The introduction normally consists of two parts: the placing of one's subject into its context in the rest of physics, and a description of the talk itself.
Start on as general a level as is feasible. If it's convenient (with a general audience), you might begin with some remarks about a recent trend in physics as a whole. Or (with a more specialized audience), you might begin with something about the direction of recent research in a broad area of physics (e.g., General Relativity), or some very general problem toward which a substantial research effort has been directed. Then, very slowly, increase the specialization until you get to the specific subject you're going to talk about. There may be three or four transitions between your starting point and the arrival at your subject. (Even if you feel your audience knows this material already, it is still worth repeating. You must fix in their minds the broad framework into which your subject fits.). If you don't know of any single, natural context for your subject, make one up.
Throughout this discussion, emphasize the types of problems under attack, why they are being attacked, the methods one uses in the attack, the reasons one thinks along these lines, etc. Why does one think about this subject at all? Why is it interesting? What has it contributed to our understanding of Nature? What is the present state of the subject? Where is it going; what can we expect in the future? (Predictions about the future are always a good way to generate enthusiasm.). It is here that one sets the mood for the entire talk. As clearly and forcefully as you can, state what the scope of your subject is, and set that subject in its broader context.
The next step is to reveal the plan of the talk. That is, one says what his three or four messages will be. You might give the title of each message (and, perhaps, write these titles on the board, so you can check them off as each message is delivered), and a few descriptive sentences on each one. Furthermore, one wants to tie all these messages together. How do the various messages relate to each other, and how, taken together, do the messages constitute a summary of your subject? If it can be done conveniently, you might also reveal here what your general conclusions will be. In short, one gives a short talk on the structure of his talk.
This introduction normally consumes about one-fifth of the time available.