Open Access
Add to BibSonomyAbstract
Femtosecond (fs) laser propagation and fluorescence of dense potassium vapor was studied, and the spectral region around the first and the second doublets of the principal series lines of potassium atoms was investigated. In our search we did not observe the conical emission in the far field, although it was previously observed in the case of rubidium. We discuss the possible reason of this unexpected result. The fluorescence spectrum revealed Rb impurity resonance lines in emission due to the collisional redistribution from the K(4p) levels into the Rb(5p) levels. In the forward propagation of 400 nm femtosecond light we observed the molecular band red shifted from potassium second doublet. However, no molecular spectrum was observed when the mode-locked fs laser light was discretely tuned within the wings of the first resonance lines, at 770 nm.
©2013 Optical Society of America
1. Introduction
Spectroscopy with dense alkali vapor has a long history [1], but before transparent sapphire cells became available extreme temperatures were possible only by using heat-pipe ovens [2,3]. However, the length of the homogeneous vapor was not well defined in such cases, unless special precautions were made [4]. Laser spectroscopy of dense potassium vapor in the heat-pipe ovens brought about many interesting spectral features connected with either atomic line broadening [5], satellite bands [6] or molecular features [7].
Femtosecond (fs) lasers have recently become available in many laboratories. Classical spectroscopy in which such a laser is used for the fluorescence measurements is rare, but the pump and probe methods have become the most frequently used method of research. However, if used just as a background light source the femtosecond laser could reveal several interesting features. It has a broadband emission, high intensity, sub-picosecond pulse duration and a relatively high repetition rate (pulse train) usually in MHz range.
In the present paper we have studied spectral features due to laser induced fluorescence and propagation through potassium vapor. Possible new effects and possible applications are studied. One of the tasks was the searched for the conical emission in dense potassium vapor induced by the femtosecond laser tuned into the blue wing of the D2 resonance line at 766 nm. In the region of the first resonance lines the nonlinear behavior of the refraction index may cause self-focusing (conical emission) and self-defocusing [8,9]. Conical emission in alkali vapors was usually observed in nanosecond laser experiments. Even in the picosecond laser experiment, with potassium vapor, a conical emission was clearly observed, but it was not observed with 150 fs laser pulses [10].Therefore, in an experiment with shorter and stronger fs pulses propagating in potassium vapor we hoped to resolve this apparent paradox.
We also investigated the energy transfer from the potassium first excited levels (excited by the femtosecond laser) toward the rubidium first excited levels. This may point to the realization of conditions for the Rb laser emission at 780 nm and 794 nm [11,12]. High density potassium vapor may also be used in the formation of an atomic filter used in many different applications [13–15]. The latter could be also investigated under the illumination of the femtosecond laser for different potassium vapor temperatures (usually high atom densities, usually above 1016 cm−3).
2. Experiment
In the first experiment we performed a series of absorption measurements using halogen lamp as a background light source (Fig. 1). For the evaluation of the absorption coefficient profiles we used Beer-Lambert expression where L is the length of the absorption cell (L = 11.3 cm), Io is the incident light and I the transmitted light. The continuum light was transmitted through the hot potassium vapor and was analyzed by digital spectrometer (OceanOptics HR4000CG-UV-NIR). The absorption profiles at different temperatures exhibit spectral features useful for the laser excitation in the region around 800 nm and 400 nm [16–19].The edges of the spectra are very noisy because of the small sensitivity and/or small intensity of the background light source used for the absorption measurements.
Fig. 1 Absorption coefficient profiles from 350 nm to 1000 nm at different temperature and/or the densities of potassium vapor.
