1. Introduction

Laser ablation (LA) is a powerful technique that has been implemented in a number of material processing and analytical applications. Some of the well-known material processing applications include micro-fabrication/machining, additive manufacturing, and thin-film deposition [1]. Similarly, LA is an integral component of common analytical techniques like laser ablation inductively coupled mass spectroscopy (LA-ICP-MS), laser ablation laser absorption spectroscopy (LA-LAS), and laser-induced breakdown spectroscopy (LIBS) [2–4]. LA is a simple technique that requires a limited number of optical components – a pulsed high-power laser and a focusing lens. However, while the application of LA may be simple, the LA process is largely unexplored at the fundamental level since the underlying physics is exceptionally complex. The properties of the laser-induced plasma that results from LA depend on the laser parameters such as focal spot size, wavelength, pulse duration, and intensity [2]. It has been widely reported that LA with short-pulse lasers (e.g., ps and fs pulse durations) exhibits reduced laser-plasma coupling (inverse bremsstrahlung), a smaller heat affected zone, and a reduced magnitude of bremsstrahlung emission compared to LA with long (ns) laser pulses [5]. Further, it has been reported in the literature that ablation with shorter wavelength laser pulses exhibits a lower ablation threshold and reduced elemental fractionation compared to longer wavelength laser pulses [6]. Within the available literature, a number of studies have considered the effect of laser pulse duration, wavelength, and intensity on LA and the resulting laser-induced plasma at a fundamental level [7].

LIBS is a powerful all-optical technique for direct and rapid elemental analysis [2]. The technique involves the collection and spectral analysis of the emitted light from the luminous plasma produced following LA on the surface of a sample. One benefit of the LIBS technique is its inherent compatibility with standoff measurements due to its all-optical nature [8, 9]. LIBS can be used not only for elemental analysis, but isotopic analysis as well, and has been applied to measure several isotopes of B, Li, Zr, Pu, and U, for example [10–14].

The formation and emission of molecular species within laser-induced plasmas has been the subject of a number of recent studies [15–20]. The increased interest in molecular emissions has been largely driven by the notion that their measurement may provide more robust isotopic identification and discrimination [12, 14]. Vibronic emission spectra from diatomic molecules exhibit an isotope dependence, and the isotope shifts of the molecular emission lines are often greater than the isotope shifts of atomic and ionic emission lines especially for lighter molecules [16]. Molecular emissions from BO have been studied extensively for measurement of the B isotopes in B-containing samples [12, 14, 21]. In laser-induced plasmas molecules are formed through a number of reaction pathways: combustion or oxidation by direct interaction with the ambient atmosphere, recombination between species present within the plasma, and fragmentation of larger molecular clusters [18,20,22]. The laser pulse duration significantly impacts the fundamental properties of laser ablation plumes, which in turn affects the resulting emission properties of the expanding plasma plume. The molecular emissions are observed at longer delays after the arrival of the laser pulse (>1 μs for fs LA), when the plasma has cooled and undergone significant expansion into the ambient atmosphere, and persist for longer periods compared to the excited atomic and ionic emissions [16, 18, 19, 23]. When compared to ns-LA, fs-LA exhibits a number of distinguishing characteristics, including a lower plasma temperature, rapid plasma temperature decay, increased ablation efficiency, and the presence of a primarily atomic plume, resulting in more favorable conditions for molecular formation [7]. The emission intensity, delay, and persistence of the atomic, ionic, and molecular emissions are influenced by the plasma chemistry. Within the available literature, only a few works are available that have reported on the formation and evolution of the molecules and their emission, respectively, within a laser-induced plasma [15, 20, 22, 24].

In the context of molecular emission, significant work has been reported in the available literature on molecules comprised of low-Z elements such as B, Zr, Sr, and Al [14, 18, 19, 23]. The majority of these studies have been focused on the diatomic oxides that low-Z elements form in the ambient atmosphere. Even though a number of studies are available on the emission of diatomic molecules, little is known on the formation of more complex polyatomic molecules in a laser-induced plasma. The formation of polyatomic molecules is likely for elements like U and Al, which would deplete the population of diatomic molecules. Further, the formation of molecules within the plasma is dependent on the plasma chemistry, which is influenced by the ambient atmospheric conditions, and has not been studied for high-Z elements like U. Wormhoudt et al. reported on the reaction of uranium with ambient oxygen and showed that gaseous uranium readily reacts with oxygen to form uranium oxides [25]. Kaledin et al. reported on the emission of gaseous UO that was cooled through supersonic expansion of the UO gas to temperatures of ~130 K [26]. The collection of high-resolution gas phase electronic spectra of UO enabled the determination of molecular constants for low-lying electronic states of UO. While extensive literature exists on low-Z diatomic molecular emission in laser-induced plasmas, the understanding of the characteristics of uranium molecular emissions from laser-induced plasmas is lacking at the present time.

