Open Access
Add to BibSonomyAbstract
We demonstrate backwards lasing in atomic argon directly excited via a three–photon pumping in air mixtures with argon mole fractions down to 10%. We achieve well collimated, narrowband coherent emission at 1327nm by using both broadband femtosecond excitation and narrow linewidth picosecond excitation in the vicinity of 261nm. This approach shows promise for standoff trace detection in the atmosphere.
© 2016 Optical Society of America
1. Introduction
In order to achieve true single-sided remote trace species detection in atmosphere, the light emitted or scattered from the target needs to be collected in the backward direction. Since for incoherent light the emission is non-directional, standoff detection at large distances is very hard to achieve by conventional light scattering or fluorescence approaches. Thus, there is a need for coherent light in the backwards direction. Recent demonstrations of remote air lasing in atomic oxygen [1,2] and nitrogen [2–4] provide such a source, but this approach requires dissociation of the molecular species followed by the excitation of the resulting atomic species. This dissociation step leads to significant fluctuations in the backward lasing amplitude and mode structure due to the highly nonlinear nature of the dissociation process. In order to obtain lasing in air without the need for a preliminary dissociation step, one needs to rely on the existing atomic species in atmospheric air. The best atomic candidate for generating remote lasing in the atmosphere is argon, because it is the next populous species in air after nitrogen and oxygen (almost 1%). Once backward lasing is acheived, trace gases can be detected by sending a copropagating second laser tuned to a transition in the trace species. This resonant laser will modulate the complex refractive index (refraction and/or absoprption) for the pump beam. Since the backward lasing response is highly nonlinear to any modulation of the forward propagating pump beam, recording the fluctuations of the air laser can then indicate the presence of the trace species.
Stimulated emission in the VUV spectral region has been observed in noble gas mixtures [5], in argon dimers via multi-photon excitation [6] or pulsed discharges [7,8], and the lasing in argon ions led to the discovery of one of the first gas lasers [9]. However, for remote lasing in air we are interested in optically induced stimulated emission from argon atoms. Given the high energy levels of the excited states in argon, a two-photon excitation similar to oxygen or nitrogen would require the use of photons with wavelengths below 200nm, which makes propagation in atmospheric air problematic. In order to excite the argon atoms in air we have to resort to a three-photon excitation. While no stimulated emission has been reported from optically excited atomic argon to date, three photon excitation has been used for the ionization of argon [10]. One can identify attainable excited states via multiphoton excitation by using resonantly enhanced multiphoton ionization (REMPI) spectroscopy [11]. In particular, we have shown that using a 3 + 1 REMPI excitation with 261nm photons, the argon atoms can be excited and further ionized [12,13].
2. Experimental setup
It is this three-photon absorption at 261nm that we have used to demonstrate lasing in atomic argon [14]. As shown by the energy level diagram in Fig. 1, the three- photon pumping at 261nm brings the argon atoms from the ground state 3s23p6 to the upper lasing state 3s23p5(2P˚1/2)3d. This three-photon excitation is followed by spontaneous emission from the 3d state to the lower laser state 3s23p5(2P˚1/2)4p. The 1327nm emission brings the atom into the 4p state, which cannot decay radiatively to the ground state because of the parity rules.
Fig. 1 Argon three-photon excitation using 261nm photons is followed by emission at 1327nm. The suggested 6-photon mixing generates 145nm photons.
Following the same recipe used to create air lasing in atomic oxygen and nitrogen [1,2], we expect to obtain forward and backward stimulated emission by creating a gain region in air where the spontaneous emission at 1327nm gets amplified and becomes coherent emission along the pumping laser propagation direction.
