Collisional Narrowing of Lines

In the infrared and microwave ranges, collisions may sometimes cause a narrowing of the linewidth instead of a broadening Dicke narrowing) [3.34]. This can be explained as follows if the lifetime of the upper molecular level (e.g., an excited vibrational level in the electronic ground state) is long compared to the mean time between successive collisions, the velocity of the oscillator is often altered by elastic collisions and the mean velocity compo- 

There is a second effect that causes a collisional narrowing of spectral lines. In the case of very long lifetimes of levels connected by an EM transition, the linewidth is determined by the diffusion time of the atoms out of the laser beam (Sect. 3.4). Inserting a noble gas into the sample cell decreases the diffusion rate and therefore increases the interaction time of the sample atoms with the laser field, which results in a decrease of the linewidth with pressure [3.36] until the pressure broadening overcompensates the narrowing effect. 

There is a second effect which causes a collisional narrowing of spectral lines. In case of very long lifetimes of levels connected by an EM tran- 

---

Figure 2.46 illustrates the advantages of this technique. The upper spectrum represents a Lamb peak in the intracavity saturation spectrum of the neon line (l 2p) at A. = 588.2 nm (Sect. 2.3.3). Due to the collisional redistribution of the atomic velocities, a broad and rather intense background appears in addition to the narrow peak. This broad structure is not present in the dichroism and birefrin-gent curves (Fig. 2.46b, c). This improves the signal-to-noise ratio and the spectral resolution. 

In passing it is interesting to note that Fig. 5 qualitatively explains the reason for the difference in the effect of motion on spectral lines in radiofrequency and optical spectroscopy. In radiofrequency spectroscopy one refers to motional narrowing, while collisional broadening is used to... 

As before o is the cross section in the absence of a microwave field. Naturally <70(0) = o. This expression is the origin of the lines drawn in Fig. 9, which evidently match the experimental cross sections. Similar results have also been observed with the K system of Eq. (10) using velocity selected beams to obtain narrower collisional linewidths allowing, the use of rf frequencies of 4 MHz, instead of 15 GHz [Thomson 1992], Since the collisions last longer, the rf fields can be very weak, < 0.1 V/cm. An interesting aspect of both the 15 GHz and the 4 MHz measurements is that the Bessel function oscillation of the cross section with microwave or rf field amplitude is observed indicating that the coherence of the colliding atoms is maintained over multiple field cycles. 

Line sources are capable of producing the best linear relationship between instrument response and concentration. For optimum linear response, the halfwidth of the source line used for excitation should be less than the half-width of the absorption line of the sample. This requirement is met by most line sources, since at temperatures of the flame, absorption lines of the sample undergo substantial Doppler and collisional broadening whereas the corresponding source lines remain narrow. 

In addition to the rotational structure 16.20, the inversion spectrum has a hyper-fine structure. For the main nitrogen isotope thehyperfine structure is dominated by the electric quadrupole interaction ( 1 MHz) [69]. Because of the dipole selection rule, AAl = 0 and the levels with 7 = AT are metastable. In beam experiments, the width of the corresponding inversion lines is usually determined by collisional broadening. In astrophysical observations, the lines with 7 = AT are also narrower and stronger than others, but the hyperfine structure of the spectra with high redshifts is unresolved. 

Often the narrow Lamb dip at the center of the gain profile of a gas laser transition is utilized (Sect. 7.2) to stabilize the laser frequency [5.78,5,79]. However, due to collisional line shifts the frequency vq of the line center slightly depends on the pressure in the laser tube and may therefore change in time when the pressure is changing (for instance, by He diffusion out of a HeNe laser tube). 

With techniques of sub-Doppler spectroscopy, even small collisional broadening effects can be investigated with high accuracy. One example is the measurement of pressure broadening and shifts of narrow Lamb dips (Sect. 7.2) of atomic and molecular transitions, which is possible with an accuracy of a few kilohertz if stable lasers are used. The most accurate measurements have been performed with stabilized HeNe lasers on the transitions at 633 nm [13.17] and 3.39 p.m [13.18]. When the laser frequency co is tuned across the absorption profiles of the absorbing sample inside the laser resonator, the output power of the laser Pl co) exhibits sharp Lamb peaks (inverse Lamb dip) at the line center of the absorbing transitions (Sect. 7.3). The line profiles of these peaks are determined by the pressure in the absorption cell, by saturation broadening, and by transit-time broad-... 

A CO2 laser operates on the emission bands between vibrational combination states generating emission on discrete rovibrational transitions in the i>i and 2 2 3 bands, centred around 10.6 and 9.6 pm, respectively. Population inversion is achieved by collisional energy transfer from plasma-excited N2 to CO2, usually in a mixture with He. A particular rovibrational emission line can be selected using a rotatable diffraction grating incorporated in the laser cavity. CO2 lasers can achieve very high continuous-wave (cw) power levels of up to 100 W from commercially available systems. In addition, CO2 lasers are robust, narrow-bandwidth and low-cost systems well able to induce IRMPD, but a disadvantage is clearly its limited tunability. It should be noted that fixed-frequency CO2 lasers are used routinely in commercial MS platforms to induce dissociation as an alternative to CID. 

---