How will it make a difference if gamma spectroscopy is carried out using powder rather than crystal. I think resolution will be poor but why I am not able to make it out.
A short try of an answer. If you use scintillation detectors, its important not to reabsorb or to scatter the emitted light. The statistical losses will broaden your peaks, ergo worse resolution.
In normal gamma spectroscopy one uses a single crystal placed inside a container with walls which are excellent reflectors or scatterers of the emitted light by the crystal. The light is emitted by the impinging gammas on the crystal. Here, the idea is to collect all the emitted light to strike the photo cathode in an attempt to produce the highest possible electrical pulse. However, when using a powder you certainly produce the same amount of light but this light will diffuse and get adsorbed before reaching the photocathode, so practically speaking you are drastically attenuating your pulse. So no wonder if you do not measure any reasonable pulse when using a powder.
As Dr. Moreh stated you want a single crystal. If you have a set of crystals, either in air or vacuum, you are adding a different index of refraction, N, and thus reflection will occur. A great example of this is BGO wherein its high N does not match the N of the glass of the photomultiplier tube and you have losses. In an ideal world, you want everything to match N's (see N of CaF(Eu) for example). Finally, the only reasonable powder is ZnS(Ag) or better ZnS(Cu) which has self-absorption problems.
The answers by Hanno, Erik, Raymond, and Phillip are correct. The multiple surfaces of a powder cause multiple refections of the emitted light at the multiple surfaces. Note that a black powder can appear to be white, that is why a scratch on a black surface is particularly visible. Many single crystal detectors use a powder as the reflection medium.
Bicron made shock resistant NaI detectors of various shapes by forming the detectors using granules of NaI. The gaps between the granules were filled with a plastic having the same index of refraction as NaI. The resolution and efficiency was very close to single-crystal NaI.
A scintillator powder must have a filler with matching index of refraction if used as more than a thin layer.