Recent work by our group has shown that the uranium plasma is very sensitive to the composition of the ambient atmosphere; a significant reduction in persistence of a U atomic ground state transition between ambient air and N₂ environments has been observed in absorption studies [15]. The abundance of reactive species such as O₂ in air leads to the formation of uranium oxides that may be responsible for the quenching of the atomic absorption signal. Limited studies exist on the effect of ambient atmospheric conditions on uranium laser-induced plasma emission and chemistry [15, 17, 27], and even fewer of them consider the formation of uranium oxides within LA plasmas [15, 17]. Due to the inherent nature of standoff measurements, LA is performed under ambient atmospheric conditions, which is why the formation and chemistry of uranium oxides within the plasma formed in air and their effect on the measured emission signal must be elucidated.

In this work, we report on the formation and dynamics of uranium molecules (uranium monoxide) in a laser-induced plasma. The plasma is formed following fs LA of a solid uranium metal sample (natural isotopic composition) contained within a vacuum chamber where the pressure and composition of the ambient atmosphere is varied. Uranium atomic and molecular emission features over a range of different air pressures and emission wavelengths are studied.

2. Experimental setup

The sample used was a uranium metal foil (natural isotopic composition, 0.7% ²³⁵U), with a size of approximately 0.5 cm x 2.0 cm and was lightly oxidized with minor surface defects. The sample was provided by The Pennsylvania State University’s Radiation Science and Engineering Center.

A schematic of the experimental setup is shown in Fig. 1. The vacuum chamber was evacuated to a residual pressure of ~0.5 Torr and backfilled with argon or air through an installed tube feedthrough. The sample was mounted to a holder, which is capable of translating the sample along the x- and y-axes perpendicular to the laser beam. The fs laser was a commercial Ti:sapphire chirped-pulse amplification system (Trident X, Amplitude Technologies), from which ~6 mJ, 800 nm pulses were split at a minimum pulse duration (FWHM) of 42 fs and at 10 Hz repetition rate.

 

Fig. 1 Simplified schematic of the experimental setup utilized in this work, which is licensed for laser ablation analysis of radioactive and special nuclear materials (SNM).

Download Full Size | PDF

The fs laser pulses were focused using a 30 cm focal length BK7 lens normal to the sample surface, producing a spot size of ~200 μm. The emission from the luminous plasma was collected and coupled into an optical fiber by a 5.0 cm focal length achromatic lens (400–700 nm) mounted at an angle of ~45° with respect to the laser beam path; hence the spectral emission from the plasma is spatially integrated. The optical fiber (a 1 mm diameter fiber bundle) was coupled out of the vacuum chamber through a fiber feedthrough and terminated at the spectrometer in a linear arrangement aligned to the entrance slit. The spectrometer (Horiba Jobin Yvon iHR 550) had an 1800 grooves/mm grating coupled to an Andor iStar 334t intensified CCD (ICCD). The plasma emission collection optics and spectrometer system were intensity calibrated using an Ocean Optics DH2000 Deuterium Tungsten calibration lamp.

For LIBS measurements, the plasma was generated over a range of ambient air and inert gas pressures measured using a digital manometer connected to the vacuum chamber. The ICCD acquisition gate delay and width was varied according to the measurement requirements, and each measurement was synchronized to the laser pulse using a Stanford Research Instruments (SRS DG645) Digital Delay Generator. Each spectrum is the accumulation of 10 individual laser ablation spectral measurements. Two cleaning shots were performed before each spectral measurement, which served to remove the thin oxide layer on the uranium metal sample.

3. Results

3.1 Spectral analysis of uranium plasmas

LA of uranium samples allows for the study of the atomic, ionic, and molecular species that are formed in the expanding plasma plume and through interaction of the plume with the ambient atmosphere. Depending on the irradiation conditions (pulse duration, laser wavelength, focusing configuration, etc.), the temporal behavior of the emitting atoms, ions and molecules can change drastically. Typically, LA plumes are hotter and denser at the earlier times of plasma evolution and become cooler and less dense as the plasma expands into the ambient atmosphere [28]. Due to high temperatures at early times of plasma evolution, free-free (bremsstrahlung) and free-bound (recombination) emissions dominate over bound-bound (characteristic spectral lines) emissions. Even though it is well known that ultrafast LA reduces the magnitude of continuum emission, the size of the heat affected zone, and the plasma temperature in comparison with ns LA [7] (which should create a favorable environment for molecular formation), limited experimental studies are available that provide evidence of these characteristics of fs LA of U metal.

