for sure, yes. But be aware of the following. The emission signal level (or amplitude) on the transducer output would be caused by different attenuation frequency functions in your steel and your composites. In the steel attenuation is low, the emission high frequencies would be attenuated weakly if compared to emission lower frequencies (due to weak attenuation frequency function dB/(cm MHz)). Therefore emission source and emission pik-up transducer transducer could be separated by longer distance. In composite attenuation is high, the emission high frequencies would be attenuated strongly if compared to emission lower frequencies (due to strong attenuation frequency function dB/(cm MHz2)). So in composite: firstly the emission signals would be attenuated after short propagated distance, and the secondly the higher frequencies would be attenuated more rapidly than lower frequencies of broadband emission spectra. These phenomena You should consider when choosing the frequency of transducer.
I agree with Rytis and I would like to add the following; In case you want to identify failure mechanisms (matrix cracking, fibre breakage etc) in composites you may need to use resonant sensors in order to amplify your resolution in a specific narrow band frequency.