To my family and teachers.
iii ACKNOWLEDGMENTS I would like to express my sincere gratitude to my advisors, Dr. Vladimir V. Khmelenko and Prof. David M. Lee, for their enormous help in planning and conducting experiments, which allowed for this thesis. As well as their help and mentoring during the preparation of the manuscripts, which laid the foundation for this thesis.
I would like to thank my colleague, Cameron Wetzel, for his dedication to improving and maintaining the experimental setup, as well as being my guide in this lab.
The physics machine shop provided essential experience in scientific instrument manufacturing and precisely fabricated many parts that allowed our experimental setups to function correctly. Garrick Garza, the shop supervisor, and Research Instrumentation Specialist John Lopez consistently offered advice and technical assistance in manufacturing, designing, and repairing equipment critical for conducting the experiments described in this thesis. I deeply appreciate my committee members, Ivan V. Borzenets, Artem G. Abanov, and Andrew Comech, for their valuable advice during my thesis defense preparation and for proofreading this thesis.
Finally, I would like to acknowledge the hard work of the administrative staff in this department: Heather Walker, Elise Zwahr, Robert Gunn, René Gunn, Frances Ellison, Veronica Rodriguez, and Department Head Grigory Rogachev. Without their patient assistance, this thesis would not have been possible.
David M. Lee and Vladimir V. Khmelenko made significant contributions to all the experiments, papers written, and this thesis throughout my Ph.D. program. Cameron Wetzel assisted with the experiments and implemented numerous improvements to the experimental setup discussed in this work, while also doing an excellent job of maintaining the experimental equipment.
Ivan Borzenets dedicated his time and experimental skills to co-authoring one of the papers that serves as a foundation for this document.
I would also like to extend my appreciation to Garrick Garza for his invaluable contributions, as he has rescued our experiments multiple times with his quick and precise repairs of our equipment in the machine shop.
2.1 Schematic of the experimental setup for the optical study of the luminescence of gas jets and impurity-helium condensates. On the figure, (1) quartz capillary, (2) bifurcating optical fiber, (3) RF discharge cavity, (4) digital camera, (5) convex lens (6) optic fiber, (7) motorized lab jack, (8) Liquid helium Dewar, (9) LN 2 Dewar, (10) fountain pump, (11) first thermometer, (12) second thermometer, (13) sample accumulation beaker, (14) gas jet, (15) 0.75 mm orifice, (16) bifurcating optical fiber input, (17) impurity-helium condensate (18) LN 2 refill system. . . . . . . . . . . . . . . . . . . 17 In frame 1, one can see the usual regime after some time after the explosion. In frame 2, the bottom of the beaker is heated above 36 K (registered by the thermometer, one can see), and the β-group emission is weak. There is no green glow in the region directly under the jet at the bottom of the beaker. In image 3, the fountain pump is turned on and starts to supply HeII into the beaker. In image 4, the bottom of the beaker is cooled down, and the intensity of β emission is higher than in image 1. The gas mixture 4.9 Time dependence of α-group component amplitudes during decay process for collection of nanoclusters formed by N 2 :Ne:He = 1:50:1000 gas mixure: line at 519.92 nm -green circles, 520.14 nm -blue squares, 520.2 nm -purple up triangles, 520.63 nm -orange down triangles, 521.19 nm -cyan crosses, 521.87 nm -magenta stars, and 522.86 nm -maroon rhombuses. Decay times for each line are listed in Table 4.2. 4.10 Dynamics of the α-group narrow component (λ = 519.9 nm) weight. Weight is calculated as a ratio of the narrow component integral luminescence to the entire α-group luminescence. Presented data were obtained in experiments with different gas mixtures: N 2 :Ne:He = 1:20:400 -green circles, N 2 :Ne:He = 1:50:1000 -blue squares, N 2 :Ne:He = 1:50:2500 -purple up triangles, N 2 :Ne:He = 1:1000:50000 -orange down triangles, Ne:He = 1:50 -teal crosses, and N 2 :Ne:He = 1:100magenta stars. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.11 Integrated spectra of α-group luminescence decay at T ≈1.3 K for collection of nanoclusters formed by gas mixture N 2 :Ne:He=1:50:2500 (black line) and integrated spectra of α-group during explosive destruction of this collection of nanoclusters during warming to T≈13 K (red line). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiv LIST OF TABLES
