1. Introduction

There is a significant interest in understanding the physics and chemistry of molecular formation in laser ablation (LA) plumes which is relevant to a large number of LA applications such as reactive deposition, nanoparticles synthesis, combustion and analytical tools, fireball plasma chemistry and isotopic analysis [1–5]. The molecules in a laser-produced plasmas (LPP) are formed through a number of pathways and prominent among them are oxidation through interaction with ambient O₂ (combustion), recombination with other elements present in the matrix in the gas-phase, and fragmentation/dissociation of larger clusters/molecules, and all of these processes are directly related to the physical conditions of the plasma plume. Typically, a laser-plasma is hotter in the early times of its evolution with temperatures in excess of 10,000 K that inhibit molecular formation. In addition to this, the shock waves formed in the plume-ambient interface will act as a barrier for oxidation by separating the plume from the atmosphere at early times of its evolution [6]. However, the rapid decay of temperature due to expansive cooling, combined with the weakening and collapse of the shock front due to expansion, results in an increase in collisions of atomic plume species with atmospheric constituents. Even though extensive studies are available in the literature, still there exists a lack of knowledge about when and where the molecules are formed in a transient plume system like an LPP, and this can be related to complex plasma chemistry and transport processes involved in a transient laser-plasma system. Recent studies have shown that there exists a correlation between the target molecular structures and emission from LA plumes [7, 8].

A large number of applications of LA are performed in ambient air where the presence of oxygen will result in the formation of molecular oxides [5, 9, 10], It is well known that oxygen present in the ambient gas affects all LA-based optical analytical tools such as laser-induced breakdown spectroscopy (LIBS), laser-absorption spectroscopy (LAS), and laser-induced fluorescence (LIF) spectroscopy through a reduction in analyte signal intensity and its persistence [11, 12]. However, so far the reported studies have been focused on the generation of lighter molecules (BO, CaO, AlO, SrO, SiO, AlN) [5, 6, 13–15], and limited knowledge exists on the formation of diatomic molecules of high-Z elements and polyatomic molecules in an LPP. For example, the understanding of formation channels for U molecules in an oxygen rich and high-temperature environment is very limited. U plasma chemistry has significant importance for the forensic analysis of nuclear samples as required for nuclear reactor diagnostics, fuel fabrication, waste treatment, and treaty verification. LPP also provides an excellent lab-scale platform for studying actinide gas-phase plasma chemistry, which is beneficial for non-proliferation monitoring and nuclear forensics. Gas-phase U chemistry is also important for understanding the U nucleation physics and debris generation.

Detailed spectroscopic studies exist in the literature on UO emission via LIF of super-cooled supersonic expansions of laser-produced uranium plasmas at low temperatures (~130 K) and from UO thermal emission recorded from a high-temperature furnace [16, 17]. These fundamental spectroscopic studies have identified numerous UO electronic transitions in the UV-VIS spectral regions. Despite the identification of numerous UO emission bands, almost all emission studies in LA plumes focus on identification of UO based on a spectral emission feature around 593.55 nm [9, 10, 12, 18]. The spectral features of U containing plasmas in air are typically reported to present a large poorly-resolved background signal, and several articles relate its origin to overlapped and crowded spectral features of U atomic transitions [19, 20], however, no systematic studies at high-resolution have attempted to elucidate the identity of these emission features, nor have there been studies exploring the influence of the plume dynamics due to confinement from ambient gas on spatial extent of U_xO_y formation. The present studies are directed at understanding the complex structure of the broadband uranium oxide emission in a uranium metal LPP, and attempting to identify multiple UO spectral features in high-resolution spectra collected in the visible spectral region. Initially, laser-ablation studies of U metal foils in near-vacuum conditions are carried out to evaluate the effect of surface oxidation on the emission features from the uranium metal plasma. This is followed by studies performed in atmospheric air or in 100 Torr Ar with trace O₂ to suppress the formation of larger oxides and to help isolate UO emission from U atomic emission. Multiple UO emission features are identified based on the comparison to furnace emission data from Kaledin et al. [16, 17]. Our studies also highlight that the observation of broadband emission in the visible spectral region from a laser-produced uranium plasma is contributed by emission from UO and higher oxides.