Figure 2 shows the LIBS emission spectrum in the range of 400–600 nm from fs-LA of a uranium metal sample recorded with an ambient Ar pressure of ~82 Torr. Since U is a high-Z material, its plasma emits a very crowded emission spectrum. Many of the strongest optical transitions and energy levels for U atoms and ions have been reported in the literature [29, 30]. Selected uranium emission lines are indicated in the spectrum (Fig. 2) for both neutral (blue) and ionic (red) atomic emissions.

 

Fig. 2 Intensity calibrated LIBS emission spectrum of natural uranium metal sample following femtosecond laser ablation in the range of (a) 400 – 500 nm and (b) 500 – 600 nm with a 1.0 μs gate delay and 10 μs gate width averaged over 10 laser shots. After the vacuum chamber was evacuated, it was filled with argon to a pressure of 82 Torr; however, 3 Torr of ambient air leaked into the chamber during the measurement. Strong and commonly referenced uranium emission lines are labeled in the spectrum for neutral (blue), ionic (red), and molecular (black) emissions.

Download Full Size | PDF

3.2 Identification of uranium molecular emission from fs-LIBS plasmas

The fs LIBS spectrum shown in Fig. 2 was acquired under a rarefied Ar atmosphere of 82 Torr; however, due to the use of HEPA filters and additional fittings on the vacuum chamber to satisfy the mandated experimental safety procedure, a small but constant air leak into the chamber was observed. For the case of Fig. 2, the air leak caused approximately 3 Torr partial pressure of air to be present by the end of the spectral scan. The inset of Fig. 2(b) shows the uranium fs LIBS emission spectra for the spectral range studied by Kaledin et al. [26], where the authors identified a number of uranium monoxide (UO) emissions, including a strong Q branch emission centered at 593.57 nm. From inspection of the inset in Fig. 2(b), an emission feature is observed at 593.57 nm, and a review of the Los Alamos Scientific Laboratory Uranium Atlas [30], the Kurucz Database [31], and the NIST Atomic Line Database [32] reveals that this emission feature is not associated with either U I or U II emissions. Additional emission bandheads reported by Kaledin et al. [26] were not observed in the measured spectra, which could be due to the low intensity of the molecular emissions and spectral interference from nearby U I and U II emissions.

Uranium is pyrophoric and will oxidize when heated in an ambient air atmosphere [33]. In previous work by Kaledin et al. [26], Heaven et al. [34], and Wormhoudt [25], the emission of UO from the reaction of U with O₂ has been identified, 38 vibronic bands analyzed including the ²³⁵U and ¹⁸O isotopologues, and the center of one strong band for ²³⁸U¹⁶O is reported at 593.52 nm. Also, in emission at approximately 2500 K a strong band is reported at 593.48 nm. The latter study showed the dependence of the UO formation rate and emission on the concentration of O₂ available for the U to oxidize with. As the O₂ concentration was increased, uranium in the reaction flow tube readily formed UO₂ and UO₃ that reduced the population of UO. Moreover, Wormhoudt et al. showed that U oxidizes in the presence of O₂ and forms UO even when only trace concentrations of O₂ are present [25]. Based on these prior findings, it is hypothesized that the 593.57 nm emission line observed in Fig. 2 is due to UO emission. To confirm that this emission line is associated with a molecular transition, temporally resolved emission measurements and pressure dependent studies were performed.

3.3 Dynamics of uranium molecular emission in fs-LIBS plasmas

The ambient air pressure influences the physical and chemical conditions of the plume, which in turn affect the temporal history of the population of excited neutral atoms, ions, and molecules [8, 15, 23]. For example, the formation of UO within the plasma is dependent on the oxidation of uranium atoms in the plume in the ambient atmosphere. To further examine this dependence, we assessed the time evolution of U plasma spectral features at various ambient air pressures in the spectral region of 590–597 nm, where the UO band resides. Figure 3 shows the temporal behavior of the uranium emission spectrum as for several gate delays and widths at three different pressures (95, 440, and 730 Torr). At early times (100 ns delay), the spectrum is dominated by the bremsstrahlung and recombination emissions for all three pressures; however, several uranium atomic emissions are also visible, for example at 591.53 and 593.38 nm. At 1 μs delay the uranium atomic and molecular emissions are clearly visible for the 440 and 730 Torr pressures. As the plasma evolves (2 μs delay), the uranium molecular emission intensity increases above the nearby 593.38 U I emission. At late times (8 μs delay), the uranium molecular emission becomes isolated for the 440 and 730 Torr pressures, and only the strong 591.53 nm U I transition to the ground state is visible.

