The 'photon energy' is hn, and the momentum is hk, where n is the frequency, and k is the wave vector. This is the standard association to plane electromagnetic waves. Since a definite frequency is a property of an infinite monochromatic wave, and the definite k-vector is the property of a wavefront of infinite transverse extent, this association defines an extremely nonlocal object. I am happy to see Lamb's paper quoted, where several misconceptions concerning the word 'photon' is discussed. To be very short, the basic point is that, to discreteness many physicist associate 'granularity' and sharp localization in space. (The temporal extent then is connected to the spatial one, by dividing the latter by c, the velocity of light.) A pulse of, say, finite duration is a many-mode and many-photon field, containing Fourier components, whose (disrete) excitation degrees are the photon population numbers. Lamb stresses to keep in mind that photons are not just localized packets, but in general they are excitation of the natural modes satisfying the proper boundary conditions, set by the experimenter (experimental apparatus). My answer is NO, to the question >> "pulses of single photons" is it a correct use of the word photon?

Thanks Mr. Sandor Varro for your nice comment. Surely I will go through the references suggested by you for more detailed understanding.

Please check (http://www.pivotalconceptsinscience.com)

for a proposed structure for the photon phenomenon.

Thanks Mr. William for providing link.

You have to read about poissonian, sub-poissonian and super-poissonian distributions of photons to introduce your work conditions much more precisely. This is a matter of quantum optics and some detailed discussion before making such question would be of benefit

I must confess to having some difficulty with this question, so I'll provide some answers that might help.

First of all, "pulses of single photons" is absolutely correct terminology, and refers (as Oday Hamadi alluded to) to sources that have sub-Poissonian statistics. The conventional meaning is that at any given instant of time, the maximum number of photons that can be detected is one. This is what one expects for a source that continuously emits single photons, for example a single atom that is continuously pumped and can never emit more than one photon at a time.

However "pulses of single photons" is more likely to refer to a pulse train where every pulse (that is finite in time) contains one and only one photon. Again, such sources exist, and new technology is making them more efficient. (In practise such sources usually have negligable two photon rates to be useful, but will often have large zero photon probabilities).

Now the above is all a statement of the number of photons. This is actually different from the next part of your question, about time/bandwidth effects. Even classically, time-bandwidth plays an important role. So it is certainly true that a single pulse of light, irrespective of the number of photons it contains, cannot be single frequency. Naturally this also means that the energy of a given photon will also have an uncertainty associated with it, but this does not mean that the number of photons has somehow an increased uncertainty from what the pulse would have if the temporal extent of the pulse were altered.