2. Experimental details

For laser ablation, pulses from a Q-switched Nd:YAG laser (Surelite I, 1064 nm wavelength, 6 ns full-width half maximum (FWHM), 0.5 mrad divergence) were focused on a 1 mm thick U metal target using a f = 15 cm plano-convex lens with an estimated spot size ~1 mm, providing a laser fluence on the target ~12 J/cm². The sample used was a uranium metal foil (99.9% purity, natural enrichment, 99.3% ²³⁸U and 0.7% ²³⁵U), with a size of approximately 1.0 cm × 1.0 cm. The sample was mounted in a 6” cube-shaped vacuum chamber and pumped using a dry pump which provided a base pressure of ~4 mTorr. The vacuum chamber had optical windows for laser, emission collection and ports for gas feed-through, and pressure gauge. The gas feed-through is connected to a gas manifold so that the nature and pressure of the ambient medium in the chamber can be controlled. The entire chamber was mounted on a motorized x-y-z translator so that it can be moved to avoid the effects of sample drilling and cratering.

For performing emission spectroscopy, the self-emission from the plasma was collected at right angles to the plasma expansion direction using appropriate optics and imaged onto the slit of a 0.5 m triple grating spectrograph (Princeton Instruments, Spectrapro 2500i). The wavelength-dispersed light from the spectrograph was recorded using an intensified CCD camera (ICCD, Andor iStar 334T). The spectra were recorded either using a 2400 grooves/mm grating in the UV-VIS region providing a spectral resolution ~0.04 nm with a 30 µm slit opening, or using a 1200 grooves/mm grating (spectral resolution ~0.08 nm). The instrumental broadening was measured using a frequency stabilized He-Ne laser. The wavelength calibration was performed using U spectral lines. The ICCD acquisition gate delay and integration times were varied according to the measurement requirements, and each measurement was synchronized to the laser pulse. A dove prism was inserted into the optical path for capturing 2D spectral imaging. The ICCD camera was also used for self-emission images of U plasma at various pressure levels which provided information about the morphology of the plume and emission persistence.

3. Results and discussion

Uranium, as a high-Z material, presents a very crowded emission spectrum in the UV-VIS. According to the literature, there are more than 92,000 neutral and singly ionized emission lines in the visible region originating from nearly 1600 energy levels [21–23]. In the present study, ns laser pulses with energies ~75 mJ and with a fluence ~10 J/cm² (~2 GW/cm²) are used to generate laser-produced plasmas from the U metal foil. Although the presence of high-temperature conditions in the plume hinder the molecular formation at early times of plasma formation, the use of significantly higher laser intensity compared to the ablation threshold will be useful for increasing the persistence of atoms and molecules which is essential for segregating the molecular spectral emission features from the crowded and overlapped spectral features of U atomic lines. The emission from the molecules in a LPP typically appears at later times of its evolution compared to the excited atomic and ionic emissions [24]. This can be related to the requirement of cooler conditions in LPP for enhancing molecular formation as well as the shock wave mediation of plasma chemistry in the plume [6]. In addition to this, the common line broadening mechanisms, such as Stark and Doppler, will be negligible at later times of plasma evolution [4] where molecular emission is prevalent.

The formation routes for gas-phase uranium oxide species in a plasma from an oxidized sample could be through gas-phase oxidation with ambient oxygen or through gas-phase interaction between U atoms and oxygen from the sample. For differentiating UO formation routes, ablation experiments were performed in 4 mTorr pressure where the oxygen supply is limited to sample oxide contamination or at 100 Torr ambient pressure where UO formation through gas phase collisions with ambient species will be predominant. At ~4 mTorr pressure, the number densities of N₂ and O₂ are ~5 × 10¹⁴ /cm³ and ~1 × 10¹⁴ /cm³ with respective room-temperature mean-free paths (among the ambient gas species) ~1.5 cm and ~5.6 cm respectively. Hence at this ambient pressure range, the laser-plasma is in the free-expansion regime [25]. It is well known that the presence of an ambient gas during laser-plasma expansion leads to plume confinement which in turn increases collision rates of atoms and molecules in the plume. The ICCD fast-gated images of the U plume obtained at 4 mTorr and 100 Torr air pressure levels given Fig. 1 attest the expansion and confinement properties of plasma plumes at varying ambient pressure levels. At 4 mTorr, the plume expands freely, and at 100 Torr the plume expands at a reduced velocity up to ~15 µs and then collapses because of the pressure exerted by the ambient gas at later stages of plume expansion. The plume showed an elliptical morphology which can be related to the use of large spot size (~1 mm) [26]. The images obtained at 100 Torr pressure show a collimation along the plume expansion direction at times ≥ 1 µs with the most intense emission appearing away from the target surface, and this is due to the deceleration of energetic U atoms/ions occupied in the plume front positions by the ambient gas species. The asymmetric shapes of plasma plumes at times 0.5-6 μs are due to the presence of instabilities generated in the plume front by the ambient gas medium [25]. The role of ambient gas on U plasma is studied at 100 Torr pressure levels because at these pressure levels the plasma plume is expected to provide the highest emission signal to noise ratio (SNR) and/or signal to background ratio (SBR) [27].

