CHAPTER 1
Gas state lasers
Terry A. King
Table of contents
Abstract 05
1.1 Introduction 05
1.2 Basic principles 06
1.2.1 Line broadening mechanisms in gases 07
1.2.2 Saturation irradiance 09
1.2.3 Threshold operation 09
1.3 Basic structures of gas state lasers 10
1.3.1 Gas laser media 10
1.3.2 Optical resonators 12
1.3.3 Pumping techniques 13
1.3.4 Emission characteristics 13
1.3.4.1 Coherence 13
1.3.4.2 Divergence 13
1.3.4.3 Coherence length 14
1.3.5 Pulsewidth control 14
1.3.6 Wavelength selection and frequency control 15
1.4 Types of gas lasers 16
1.4.1 Helium-neon 16
1.4.2 Ion gas lasers 16
1.4.3 Excimer lasers 20
1.4.4 Molecular lasers 23
1.4.4.1 Carbon dioxide 23
1.4.4.2 Carbon monoxide 25
1.4.4.3 Nitrogen 26
1.4.4.4 F2 27
1.4.5 Metal vapour lasers 27
1.4.5.1 He-Cd 27
1.4.5.2 Copper and gold vapour 27
1.5 Perspectives 29
References 30
Abstract
A brief review of the structures and operation of gas lasers of particular interest in photobiology is given in this chapter along with details of their emission characteristics. Gas lasers provide laser wavelengths over a very broad range from the vacuum UV to the far-IR in continuous wave and pulsed operation and from low to high powers. Many of these wavelengths have found valuable application in photobiology, particularly in the several forms of fluorescence technique, microscopy, imaging and Raman spectroscopy, as well as in such techniques as optical trapping and tweezers, micromanipulation and microdissection. Extensive use has been made of the He-Ne laser (wavelength 632.8 nm), argon ion laser (488.0 and 514.5 nm), krypton ion (647.1 nm), He-Cd laser (441.6 and 325.0 nm), excimer lasers (ArF 193 nm, KrF 248 nm, XeCl 308 nm) and nitrogen laser (337.1 nm). The ion and excimer lasers have application in the pumping of tunable dye and titanium-sapphire lasers. Since the discovery of the first gas laser in 1961 many gas lasers have been devised, a detailed understanding built up of their operation and performance and the commercial technology has reached a highly developed and mature state.
In recent years solid-state alternatives to several of the common gas lasers have been developed based on diode lasers, diode pumped solid-state lasers and fibre lasers which offer advantages of compactness and efficiency and operation from low voltage power supplies. However, solid-state substitutes are not presently available at many of the useful gas laser wavelengths, such as the high power pulsed output from excimer lasers in the UV, high power continuous wave output in the near UV, visible and near IR, and mid-and far-IR wavelengths from molecular gas lasers.
1.1 Introduction
The interaction of laser light with biological samples gives an optical signal which carries information on the composition, structure and function of the sample. The probe laser light may be used in the basic measurements of absorption, fluorescence and phosphorescence emission, and Rayleigh and Raman scattering. In this there is a remarkable variety of techniques which have evolved with important applications in photobiology: spectral microscopy and imaging, scanning confocal microscopy, optical labelling, ultrafast pulse and fluorescence lifetime spectroscopy, fluorescence probe (marker) spectroscopy, fluorescence recovery after photobleaching (FRAP), fluorescence in situ hybridization (FISH), fluorescence resonance energy transfer (FRET) and optical trapping and tweezers [1–4]. Special attention can be drawn to the use of fluorescence in several of these techniques with the use of spectral (continuous), dynamic (pulsed) and imaging methods. These applications include laser scanning confocal fluorescence microscopy, time-correlated single-photon counting lifetime studies, total internal reflection fluorescence, near-field scanning fluorescence optical microscopy (NSOM), fluorescence correlation spectroscopy and multi-photon fluorescence microscopy [4]. The Raman scattering technique in its various forms has widespread application in photobiology through its sensitivity to molecular vibrations. Surface enhanced Raman scattering derives an enhanced Raman signal from molecules attached to surfaces to give molecular structural information. Coherent anti-Stokes Raman scattering microscopy enables the identification of molecular constituents of cells without the use of dye markers.
Gas lasers provide source radiation for many of these techniques. In particular, these include the argon (514.5 and 488 nm) – and krypton (647.1 nm) – ion lasers at medium power levels and which provide UV, visible and near-IR wavelengths; the He-Cd laser (441.6 and 325.0 nm), excimer lasers (ArF 193 nm, KrF 248 nm, XeCl 308 nm), N2 lasers (337.1 nm) and F lasers (157 nm). The ion lasers and excimer lasers are also used as pump sources for the tunable liquid dye and solid-state titanium-sapphire lasers. In addition there are several gas lasers which have more specialised applications in photobiology.
