On the reliability of determination of energies of HOMO levels in organic semiconducting polymers from electrochemical measurements

Relationship of Ionization Potential and Oxidation Potential of Organic Semiconductor Films Used in Photovoltaics

“…The scatter in the data is also large, with R 2 values of 0.84 and 0.86 for first and second scan data plots, respectively. Again, this is better than the literature study of conjugated polymers ( R 2 = 0.72) as expected, but significantly worse than for the metal‐organic complexes ( R 2 = 0.98) in the study by D'Andrade et al that followed the same comprehensive experimental approach. The latter points to a larger variation in polarization energies between the different conjugated polymers as compared to the conjugated metal‐organic complexes following the model of Sworakowski and co‐workers, and hence suggests a more limited use of CV‐derived energy levels in organic solar cell design compared to organic light emitting diodes.…”

“…The latter finding is very important, as in order to convert CV values to UPS IP (or IPES EA), one not only needs to obtain the slope and intercept for the particular class of organic molecules or polymers, but also evaluate how closely they follow the linear dependence, as large (random) deviation would make CV data unreliable for use in energy level tuning of heterojunctions. In a recent paper comparing literature values of IP and E ox values for a series of conjugated polymers, a linear fit with α + = 1.7 ± 0.2 and β + = 4.6 ± 0.1 eV was constructed, but with significant scatter in the data points ( R 2 = 0.72), clearly too large to be of use for detailed energy level alignment design in bulk heterojunction solar cells. The large scatter/poor correlation is partially expected as both the determination of the IP from UPS data and E ox from CV data vary significantly between groups due to e.g., choice of calibration, experimental settings, sample preparation, and data analysis methods, so vastly more reliable data and better fits likely are to be obtained if the same researchers obtain the IP and E ox values on the same materials, following the same procedures and under the same conditions.…”

“…It has been found that for molecules dissolved in solution, the slope ( α + ) is solvent dependent and is smaller in low dielectric solvent . In general, it is reasonable to assume that polarization effects and hence the slope in the CV experiment differs for a solid polymer film compared to molecules/polymers dissolved in solution . Nevertheless, to better understand the processes involved, we take the data set from D'Andrade and fit the purely organic molecules and metal‐organic complexes separately, see Figure S9a–c, Supporting Information.…”

Relating Frontier Orbital Energies from Voltammetry and Photoelectron Spectroscopy to the Open‐Circuit Voltage of Organic Solar Cells

“…Fc/Fc + is one of the redox couples recommended by the International Union of Pure and Applied Chemistry for reporting energy levels in nonaqueous solvents. [36][37][38][39][40] This would alter the proportionality constant between E CT and (E HOMO,D − E LUMO,A ). [35] Recently, several experimental studies of the oxidation potential as obtained from CV (E ox,CV ) and ionization potential from UPS (IP) have revealed that both quantities are linearly correlated, i.e., IP = α + qE ox,CV + β + , but that the best-fit slope varies considerably from α + = 0.9 to 1.5, depending on the material class.…”

“…Recently, several experimental studies of the oxidation potential as obtained from CV ( E ox,CV ) and ionization potential from UPS (IP) have revealed that both quantities are linearly correlated, i.e., IP = α + qE ox,CV + β + , but that the best‐fit slope varies considerably from α + = 0.9 to 1.5, depending on the material class . This would alter the proportionality constant between E CT and ( E HOMO,D − E LUMO,A ).…”