Figure 2 shows the experimental setup used for lasing in argon. For the picosecond experiments described in Section 3 we used the third harmonic of a Coherent Hidra amplified laser system capable of delivering 120ps (FWHM) at 783nm. For the measurements described in Section 4 we used a Spectra-Physics Spitfire femtosecond amplified laser system equipped with an Optical Parametric Amplifier capable of delivering 261nm pulses with pulsewidths less than 100fs. After focusing with a 30cm focal length lens, we obtain the three- photon excitation of the argon atoms contained in a glass cell. The maximum energy of the pump laser pulses is 1mJ for the ps system, and 10μJ for the fs system, leading to peak intensities achieved in the gain region of 5x1011 W/cm2 and 8x1012 W/cm2, respectively.
Fig. 2 Experimental setup for forward and backward lasing in argon via three-photon resonant optical pumping.
The cell placed in the focal region of the pump beam contains the gain medium which can be pure Ar gas at different pressures, or a mixture of Ar and atmospheric air.
As shown in Fig. 2, following this excitation, we obtain well-collimated backward and forward propagating laser beams at 1327nm. As in the case of oxygen and nitrogen atomic emission [1,2], due to the highly elongated gain region (we estimate a Rayleigh range of 8mm), the 1327nm photons are experiencing gain predominantly along the direction of propagation of the pump beam. This leads to strong stimulated emission in the forward and backward direction, as illustrated in Fig. 2. In order to detect the 1327nm emission we used a fast detection system (~20ps resolution) comprised of New Focus 1454 detectors and a Tektronix DPO73304D 33GHz oscilloscope for temporal characterization, a slower detector (Thorlabs DET10C) for energy measurements, and an Ocean Optics NIRQuest spectrometer for spectral measurements.
3. Lasing in argon
The fast detection system described above allows to directly measure the temporal behavior of the forward and backward lasing from argon. Figure 3(a) shows that the pulses emitted in both directions are simultaneous and have a pulsewidth of 50ps (FWHM). The spectra shown in Fig. 3(b) show the 1327nm narrow line emission with a linewidth limited by the 1nm resolution of the spectrometer. Both the forward and backwards emission are recollimated and the obtained beams have been monitored over several meters distance to ensure the low divergence (less than 10mrad) associated with stimulated emission.
Fig. 3 Temporal (a) and spectral (b) shape of the 1327nm emission from Ar in the forward (squares) and backward (circles) direction.
Using two fast detectors simultaneously allows for recording the emission in both directions at the same time, while varying the argon pressure in order to change the density of the gain medium. As shown in Fig. 4, at higher densities the backwards emission does not increase as fast as the forward emission. We can interpret this result by considering the interplay between the stimulated emission and a wave-mixing process that takes place only in the forward direction due to phase-matching. Since we are using a three-photon pumping process at 261nm followed by the emission of a photon at 1327nm, we are suggesting a 6-wave mixing process, as depicted in Fig. 1. The transition from the state 3s23p5(2P˚1/2)4p to the ground state must involve two photons (for parity reasons), and the most probable combination is using another 261nm from the pump, which results in the generation of a 145nm photon. As the argon density is increased, the 6-wave mixing process is more probable, and this reduces the density of the excited states needed for the stimulated emission process. This is accord with results we have obtained while studying the interplay between forward and backwards emission of other atomic lasing systems, and further studies on the dependence of argon pressure are needed to quantify the effect.
Fig. 4 Backwards (squares) and forward (circles) stimulated emission as function of the Ar pressure.
Figure 4 also shows that at lower pressure, before reaching saturation and the above-mentioned 6-wave mixing process, the backwards emission is actually stronger than the forward emission. We attribute this effect, also observed in the other optically pumped atomic lasers (O and N), to the asymmetry of the gain region. Due to the fact that the atoms are optically excited via multiphoton absorption, as the pump beam propagates through the gain region it becomes less intense, and hence the gain is slightly stronger in the region closer to the focusing lens than in the region further away from it. Furthermore, since the multiphoton excitation is accompanied by photo-ionization of some of the excited atoms, this leads to a negative change in the refractive index, which defocuses the beam as it propagates away from the focusing lens. This also results in less intensity in the further part of the focal region, as opposed to the closer one. Since the gain grows exponentially, even a very small asymmetry in the spatial distribution of the gain leads a stronger beam in the backwards direction, which experiences the stronger gain region towards the focusing lens, as opposed to the forward propagating beam, which experiences slightly weaker gain in the region further away.
