Amorphous oxide semiconductors are important materials for large-area thin-film devices, where uniformity and low-temperature processing are valuable. Their transport behavior depends not only on which elements are present, but also on how their atomic orbitals connect through a disordered structure. A computational comparison of amorphous indium gallium zinc oxide (a-IGZO) and amorphous zinc oxynitride (a-ZnON) illustrates how different material designs can create different electron-conduction pathways.
The findings are relevant to engineers and technical decision-makers assessing semiconductor materials for thin-film transistors (TFTs), display backplanes, and other large-area electronics. They also demonstrate how structural calculations can help screen amorphous materials, while making clear that calculated transport indicators are not direct substitutes for measurements on fabricated devices.
Why amorphous semiconductors matter
Large-area electronics place competing demands on active-channel materials. Hydrogenated amorphous silicon has historically offered low-cost fabrication and good areal homogeneity, while low-temperature polysilicon can provide higher mobility but may have less uniformity. Metal-oxide semiconductors offer another approach: their largely non-directional bonding can help preserve useful electronic transport even when the material lacks crystalline order.
In a-IGZO, a multication oxide, the conduction band is associated with metal-cation s orbitals. These orbitals are relatively insensitive to changes in bond angle, a feature that supports transport in an amorphous network. However, the presence of several cation species may affect the continuity of the conduction path. A different strategy is to keep one metal cation while combining multiple anions. Zinc oxynitride is an example of this multianion approach.
Different conduction paths in a-IGZO and a-ZnON
Calculations comparing a-IGZO with two a-ZnON compositions show a key distinction in the orbitals that contribute near the conduction-band minimum. In a-ZnON, the conduction-band edge is primarily associated with zinc 4s orbitals. Because zinc is the only metal cation in the material, the proposed transport path runs through overlapping Zn-4s orbitals.
For a-IGZO, indium s orbitals contribute strongly to the conduction-band edge, but they are not the only contributors. The calculations also find significant contributions from gallium and zinc s orbitals. This supports a picture in which electron transport can involve overlapping outer s orbitals from different cations, rather than relying exclusively on indium orbitals.
This difference matters when interpreting composition. In a-IGZO, the mix of metal cations shapes the available connections along the conduction path. In a-ZnON, the metal cation remains zinc while the oxygen-to-nitrogen ratio varies. The study found that the zinc-based conduction path was present in both of the examined anion compositions.
Using orbital overlap to assess transport
An orbital overlap integral estimates how much two electron wavefunctions associated with neighboring atoms overlap. In the materials studied, the authors calculated pairwise overlap between the outer s orbitals of metal cations and combined the results into a volume-normalized value. This provides a structural indicator of how extensively cations may contribute to connected conduction pathways.
For a-IGZO, multiple types of cation pairs contribute to the total. The prominent contributions vary with composition, reflecting the different numbers and arrangements of indium, gallium, and zinc atoms. For a-ZnON, the relevant metal pairs are Zn-Zn. The calculated total normalized overlap was more than twice as high for the examined a-ZnON structures as for the a-IGZO structures.
The comparison is consistent with a-ZnON offering a more favorable calculated pathway for electron transport. The paper also relates this result to previously reported Hall-mobility measurements, which are higher for a-ZnON than for a-IGZO. That correspondence should be interpreted as a trend, not a direct prediction of device mobility: measured mobility also depends on factors such as carrier concentration, processing conditions, and scattering.
What effective mass adds to the picture
Electron effective mass, estimated from the curvature of the conduction band, is another indicator used to discuss transport. In the representative calculations, a-IGZO effective masses were approximately 0.21 to 0.22 times the free-electron mass, compared with approximately 0.16 to 0.17 for a-ZnON. The lower calculated values for a-ZnON point toward faster electron response within the band-structure description.
Effective mass does not explain the full difference in reported mobility. Mobility also depends on scattering, and those mechanisms in amorphous semiconductors are difficult to characterize. Across multiple modeled structures, the researchers observed a strong correlation between effective mass and total normalized orbital overlap. The overlap measure also tracked the mobility trend more closely, though the relationship remained qualitative rather than definitive.
Implications for materials evaluation
First-principles density functional theory (DFT) calculations can provide detailed information about electronic structure, but applying them to complex amorphous systems is computationally demanding. An overlap-based measure can be calculated from structural information at comparatively low computational cost. It may therefore help compare candidate structures or compositions before more intensive calculations and experimental validation.
For engineering teams evaluating materials, the practical value is a clearer link between atomic-scale structure and transport hypotheses. The approach can help identify which cation pairs contribute to a conduction network and compare how changes in composition or structure may affect that network. It does not replace testing thin films under relevant deposition, annealing, and device conditions.
Conclusion
The comparison of a-IGZO and a-ZnON shows that amorphous semiconductors can support electron transport through distinct orbital networks. Mixed-cation a-IGZO involves contributions from the s orbitals of multiple metals, while a-ZnON relies on overlapping zinc s orbitals. Normalized orbital overlap offers a lower-cost structural indicator that correlates with calculated effective mass and follows the reported mobility trend. Used alongside DFT and experimental characterization, it can support more informed screening of materials for large-area thin-film electronics.