 

Fig. 1 The temporal evolution of U plasma in (a) 4 mTorr and in (b) 100 Torr air pressures. The timings given in the images represent the time after the onset of plasma formation. Each image is obtained from a single laser shot and normalized to its maximum intensity. The gating times used for capturing self-emission images in vacuum were 2 ns until 100 ns delay and 5ns for times > 100 ns. For recording the images at 100 Torr air, 2 ns gate widths were used until 500 ns, and 10% of gate delay times were used at times > 1µs. The arrow mark represents the laser direction.

Download Full Size | PDF

The U metal sample used in the present study is exposed to air, therefore surface oxidation of the U sample cannot be avoided [28]. The surface oxidation mechanisms of uranium in the presence of oxygen or water vapor are well-known, and the thickness of the oxide layer on the U metal target depends on the duration of its exposure to ambient air and environment (e.g. humidity). The LA process can be used for removing the oxide contamination from the U metal sample, and the thickness of oxide layer can be measured by monitoring UO emission signatures. The measurements are performed at 4 mTorr pressure (free expansion regime), and hence the UO formation is dictated by the gas-phase interaction between U atoms and oxygen from the sample. The spectral region selected for this study is 591-598 nm where several U atomic lines are present in addition to the well-known UO band at 593.55 nm [9]. Figure 2(a) gives the spectral features of the U plasma obtained during a sequence of laser shots hitting at the same location on the target. The laser delay and gate width used were 0.5 µs and 5 µs, and the spectral measurement was performed at a distance 2 mm from the target. The prominent UO peak at 593.55 is marked along with other major U atomic lines. The spectral features show the disappearance of the UO peak with an increasing number of laser shots, and negligible UO emission is observed after 37 laser pulses. Assuming 50% laser energy coupling to the target, a rough estimate of the ablation efficiency [29] showed ~0.4 ± 0.2 µm/pulse.

 

Fig. 2 (a) The spectral features collected from U plasma in vacuum during the surface cleaning process. The numbers given in the plot represent the sequence of laser shot used. The laser delay and gate width used were 0.5 μs and 5 μs. The emission analysis was performed at a distance 2 mm from the target surface. In (b), the changes in the spectral intensity of U I 597.15 nm, UO 593.55 nm and background signal intensity (595.05 nm) are given.

Download Full Size | PDF

The spectra given in Fig. 2 were each obtained from single ablation shot with no averaging. An interesting observation from the spectral features during the cleaning process is nearly an order of intensity reduction of the background signal along with the disappearance of UO peak at 593.55 nm with an increasing number of laser shots. Since the spectral information is recorded under the free-expansion regime [see Fig. 1(a)] and at a time delay of 500 ns, the contribution from the plasma continuum (free-free or free-bound radiation) is negligible. This indicates that the background emission could be contributed by some other mechanisms. The changes in the intensity of background-subtracted U I 597.15 nm, UO 593.55 nm and background signal intensity obtained from 595.05 nm, where U I spectral lines are absent, are given in Fig. 2(b), and it shows both the background intensity as well as UO emission peak intensity decrease with reduction in oxide layer thickness from the sample while the U I intensity showed more or less similar intensity during the cleaning process. This observation indicates that the strong background-like emission present in the U spectral features could be caused by oxide spectral emission. An enhancement in U I signal intensity can be seen in Fig. 1(b) for pulses up to ~15 shots, and this is due to a reduction in oxygen scavenging of U atoms for UO formation. A slight reduction in the U I emission intensity is observed beyond 15 shots, and this may be due to crater effects.

