Our Program
This program was initiated by the U.S. Air Force and explores the
the way in which meso-strcutured nanocomposites consisting of a nanophase and a conducting polymer host phase, might be utilized within a photovoltaic device. While this has been the goal of many research groups world-wide, our program is distinguished in several ways. The first is the use of nonequilibrium thermodynamics to control the order within an all-organic absorber, and in hybrid nanocomposite absorbers. We have used these methods to create high mobility pathways in the active layer allowing for dramatic increases in performance; with first publications of polymer devices exceeding 4.8%, 5.2% and 5.7% PCE. Most recently our program has teamed with U. South Carolina to integrate multiple exciton generation (MEGS) technology into our thin film devices.
The second area of distinction for our program is the use of novel concentrator systems based on fibner optics that couple confined radiation modes into photovoltaic devices and greatly enhance internal conversion effeciencies in the thiophene systems. FiberCell, as it has come to be called, has the potential to revolutionize organic devices since it is highly compatible with multiple absorbing layers/multiple junction, and multiple quasi-fermilevel approaches to ultra-high effeciency in organics. The Center holds several patents on the architecture as well as the first publications of its performance. FiberCell Inc is a spinoff company from the NANOTECH Center and is funding this work.
Our Facilities
The Nanotech Center maintains 2 solar simulators: an AM1.5g and an AM1.5g Class A solar standard for simulating the solar spectrum. Our standards are manufactured by ORIEL and utilize notch filters to remove atmospheric absorption components of solar spectrum. Both standards are regularly calibrated and are found in the facility's cleanroom. Naturally, the effeciency of the system is only one of a number of performance metrics that need be considered in testing of solar cells. Our facility is setup to simulate angular light gathering power, lifetime and accelerated lifetime of performance, EQE, IPCE, and other metrics.
Our Methology
While there are several standards testing facilties world-wide, such as NREL and Freiburg in Germany, the relatively short lifetime of unencapsulated OPV devices typically studied in labatory settings, along with the quick turn around nature of long term, high volume studies, make it more adventageous for most researchers to rely upon their own calibration methods and then compare results periodically with these national standard sources. How a research group does this is essential to how believable their results might be. We use the method outlined by Emry et al. from NREL.
First we determine the spectral response of our light source using a calibrated photodiode, and a monochromator with a known grating blaze (or response). The measured spectral response is convoluted with the response of the optical instruments to get how much light actually occurs at each wavelength over a spectral window of 200 nm to 1500 nm. This now represents our "standard light curve". The integrated power of the simulator is set to match that of the integrated power of 1 sun illumination for a 1.5 air mass; 1000mW/cm2.
The using our standard response curve as described above, we integrate the power over the absorption window of the device to be tested. Again, this is determined using a calibrated spectrophotometer so that there is little error in the width of the absorption window. The ratio of this power to that of the true AM1.5g (as published) is then taken to yield the spectral mismatch factor. Notice that taking the micmatch facotro only over the part of the spectrum that is in fact absorbed, is important. For the example given below, mismatch between the solar spectrum and the simulator is pretty good up till about 600 nm. For devices that absorb at wavelengths longer than this, the simulator begins to deviate strongly.
Taken from Yang Yang UCLA
If the mismatch is too great (>5%) then it is necessary to perform the test as a function of the integrated power of the source so that the Voc dependence on photon flux can be estimated.
The size of the active area of the device also plays an important role in error. Masking the device such that only the cointacted active area is illuminated, or alternatively estimating the total internal waveguiding of the thin film system using angle dependent measurements is necessary for all de vices under about 1 cm2.
Notice in this example, each strip on the substrate is a different device, thus the active area is quite small in our tests.
Even taking these steps, there are of course errors which are particularly troublesome for devices in which the performance is low (<10%). Many of the sources of these errors are additive in nature, and so relative measurements between standard solar cells (so called reference cells) and the test cells are appropriate when each of the cells are utilizing the approximately same absorption window.
However, when it is estimated that error can be systemic and proportional to an extrinsic property, then absolute measurements can only be meaningful when compared between standards such as those at NREL etc.
Example: An example of these issues found in the use of our standard is when we measure a mismatch of 15% or so over the spectral region of the active layer we are studying. By running the PCE at 0.25, 0.50, 0.75, and 1 sun, we can determine how the Voc might behave as a function of photon flux. For relatively high mobility devices, we see little change in the Voc and so report the number we achieve for effeciency with one sun illumination after the current value has been corrected by the required amount. This number is of course in error slightly, but represents a reasonable number with which to compare to other devices within our study. However, for lower mobility devices, the Voc value might vary significantly, and these measurements can not be reported directly. A more accurate simulator such as found at NREL, but not typically found at universities, must be used.