Www.cell.com



PV Reporting ChecklistThe following document provides some considerations followed by a “checklist”. Individuals should consider including the details of a completed checklist as supplemental information for the paper all in one place for the interested readers. Of course, anything that individuals feel is unnecessary or not relevant should be explained. The purpose behind this can generally be summarized by the points below.Want to allow groups to fairly and completely disclose and be able to compare performance of devicesWant to be able to screen known bad practices/sources of error and uncertaintyWant to make sure that critical readers and reviewers can fully understand the resultsIn the event one of the major reasons why the work is of particular interest is due to high (record) performance, then the community should have confidence in the numbers generated from measurements and way in which they were performedDevice area considerations:An aperture measurement should yield the correct Jsc. This value should be reported in a clear manner.Shadowing effects from the aperture will lead to a calculable reduction (ignoring temperature effects) in the Voc and can also alter the fill factor.Mesa isolation negates the need for apertures.Jsc difference with and without aperture establishes how large of a problem outside area collection may be. Multiple device areas can also be used to help explore this issue. Making larger devices should minimize this effect (see Figure 2B).?Evaporation through shadow masks as an area definition method can lead to judgement calls on area due to diffuse edges. Apertures should be opaque and larger than a substrate to prevent waveguiding through the substrate. Ideally an aperture should be made from a rigid material such that can have a well-defined, non-changing opening. Vinyl (e.g. electrical) tape stretches and forms a poor aperture. Device area checklist:What method was used to establish Jsc (e.g. aperture, multiple device areas, mesa isolation)?What are the aperture area and unapertured device areas (with error bars)? How were areas determined? Reference cell considerations:A good reference cell should:Have a well-known QE that is closely matched to the cell under test.Have a light impervious package that minimizes edge and backside light collectionBe stable over timeHave been calibrated by an independent certification labBe recalibrated every year to verify stabilityFor groups testing multiple bandgaps multiple reference cells and/or filter combinations should be strongly considered.A packaged test cell that has been measured by a calibration lab might be used as a way of cross-checking measurements. Such a cell might be stored in the dark and measured periodically (e.g. once a month) against the reference cell.Reference cell checklist:What was used as the reference cell(s)?Was the reference cell calibrated by a certification lab (which one and when)?What were the QE’s of the reference and test cells (plotting them together would be most convenient)?Quantum efficiency considerations:Quantum efficiency should be measured with a white light bias. In the event non-linearity is not measured establishing that the white light bias yields ~0.3-0.4 of the cell’s Isc is a good guess to minimize errors.Integrated QE should be compared to the Jsc as measured by JV. This should match within a few percent Reports of JV characteristics and QE should be from the same cell rather than “representative” cellsThe chopper on a QE system should be placed such that the white light bias does not interact with it.Quantum efficiency should be chopped at a value that will reject line frequency noiseAbsolute QE can only be measured if the incident monochromatic light for all wavelengths is fully contained by the cell areaQuantum efficiency checklist:What white light bias level was used in the quantum efficiency measurement and why was that level selected?What is the integrated QE compared to the Jsc from current-voltage measurement of the same device?Solar simulator calibration considerations:Spatial uniformity calibrations must be performed at a minimum with each bulb change/adjustment. Spectra of the simulator should be measured at different times over the life of bulbs. At minimum representative spectra characteristic of bulbs in the specific (model) simulator with that filter set should be used as guides to understand the level of uncertainty in the measurement. Such spectra can be obtained from the solar simulator vendor if not measured by a particular lab.A spectral mismatch factor should be calculated. In the event that representative lamp spectra are being used, how much the mismatch factor might change could be included.Groups should strongly consider including reference cell QE, solar simulator spectra used to determine mismatch factor (and how the spectra were obtained), and cell QE in supplemental information. Including this information not only makes everything more transparent, it encourages group members to understand the contributions and import of each of these to accurate efficiency measurements.Solar simulator calibration checklist:What error level is associated with the spatial uniformity over the measurement area?What is the source of the lamp spectrum used for spectral mismatch corrections (e.g. representative spectra from a new/old lamp obtained from a vendor, spectrum measured on that specific lamp)?What is the spectral mismatch factor?Current-voltage sweep parameter considerations:Perovskite solar cells often exhibit hysteresis, which affects the determination of the cell efficiency based on current-voltage scans. Hysteresis can have multiple sources such as capacitive charging/discharging, defect related trapping/detrapping, change of field distribution from ion migration. These factors can play individually or collectively to different degrees depending on the actual device stack and materials selection. Conducting current-voltage scans with different sweep directions and rates can provide valuable information about hysteresis. Hysteresis is associated with transient behaviors on time scales over a wide range. Thus, the history of preconditioning prior to current-voltage measurement can affect the results. Devices with no hysteresis can also have a cell efficiency that is not consistent with the result from the current-voltage sweep. Thus, it is recommended to use stabilized power output to check cell efficiency at individual voltage bias point or to use asymptotic sweep to determine the efficiency at the maximum power point. Maximum power point tracking is also useful to avoid the impact from hysteresis to determine the cell efficiency. Current-voltage sweep parameter checklist:What method was used to establish the efficiency (e.g. asymptotic sweep, stabilized power output, low-hysteresis scan)?What were the conditions used for this measurement (e.g. scan direction, rate, step size, stabilization time)?What test conditions were used for this measurement (e.g. ambient, temperature, irradiance)?What was preconditioning (e.g. heat, light, voltage bias, time) were performed and what is the timescale associated with any (meta)stability associated with this process?Describe any other noteworthy behavior (e.g. hysteresis at different scan rates).Certification:If any independently certified results are presented, which lab and provide all relevant information about the test conditions and results. Statistics considerations:Processing and characterization discussions may be most representative of the mean device performance. Hero cells that are outliers may not be representative of this. Sample sizes should be sufficiently large (e.g. sqrt(N) > 3 ) such that the standard deviations is smaller than the difference in reported performance parameters. Statistics checklist:How close to the mean is device data presented and from what size sample set? ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download