If not mistaken, it might be out-of-phase stretch of P-O-C in a phosphorylated starch (P-O-C6H5)!

but there is no addition of phosphate in grafting reaction. I'm only used KPS as radical initiator and Diallyldimethylammonium chloride as monomer. is there possibility that these peaks are peaks of C-O bond streching towards all grafted samples indicate formation of bond between starch and polyDADMAC resulted from grafting process

Yeah it can be due to grafting because P-O-Ar (aromatic) gives two bands: O-C at 1242-1110 cm-1 and P-O at 996-905 cm-1. So the first peak in your spectrum (916 cm-1) is due to grafting and the second one (1016 cm-1) is related to starch molecule itself.

Check the attached spectrum of a dye (Polydiallyldimethylammonium chloride) grafted with polyDADMAC.

Thanks Abdulrahman for the attachment. I think (1016 cm-1) is due to grafting while (916cm-1) is related to starch. But in starch, there also have ether unit as shown in attachment. I'm also observed that there is absence of peak at 1337 cm-1 for OPT sample and appeared of peak 1365 cm-1 (second attachment). seem it shifted. Do we need to consider/highlight this difference in our discussion?

Second attachment

Check out this paper "Application of response surface methodology (RSM)

for optimization of cassava starch grafted

polyDADMAC synthesis for cationic properties" ... I have attached it, it contains similar discussion. According to the paper: 1016 is related to pure starch while the other one 916 doesn't appear in pure starch spectrum.

Could those be from HOOPs (hydrogen out of plain) modes? (Sorry never worked on starch :-)

Might show the crystallinity index