Another reason is that except Iron, most of the metals do not show fluorescence with x-ray. For steel we do change the target from Copper to Chromium or any thing that don't show fluorescence.
sorry for correcting you: all elements starting from Li (Z=3) show x-ray fluorescence caused by x-ray excitation.
The fluorescence wavelength/photon energy depends on the atomic number. The question here is, which x-ray target material fluorescence line does not bother the XRD pattern via fluorescence excitation of the sample. This guides us for example to Co or Cr as x-ray tube target material, when Fe is being part of the sample.
This is a fundamental question.... Above mentioned answers are very much correct, but I would like to add a few more physics to it. I think this would bring more clarity.
In X-Ray diffraction, the X-rays are generated by hitting high energy ((45-50)kV & 40mA) and highly accelerated electrons produced from the cathode on a target material known as the anode. As such these highly energetic electrons transfer a lot of energy to the target material and heat it up. You can make a rough calculation that approximately 1.2kW of heat is generated in that process. This limits the choice for the target material to be of high melting point with excellent electrical conductivity.
Only transition metals in the period table fulfill the above requirement. They are highly conductive and are having a high melting temperature.
But for diffraction purposes, it's the wavelength of K-alpha which is used. The K-alpha values (Angstrom) for various transition metals as target materials are,
Cr Fe Co Cu Mo Ag
2.29 1.94 1.79 1.54 0.71 0.56
You can see here the wavelength of Mo and Ag are normally too short for most powder XRD. And if we use short-wavelength, they would scatter weakly and would shift the pattern towards lower Bragg angles which may result in loss of d-spacing accuracy. Moreover, the penetration depth would also be large.
For targets such as Mo and Fe, the wavelength is large which decreases the energy and intensity as well. Further, due to the higher wavelength, the XRD pattern shifts towards a larger Bragg angle which increases the acquisition time. As such due to low energy, the incoming rays may get absorbed by the specimen.
Thus considering the above factors, it is optimized neither to have a high wavelength nor to have a short wavelength. That's why copper is being chosen as the target material. Its melting point is 1093 degrees and is highly conducting.
Again as like mentioned by others, the fluorescence effect plays a role when a copper target is used for iron (Fe) and Manganeese(Mn) containing materials. In order to avoid that, a Co target is used which has also a high melting temp with a K-alpha value of 1.79, pretty close to Copper.
There are couple of reasons for Cu being majorly used as target material compared to any other material.
Firstly, it has low cost. Secondly, copper has very high thermal conductivity. When electrons strikes the target , only small fraction is of it gets converted into X -ray and most of it is lost in heat.Due to high thermal conductivity, copper will disperse the heat in more efficiently and prevent melting.