Design Meets Nature: Tetrahedrite Solar Absorbers

Abstract

used to guide the selection and design of new absorbers using computational techniques. This metric captures the leading physics of absorption relevant to PV efficiency, and it improves upon the simple Shockley-Queisser bandgap model [ 7 ] by considering the full absorption spectrum of the absorber. Hence, SLME should be effective for guiding the selection and design of new ultrathin absorber materials for study and use in drift-aided cells. We have applied the SLME approach to the analysis of all ternary chalcogenides containing Cu, including materials in the system Cu-V-VI. [ 9 ] Several materials were found to exhibit SLME values much higher than those of conventional thin-fi lm absorbers.Subsequently, two general principles were formulated to guide the selection and design of new high-performance absorbers. [ 9 ] In the fi rst case, structural isolation of a Group-V atom in its 5+ oxidation state, e.g., Sb 5+ , within a Cu-rich matrix leads to a narrow Group-V-derived s-orbital band at the conduction band minimum (CBM). In combination with a narrow Cu d band at the valence band maximum (VBM), a high joint density of states and a high absorption coeffi cient ensue. In the second case, if the Group-V atom exists in the lower 3+ oxidation state, e.g., Sb 3+ , it adopts an ns 2 valence electron confi guration. In this 3+ oxidation state, highly asymmetric coordination environments can lead to long distances between the Group-V atoms, resulting in fl at bands and attendant high joint densities of states. These features contribute to an abrupt absorption onset and a high absorption coeffi cient. Strong absorption is reinforced by the parity-allowed nature of the electric-dipole V s or Cu d → V p transition. Hence, the second design principle focuses on structurally isolating the lower oxidation state Group-V element in a semiconducting matrix. Results and DiscussionThese considerations have led us to consider the mineral tetrahedrite, i.e., Cu 12 Sb 4 S 13 , as the basis for a new family of solar absorbers. This material was not revealed as a candidate in the initial SLME search [ 9 ] of Cu-V-VI materials because it was computed to have a zero bandgap, i.e., it is a metal and not a semiconductor. This metallic behavior can readily be appreciated by considering the mixed oxidation-state nature of Cu 12 Sb 4 S 13 . A mixture of formal oxidation states Cu 2+ (d 9 ) and Cu 1+ (d 10 ) is required for charge neutrality, i.e., 10 of the 12 Cu atoms in the chemical formula Cu 12 Sb 4 S 13 are monovalent, while the remaining two Cu atoms are divalent. An incompletely fi lled d-valence band is expected to give rise to the observed degenerate (metallic) behavior, [10][11][12][13] consistent with the Computational inverse design and consequent experimental results allow for the identifi cation of new tetrahedrite-mineral compositions as promising absorber candidates in drift-based thin-fi lm solar cells. In device simulations, cell effi ciencies above 20% are modeled with absorber layers as thin as 250 nm. These new compositio...

Thermoelectric properties of Co substituted synthetic tetrahedrite

“…with H D is Debye temperature, m * (=1.3 m e which was estimated from the band structure as reported [32]) being the carrier effective mass, e the carrier charge, k B the Boltzmann constant, ⁄ the reduced Planck's constant, T the temperature, and n the charge carrier density, respectively. The carrier concentration values are estimated using the room temperature Seebeck coefficient data for all the compounds.…”

Selection Metric for Photovoltaic Materials Screening Based on Detailed-Balance Analysis

“…This has frequently not been implemented correctly and has led to confusion and mistakes in the past [37][38][39][40][41][42][43][44][45][46][47][48][49]. The internal and external parameters used in the scope of this work are listed in Table I as well as the equations that connect an internal parameter with its external counterpart.…”

“…One of the first and maybe most prominent examples of such a selection metric proposed for computational materials screening is presented by Yu and Zunger in 2012 [37] and has been widely used to estimate efficiency limits in the last years [38][39][40][41][42][43][44][45][46][47][48][49]. In their paper, they proposed a "spectroscopic limited maximum efficiency" SLME selection metric that aims to calculate efficiency limits for non-step like absorption coefficients beyond the radiative limit.…”

Accelerated Discovery of Efficient Solar Cell Materials Using Quantum and Machine-Learning Methods

“…Note that the SLME uses an exponential function to model the fraction of radiative recombination as given in Eq. (6). As a consequence, the SLME quickly goes to zero as the difference between the direct allowed and the fundamental band becomes larger.…”