Introduction


This rotation is a continuation of the work done by Andres Cimmarusti.We were working with a Ytterbium discharge lamp in order to create a laser lock using the absorption signal from Ytterbium.For this,we used a laser scan around 399nm.For the initial work done to estimate the Yb atom density,please look up Andres' work.I have worked with Matthew Adams for this part of the project.For discussion on the unsaturated absorption spectrum that we obtained,a simulation of the spectrum,doppler and pressure broadening,please look up Matt's work. In the part that I present here,I will discuss our efforts to perform a saturation spectroscopy with the Yb discharge lamp in an earnest effort to get around the doppler broadening.However,we are still far from a fitting conclusion.

A general discussion on saturation spectroscopy


A schematic representation of the basic experimental setup for doing Doppler-free saturation spectroscopy is shown here.

Two counter-propagating laser beams derived from a single laser beam are sent through an atomic vapour cell. While the "pump" beam has a high intensity and serves to bleach the atomic gas, i.e. to make the gas transparent, the transmittance of the "probe" beam through the atomic vapour is recorded with a photodiode and gives the actual spectroscopic signal. Typically the probe beam has a factor of ten smaller intensity. If the pump beam is blocked and only the probe beam goes through the vapour cell, one obtains a simple absorption line exhibiting a strong Doppler broadening. Now, if in addition the pump beam is sent through the vapour cell, a very narrow spike appears in the probe-beam signal at the resonance frequency of the atomic transition.This is the so-called Lamb dip, named after W. Lamb who first recognized the huge potential of saturation spectroscopy.

The reason for this spike to show up is the following: only for atoms having zero velocity along the long axis of the vapour cell the two counter-propagating laser beams have the same frequency. Atoms moving e.g. to the right see the frequency of the pump beam shifted into the blue whereas for the same atoms the probe beam is red-shifted. Hence, whenever atoms with non zero velocity see the probe beam shifted into resonance due to the Doppler effect, at the same time the pump beam is shifted further away from resonance. Atoms at zero velocity, however, do see both the pump and the probe beam. The high intensity or equivalently the high photon fux in the pump beam, leads to a high absorption rate. The atoms absorb photons from the pump beam and undergo a transition to the excited state, from where they eventually fall back to the ground state by spontaneous emission. If the pump-beam intensity is high enough, the ground state is significantly depleted and therefore the absorbance of the probe beam is reduced compared to the case without pump beam: a spike appears in the transmission spectrum of the probe beam. Note, that for infinite intensity of the pump beam, on average half of the atoms would be in the excited state and no probe absorption would occur. This is called "saturation" of the atomic transition. Hence, the name "saturation spectroscopy". Obviously the dip is much narrower than the Doppler width. If the intrinsic frequency width (linewidth) of the laser used in the experiment is small enough, the observed width of the Lamb dip can be as small as the natural linewidth of the atomic transition.

A schematic of the probe beam signal that we would obtain with and without the pump beam is shown here.

Laser power required to perform saturation spectroscopy


To perform saturation spectroscopy,we need to know the minimum laser power required to saturate the absorption spectra due to the pump beam.For this,we need to know the diffraction limited spot size of the laser beam inside the discharge tube.
The radius of the spot size is determined from the equation:
r=1.22*wavelength*N
where wavelength refers to the wavelength of light and N refers to the f-number.
In our present setup,the plates which contain the discharge has a radius of about 25mm and they are about 3mm apart. Using these values gives us a spot size of
r=1.22*399nm*(25/3)~4E-04 cm.
Thus the area of the spot size is approximately 5E-07 cm^2.
In order for the pump beam to saturate the absortion spectrum,we need 66mW/cm^2.Hence,the power that we need to saturate Yb is
P=66*5E-07mW=33nW.
Thus if we are above this intensity we can saturate the Ytterbium.The measured power of the laser when it is at full intensity is near 17 microWatt.Hence,we can quite easily saturate the Ytterbium atoms.

Procedure and Experimental setup


Photographs of the experimental setup that we have used can be found here.
In the photograph,the laser enters from the left and is split up by using the beam splitter,one of them acting as the pump beam and the other as the probe beam.We could separately control the intensity of both the pump and the probe beams.The pump beam goes straight through the discharge tube while the probe beam is deflected by the use of mirrors to enter the discharge tube in an almost diametrically opposite direction than the pump beam.Since the discharge region is small and has a weird geometry,we had a hard time in lining up the setup.
The probe beam after passing through the discharge tube is now deflected by the beam splitter into the photodiode. In order to be able to indovidually control the beams,the pump and the probe beams are not exactly in the same line but are at a slight angle from each other.During the experiment,we perform a laser scan and look for any effect of saturation spectroscopy on the CRO.

Conclusion


It is unfortunate that we were unable to get any reasonable result for the saturation spectroscopy.We were able to retrieve the absorption spectra by increasing the intensity of the probe beam.However,at low intensity of the probe beam in comparision to that of the pump beam,we were unable to get any reasonable result.
The most simple reason for this could be miss-alignment,however this does not seem to be a very reasonable explanation since we have aligned and re-aligned the setup on several different occasions without any noticable difference.
We later observed that beam splitter that we have been using,designed for 400-600nm,was not working properly in our case with the 399nm laser.Only last week we replaced it by a 50-50 beam splitter,however,till date,with not much success.

Acknowledgements


I express my sincere thanks to Dr. Wes Campbell for his immense help in setting up the experiment.It was a joy working with Matthew Adams,my partner in this project.I also pay my warm regards to Prof.Chris Monroe for providing me the opportunity to work on this project and Dr.Luis Orozco for organizing this course.