4. Broadband femtosecond pumping for narrowband atomic lasing
4.1 Femtosecond pumping
Since we are interested in lasing in atmospheric air, where the concentration of Ar is just below 1%, we need to optically pump the Ar atoms via the three-photon excitation very efficiently, which requires a high pump intensity. At the same time we do not want significant ionization, which means that we need to achieve high intensity without high energy and with pulses short enough as to not allow avalanche ionization. The obvious answer is the use of ultrafast, femtosecond pulses at 261nm. The only problem with using 100fs pulses to excite atomic species is the fact that the broadband nature of ultrafast pulses is at odds with being resonant with the narrowband transition between the ground and excited states of the atomic species. However, multi-photon transitions can be achieved resonantly with much broader excitation due to the fact that the non-degenerate multi-photon transitions allow for all the photons to contribute to the excitation. For example, atomic hydrogen can be efficiently excited using a two-photon transition, and luminescence from the excited state can be obtained, when the optical pumping is realized with broadband, femtosecond pulses [15].
Using the same optical setup shown in Fig. 2 with the femtosecond system described in Section 2, we have obtained lasing at 1327nm in the forward and backwards direction using three-photon femtosecond pumping. In this case, however, we are able to achieve narrowband emission (estimated ~10GHz, corresponding to 50ps pulses) at 1327nm, while using broadband pumping (~10THz, corresponding to 50fs pulses) at 261nm.
In Fig. 5 we show the three-photon excitation spectra obtained by monitoring the stimulated emission in the forward (circles) and backward (squares) direction while tuning the pump laser wavelength. The spectra are obtained for pumping with linear (top figure) and circular (bottom figure) polarization. To explain these spectra we have to note first that the broadband pumping allows for excitation of more excited states than the energy level depicted in Fig. 1. In fact, Fig. 5 shows the convolution between the bandwidth of the UV pump laser (narrowed by the fact that we are investigating a three-photon transition), and the transitions to the following 3s23p5 excited state levels: (2P˚1/2)3d[5/2] (J = 3), (2P˚3/2)3d[5/2] (J = 3), (2P˚3/2)5s[3/2] (J = 1), and (2P˚3/2)3d[3/2] (J = 1). According to the NIST database, the three-photon transitions to these levels require 261.27nm, 263.81nm, 263.98nm, and 262.82nm, respectively. Since only J = 1,3 transitions are alowed for 3-photons, and only J = 3 are allowed for circular polarization [16], the J = 3 transitions are depicted in Fig. 5(b). From comparing the spectra in Fig. 5 it follows that circular polarization has a higher cross-section than linear polarization, such that the 261.27nm has a lower threshold for circular polarization. The other transitions are all covered by the borad pulse around 263nm, so both linear and circular polarizations can excite the stimulated emission in that spectral region.
Fig. 5 Backwards (squares) and forward (circles) three-photon excitation spectra of argon while pumping with linear (a) and circular (b) polarization.
As in the case of ps pumping, the emission is highly collimated. Is also noteworthy that the 1327nm emission shown in Fig. 3 to be around 50ps is an indicative of the gain length, and is independent of the pumping pulsewidth (shown here to be either ps of fs). Another indicative of stimulated emission is the fact that the emitted pulses are orders of magnitude faster than the spontaneous emission lifetime (usualy nanoseconds).
4.2 Polarization dependence for argon lasing
While performing the measurements shown in Fig. 5, we have also recorded the polarization of the emitted laser beams at 1327nm. Table 1 shows the results of this polarization study.
