I am using a femtosecond laser (Ti-sapphire) which has pulse width 200 fs. Now, I want to change the pulse width from 200 fs to ps range using glass rods for our microscopy application. But I do not have any idea how to check the pulse width?
You need an autocorrelator which covers the range you are working at. Many cover wavelengths from a few hundred nm to a few microns (Your TiS wavelength is about 800nm) and as for the pulse width. The ranges can be from a few fs to ps. You may need to use different optical elements as the pulse width goes up.
An example can be seen at: http://www.lightcon.com/product/tipa.html. Frequency Optical Gating devices (FROG, go and google Trebino and FROG) can be used as well. Most solid state electronic systems such as oscilloscopes etc. will not be able to do this. In some cases, you may use an extra parametric analyser for such time domains but this is nowhere near the Autocorrelator approach performance for laser beams. Should you use the TIPA I can send my own developed software which produces pulse statistics as an add-on to their software. This is useful for observing trends in pulse width and checking environmental stability for the pulse width over time periods.
The most important thing is that the oscilloscope bandwidth must be greater than 1/T where T is the pulsewidth. So you need a scope with about 10 GHz bandwidth. You need also that photo detector capacitance Cd to be very small to make with the transimpedance amplifier resistance Rf a time consatnt Cd Rf that must be smaller than the pulse width of the optical narrow pulse. So, Also about i ns time constant may be suitable to great extent. The more safe condition not to shape the pulse is a rise time about T/10.
To see the effect of the Cd and Rf on the optical pulses please see the paper in the link:Conference Paper A comprehensive study of an optical transceiver