In this chapter the basic principles of laser operation are briefly reviewed and some of the laser radiation characteristics. Those gas lasers of particular value in photobiology are described to give a survey of their design features and emission properties.
1.2 Basic principles
Before the invention of the laser, available light sources emitted light from thermal excitation such as a tungsten filament lamp with a spectrum corresponding to a quasi-black body emitter at the temperature of the emitter, or by spontaneous emission from atoms or molecules as in a gas discharge arc or fluorescent tube. Their brightness is limited by the temperature of the emitter. In the laser the essential process of stimulated emission was introduced, so that, in conjunction with population inversion, a net optical amplification (or gain) could be created. A brief description of the principles of operation of lasers is given here; extensive discussion can be found in various texts [5–8].
Stimulated emission was first used in 1953 to create a high brightness source from transitions between the two lowest levels of ammonia molecules, giving a very narrow emission line at a wavelength of 12.6 mm; this was termed the maser since the wavelength was in the microwave region.
The basic elements of a laser are an active medium with suitable energy levels, a means of injecting energy into the medium (a process known as pumping) and a resonator in which the amplification process can occur. In a 2-level system interacting with light radiation with which it is in resonance, the processes of absorption, spontaneous emission and stimulated emission can occur. The light wave can be absorbed or be amplified depending on the population densities N1 and N2 of the two levels. For a wave travelling in the z direction the change in irradiance of the wave over a length dz is dI = γIdz. For an initial irradiance I0 at z = 0, the beam intensity varies along the propagation distance z exponentially as I(z) = I0eγz. The unsaturated gain or loss coefficient γ of the medium at frequency v is
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]
Here g1and g22 are the degeneracies of the two levels and A21 is the Einstein A-coefficient, related to the lifetime τ of the upper level as A21 = 1/τ. The quantity g(v) is the normalized spectral function (or lineshape function) for the transition, such that
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]
The gain coefficient γ(v) can be > 0 if N2 >g2/g1N1. In this case the irradiance grows exponentially. The gain coefficient varies as v-2, making it easier to achieve higher gain in the infra-red than in the ultra-violet, although some cw UV lasers and pulsed UV lasers are available. Spontaneous emission which scales as v3 competes with stimulated emission.
A laser species may be characterized by its stimulated emission cross-section σ0. This is related to the gain coefficient γ(v) as
γ(v) = (N2 - g2/g1 N1)σ0
This leads to
σ0 = c2A21g(v)/8πv2
i.e. the stimulated emission cross-section depends on the characteristics of the emitter.
1.2.1 Line broadening mechanisms in gases
The spectral line of the transition will be broadened by various processes. There is a natural linewidth from the spread in energy for transitions between energy levels in atoms or molecules. For a natural broadened transition the lineshape follows a Lorentzian function, when normalized as
g(v) = ΔvN/2π [(v - v0)2 + (ΔvN)/2)2]-1
Here the linewidth ΔvN is related to the lifetime of the transition as ΔvN = 1/2πτ. In the He-Ne laser the natural linewidth is about 20 MHz.
However, in a gas laser, natural broadening is rarely the dominant lineshape determining mechanism. Collisions or electrostatic forces between neighbours induce interactions which equally affects all the emitters. For collisionally broadened transitions the lineshape is also Lorentzian and has a linewidth Δv = Δvc with
Δv = NQ/π (16/π kT/m)½
where N is the number density of atoms or molecules, Q is the collision cross-section and m is the atomic or molecular mass. This also is described by a Lorentzian function and occurs in gases due to collisions of the emitter and may be the predominant broadening mechanism at higher pressure.
The natural and collisional broadening mechanisms are termed homogeneous broadening in which all the emitters are affected equally. Alternatively, when the emitters have a distribution of broadened components this is termed inhomogeneous broadening. In gases the range of thermal velocities of the atoms or molecules leads to Doppler broadening where there is a distribution of Doppler shifts among the emitters. This is usually the dominant broadening mechanism in low density gases.
The Doppler broadened transition has a Gaussian lineshape. For an emitter of mass m with a centre frequency v0, the distribution of frequencies is
gD(v) = 2/ΔvD(In2/π) 1/2 exp[-(2(v - v0)/ΔvD) 2)2 ln 2]
The linewidth (full-width at half-maximum) is
ΔvD = 2v0(22kT In 2/mc2) 1/2
A contribution to line broadening also occurs dependent on the light intensity within the laser cavity (power broadening). For the He-Ne laser a power broadening of up to 100 MHz is found.