 

Fig. 3 Uranium LIBS spectra at 95, 440, and 730 Torr air pressure for several gate delays and widths. The spectra were intensity corrected for a range of ICCD gains used during the measurement. A vertical dashed line has been placed at the center of the observed UO emission line to help guide the eye.

Download Full Size | PDF

Figure 4 shows the temporally resolved peak emission intensity for the U I 591.53 nm and the UO 593.57 nm band. An exponential decay function was fit to the temporally resolved peak intensity of the two emission features from which we determine that the UO emission persists ~2 × longer compared to the ground state 591.53 nm U I emission. The measured persistence (1/e²) for U I and UO emission features are 7.3 µs and 13.5, µs respectively. The UO emission feature was observed at early times with significant intensity and continued to decrease (except for two outliers at 6 and 7 μs). The presence of UO molecules during the early times of plasma evolution could be related to the lower temperature of the plasma in fs LIBS in comparison to ns LIBS [23]. Figure 5 shows the ratio of the peak intensity of the 591.53 nm U I emission to the UO 593.57 nm peak intensity. The persistence of UO emission over the U I emission is highlighted in Fig. 5, where the ratio is below unity for the first 12 μs and then increases above unity as the hypothesized molecular emission dominates the emission spectrum.

 

Fig. 4 Temporally resolved peak intensity of (a) U I 591.53 nm and (b) UO 593.57 nm emission lines with a constant gate width of 1.0 μs. The smooth curves in the figure are the exponential decay fit. The measurements were taken at atmospheric pressure.

Download Full Size | PDF

 

Fig. 5 Temporally resolved molecular-to-atomic emission ratio for the emission lines referenced in Fig. 3 with a constant gate width of 1.0 μs under ambient atmospheric conditions.

Download Full Size | PDF

The effect of ambient pressure on the U and UO emission from the uranium plasma can be seen in Fig. 6. Both the U I emission and the uranium molecular emission intensity increase as the pressure initially increases, as shown in Fig. 6. However, at ~500 Torr the emission intensity begins to decrease for both species. The UO emission intensity at low (< 100 Torr) air pressures rapidly increases compared to the U emission intensity. For the lowest air pressure studied in this work (10 Torr), the UO emission intensity was less than the U emission intensity; however when the air pressure was increased to 50 – 100 Torr the UO emission intensity increased to about ~1.5X that of the U emission intensity. The increase in the UO emission intensity compared to the U emission intensity at low pressures suggests that an increase in the oxidation of U in the LA plasma occurs independent of confinement effects.

 

Fig. 6 U I 593.38 nm and UO 593.57 peak emission intensity as a function of the ambient air pressure in the vacuum chamber.

Download Full Size | PDF

4. Discussion

A number of studies have reported on the LIBS emission spectra of uranium and uranium containing samples [10, 15, 35–37]; however, only one study to date has reported the uranium LIBS emission spectrum over a range of 100’s of nm [36]. From inspection of the available spectroscopic databases for U I & II emission, over 92,000 spectral lines are observed originating from ~1600 energy levels [29], and according to Palmer et al. [30] there are at least ~2800 emission lines in the 400–600 nm region shown in Fig. 2. In the work by Palmer et al., the authors limited the number of emission lines reported for U I & II, and a number of additional low intensity emission lines are observed in the reported spectra that are not tabulated in the text. The presence of a large number of emission lines results in a crowded emission spectrum, where interference between individual emission lines is possible [27, 35].

The fs laser plasmas typically exhibit reduced continuum emission with limited persistence in comparison with ns laser plasmas [7]. The spectral features given in Fig. 2 have been recorded at a delay of ~1 µs. However, even at this relatively long delay the background emission is significant, indicating that a large fraction of the baseline is in fact due to the overlap of the wings of interfering uranium emission lines and possible weaker unresolved uranium emission lines [35, 37]. Strong U I emission lines are observed for fs LA at 462.02, 502.74, 528.04, 562.07, and 591.53 nm. The most commonly reported uranium emission lines in the 400-600 nm region occur at 409.01, 424.17, 424.43, and 424.62 nm; the latter three emission lines are shown in the inset of Fig. 2(a). The 424.43 nm uranium ionic emission is commonly studied due to its relatively large isotope shift (~25 pm) [38]; however, as can be seen from Fig. 2, the signal-to-background ratio (SBR) is low compared to the strong emission lines mentioned above. This result highlights the need for improved spectral modeling and analysis of the uranium emission over large wavelength ranges in ns & fs LA plasmas.

