We observed that the annihilation radiation photopeak in a Ge(Li) detector is considerably broader than that of a $\gamma$ ray of the same energy. It seems reasonable to assume that the increased width is the result of the Doppler broadening of the annihilation photopeak, i.e., the longitudinal Doppler shift of the radiations is measured, while the transverse shift is measured in the usual angular-correlation experiments. By using a computer stripping program to remove the distortion produced by the finite energy resolution of our detector, we obtain momentum distributions in agreement with those which have been published. Only one detector is necessary for these measurements, and all momentum channels are detected at once. However, the detector energy resolution severely limits the momentum resolution. Arguments are presented which indicate that no very large improvement in Ge(Li) detector resolution can be expected.

• Received 10 November 1967

DOI:https://doi.org/10.1103/PhysRev.170.351