Our sapphire cell was heated with an oven consisting of two parts as shown in Fig. 2. These two parts were separated by about 5 mm in order to define the lowest temperature, which was measured by using a thermocouple. Within the heaters a second thermocouple measured about 150 °C higher temperature. However, the potassium atom density was determined only by the thermocouple located in between the heaters, where the temperature was the lowest. We used the vapor pressure curve for potassium suggested by Nesmeyanov [20], which was best given by the following formula in the range of our temperatures:
The potassium atom density is then given by:In addition, potassium densities can be directly determined from the data in Fig. 1 using the measured reduced absorption coefficient for temperature independent spectral features. Those features were experimentally and theoretically determined by Vadla et al. [16]. They have obtained the value of for the triplet satellite band at 721.5 nm. The reduced absorption coefficient to obtain the corresponding atom density is defined as: , where k is the absorption coefficient in cm−1. This triplet satellite band is independent of temperature, because its lower potential curve is very weakly bound; the above value was obtained for the temperature interval from 700 K to 1100 K. We checked only the potassium atom concentration at the highest temperature in Fig. 1 because it falls into the given temperature interval. The measured value of the absorption coefficient at 721.5 nm at 460 °C (733 K) was 0.08 cm−1. This value was measured from the background molecular continuum to the peak of the satellite band. We obtained, which compares well with determined by Eq. (1). The corresponding potassium atom densities are also provided in Fig. 1, which we consider as very high atom concentrations. The dimer concentration is usually about a few percent of the atom concentration.Fig. 2 Experimental setup with oven and the sapphire cell containing dense potassium vapor. Laser induced fluorescence was observed side-on by means of the digital spectrometer. NDF is the appropriate combination of the neutral density filters.
In the second experiment we illuminated the hot potassium vapor with the mode-locked femtosecond laser (Millennia pumped Tsunami, Newport, Spectra-Physics, repetition rate 80 MHz, pulse duration 100 fs). By changing the wavelength, in steps of about 10 nm, from 750 nm up to 835 nm, we observed laser induced fluorescence at a right angle from the middle portion of the sapphire cell not covered by heating elements. The profiles of the incoming laser beams are shown in Fig. 3.
Fig. 3 Profiles of the incoming femtosecond laser light before the dense potassium vapor in the sapphire cell (750 nm gray, 760 nm green, 770 nm pink, 780 nm blue, 785 nm red, 800 nm black, 830 nm purple).
3. Results
We spectrally analyzed the fluorescence induced by the fs laser beam passing through the hot potassium vapor at 380 °C, as shown in Fig. 4. For this observation the heaters were separated by 5 mm, which causes a certain temperature gradient. However, the particle density is mainly determined by the lowest temperature measured at the cell wall in between slightly separated heaters. The LIF spectrum when the fs laser was centered at 780 nm, 785 nm and 800 nm exhibited strong absorption of both potassium resonance lines at 766 nm and 770 nm. However, the impurity rubidium resonance lines at 780 nm and 794 nm were observed in emission. This indicates the possibility of using it for the amplification of Rb resonance lines in the K-Rb mixture. This may lead to laser emission at Rb D1 and Rb D2 lines, provided the experiment is performed using appropriate laser cavity. Collision cross section for the excitation energy transfer is relatively large [21]. The efficiency of this hypothetical laser might be close to the efficiency of the heavy alkali vapor lasers [22, 23]. We assume, by inspection of Fig. 1, that the Rb atom density in the present experiment is 104 time smaller (or even less) than the density of potassium atoms. We are currently preparing the experiment with a potassium cell inside the optical cavity to investigate further the possibility of this novel K-Rb laser. Of course, we shall certainly try the K D1 and K D2 excitation with cw and/or nanosecond lasers in order to further prove the realization of the new laser at Rb D1 and Rb D2 lines.
Fig. 4 Side-on observed laser fluorescence induced by the femtosecond laser peaking at 750 (gray), 760 (blue), 780 (green), 785(orange) and 800 nm (pink) through potassium vapor column at 380 °C.
When we connected the two heater elements the temperature in the whole sapphire cell was uniform and it was much higher. We set the heating power with a variable transformer so that the potassium vapor temperature was between 600 °C and 700 °C.