In previous work by Barefield et al. [36], long pulse (ns) lasers pulses were used to ablate mixed oxide uranium containing samples over an acquisition spectral range of 300–600 nm. The acquisition of the spectra was delayed over the range of 1–4 μs following the laser pulse in order to reduce the influence of the prominent bremsstrahlung and recombination emission in the measured LIBS spectra. Compared to the previous ns LIBS spectra, the fs LIBS spectra in Fig. 2 exhibit a lower background and an overall reduced characteristic emission intensity; however, comparison of emission line intensities and behavior over the measured spectral window is not possible due to the lack of intensity correction in the ns LIBS results.

Compared to ns LA, fs LA results in a cooler, less dense plasma that exhibits reduced bremsstrahlung and recombination background emission as well as reduced spectral broadening [7]. The reduction in the background emission allows for the use of non-gated detectors and improved SBR for the measured emission lines [21, 39]. Due to the large number (1,000’s) of emission lines in the spectra of high-Z elements relevant to nuclear security, a reduction in the spectral broadening of the emission lines in fs LIBS measurements improves the potential for discerning the individual emission lines that may otherwise be convolved or overlapped in ns LIBS measurements. Further, as the isotope shift of both atomic and molecular emissions are of interest for uranium and other high-Z elements, the isotope shift-to-linewidth ratio must be considered as a figure of merit, quantifying the ability to resolve an isotope shift in the measured LIBS spectra. Based on the above discussion, fs LA offers a number of analytical merits compared to ns LA for the analysis of nuclear security relevant materials.

Figures 3 and 4 show the UO emission appears at later times when compared to atomic emission, and UO is predominant at later times of the plasma lifecycle. In laser-produced plasmas, the ionic species exist only at early times of the plasma evolution when the plasmas are hotter and hence the persistence of their characteristic emission is limited. Neutral emissions appear when the plasmas are cooler and can persist for longer times in comparison with ions. The plasma-assisted chemical reactions occur in the presence of an ambient atmosphere once the plasma has sufficiently cooled (~5000 K), which leads to production of excited molecular radicals [11, 25, 26]. Hence the appearance of molecular emission is delayed and the molecular emission can persist for longer times in comparison with emission from neutral and ionic species [7, 26]. The delay in the formation and observation of the radiating molecular species is shorter (~300–500 ns) for fs LA compared to ns LA (~5–10 µs) which could be the consequence of a range of conditions in fs LA, including the lower plasma temperatures, reduced thermal effects during ablation, increased ablation efficiency, weaker shock wave, etc [7, 26].

The intensity of both the U I and UO emission increases as the pressure initially increases and is shown in Fig. 6. The observed increase in the emission intensity as the gas pressure is increased has been explained as a result of a simultaneous increase in the number density and excitation temperature of the emitting species due to collisional confinement of the plasma by the ambient gas [40]. At $\gtrsim$500 Torr the emission intensity begins to decrease for both U I and UO species, which is partially attributed to rapid collisional cooling of the plasma and self-absorption in the high density U plasma at higher ambient air pressures [41]. The decrease in the uranium atomic and molecular emission as the pressure was increased above ~400 Torr can be further explained by the preferential formation of complex uranium oxides (U_xO_y) at higher air pressures [25].

The oxidation of uranium and the formation of complex uranium oxides in LA plasmas has not been studied extensively in the literature, and the formation and depletion mechanisms of UO in a laser produced uranium plasma is unknown. It has to be mentioned that the oxidation in a transient laser-plasma system expanding into an ambient is dictated by diffusion of oxygen molecules into the plasma, strength of the shockwaves which hinder plasma-ambient interaction and plume thermodynamics. A simplified version of the gas-phase chemistry for the reaction of atomic U with O₂ can be represented by the following reaction pathways:

The reaction of U with N₂ present in air to form NUN is possible through an insertion mechanism [45]. The dissociation bond strength for UO compared to O₂ favors reaction of U with O₂ to form UO; however, atomic uranium cannot abstract a nitrogen atom from N₂ as the reaction is endothermic and the required activation energy for this reaction (>99 ± 5 kcal/mol) is not likely to be provided by hypothermal uranium atoms in the plasma plume [45]. Further, the insertion reaction required to form NUN is only slightly exothermic compared to the significantly exothermic uranium oxidation reaction (222 ± 7 kcal/mol) shown above in reaction (1) [45]. This is also supported by the fact that our recent laser absorption studies showed that the persistence of the U ground state population is significantly reduced in the presence of air compared to nitrogen environment [15].