TABLE Page 3.1 Wavelengths of the transitions of molecular nitrogen, which were identified in the obtained spectra. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 3.2 Wavelengths of the transitions of atomic nitrogen, atomic oxygen, and helium, which were identified in the obtained spectra. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 3.3 Parameters of Lennard-Jones interaction potentials, minimum of energy ϵ in K, and distance between atoms, which corresponds to the potential minimum, R min in Å of rare gas atoms, oxygen atoms, and nitrogen molecules . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 3.4 Wavelengths of the transitions for molecular nitrogen, which were identified in the spectra presented in Fig. 3.10. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 3.5 Wavelengths of the transitions of atomic nitrogen, oxygen, krypton, and helium, which were identified in the spectra presented in Fig. 3.10. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 3.6 Positions of Vegard-Kaplan bands observed during enhanced emission phase upon different N 2 -Kr-He gas mixtures condensation. The position of V-K bands in solid N 2 and solid krypton are also listed for comparison. Wavelengths are given in nm. . . 48 4.1 Wavelength (λ), widths (σ) and decay times (τ ), of the α-group spectra components for nitrogen-neon nanoclusters formed by different nitrogen-neon-helium gas mixtures. The spectra of the α-group were fitted as the sum of five components represented as Gaussian functions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 4.2 Parameters of α-group spectral components (centers, widths, decay times) extracted from the spectra fits of the nitrogen-neon nanoclusters luminescence decay process. For each mixture, the sum of seven Voigt functions was used. The parameters extracted from those fits are presented in the table. The amplitudes obtained from fits at different stages of the sample luminescence intensity decay were used to determine the characteristic decay times. For components 3-7, the exponents with two different characteristic times have been used. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 1. INTRODUCTION AND LITERATURE REVIEW 1.1 Introduction to the nanocluster immersed in bulk superfluid helium study
The initial investigations of luminescence of solid nitrogen subjected to bombardment by electrons, protons, and X-rays were conducted in 1924 by Vegard [1,2]. These experiments were intended to support his hypothesis that the auroral spectrum could be accounted for by emissions from clusters of solid nitrogen. Vegard suggested that the majority of the auroral spectral lines could be attributed to the emissions of nitrogen molecules and ions. Notably, he associated the prominent auroral green line at λ = 5577 Å with the emissions from solid nitrogen when exposed to electrons. Subsequently, McLennan et al. [3,4] which was unexplained at the time. In the laboratory experiments involving electrical discharges in gas mixtures, they discovered that the 5577 Å line was greatly enhanced when a trace amount of oxygen was added to helium gas. They measured the line's wavelength at 5577.35 Å and concluded it is identical to the auroral green line. The authors determined that the line originates from a spectrum of oxygen and not from nitrogen or the known molecular oxygen band system.
They propose that metastable helium, present in the Earth's upper atmosphere and excited by solar rays, acts as the agent that provides the energy to dissociate and excite oxygen atoms, which emit this specific wavelength. By using gaseous mixtures of helium, oxygen, and nitrogen, they were able to reproduce the entire auroral spectrum in the laboratory. The emission at λ = 5577 Å was assigned to the forbidden transition 1 S → 1 D of the oxygen atoms in the gas phase [5,6].
In 1950, researchers became interested in studying the luminescence of solid nitrogen for the second time when a program aimed to achieve high concentrations of stabilized atoms in solid matrices at low temperatures was initiated [7]. Although the program aimed to achieve high concentrations of stabilized radicals for practical applications, this goal was not achieved. However, comprehensive studies were conducted on the properties of various matrix-isolated species, including oxygen and nitrogen atoms in solid molecular nitrogen.
When nitrogen gas discharge products condensed onto cold surfaces at approximately 4.2 K, and during the bombardment of solid nitrogen by electrons, a bright green emission was observed.
The most intense luminescence occurred at wavelengths of approximately 523 nm and 560 nm.
These emissions were attributed to the nitrogen atom transition 2 D → 4 S (referred to as the αgroup) and to the oxygen atom 1 S → 1 D transition (so-called β-group), correspondingly [8,9,10].