In Figs. 5(a)–5(c) we present three transmission profiles together with 5(d) the profile of the incoming fs laser, which spans the spectral region between 760 nm and 780 nm. From inspection of Figs. 5(a)–5(d), we may see that the transmission spectrum at higher potassium vapor temperatures is extended between 750 nm and slightly beyond 790 nm. This means that very dense potassium vapor is actually generating frequencies in the two spectral regions 750-760 nm and 780-790 nm, which lie outside the femtosecond laser profile. It is reasonable to assume that the self-phase modulation causes the laser pulse spectral broadening outside the 760-780 nm intervals. At the highest temperature of 700 °C we observe appreciable absorption of the Rb D2 resonance line, stemming from the rubidium impurity. In addition, close inspection reveals that the red wing intensity is appreciably enhanced, but the corresponding conical emission was not observed.
Fig. 5 Transmission of the femtosecond laser peaking at 770 nm, (a) The absorption of the first potassium doublet lines at 766 nm and 770 nm observed at 560 °C (). At higher temperatures the absorption of the rubidium impurity resonance line at 780 nm is observed at (b) 675 °C () and (c) 700 °C (). In (d) we show the spectral profile of the incoming fs laser beam. The self-phase modulation causes the laser pulse spectral broadening outside the 760-780 nm interval.
Finally, we observed the laser induced fluorescence spectrum at a right angle (split heating elements) of the fs laser beam at 406 nm (Femtosecond Frequency Doubler, Spectra Physics, Model 3980) for the vapor temperature of 475 °C. We observed the molecular band peaking at about 422 nm. This corresponds to the band at about 430 nm observed in the absorption spectrum of Fig. 1. The shift is caused by the absorption of fluorescence light until it exits the sapphire cell.
With two heaters connected the temperature in the whole sapphire cell was uniform and it was much higher. We set the heating power with a variable transformer so that the potassium vapor temperature was between 600 °C and 700 °C. In such cell conditions the transmission spectrum of the 404 nm laser beam revealed relatively weak molecular band peaking at about 415 nm, as shown in Fig. 7. This shift in wavelength, compared to Fig. 6, is caused by even stronger absorption of the K2 band peaking at 430 nm (compare with the absorption spectrum in Fig. 1). The absorption of the second potassium doublet at 404 nm is readily seen in Fig. 7.
Fig. 6 Femtosecond laser induced fluorescence spectrum of potassium vapor at 475 °C (femtosecond laser had a power of 50 mW peaking at 406 nm). Potassium dimer band had a maximum at 422 nm.
Fig. 7 Transmission spectrum of the femtosecond laser at 404 nm (dotted line). The absorption of the second potassium doublet at about 404 nm could be observed together with potassium dimer band in emission with a peak at 415 nm, for two temperatures at 640 °C (N = 2.3×1018 cm−3, red line) and 670 °C (, blue line).
4. Discussion
Our search for the conical emission in dense potassium vapor by means of the femtosecond laser transmission did not reveal any positive results. Just like in the experiment by Sarkisyan et al. [10] with 150 fs pulses, we could not observe the conical emission in the far field, even though stronger and shorter fs laser pulses were used. This is in contrast to the cases of denserubidium [9] and cesium vapors [8]. Two different phenomena were found in those two experiments. In the case of rubidium conical emission in the far field was observed when the fs laser (Tsunami, Spectra Physics) was tuned in the blue wing of the D2 resonance line, with increasing apex angle from 730 nm up to about 765 nm, and at atom densities above 5x1016 cm−3. However, in the case of cesium a cone emission was observed when the fs laser was tuned to the blue side of the Cs2 B-X molecular band. In both Rb and Cs2 cases the self-phase modulation caused effective self-focusing which was directly seen near the exit of the all-sapphire cells. No such self-focusing was observed in the present experiment with fs laser tuned in the blue wing of potassium D lines. Unfortunately, in the present experiment we could not tune the fs laser on the steep blue side of the K2 or Rb2 B-X bands because the gain curve of our femtosecond laser does not perform below 700 nm. The wavelength where the possible conical emission could occur in the case of K2 B-X band is approximately 625 nm, whereas for the Rb2 B-X band the analogous wavelength for inducing the conical emission should be at about 645 nm. The case of the blue wing excitation of the cesium resonance lines at 852 nm and 894 nm has not been observed yet.
