Electrochemistry of semiconductors & metal-macromolecule interfaces

Our goal is to provide industry with test structures and models of next-generation photovoltaics, with an initial focus on CdTe and CuInxGa1-xSe2 (CIS or CIGS) materials. These tools will enable interpretation of measured external properties affected by geometry, grain structure, and nanoscale phase separation, which will support improved processing and design of Second Generation (thin film) and Third Generation (nanostructured) photovoltaic devices.

One of the most attractive properties of CdTe and CIGS for photovoltaic applications is that they are amenable to a variety of cost-effective fabrication methods, including sputtering, printing, and electrodeposition.  We are adapting our experience with the electrochemical “superfill” of metal in sub-micrometer-scale trenches and vias to the fabrication of novel photovoltaic structures.  We are developing instrumented photovoltaic test structures using electrochemical deposition to determine the role that various geometries and internal defect structures have on efficiency and other photoelectrical properties.

To guide and interpret the experimental measurements, we are developing models of semiconductor carrier transport that allow for arbitrary 2D and 3D geometries.  We will model the phase separation of CIGS photovoltaic alloys under different deposition conditions and simulate the photovoltaic behavior of devices made with these predicted microstructures.  Atomistic simulations are underway to gain an atomic scale understanding of measured properties and provide key information about local structural changes, preferred adsorbate locations, and surface stress due to interactions between electrodes and charged particles.  They will also provide interfacial materials properties that are unavailable or unobtainable by experimental measurement.

CdTe and CIGS are some of the most stable and efficient photovoltaic materials. A wide variety of deposition methods enable novel device structures at previously unobtainable dimensions, but optimal structures and dimensions are unknown.  Sensitivity of the micro-structure (and, ultimately, the device efficiency) to deposition methods and processing conditions varies, and potential new and cheaper fabrication methods have not been verified. Furthermore, much of the efficiency of CIGS is due to microstructure (nanoscale phase separation and grain boundaries) that is not well understood.  Quantities as fundamental as the position of the pn junction, the electrical characteristics of grain boundaries, and even the existence of some thermodynamic phases, are as yet unmeasurable.  

We want to examine the effect of microstructures obtained from rigorous thermodynamic and kinetic modeling. As a case in point, Stanbery et al. have proposed an “intra-absorber junction” model to explain the observation that defective CIGS material has better photovoltaic properties than perfect crystals. The notion is that nano-domains of Cu-rich and In-rich material form a distribution of pn junctions throughout the absorber, permitting electrons to be collected before being lost to recombination. We performed preliminary simulations of segregated n and p nano-domains that illustrate the plausibility of the mechanism. Future work will use information from the Thermodynamic & Kinetic Data for Sustainable Energy project to evolve realistic phase-separated structures and then calculate photovoltaic properties based on the observed phases, structures, and alloy concentrations. We will work with colleagues in industry and at the National Renewable Energy Laboratory to obtain the crucial materials parameters and to validate the model against experimental observations.

We have implemented a photovoltaic device model that gives us complete control over geometry and microstructure. Studies are underway to determine the optimal trench geometry for photovoltaic efficiency before undertaking the costly and time-consuming process of fabricating actual devices. Where available, we are validating our predictions against other simulations and experimental data.

We have developed a CdTe electrodeposition capability and are currently refining the incorporation of dopants while plating. Through a collaboration with CNST we are fabricating devices that use interdigitated electrodes both for independent electrode-position of n- and p-type material during fabrication and for collecting the photo-generated current in the fabricated devices. The devices will enable quantitative evaluation of bulk and interface properties of 3-d devices through controlled variation of length scales independent of the absorber

thickness including electrode pitch and cross-section. X-ray diffraction and scanning electron micros-copy reveal highly textured, crystalline CdTe. Optical and electrical measurements reveal photovoltaic behavior consistent with the irregular pn junction in these preliminary materials. Ongoing work will focus on improving deposit morphology and coupling experimental results for the impact of the different length-scales enabled by these template substrates with analysis and modeling.

current transients for electrodeposited CdTe

EQE of interdigitated CdTe photovoltaic devices