In a 1981 study, Sayer et al. [11] examined the crystal structure of solid nitrogen-neon (N 2 /Ne) mixtures at low temperatures, analyzing the emission spectra of trapped nitrogen and oxygen atoms. Deposition of the nitrogen-neon gas on the substrate at 7 K does not create well-defined crystalline structures except for the almost pure solids. Low-energy electron (500 eV) excitation was used to obtain spectra of atoms trapped in those solids. The study focused on the spectral structure and lifetimes of the α-group of nitrogen atoms and the β-group of oxygen atoms, providing valuable insights into the solid's structure, building on findings from earlier electron diffraction studies [12]. A key result shows that N 2 -Ne solids, formed through gas-phase deposition, can solidify into one of three distinct types of binary solids, depending on nitrogen concentration. The study identifies three different solid structures determined by N 2 concentration. First, for mixtures with 0.1% or less of the nitrogen (like 0.1% N 2 in Ne), the films exhibit a well-ordered face-centered cubic (f.c.c.) structure, similar to pure Ne solids. Second, for higher nitrogen concentrations, from 1% to 75% N 2 , the mixtures form monophase amorphous solids where N 2 and Ne are randomly distributed. In these amorphous solids, the spectral features are broad and lack fine structure, which is attributed to the disordered lattice and the presence of a net static field throughout the binary solid. For mixtures containing approximately 50% N 2 in Ne, the spectral data indicate that phase separation occurs, resulting in a solid with large crystallites of nearly pure nitrogen, while neon is expelled.
In experiments with solid nitrogen, only trace amounts of oxygen (1-10 ppm) were typically present. However, the significantly shorter radiation lifetime of the O ( 1 S → 1 D) transition in the solid N 2 matrix (with a lifetime, τ β , of 0.2 ms [10]) compared to the N ( 2 D → 4 S) transition (with a lifetime, τ α , of approximately 25-37 seconds [8,13,14]) results in that even small amounts of oxygen contamination in solid molecular nitrogen produce bright emissions from oxygen atoms, comparable to those from the nitrogen atoms.
Recent studies have examined the radiation effects and relaxation processes in solid N 2 that has been pre-irradiated by an electron beam, as well as in solid N 2 doped with rare-gases (RG) matrix [15,16,17,18,19,20]. It was found that at low temperatures, the thermoluminescence of preirradiated solids occurs due to the neutralization of ions by de-trapping electrons. In contrast, at temperatures of 20 K or higher, neutral atom diffusion and their recombination become significant.
In 2017, Savchenko et al [21] investigated electronically induced processes in solid nitrogen (N 2 ) films, with a specific focus on the desorption, or ejection, of atoms and molecules from the surface of solid samples. Using a low-energy electron beam to avoid physical sputtering, the researchers employed luminescence spectroscopy to monitor the final electronic states of the desorbing particles. A key finding was the first-ever observation of vacuum ultraviolet (VUV) emissions from desorbing nitrogen atoms. The spectral lines of these emissions matched their gas-phase counterparts, confirming the atoms had left the solid surface. The study also provided new data on the desorption of excited N 2 molecules. By varying the electron beam energy and film thickness, it was possible to distinguish between surface and bulk processes, confirming that desorption is a surface-related phenomenon.
To understand the underlying mechanisms, the researchers implemented a novel method called nonstationary desorption (NsD), where a pre-irradiated film is gently heated under the electron beam while the desorption yield is measured. The results showed a strong correlation between peaks in the desorption yield and the yield of thermally stimulated exoelectron emission (TSEE), particularly a prominent peak around 10 K. This correlation provides direct evidence that charge recombination is a crucial driver of the desorption process: electrons released from traps recombine with positively charged centers (like N + 3 and N 4 +), releasing the energy that ejects particles from the surface. The study suggests that different recombination reactions are responsible for the desorption of atoms and molecules, highlighting the complex dynamics of electronic excitations in solid nitrogen.
In a 2015 paper, Savchenko et al [17] investigated the effects of electron beam radiation on solid nitrogen (N 2 ) and nitrogen-doped neon (Ne) matrices using luminescence and activation spectroscopy methods, including cathodoluminescence (CL), thermally stimulated luminescence (TSL), and TSEE. In nitrogen-containing Ne matrices, irradiation leads to the accumulation of N atoms, ions such as N + and N + 2 , and trapped electrons. The relaxation processes following irradiation are shown to be temperature-dependent. At low temperatures, neutralization reactions between ions and electrons are the dominant mechanism, while at higher temperatures, diffusion-controlled reactions involving the recombination of neutral species also contribute significantly.
Abovementioned research provides strong evidence for the formation of the tetranitrogen cation (N + 4 ) in pure solid nitrogen through a "hole self-trapping" process, where a N + 2 ion reacts with a neighboring N 2 molecule (N + 2 +N 2 → N + 4 ). This N + 4 cation is identified by its characteristic fingerprint, a dissociative recombination reaction with an electron that produces an excited nitro-
The formation of N + 4 is supported by the growth of this specific emission's intensity during irradiation and its appearance in TSL and nonstationary luminescence (NsL) experiments as the sample is heated, which releases trapped electrons and initiates the neutralization reaction.