It seems, on one hand, that nonlinear refractive index characteristics of dense potassium vapor are very small. It means that the intensity dependence of the nonlinear refractive index is inefficient to produce self-focusing [24]. In order to check this we shall employ in the near future a longer all-sapphire cell filled with potassium (L = 16 cm). On the other hand, it could be that the self-focusing might be precluded by efficient ionization of potassium atoms by three photon ionization. Pendrill et al. [25] observed directly an appreciable ionization whenever they did not observe the conical effect in the nanosecond laser excitation of potassium resonance lines. In a different experiment Pentaris et al. [26] observed conical emission at 5p-4s transition, after the 4s-6s two photon fs excitation, whereas they reported no conical emission at 4p-4s potassium resonance transition. However, their fs laser had much larger power and smaller repetition rate, which points to the different conical emission mechanism. The interplay between strong nonlinearity in the index of refraction and the three photon ionization may cause the formation of the soliton wave or light bullets in the dense potassium vapor [27]. It is a big contrast to dense rubidium and cesium vapors illuminated by the red femtosecond lasers. The question is why dense potassium vapor exhibits such an effect in which three photon ionization may be much more effective than in the case of rubidium and cesium. One of the possible answers could lie in the existence of the Cooper minimum [28] in the photoionization which is well known to be very deep in the case of potassium, rubidium and cesium [29]. Unfortunately, at present we do not know the positions and depths of the Cooper minima for three photon ionization in K, Rb and Cs atoms. We shall check in the near future whether in the case of potassium there exists appreciable ionization by three photon excitation in the experiment using thermionic detection [30,31]. It would be interesting to extend our studies to lighter atoms and their corresponding dimers like Li and Na, where in the latter case a conical emission was clearly observed with much more powerful femtosecond laser [32]. In the case of a positive outcome we might be able to use light bullets in establishing laser guided discharges [33,34]. In Table 1 we present the state of art in the observation of the molecular and atomic conical emission for K, Rb and Cs cases.
Table 1. Femtosecond Laser Induced Conical Emission in Dense Alkali Vapors
Dense alkali vapor generated in all-sapphire cells offer many interesting applications. However, they were not investigated by means of the mode-locked femtosecond laser in the blue and red spectral regions, where the first and the second principal series doublets are usually located. Potassium vapor at high temperatures offered us a unique opportunity to use the fs laser in the regions around 770 nm and 400 nm using fundamental and second harmonic beams. Our results point to a possibility that potassium molecules were destroyed by the strong femtosecond laser at 700 nm, but at 400 nm some molecules did survive along the laser beam.
Violet femtosecond laser was exciting the second principal series doublet, but again we did not observe any conical emission. However, since the power of the violet fs laser was much reduced, it did not destroy potassium dimers completely. Therefore, we were able to observe the molecular band in emission peaking at 422 nm and 415 nm depending on the potassium vapor density. The later was responsible for the absorption which shifted the peak of observed molecular band.
5. Conclusion
Hot potassium vapor was studied with femtosecond laser light at the second and the first resonance line doublets of potassium at very high densities and temperatures, enabled by means of the all-sapphire cell. We observed several interesting spectral features which point to a possible new laser effect if the cell were situated in the laser cavity. We also looked for conical emission when the fs laser was tuned to the blue wing of the K D2 line and the effect was extremely small. This indicates that self-focusing in dense potassium vapor is almost negligible or it is counteracted by the efficient three photon ionization. Therefore, we certainly hope to pursue additional experiments in order to clearly resolve the problem why femtosecond laser in the case of potassium does not exhibit conical emission.
