Why is nitrogen not used as a pentavalent impurity? I don't understand why nitrogen though a fifth group element is not used as a pentavalent impurity in doping a semiconductor.
 A: A good dopant in a semiconductor requires a suitable energy level (donor/acceptor), high solubility (so you can put lots in), low vapor pressure (so it doesn't disappear during processing), and suitable diffusivity (so it stays generally where you want it during processing).  
Table 5.2 in Shimura's 'Semiconductor Silicon Crystal Technology' indicates that the maximum solubility of N in Si is $5 \times 10^{15} / cm^{3}$.  Compare that to the commonly used dopant P, which has a maximum solubility of $1.3 \times 10^{21} /cm^{3}$.  Those 6 orders of magnitude in solubility pretty much tell the whole story - you cant get enough in to be technologically useful, even before considering the energy level.
For completeness, however, lets look at the energy levels. Murakiami et al, Jpn. J. Appl. Phys. 27, L1414 (1988) measure that off-center substitutional N has an energy level 0.33 eV below the conduction band. Again, compare that with P where the donor level is 0.045 eV below the conduction band (Fig. 13 in Sze). 
All together, N is a horrible choice for a dopant in Si.  Sulpher would be a better choice than N, having both higher solubility (by an order of magnitude) and a donor state closer to the band edge (0.26 eV below). 
A: Nitrogen-vacancy (NV) point defects are e.g. known in diamond. These create luminescence in the visible.
I think it's more a question of the technology. N2 usually is not very reactive. You would have to use other precursors or crack N2, which requires high temperatures. Otherwise N is not built into the crystal.