Table 1. Polarization dependence of optically pumped argon lasing
The result is that the Ar laser maintains the polarization of the optical pumping laser. A linear polarization of the pump laser results in the same linear polarization imprinted upon the emission, and the circular polarization for the 261nm pump results in a circular polarization for the 1327nm emission. Under circular polarization, the forward propagating beam retains the same direction for the rotation, while in the backwards direction, the direction for the circular polarization is reversed. This is due to the fact that the propagation vector k is reversed for the backwards propagating beam, while the rotation of the electric field is conserved. This effect is similar with the reflection on a mirror, where the circular polarization changes from left to right, and vice versa.
5. Lasing in argon-air mixtures
Using femtosecond pumping allows us to obtain much stronger three-photon excitation with less energy per pulse. Considering the response of the InGaAs detector we estimate around 10nJ/pulse at 1327nm were obtained using 10mJ/pulse at 261nm. In order to achieve our goal of obtaining Ar lasing in atmospheric air we have reduced the pressure of the Ar gas and monitored the emission in the backwards direction.
As seen in Fig. 6, in pure Ar the stimulated emission is still strong even when the density of Ar is reduced to less than the density of Ar atoms found in atmospheric air (7 Torr). The backwards lasing signal is much stronger than the noise level indicated by the solid line in Fig. 6.
Unfortunately, adding atmospheric air drastically changes the lasing efficiency. In Fig. 7 we show the backwards lasing at 1327nm obtained while varying the partial pressure of Ar with (squares) and without (circles) atmospheric air.
Fig. 7 Backwards lasing in pure Ar (circles) and in Ar in atmospheric air (squares) as function of the partial pressure of Ar.
As seen in Fig. 7, while lasing in pure Ar is still strong, the air-Ar mixture with the same density of Ar atoms shows less stimulated emission. With the current setup (laser intensity, near-infrared detector) we have achieved backwards lasing from air containing as low as 10% Ar, which is ten times more Ar than we find in atmospheric air. The higher lasing threshold in atmospheric air can be attributed to the rapid collisional transfer of energy from the excited Ar states to the electronic excited energy levels of nitrogen and oxygen. The goal of obtaining backwards lasing from Ar in atmospheric air can be reached by further improvements on detection and by increasing the pump energy to allow for longer gain paths and thus lasing at lower Ar partial pressure. For this purpose using femtosecond pulses is beneficial, because it avoids the potential problem of avalanche ionization. Additionally, femtosecond pumping occurs faster than the collision time, so collisional deactivation of argon during pumping does not occur.
We note that while using a minor species such as Argon as the gain medium for air lasing might be seen as a difficulty due to a reduced density, as opposed to the molecular oxygen or nitrogen, we have to consider that in order to achieve air lasing in the molecular species, efficient dissociation is required. In order to obtain atomic oxygen and nitrogen at the same levels or higher when compared with argon, a strong laser pulse is required (either the UV pump or a pre-dissociating pulse), much stronger than what is required for the multiphoton atomic excitation. Furthermore, atomic species such as argon do not require time for dissociation, and a single femtosecond laser can be used with the additiojnal advantage of having high intensity with less energy per pulse.
6. Conclusion
In conclusion, we have obtained three-photon induced stimulated emission from atomic gaseous argon. We have achieved narrowband emission from atomic Ar while pumping with broadband, femtosecond pulses. We have demonstrated both forward and backwards lasing from argon mixed with atmospheric air, and shown backwards remote lasing from 10% atomic argon in air. Argon backwards lasing is a significant advancement in air lasing because it does not require photo-dissociation of any molecular species, and it is expected that argon lasing in atmospheric in air will greatly aid the standoff trace species detection.
Acknowledgment
We would like to acknowledge the support from the US Office of Naval Research (ONR) under grant number N00014-15-1-2656, “Stand-off detection of trace species by simultaneous Radar REMPI and backward lasing”, program manager Dr. Joong Kim.