Previous studies have shown that the presence of strong background-like emission from U plasma and attributed this to the overlapping of crowded U atomic spectral features [9, 19, 20]. However, the present studies clearly demonstrate that the background-like emission from the U plasma is caused by the emission from unresolved gas-phase U oxides, and in the following, we define this background-like emission as the uranium oxide transition array (UOTA). This is also consistent with our previous studies on LA of U samples using LAS and LIF which have shown that the intrinsic absorption linewidths of U transitions in LPPs are < 2 pm and that U atomic absorption spectra can be measured without a background absorption [30, 31], which suggests that the background observed in emission spectra is due to overlapping spectral lines broadened by the spectrograph resolution.

The self-emission images given in Fig. 1 show that the persistence of the plasma increased significantly with the addition of an ambient gas for confinement. Thus, conducting spectral studies of uranium plasmas in an oxygen-rich ambient environment is useful for increasing the gas-phase collisions and promoting UO formation. However, identification of molecular features in a crowded spectrum like a U plasma is always a challenge when considering that the molecular emission features will co-exist with emission from neutral U atoms originating from near resonance or resonance transitions even at late times of the plume evolution. The molecular emission is favored at late times during plasma evolution as well as farther away from the target surface where the plasma is relatively cooler. Considering the composition of air, both nitrides and oxides can be formed during the interaction between the U plume and ambient air species. However, the formation of nitrides may be insignificant due to extremely high bond dissociation energy of N₂ molecules in comparison with uranium reaction pathways with oxygen molecules. This is consistent with our previous reports on LAS and LIF of a ~1% U containing plasma where a significant reduction in U ground state population was noticed in the presence of oxygen-rich environment while the persistence of signal was enhanced in the ambient N₂ environment [12, 32].

Time-resolved spectral features are recorded to evaluate the evolution of atoms and molecules in the U plasma in 100 Torr air, and results are given in Fig. 3(a). As the figure shows at early times, the emission from U atoms dominates while UO emission predominates at later stages of plasma evolution. The recorded spectral features at a 40 μs delay time with a 30 µs gate width show the prominent UO band at 593.55 nm and U I resonance transition at 591.539 nm. The recorded persistence (τ_p) of UO emission (⪞ 40 µs) is significantly higher compared to plasmas generated using 40 fs, 6 mJ pulses (τ_p ~8 µs) [9] and 6 ns, 2 mJ pulses (τ_p ~1 µs) [18]. It should be noted that the spectral features recorded from the U plasma even at late times of its evolution show substantial UOTA emission. The 2D spectral image given in Fig. 3(b) shows the spatial extent of UO emission at 593.55 nm along with U I at 591.539 nm and UOTA emission.

 

Fig. 3 (a) Time-resolved spectral features recorded from U plasma at a pressure of 100 Torr air and at a distance 1.5 mm from the target surface. The sample surface oxidation was cleaned using multiple laser shots before acquiring the spectral features. The gate widths used were 200 ns, 2 μs, 20 μs and 30 μs for delays 2 μs, 10 μs, 20 μs and 40 μs respectively. (b) 2D spectral image obtained with 20 µs delay and 40 µs gate width is given.

Download Full Size | PDF

Though it can be ascertained that the background-like emission in a U plasma is contributed by uranium oxides, the time-resolved spectral features show the appearance of one prominent UO band (593.55 nm). Kaledin and Heaven [16] identified 180 UO bands during their furnace studies in the visible spectral region at temperature ~2400K. They used a controlled thermodynamic condition for optimizing the steady-state concentration of UO, while in the present experiment the presence of high-temperature environment along with transient nature of LPP will not help to elucidate the crowded ro-vibrational bands of UO. In addition to this, the plasma chemistry of U shows the UO is acting as a precursor for higher uranium oxides (U_xO_y). Finko et al. [33] recently developed a reaction mechanism to describe U_xO_y formation with initial conditions relevant to a plasma generated by ns laser ablation in the air. Their 0-D simulations demonstrated that the consumption of UO and formation of UO₂ and UO₃ occurs on the timescales of 10 - 100 µs. While 0-D simulations ignore the transport properties of the plume, they may be similar to the physical conditions at the interface of the laser ablation plume and the atmosphere.

