If the laser you're using has a lot of pulse to pulse amplitude instability (from longitudinal mode-beating or other effects) or is multi-transverse mode such that the beam profile is not uniform on a shot to shot basis, then you'll have very variable ablation conditions shot to shot. If the pulse is too long then you'll have a randomly varying phenomenon called debris screening - the later part of the pulse will be scattered or absorbed in the layer of material that is starting to come of the surface from the impact of the front of the pulse. The beam profile will also be a key parameter - a flat-top is ideal for uniform ablation. A Gaussian pulse has significant peak to average power variation so your ablation dynamics will vary over the beam profile, especially if the profile is varying shot to shot in a multi-mode laser. However if you use a laser with good mode quality, for example a super-Gaussian coupled resonator (QS-Nd:YAG), then you should get excellent uniformity from shot to shot from both a temporal and beam profile point of view. A super-Gaussian beam has a flatter top and will do better in terms of uniformity. If you're using a lamp-pumped Q-switched Nd:YAG then make sure the flashlamps are simmering properly. This will help the pulse to pulse energy stability a lot. If your QS Nd:YAG is not simmered the lamp discharge will be varying on a pulse to pulse basis which will change lots of things (shot to shot stability, pulse width, mode etc).

If the laser is good then the next thing to look at is whether the pulse rep rate is too high, i.e. you are trying to ablate the substrate through the debris cloud of the previous pulse, in which case your beam will be locally and randomly obscured or absorbed. You should arrange to purge the surface with a Venturi-like effect to make sure you are not shooting through debris or bubbles or excessive refractive index variations (from heating the liquid you are ablating in). Also you'll have to examine the mechanism carefully to determine if you need to scan the ablation beam around on the target so that you're always dealing with a surface at equilibrium. Surface prep prior to ablation may be important. The surface will be modified by a shot, so the next shot will not interact with the surface the same way.

There's a way to take a polychromatic distribution of NPs and knock then down in size and improve the width of the distribution by a second step, irradiating in the plasmon resonance. See for example, http://www.sciencedirect.com/science/article/pii/S0169433213007836 Modification of the silver nanoparticles size-distribution by means of laser light irradiation of their water suspensions Original Research Article Applied Surface Science, In Press, Corrected Proof, Available online 29 April 2013 A.S. Nikolov, R.G. Nikov, I.G. Dimitrov, N.N. Nedyalkov, P.A. Atanasov, M.T. Alexandrov, D.B. Karashanova

There are also non-laser methods of producing relatively tight distributions of nanoparticles without chemicals. See for example "Investigation of size distribution of silver nanoparticles" R. Khordad c,⇑, M.R. Vakili b, A.R. Bijanzadeh, doi:10.1016/j.spmi.2011.11.007, Superlattices Microstruct. (2011). You could start with this and use your laser to re-size the particles?

Hope this helps. Best regards, John

Thank you very much for your information. I'll study the papers you proposed and see what helps mostly. The laser i use for my experiments is an Q-Switched Nd:YAG pumped with flashlamps with tunable repetition rate from 1 - 10 Hz.

I recommend to download the following papers available in my homepage in researchgate:

1- Time-resolved evolution of metal plasma induced by Q-switched Nd:YAG and ArF-excimer laser, S.Z. Mortazavi, P. Parvin, M.R. Mousavi Pour, A. Reyhani, A. Moosakhani, Sh. Moradkhani, Optics & Laser Technology 01/2014; 62:32–39.

2- Two-dimensional simulation of laser ablation with 235 nanosecond pulses

Sousan Ghalamdaran, Parviz Parvin, Mohamad Javad Torkamany, Jamshid Sabbagh Zadeh, Journal of Laser Applications 12/2013; 26(1).

3- Generation of various carbon nanostructures in water using IR/UV laser ablation, Seyedeh Zahra Mortazavi, Parviz Parvin, Ali Reyhani, Soghra Mirershadi, Rasoul Sadighi-Bonabi, Journal of Physics D: Applied Physics,

4- Effect of Laser Wavelength at IR (1064 nm) and UV (193 nm) on the Structural Formation of Palladium Nanoparticles in Deionized Water, Journal of Physics Chemistry-C, 01/2011; DOI:10.1021/jp1091224

Thank for the papers you proposed but my main goal is to narrow the size distribution not by changing the laser wavelength at first step. I tried centrifuging then colloidal but it didn't help enough. My basic characterization technique is Dynamic Light Scattering so polydispersed solutions aren't suitable for this kind of characterization .

I am not sure the wavelength can easily narrow the size distribution. The mechanism is complex based on ablation and aggregation and condesation etc according to the papers, I have listed. To my knowledge, the plasma created by IR laser is more enegetic than second hremonic or third harmonic generation of Q switched Nd:YAG laser leading to larger size. The colder plasma may genrate smaller NPs, however the size distribution arises from the whole laser properties and not only the laser line.

Yes, i understand what you mean... There are many different parameters to be examined in the process. Thank you for your information.