This was my second time attending the Photovoltiacs Specialists Conference. My PhD research focus is the use of lifetime measurements to characterize and identify PV efficiency-limiting defects in PV materials, and so I was excited last year to learn that PVSC had Area 8 dedicated to talks and posters on emerging characterization methods. I was anticipating another year of great characterization content, and the  PVSC did not disappoint.

The characterization sessions at this year’s event were well-attended, and not only amongst the people developing new characterization methods. Area 8 served as a valuable source of information for researchers looking to find a useful characterization technique for a problem that they are facing. Area 8 also provided a platform for interaction between researchers developing different characterization methods. There are always many different approaches to the measurement of a material property, and it is important that we can reconcile the results from various techniques.

In his talk, “Dual Sensor Technique for the Advanced Characterization Of Recombination Parameters in Photovoltaic Materials“, Dr. Richard Ahrenkiel gave insight into the nature and origin of discrepancies between lifetime measurements made using photoluminescence and conductance decay methods by simultaneoulsy applying both techniques to the same sample. Dr. Ahrenkiel showed that the mobility is not a constant with injection level, and that it introduces significant systematic error to lifetimes measured using photoconductance decay. By using this discrepancy between the two techniques, and combining measurements from both methods, Dr. Ahrenkiel was able to calculate the changing mobility with injection level.

Dr. Darius Kuciauskas expanded on the technique he proposed at last year’s  PVSC using two-photon absorption to achieve better estimates of the bulk carrier lifetime in his talk in Denver, “Analysis of Minority Carrier Lifetime and Surface Recombination Velocity in CdTe by Using Time-Resolved Photoluminescence with One-Photon and Two-Photon Excitation“. In this talk, he explained how effective lifetimes in CdTe films were measured to be only several nanoseconds long; however, using this new method he was able to show that the bulk lifetime is actually 350ns. This improvement in spatial specificity of the photoluminescence decay lifetime technique arises from the difference between one- (above bandgap) and two-photon (sub-bandgap) absorption.

There are problems with the traditional one-photon absorption lifetime method that has been conventionally employed. Since absorption occurs so close to the front surface, the measured lifetime is very sensitive to surface recombination, and also screening by diffusion away from the front surface. In order to generate carriers away from the front surface, Dr. Kuciauskas uses the simultaneous absorption of two sub-bandgap photons as his excitation mechanism.

Typically, sub-bandgap photons transmit through a material; however, there is a probability that two photons can be simultaneously absorbed to generate an electron-hole pair across the bandgap. The probability of this process is typically very low; however, it is a nonlinear process and the probability increases with the square of the light intensity (since two photons are involved). Greater homogeneity can be attained using this weaker type of absorption, yielding significantly better estimates at the bulk lifetime.

Since the rate of two-photon absorption depends on the square of the light intensity, the excitation volume can be localized within the material by focusing the laser to a controlled depth. Since the excitation volume is compressed spatially by this focusing, there is a tradeoff between the resolution of this technique and the degree of diffusion that needs to be accounted for. Fortunately, Dr. Kuciauskas showed that diffusion can be modeled fairly reliably in the bulk (due to approximate spherical symmetry) than at the surface, and the local lifetime can be measured by fitting the photoluminescence decay results with a model that accounts for diffusion and recombination.

When dealing with the measurement of an effective lifetime that combines the surface and bulk lifetimes, it can be very hard to separate the participating mechanisms. This becomes increasingly difficult when trying to study materials with high surface-area to volume ratios (e.g. materials with small grains) as it is harder to place your sampling volume away from the surfaces. Dr. Kuciauskas’ work is focusing on overcoming these barriers, so that we can separate the characteristics of these mechanisms. Other physical processes, like the injection-level dependent mobility that Dr. Ahrenkiel explained, screen our ability to measure the lifetime with accuracy, and being aware of these processes we can reduce systematic error in our measurement systems. By improving the specificity and accuracy of our measurements of carrier dynamics, we make progress in understanding and probing the physical mechanisms that hold us back from reaching the maximum photovoltaic efficiency limits.

Martin Gerber 5^th Year PhD Student Engineering Physics McMaster University

Martin Gerber
5th Year PhD Student
Engineering Physics
McMaster University