Acknowledgments
We would like to acknowledge the financial support of the Research Administration of Kuwait University through research project GS 01/08 and GS 03/01. We are also deeply grateful for the support of the Kuwait Foundation for Advancement in Sciences (KFAS, Project code 2012151301). We greatly acknowledge the discussion with Dr. V. Vaičaitis.
References and links
1. D. E. Johnson and J. G. Eden, “Continua in the visible absorption spectrum of K2,” J. Opt. Soc. Am. B 2(5), 721–728 (1985). [CrossRef]
2. C. R. Vidal, “Vapor cells and heat-pipes,” Exp. Methods Phys. Sci. 29B, 67–83 (1996).
3. B. Steffes, X. Li, A. Mellinger, and C. R. Vidal, “Heat-pipe oven for large column densities with a well-defined optical path length,” Appl. Phys. B 62(1), 87–90 (1996). [CrossRef]
4. Y. Tamir and R. Shuker, “Novel design of hot windows in saturated metal vapor absorption experiments,” Rev. Sci. Instrum. 63(2), 1834–1837 (1992). [CrossRef]
5. A. Gallagher, “Line shapes and radiation transfer,” in Springer Handbook of Atomic, Molecular and Optical Physics, G. W. F. Drake, ed. (Springer, 2006), pp. 279–293.
6. S. Vdović, D. Sarkisyan, and G. Pichler, “Absorption spectrum of rubidium and cesium dimers by compact computer operated spectrometer,” Opt. Commun. 268(1), 58–63 (2006). [CrossRef]
7. S. Falke, H. Knöckel, J. Friebe, M. Riedmann, E. Tiemann, and C. Lisdat, “Potassium ground-state scattering parameters and Born-Oppenheimer potentials from molecular spectroscopy,” Phys. Rev. A 78(1), 012503 (2008). [CrossRef]
8. D. Aumiler, T. Ban, and G. Pichler, “Femtosecond laser-induced cone emission in dense cesium vapor,” Phys. Rev. A 71(6), 063803 (2005). [CrossRef]
9. H. Skenderović, T. Ban, N. Vujičić, D. Aumiler, S. Vdović, and G. Pichler, “Cone emission induced by femtosecond excitation in rubidium vapor,” Phys. Rev. A 77(6), 063816 (2008). [CrossRef]
10. D. Sarkisyan, B. D. Paul, S. T. Cundiff, E. A. Gibson, and A. Gallagher, “Conical emission by 2-ps excitation of potassium vapor,” J. Opt. Soc. Am. B 18(2), 218–224 (2001). [CrossRef]
11. W. F. Krupke, “Diode pumped alkali lasers (DPALs)—A review,” Prog. Quantum Electron. 36(1), 4–28 (2012). [CrossRef]
12. B. V. Zhdanov and R. J. Knize, “Alkali lasers for magnetic resonance imaging,” Cent. Eur. J. Phys. 8(2), 184–193 (2010). [CrossRef]
13. L. D. Turner, V. Karaganov, P. J. O. Teubner, and R. E. Scholten, “Sub-Doppler bandwidth atomic optical filter,” Opt. Lett. 27(7), 500–502 (2002). [CrossRef] [PubMed]
14. R. B. Miles, A. P. Yalin, Z. Tang, S. H. Zaidi, and J. N. Forkey, “Flow field imaging through sharp-edged atomic and molecular ‘notch’ filters,” Meas. Sci. Technol. 12(4), 442–451 (2001). [CrossRef]
15. A. Cerè, V. Parigi, M. Abad, F. Wolfgramm, A. Predojević, and M. W. Mitchell, “Narrowband tunable filter based on velocity-selective optical pumping in an atomic vapor,” Opt. Lett. 34(7), 1012–1014 (2009). [CrossRef] [PubMed]
16. C. Vadla, R. Beuc, V. Horvatic, M. Movre, A. Quentmeier, and K. Niemax, “Comparison of theoretical and experimental red and near infrared absorption spectra in overheated potassium vapour,” Eur. Phys. J. D 37(1), 37–49 (2006). [CrossRef]
17. G. Pichler, S. Milosevic, D. Viza, and R. Beuc, “Diffuse bands in the visible absorption spectra of dense alkali vapours,” J. Phys. B 16(24), 4619–4631 (1983). [CrossRef]
18. S. Milošvić, G. Pichler, R. Düren, and E. Hasselbrink, “Fluorescence studies of the K2 diffuse band at 572.5 nm,” Chem. Phys. Lett. 128(2), 145–149 (1986). [CrossRef]
19. R. Beuc, S. Milosevic, M. Movre, G. Pichler, and D. Veza, “Satellite bands in the far blue wing of the potassium first resonance doublet,” Fizika 14, 345–349 (1982).
