Characterizing a 473 nm DPSS Laser for use in Oblique Illumination of Fluorescently Tagged DNA

This project involves characterizing a blue diode-pumped solid state (473 nm DPSS) laser for output power, polarization and beam profile. The project came about after a chance meeting with Dr. Jonathan Sokolov of the Garcia Center, whose research utilizes techniques in modern microscopy to study the structure of DNA and other polymers. Dr. Sokolov had recently purchased the DPSS laser for experiments involving oblique illumination of DNA. His proposed project was well suited to my interests, as I had become fascinated with illumination techniques and optical characteristics of microscopes through an advanced microscopy course as a sophomore at Juniata College.

Dr. Sokolov's research involves confocal imaging of fluorescently tagged double and single stranded DNA to determine tag binding orientation. DNA-specific fluorescent tags are expected to bind to the different structures in distinct orientations; this can be determined from differing responses of tags to incident polarized light. In confocal imaging (microscopy) two dimensional optical sections of a specimen are combined to form a three dimensional representation of the object. A Laser Scanning Confocal Microscope (LSCM) is equipped with one or more lasers of various wavelengths which can be used to excite fluorescently tagged specimens for imaging purposes. Unfortunately, the laser illumination system of the Garcia Center LSCM is integrated vertically into the optical train of the microscope, inhibiting the needed flexibility for proper alignment of polarized light with respect to the orientation of the DNA strands. To allow flexibility, an off-axis illumination source is necessary, provided by the separate 473 nm laser.

Our laser [1] accepts a 0 -- 5 Vdc control voltage V_c which regulates its output power. Output power as a function of V_c was measured with a calibrated power meter in 0.5 Volt steps from 0.5 to 5.0 Volts. Power was just 1.35 mW at V_c = 0.5 Volts and approximately doubled for each 0.5 Volt increment up to 4.0 Volts, where it leveled off at 110 mW. We later found that beam quality degrades dramatically at V_c < 2.0 Volts. We conclude that the laser can provide an output beam power from ~ 16 to 110 mW with acceptable beam quality.

The laser's linear polarization was next studied with a Glan-Thompson polarizer. The orientation was found to be vertical; however the maximum extinction ratio was found to be just 35:1, much poorer than the 100:1 ratio specified by the manufacturer [1].

The final and most involved set of measurements seeks to profile (determine size and shape of) the laser beam as a function of distance z from the laser. Several methods have been employed. In the first and simplest, the diameter (width) of the laser beam was estimated from its visual appearance at distances up to z = 26 meters. At the larger distances the spot was clearly seen to be elliptical (~ 50% wider than high); all of our estimates and measurements so far are confined to the horizontal plane. The visual method overestimates the true width 2 w(z) of the beam. A correction factor to account for this was obtained by comparing the estimated width at z = 8.0 meters with an actual measurement made by scanning a 1 mm square photodetector across the beam. The profile thus measured was a very close match to the Gaussian shape expected for an ideal laser mode. At V_c = 2.0 Volts these visual width estimates correspond to a full angular width of 0.64 milliradian (mR). According to the standard theory of Gaussian beams, if the waist (minimum size) of the beam occurs at the face of the laser then this divergence corresponds to a waist radius of w₀ = 470 microns. Beam size near the laser was carefully measured by both the standard pinhole and razor-blade methods. These results for w₀ are 20% to 50% greater than 470 microns; the reason for this discrepancy is under investigation.

Research supported by NSF Grant PHY-0851594. We thank Suri Bandler and Ashish Sridhar for their assistance with the polarization measurement.

[1] Model DHL-B50N from Ultralasers, Inc. Toronto, Ontario, Canada. http://www.ultralasers.com.