rev. Mater. 5, 1 (2021).

Paper details

vlnc:W_ge885phq
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents