dear shami I suggest you take a look up at your primer pair: if they're dimming, if they have a lot of C-G concentration or the hybridization temperature is ideal because that second band looks like a non-specific amplification.

The primer dimer is probably the smallest blurred band on the gel. As you have run an annealing temperature gradient I think that you should try a DMSO gradient. Run one sample at a final dmso concentration of 0%, 1%,2%.......8% dmso and what often happens is that the non specific band gets fainter with increasing dmso and often you get a clean amplification. At very high dmso the pcr will fail but you may find an amount that works well for you. Any grade of dmso will work.

You could also try using less primer as the smallest blurred band looks like either too much primer or primer dimer

1. Non-specific binding

2. Problem in gel loading

3. Issues in primer design

You should run no templete control , positive control and also negative control.@Shami Álvarez