When obtaining an IR spectrum, a secondary amine should exhibit only one N-H stretch, however, when a student ran his sample, two N-H stretches appeared in the spectrum. What could account for this? 

Poor spectroscopic technique?

An impure sample?

 

I was thinking "A shoulder band usually appears on the lower wavenumber side in primary and secondary liquid amines arising from the overtone of the N–H bending band: this can confuse interpretation."