To extend the lifetime of UO in the LPP it is critical to limit the formation channels for higher U oxides, and this can be achieved by reducing the oxygen concentration in the gas. The U oxidation process can be controlled by limiting the number density of oxygen molecules present in the background gas by lowering the pressure of the ambient environment. But, the reduction in ambient pressure leads to a decrease in the probability of oxidation reaction routes because of the reduced plume-ambient species interactions which in turn reduces the UO emission intensity and synthesis of uranium oxides larger than UO. Instead, by adding a trace amount of oxygen to other non-reactive ambient gas such as Ar ~100 Torr is useful to improve the persistence of all species in the plume (by controlling oxidation pathways), as well as to separate the atomic and molecular emission features in time. So spectral analysis of U plasma is carried at 100 Torr pressure Ar pressure which contained trace quantity of O₂ molecules (~2%).

Figure 4 gives the U spectral features recorded at ~1.5 mm from the target from a LPP in the spectral range 591-601 nm under different delay conditions and in the presence of ~100 Torr ambient pressure (~98 Torr Ar and ~2 Torr O₂). For comparison purposes, the measured spectrum in low-pressure (4 mTorr, 1 µs delay and 5 µs gate width) conditions is also given in Fig. 4(a) which primarily contained emission from U atoms and is used for absolute wavelength calibration. The spectral features obtained at low-pressure conditions showed insignificant background, indicating that the contribution from continuum and UO emission are negligible. The spectral features given in Fig. 4(b), recorded with a 25 µs delay and 15 µs gate width, show significant UOTA along with atomic lines originating from lower lying levels, a UO spectral feature at 593.55 nm and additional spectral features presumably from UO. To help discriminate atomic from potential molecular emission, a simulated emission spectrum for U I has been calculated and plotted on a separate y-axis in Fig. 4(b). This spectrum was simulated using line parameters from the Palmer spectral atlas [21] and was broadened by a 0.05 nm FWHM Lorentzian function to account for the instrumental broadening from the spectrograph. The emission was simulated assuming a 0.2 cm slab of plasma with a uniform U I number density of 1 × 10¹² cm⁻³ at a temperature of 4000 K. While the actual temperature of the plasma and U I number density cannot be ascertained due to significant spectral congestion, the simulated spectrum serves as a qualitative tool to identify U I emission from potential uranium oxide emission. The Palmer spectral atlas only has ~4000 assigned transitions – a fraction of the estimated ⪞ 90,000 emission lines predicted in the UV-VIS spectral region.

 

Fig. 4 The U spectral features recorded from a laser-produced plasma from U metal is given. (a) The spectral features are recorded at near vacuum conditions (4 mTorr). The delay and gate times used were 1 µs and 5 µs respectively. The spectral features recorded at 100 Torr ambient (98% Ar and 2% O₂) are given in (b) and (c). The delay and gate width used for recording (b) was 25 µs and 15 µs. A spectral simulation of U I emission at a temperature of 4000 K is plotted on the right y-axis of (b). In (c), the spectral information collected with gate delay and width 40 µs and 60 µs is given. The positions and relative band intensities observed by Kaledin and Heaven [16] in the furnace studies are provided for comparison in (c).

Download Full Size | PDF

Assignment of molecular and atomic emission lines can be aided by looking at the time dependence of the emission by recording emission at longer time delays after the onset of plasma formation. A longer delay will favor formation of additional uranium oxides while reducing the number density, and therefore intensity, of the U I lines. A delay time of 40 µs with a 60 µs integration time is used in Fig. 4(c) for isolating UO emission from U atomic lines. The observed U atomic lines in this spectrum belong to resonance transitions at 591.539 nm and, 597.15 nm, and the rest of the structures seen could be assigned to UO. We see that the relative emission intensity of the U I resonance transition at 591.539 nm is roughly equivalent to the 593.55 nm spectral feature assigned to UO. This is evidence of the reduction in the U I number density due primarily to reaction with trace oxygen in the chamber. The wavelength position, as well as the intensity of UO bands reported by Kaledin and Heaven [16] acquired from emission measurements in a high-temperature furnace with uranium metal are also marked in Fig. 4(c). The most prominent UO emission lines in the spectral window given in Fig. 4(c) are 593.55 nm, 597.62 nm, and 600.57 nm. The assigned UO spectral features agree well with emission structures obtained from the LPP although a red-shift ~10-70 pm is apparent for most of the peak positions. The wavelength calibration of the spectral features given in Fig. 4 is performed using U atlas data from Palmer [21], and the measured error in the wavelength position is limited to < 10 pm. It can be speculated that the temperature of a laser-plasma system may influence the UO spectral position due to various broadening mechanisms as well as the complex overlap of multiple U_xO_y emission features. Typically, laser-plasmas are hotter at early times, and the temperature decays monotonically with time. However, the time evolution study showed an insignificant shift in the spectral position of UO band at 593.55 nm. This observation indicates that the spectral red-shift seen in the present study may not be related to differences in temperature between furnace system (~2400K) used by Kaledin and Heaven [16] and laser-plasma. Some of the plausible reasons for spectral-shifts include significantly more emission from larger uranium oxides than in the furnace studies, overlapping of congested emission from UO₂ or UO₃ which may lead to subtle shifts in the position of the peak of spectral features observed in the emission spectrum etc. Their furnace studies [34] were performed in a controlled thermodynamic condition for optimizing steady-state concentration of UO. But in laser-plasmas, because of its transient nature, the formation of larger oxides cannot be avoided.

