CpMn(CO)(3) (Cp = eta(5)-C5H5) is impregnated into polyethylene (PE) film using either supercritical CO2 (scCO(2)) or n-heptane. UV photolysis of the impregnated film at 21 K leads to formation of the coordinatively unsaturated intermediate CpMn(CO)(2), with nu(C-O) bands similar to those observed in solid CH4 matrices and frozen hydrocarbon glasses. By contrast, UV irradiation at room temperature leads to the formation of a stable species containing the CpMn(CO)(2) moiety. This compound is polymer-bound and is difficult to extract by either scCO(2) or n-heptane. On a semiquantitative level, the overall yield of this compound, identified as CpMn(CO)(2)(eta(2)-C=C), increases with the degree of unsaturation of the PE sample; larger amounts are formed in low-density PE than in Hostalen GUR 415 PE, which has a very small degree of unsaturation. The formation of the polymer-bound species can be suppressed completely by high pressures of CO. In the presence of other gases, e.g., H-2 or N-2, there is competitive formation of CpMn(CO)(2)(eta(2)-C=C) and CpMn(CO)(2)(X(2)) compounds (X = H or N). A more detailed analysis of the IR spectra, obtained with a range of selected samples of PE, shows that the precise wavenumbers of the nu(C-O) bands of CpMn(CO)(2)(eta(2)-C=C) depend on the relative amounts of vinyl (terminal), pendant, and internal C=C bands in the particular PE sample. This assignment has been confirmed by a new technique whereby model CpMn(CO)(2)(eta(2)-alkene) complexes are generated in situ in Hostalen GUR 415 PE and their nu(C-O) spectra are compared with those of the corresponding polymer-bound species. It is shown that the nu(C-O) bands of CpMn(CO)(2)(eta(2)-C=C) provide a convenient diagnostic tool for establishing the unsaturation of different PE polymers. The different CpMn(CO)(2)(eta(2)-C=C) complexes studied have significantly different reactivity toward, for example, N-2 (internal > pendant > vinyl), which can be rationalized on the degree of tau back-donation. Although PE has considerable promise as a matrix for low-temperature experiments, the unsuspected presence of olefinic C=C groups in a particular sample of PE will give rise to artifacts. Our results provide a simple and rapid means of identifying such samples (and avoiding them) before the start of more complicated low-temperature experiments.