20. A. N. Nesmeyanov, Vapor Pressure of the Chemical Elements, R. Gray, ed. (Elsevier, 1963).
21. V. Horvatic, C. Vadla, and M. Movre, “The collisional cross sections for excitation energy transfer in Rb*(5P3/2)+K(4S1/2)→Rb(5S1/2)+K*(4PJ) processes,” Z. Phys. D 27(2), 123–130 (1993). [CrossRef]
22. W. F. Krupke, R. J. Beach, V. K. Kanz, and S. A. Payne, “Resonance transition 795-nm rubidium laser,” Opt. Lett. 28(23), 2336–2338 (2003). [CrossRef] [PubMed]
23. H. Wang, Z. Yang, W. Hua, X. Xu, and Q. Lu, “Choice of alkali element for DPAL scaling, a numerical study,” Opt. Commun. 296, 101–105 (2013). [CrossRef]
24. J.-C. Diels and W. Rudolph, Ultrashort Laser Pulse Phenomena, 2nd ed. (Elsevier, 2006).
25. L. R. Pendrill, M. Pettersson, and U. Österberg, “Spectral variation of collisional ionization of excited potassium atoms,” Phys. Scr. 29(4), 313–316 (1984). [CrossRef]
26. D. Pentaris, T. Efthimiopoulos, N. Merlemis, and V. Vaičaitis, “Axial and conical parametric emissions from potassium atoms under two-photon fs excitation,” Appl. Phys. B 98(2–3), 383–390 (2010). [CrossRef]
27. Y. Silberberg, “Collapse of optical pulses,” Opt. Lett. 15(22), 1282–1284 (1990). [CrossRef] [PubMed]
28. J. W. Cooper, “Photoionization from outer atomic subshells. A model study,” Phys. Rev. 128(2), 681–693 (1962). [CrossRef]
29. I. M. Savukov, “Quasicontinuum relativistic many-body perturbation theory photoionization cross sections of Na, K, Rb, and Cs,” Phys. Rev. A 76(3), 032710 (2007). [CrossRef]
30. K. Afrousheh, M. Marafi, J. Kokaj, Y. Makdisi, and J. Mathew, “Spectroscopic studies of 5d3/2nd 1D0,2 autoionization lines of barium under collision with rare gases,” Phys. Rev. A 85(5), 052517 (2012). [CrossRef]
31. Y. Makdisi, J. Kokaj, K. Afrousheh, J. Mathew, R. Nair, and G. Pichler, “Multiphoton laser ionization for energy conversion in barium vapor,” Opt. Commun. 290, 95–99 (2013). [CrossRef]
32. V. Vaičaitis and E. Gaižauskas, “Conical fluorescence emission from sodium vapor excited with tunable femtosecond light pulses,” Phys. Rev. A 75(3), 033808 (2007). [CrossRef]
33. N. Vujičić, H. Skenderović, T. Ban, D. Aumiler, and G. Pichler, “Low-density plasma channels generated by femtosecond pulses,” Appl. Phys. B 82(3), 377–382 (2006). [CrossRef]
34. H. Skenderović, I. Labazan, S. Milosević, and G. Pichler, “Laser-ignited glow discharge in lithium vapor,” Phys. Rev. A 62(5), 052707 (2000). [CrossRef]
