The rational combination of tetracene (Tc) with crystalline silicon (c-Si) could greatly enhance c-Si solar cell efficiencies via singlet fission. The Tc/c-Si energy-level alignment (ELA) is thought to be central to controlling the required interface transfer processes. We modified hydrogen-terminated c-Si (H–Si) with 2,2′-(perfluoronaphthalene-2,6-diylidene)dimalononitrile (F6TCNNQ), C60, or NF3 and probed the effect on the c-Si surface chemistry, the Tc/c-Si ELA, the Tc morphology, and solar cell characteristics using ultraviolet and X-ray photoelectron spectroscopy, atomic force microscopy, X-ray diffraction, photoluminescence transients, device measurements, and transfer matrix-optical modeling. Submonolayer interlayers of F6TCNNQ shifted the Tc/H–Si(111) ELA by up to 0.55 eV. C60 showed no notable effect on the ELA and proved detrimental for the Tc film morphology and solar cell performance. Neither F6TCNNQ nor C60 improved the Tc-related photocurrent significantly. NF3 CVD substituted the H-termination of H–Si(100) with more electronegative species and resulted in work functions as high as 6 eV. This changed the Tc/H–Si(100) ELA by up to 0.45 eV. NF3 plasma from a remote source caused pronounced c-Si oxidation and a diminished c-Si photoluminescence lifetime, which was not observed for NF3 plasma created in close proximity to the c-Si surface or neutral NF3. We discuss possible reasons for why the improved ELA does not lead to an improved singlet fission harvest.