The peak value of the normalized lineshape function at line centre g(v0) is inversely proportional to the linewidth Δv of the transition
g(v) = 2/πΔv
Then the stimulated emission cross-section becomes
σ0 = c2A21/ 4π2v2Δv
This states that the stimulated emission cross-section for a homogeneously broadened transition is proportional to the ratio A21/Δv of the spontaneous transition rate to the linewidth.
Line broadening in gas lasers operating at relatively low pressure is mostly dominated by Doppler broadening. Collisional broadening may become dominant, e.g. in the CO2 laser, at higher pressures.
1.2.2 Saturation irradiance
The gain in the laser is reduced from the unsaturated gain as the irradiance in the cavity increases. For a homogeneously broadened line this leads to the saturated gain coefficient as
γs,h = γ/(1 + 1/1s,h)
Here Is,h is the saturation irradiance
Is,h = 4πhv/λ2g(v) (τnr)/[tauf]
where τf is the fluorescence lifetime of the upper laser level, τnr is the nonradiative decay time and h is Planck's constant. The saturated gain coefficient for an inhomogeneously broadened line is
γs,i = γ/(1 + 1/1s,i)1/2
and the saturation irradiance is
Is,i = 2π2hvΔv/λ2 (τnr/τf)
Here Δv is the homogeneous linewidth of the components of the inhomogeneously broadened transitions.
1.2.3 Threshold operation
The laser resonator (or cavity) is most usually made up of reflectors either side of the gain medium. For the laser to sustain oscillation the gain of the laser medium must be greater than the losses in the medium. The laser is a threshold device, in which the laser threshold is when the gain of the laser is equal to the losses in the laser cavity. To get a laser beam out of the laser, normally one of the reflectors is given a certain transmission. There will also be other losses in the laser due to diffraction, scattering or absorption.
For a laser cavity with two aligned mirror reflectors of reflectivities R1 and R2 spaced by a distance L, the threshold laser gain is
γth = k + 1/2L ln (1/R1) R2
Here k accounts for all the losses other than that due to mirror reflectivities.
The threshold population inversion needed to sustain laser oscillation is then,
ΔNth = (N2 - g2/ g1 N1)th = 1/σ0 (k + 1n(R1R2)/2L)
1.3 Basic structures of gas state lasers
1.3.1 Gas laser media
Gas lasers often use a mixture of gases arranged to optimize the pumping or emission properties such as discharge characteristics or excited state lifetimes. The emission may be from an electronic transition in neutral atoms (e.g. He-Ne) or ionized atoms, (e.g. Ar+, Kr+), electronic transitions in molecules (F2, N2), electronic transitions in transient excimer molecules (KrF) or vibrational or rotational transitions in molecules (CO2, CH3F) or electronic transitions in molecular ions (N2+). A summary of the pulsed or CW operation, wavelengths and energy/power of some of the most common gas lasers is shown in Table 1. The gas lasers are excited by a great variety of pumping methods, including continuous, pulsed or rf electrical discharges, optical pumping, chemical reactions and gas dynamic expansion [9].
The first gas laser, and which was also the first continuous wave (cw) laser, was based on a mixture of helium and neon excited by a radiofrequency electrical discharge [10], producing continuous wave operation at 1.1523 µ,m. Now the He-Ne laser is more usually used on the 632.8 nm line, which is one of the most common of all lasers. The laser transitions occur in the neon atoms, with the strongest lines occurring at the familiar red line at 632.8 nm, the green line at 543.5 nm and other visible lines at 594, 612 and 640 nm, and the infra-red lines at 1.15, 1.52 and 3.39 µm.
The operation of the laser can be understood by considering the energy levels of helium and neon, shown in Figure 1. The He 23S and 21S metastable levels are excited by electron collisions. Transfer of the excitation occurs efficiently in collisions of He atoms with Ne atoms due to the close proximity of the Ne 2s and 3 s energy levels to the He metastable levels.
The He-Ne gas mixture is contained in a narrow bore discharge tube, which may be excited by a dc discharge of about 10 mA or a radiofrequency discharge. The gas pressures are usually set for highest gain on the 632.8 nm transition, with partial pressures of about 2.5 mbar (He) and 0.13 mbar (Ne). The discharge tube diameter is kept quite small, since the gain is inversely proportional to the tube diameter, due to collisions of Ne atoms with the tube walls being required to reduce the population of the Ne Is level to prevent build-up of population in the lower 2p laser levels.