The UO/U line intensity ratio is found to peak at ~100 Torr air pressure (Fig. 6). The rapid cooling and expansion of the inhomogeneous plasma plume into the ambient atmosphere results in complex plasma dynamics and occurrence of plasma-assisted chemical reactions. The reaction pathways, rates, and temperatures of formation are not well understood for most of the possible plasma assisted chemical reactions that can take place within the expanding uranium plasma plume, and a limited number of studies have been performed on this topic [17, 20, 22, 23]. As the ambient air pressure is increased, preferential formation of both simple diatomic and complex polyatomic oxide molecules occurs, which depletes the atomic population density within the plasma plume [4, 15, 46].

In the fs LIBS results reported in this work, only the strong Q branch UO emission feature is observed due to the weaker emission intensities of the P and R branch emissions and the high temperatures present during the emission [26, 34]. Further, the complex spectrum consisting of atomic, ionic, and molecular emissions could lead to the additional emission branches being overlapped by atomic or ionic emission features. Lastly, the rotational structure in the Q branch is not resolved in the measured fs LIBS spectra due to the relatively strong broadening of the individual emission lines and close spacing of these emissions in the Q branch for UO.

5. Summary

We investigated the temporal evolution of uranium containing plasmas using optical emission spectroscopy under a number of ambient air pressures. A 200 nm wide spectral scan of the uranium emission originating from fs LA produced plasmas is reported and was compared to previous work available in the literature on ns LA of uranium containing samples. The spectral emission features showed the presence of atomic, ionic, and molecular uranium species from fs LA plasmas. The presence of a UO molecular emission feature at 593.57 nm in the fs LIBS plasma was confirmed through analysis of the temporally resolved emission spectra, the dependence of the emission spectra on the ambient gas pressure, and the previous work on uranium oxide emission and formation reported in the open literature. The enhancement of the uranium molecular emission intensity as the ambient air pressure was increased, along with the greater persistence of the molecular emission compared to the atomic emission, confirms that the emission originates from a molecular transition within the fs LA plasma.

Even though the fs LA provided reduced continuum emission, a significant background emission can be seen in the recorded spectral features. The presence of a large number of emission lines in the spectra results in a crowded emission spectrum, where interference between individual emission lines is possible. Hence strong background emission that appears as spectral features could be the result of the overlap of uranium emission lines. Extensive LIBS work exists in the literature on isotopic separation of the U atomic emission lines; however, the emission lines which exhibit the strongest known isotope shift have a considerably lower SBR (SBR = 1.5 for 424.43 nm) compared to the strongest observed atomic emission lines (SBR = 8 for 591.53 nm) in the 400 – 600 nm region [29, 38]. These results highlight the need for improved spectral modeling and analysis of the uranium emission over large wavelength ranges in fs LA plasmas.

Previous work by our group [15] has shown that the magnitude and persistence of the uranium absorption signal are significantly reduced in air compared to N₂ environment. The present results show that the presence of oxygen in the ambient environment promotes formation of molecular species through an oxidation process. The oxidation process removes ground state atomic uranium from the system, which explains the observed decrease in the uranium absorption signal in our previous work. The intensity of both the U I and UO emission increases as the pressure increases; however, at higher pressure ($\gtrsim$500 Torr), a reduction in signal intensity is seen for both species. Additional plasma-assisted chemical reactions can result in the formation of more complex uranium oxide (U_xO_y) molecules that could reduce the emission intensity of both the neutral atomic and UO species. Although our studies showed the presence of strong UO band in fs LA plumes and its time evolution, the detailed chemical reaction pathways for UO are still unknown. Hence further work must be conducted to investigate the formation of UO in fs LA plasmas under a range of absolute oxygen concentrations and independent of any plasma confinement effects in the measurement. Finally, the present work is very relevant to the LIF, LA-LAS, LIBS, and LA-ICP-MS communities, where generation of oxides within the LA plasma affects the population density of the species of interest.