The present studies clearly highlight that the presence of uranium oxides create a strong background-like emission (UOTA) in U spectral features. Further analyses were carried out for identifying the UO spectral line features in the 500-770 nm VIS spectral region using a 1200 grooves/mm grating (resolution ~0.08 nm), and several UO spectral features were identified. Some of the prominent UO spectral features located at 514 nm, 590 nm, 714 nm, and 727 nm are given in Fig. 5 along with U I spectral modeling data and the UO band positions identified in the furnace studies. Comparing among the UO spectral intensities, several overlapped UO emission peaks appear around 514 nm and 595 nm, however, isolated peaks can be seen in the longer wavelength side (for example 714 nm, 727.54 nm). The strong presence of UOTA is seen for all selected spectral regions used to locate UO bands identified by Kaledin and Heaven [16]. The wavelengths of the strongest and isolated UO spectral features measured and the peak position observed from ref [16]. are given in Table 1.Though the present work identified UO emission peaks, the role of higher uranium oxides on the UOTA is unclear currently, and further study in this direction is necessary.

 

Fig. 5 UO spectral features recorded with 100 Torr pressure (Ar 98 Torr and O₂ 2 Torr) at various spectral regions. For discriminating U atomic and UO molecular emission, a simulated emission spectrum for UI has been given in each plot. The positions and relative band intensities observed from furnace studies are also provided for comparison with laser-produced plasma spectrum. (a) Spectral region 510-518 nm, (b) spectral region 585 −605 nm, (c) 709- 720 nm spectral region and (d) 719-729 nm spectral region. A delay and gate times used for recording the spectral features given (a) and (b) were 40 μs and 60 μs respectively while 30 μs and 60 μs were used for measuring the spectral features given in (c) and (d).

Download Full Size | PDF

Table 1. Identified UO spectral bands peak wavelength (in air) and UO peak position reported from furnace data [16].

View Table

4. Conclusions

This article reports the complex structure of the broadband uranium oxide emission from a laser-produced uranium metal plasma. Several UO spectral features were identified in high-resolution spectra collected in the visible spectral region. The spectral emission features from a laser-produced U plasma show the presence of atomic, and molecular uranium species along with strong background-like emission named as uranium oxide transition array or UOTA. The U and UO spectral features collected during the oxide cleaning process under vacuum conditions confirm that the background-like emission in a U plasma is contributed by uranium and higher oxides.

The time-resolved spectral features in the air show the appearance of the UO spectral band at 593.55 nm along with resonance U I transitions, although its persistence is limited presumably due to the formation of higher oxides because of oxygen-rich conditions. The spectral data obtained from reduced oxygen concentration environments revealed a crowded emission spectrum presumably from uranium oxide emission, with clearly identifiable UO emission bands with finite structures in the 590-605 nm spectral regime. Several overlapped UO emission peaks appeared around 514 nm and 595 nm spectral region, however, isolated peaks can be seen in the longer wavelength side (for example 714 nm, 727.54 nm). Although several UO spectral features are ascertained in the present study, the presence of higher oxides of U and its contribution to UOTA is not clear yet. A systemic high-resolution spectroscopic study with varying oxygen concentration is essential for controlling the plasma assisted chemical reactions leading to the formation of more complex uranium oxide (U_xO_y) molecules and for capturing its spectral emission features.