Thanks Alaa A. A Alijabali for the answer. I understand that when the size of the nanoparticle increase, absorption is red-shifted. What is not so intuitive to me is, one observes a shoulder instead of a peak at bluish wavelength. 

Have a look at the attached image. for 1.9, 2.5 and 3.0 nm of gold nanoparticle, one observes is a shoulder and not the usual peak as is seen for 4.5 or 5 nm gold nanoparticle. What I intend to know is why this change in the type/ nature of the absorption spectrum.

Dear Dr. Anish Rao

Around the nano particle there is plasmon wave, and it is related to the size and shape of nano particle. With increasing the size or aggregation of nano particle the palsmon wave is altered and hence red shifted.    

Dear Rao

Please note that:

- What Ala said is correct. There is continued red shifting with larger size as appears in your own figure.

- At very small sizes (less than 3 nm) there is a shoulder as you also said. 

My comments now about the shoulder is that:

In case of smaller sizes, you measured average size I suppose.

At smaller than 3 nm, are your sure about the shape of the particles? Do smaller ones have same spherical shape?

Are there mixtures sizes? Are you sure about size uniformity? If there are more than one size, the shoulder will be due to the larger size particles present with smaller powders. 

Please try to answer these questions.

I appreciate if you could 

When we record spectra for Gold nano spheres from 200nm to 800nm we can found additional peak near 230 nm along with SPR .That 230nm peak is due to inter band absorption. To arise the 230nm peak,the SPR spectra again get increased in intensity which look like a shoulder peak. For clarity,Dilute the GNP and record the spectra,you can get definite 230 spectra along with SPR