Quantifying triplet formation in conjugated polymer / non-fullerene acceptor blends

Triplet formation is generally regarded as an energy loss process in organic photovoltaics. Understanding charge photogeneration and triplet formation mechanisms in non-fullerene acceptor blends is essential for deepening understanding of...

. Normalised intensity to show the correlation between number of charges generated (Q, acquired via integration of the JV curve) and the NFA anion peak amplitude variation with applied voltage.

Quantifying triplet populations
When quantifying the triplet formation in the blend films, the triplet molar absorption coefficient was calculated using the sensitisation method. This method involves a triplet sensitiser of known triplet molar extinction coefficient being directly excited, thereafter undergoing a triplet energy transfer to create a lower energy triplet on an acceptor of unknown triplet molar extinction coefficient. After excitation of the donor/sensitizer, the following reactions are therefore relevant: The overall decay of the donor triplet is therefore . Using microsecond transient absorption spectroscopy, the unknown triplet-triplet molar absorption coefficient can be obtained by comparison with that of the donor compound by equation 1. [1] (1) where is the triplet extinction coefficient for known donor compound, is the triplet extinction coefficient for unknown acceptor material. ΔOD D is the maximum absorbance reached for the triplettriplet absorption peak of the donor and ΔOD A is the maximum absorbance of the acceptor triplet at their respective wavelength maxima.
However, equation 1 is only relevant under conditions where k Q is much greater than k 1 and k 3 . If this is not the case, then additional corrections have to be applied, as detailed in the method of Land et al. Note that Land's additional correction for 3 A when it is also formed via direct excitation is negligible in our case. [1] If decay of the acceptor triplet occurs simultaneously with its creation via energy transfer, a maximum in OD will form, as observed in our results. In such cases, considering the kinetics of successive reactions enables the following equation: Where is the maximum OD observed in the acceptor triplet absorbance in the presence of ∆ the donor.
Finally, a correction must also be applied for 3 D decay during the energy transfer process, necessary in situations where k 1 is non-negligible. Since this is the case for ZnTPP, a correction factor of 2 1 + 2 is applied.
For this project, ZnTpp acts as the triplet sensitiser with triplet-triplet molar absorption coefficient 9500 M -1 cm -1 . The TA spectra for pristine ZnTPP and the ZnTPP/NFA sensitised solutions are shown in Figure S15, and the corresponding kinetics in Figure S16. The triplet extinction coefficients for all NFAs are calculated as shown in Table 1 (main text).
To calculate the triplet-triplet molar absorption coefficient, ITIC-2F is used as an example. Figure 15e shows the TA spectra of ZnTpp:ITIC-2F solution. The ZnTPP triplet is located below 900 nm, while the ITIC-2F triplet absorbs maximally at 1150 nm. The ITIC-2F clearly grows in over time, a hallmark of the sensitisation process. The kinetics of the ITIC-2F triplet growth and decay were probed at the maximum of 1150 nm (Figure 16c). Assuming first order behaviour such that k 2 = 1/t 2 , the time constant of the rise (3.2  10 -6 s) provides k 2 and the decay provides k 3 (1.5  10 -5 s). [2], [3] The ΔOD max of the ITIC-2F triplet is 7.8  10 -4 . Applying the corrections detailed above, the corrected ΔOD max is 1.5  10 -3 . So the triplet molar extinction coefficient is = 3.9  10 4 = ∆ × ∆ = 7.8 -4 × 9500
The triplet yield in pristine solution can be calculated using the Beer-Lamber Law A=cl, where  is the triplet extinction coefficient, c is the concentration of solution, and l is the path length of the cuvette.
The extinction coefficient of the triplet may differ in a film sample. To address this, we utilise a correction factor methodology by considering the change in absorption cross-section, , for the ground state from solution to film. The molar extinction coefficients of the ground state ( GS_soln ) of each NFA have been measured in chlorobenzene with three different low concentrations (10 -6 -10 -5 M range), where  GS_soln is established from the gradient of a plot of absorbance vs concentration. Absorption cross-section is then calculated using the known relationship: Where  GS-soln is the absorption cross-section in solution and N A is Avogadro's number; also noting that an additional factor of 1000 is required for unit equivalence. Film absorption coefficients () are also converted to absorption cross section ( GS-solid ) via  GS-solid = /N i , where N i is the number density of the initial state. N i for the ground state can be established from the mass density and molecular weight.
The following equation is then applied to establish the correction factor for each NFA: Where A is the correction factor. The values for  GS_solid have been obtained from literature, except for ITIC-Th, for which none could be found and thus an extrapolation procedure was used. The correction factor is then applied to the  of the NFA triplet, assuming that any change in  of the ground state from solution to solid will occur similarly for the triplet state. For triplet formation in the NFA blend films with PffBT4T-C9C13, we can then apply the estimated film triplet  using the following equation: 1000. .
Where n T is the triplet yield in cm -3 , OD is corrected for absorbance, A v is Avogadro's number,  T is the molar extinction coefficient of the triplet, and d is the film thickness. solution with an excitation energy at 12 µJ cm -2 , (e) ZnTpp:ITIC-2F (7  10 -6 M: 7  10 -6 M) solution with an excitation energy at 12 µJ cm -2 , (f) ZnTpp:Y6 (7  10 -6 M: 7  10 -6 M) solution with an excitation energy at 13 µJ cm -2 . All the spectra were obtained with an excitation wavelength at 426 nm.