References and links
1. A. Dogariu, J. B. Michael, M. O. Scully, and R. B. Miles, “High-gain backward lasing in air,” Science 331(6016), 442–445 (2011). [CrossRef] [PubMed]
2. A. Dogariu and R. Miles, “Lasing in atmospheric air: similarities and differences of Oxygen and Nitrogen,” in Frontiers in Optics 2013, P. Delyett, Jr. and D. Gauthier, eds., OSA Technical Digest (online) (Optical Society of America, 2013), paper LTh2H.2.
3. A. Dogariu and R. B. Miles, “Nitrogen lasing in air,” in CLEO:2013, OSA Technical Digest (online) (Optical Society of America, 2013), paper QW1E.1.
4. A. Laurain, M. Scheller, and P. Polynkin, “Low-threshold bidirectional air lasing,” Phys. Rev. Lett. 113(25), 253901 (2014). [CrossRef] [PubMed]
5. P. W. Hoff, J. C. Swingle, and C. K. Rhodes, “Observations of stimulated emission from high-pressure krypton and argon/xenon mixtures,” Appl. Phys. Lett. 23(5), 245–246 (1973). [CrossRef]
6. S. E. Harris, A. H. Kung, E. A. Stappaerts, and J. F. Young, “Stimulated emission in multiple‐photon‐pumped xenon and argon excimers,” Appl. Phys. Lett. 23(5), 232–234 (1973). [CrossRef]
7. G. Balssen, O. Dossel, H. Nahme, N. Schwentner, and H. Wilcke, “Stimulated emission in argon crystals around 10 eV,” J. Lumin. 24-25, 535–538 (1981). [CrossRef]
8. A. V. Eleskii and A. R. Sorokin, “Stimulated emission of argon dimers excited by a pulsed plasma-cathode discharge,” Tech. Phys. 42(11), 1285–1288 (1997). [CrossRef]
9. W. B. Bridges, “Laser oscillation in singly ionized argon in the visible spectrum,” Appl. Phys. Lett. 4(7), 128–130 (1964). [CrossRef]
10. S. M. Koeckhoven, W. J. Buma, and C. A. de Lange, “Three-photon excitation of autoionizing states of Ar, Kr, and Xe between the 2P3/2 and 2P1/2 ionic limits,” Phys. Rev. A 49(5), 3322–3332 (1994). [CrossRef] [PubMed]
11. M. Shneider, Z. Zhang, and R. B. Miles, “Plasma induced by resonance enhanced multiphoton ionization in inert gas,” J. Appl. Phys. 102(12), 123103 (2007). [CrossRef]
12. Z. Zhang, M. N. Shneider, and R. B. Miles, “Coherent microwave Rayleigh scattering from resonance-enhanced multiphoton ionization in argon,” Phys. Rev. Lett. 98(26), 265005 (2007). [CrossRef] [PubMed]
13. A. Dogariu and R. B. Miles, “Detecting localized trace species using Radar REMPI,” Appl. Opt. 50, A68–A73 (2011). [CrossRef] [PubMed]
14. A. Dogariu, J. Li, and R. B. Miles, “Three-photon pumped backwards lasing in argon,” in Light, Energy and the Environment 2015, OSA Technical Digest (online) (Optical Society of America, 2015), paper ETu2A.4.
15. W. D. Kulatilaka, J. R. Gord, V. R. Katta, and S. Roy, “Photolytic-interference-free, femtosecond two-photon fluorescence imaging of atomic hydrogen,” Opt. Lett. 37(15), 3051–3053 (2012). [CrossRef] [PubMed]
16. J. L. Dehmer, S. T. Pratt, and P. M. Dehmer, “Three-photon excitation of autoionizing states of atomic krypton between the 2P(3/2) and 2P(1/2) fine-structure thresholds,” Phys. Rev. A 49, 3322–3332 (1994). [PubMed]