Strong evidence for the existence of the tetranitrogen cation, N + 4 , was also obtained in other papers [16,20]. In these studies, the emission of oxygen atom β-group was observed and studied [15,16,17,18,19,20,21].
The emission from the β-group of oxygen atoms attracted researchers' attention during studies of molecular nitrogen nanoclusters immersed in superfluid helium-4 (HeII). These nitrogen nan-oclusters, which contained stabilized nitrogen and oxygen atoms, were formed by injecting into bulk superfluid helium a mixture of helium gas and molecular nitrogen gas after passing through a region of radio-frequency discharge [22,23]. In these papers, Gordon et al. explore the feasibility of stabilizing high concentrations of nitrogen atoms in a bulk superfluid helium. Previous experiments, conducted at higher temperatures, failed to stabilize chemically active atoms relative concentrations beyond a few tenths of a percent. To achieve significantly lower temperatures and improved heat dissipation, the researchers developed a novel method in which a directional beam of gas containing dissociated nitrogen atoms was introduced directly into superfluid helium at temperatures 1.2-1.5 K.
This study successfully demonstrated that the concentration of nitrogen atoms stabilized in superfluid helium can exceed 1.6%, a notable increase from previous efforts. From observations of the recombination afterglow, the authors estimated a very low activation energy for the process, approximately 0.1 kcal/mol. An important finding was the occurrence of sharp flashes of light when the liquid helium was heated through its lambda point (around 2.17 K). The authors hypothesize that this is due to the thermal explosion of solid nitrogen granules, containing stabilized nitrogen atoms, triggered by the sharp drop in the thermal conductivity of helium as it transitions from superfluid to normal state.
Luminescence spectra obtained during sample accumulation revealed the most intense luminescence from stabilized nitrogen atoms at a wavelength of approximately 522 nm (α-group) and from oxygen atoms at about 560 nm (β-group) [24,25].
In a 1994 paper, Boltnev et al present the first investigation of metastable nitrogen N( 2 D) and neon Ne( 3 P 2 ) atoms isolated within a novel quantum matrix known as an impurity-helium solid phase (IHSP). This unique solid is formed by injecting a jet of impurity particles (such as N 2
or Ne) mixed with helium gas into superfluid helium. The researchers employed luminescence spectroscopy, including laser-induced fluorescence (LIF), to study the spectral properties, decay kinetics, and light emission of the N atoms upon warming up (thermoluminescence). The primary goal was to understand the interaction between the trapped atoms and the helium matrix and to determine the energy barriers governing the stability of the IHSP.
The study revealed that the luminescence from the forbidden N( 2 D → 4 S) transition is strongly influenced by the presence of a neighboring heavy particle, like an N 2 molecule or a rare gas atoms. A key discovery was an extremely long-lived afterglow from the nitrogen atoms, which decayed for over three hours. Through thermoluminescence studies, two activation energies were determined: a barrier of E 1 =40±4 K for the fusion of neighboring impurity centers, which dictates the IHSP's stability, and a lower energy of E 2 ≈7 K for the long-lived afterglow, which is close to the energy required to form a vacancy in solid helium. Furthermore, metastable neon atoms were detected for the first time in the IHSP, and their spectral lines were found to be unshifted and unbroadened compared to their gas-phase values, indicating a very weak interaction with the helium environment. In these experiments, oxygen was not intentionally added to the nitrogenhelium mixture; rather, it was present due to a small contamination of oxygen (1-10 ppm) in the high-purity helium gas used in the studies.
The strong emission of oxygen atom β-group, accompanied by less intense β ′ and β ′′ -groups, was observed during the destruction of impurity-helium condensates containing high concentrations of stabilized nitrogen atoms [15,26,27,28,29,30,31,32]. The destruction of impurityhelium condensates is accompanied by a rapid release of chemical energy stored in the samples.
Interestingly, a substantial part of the stored energy is released through the emission of oxygen atoms surrounded by nitrogen molecules in the solid phase. The emissions of oxygen atoms βgroup were also observed in the impurity-helium samples immersed in bulk superfluid helium during the sample warming up from 1.3 to 2.16 K [33,34]. This thermoluminescence of the nanoclusters in HeII is explained by recombination reactions of nitrogen atoms residing on the surfaces of nanoclusters initiated by quantum vortices in superfluid helium. Later, rotationally induced luminescence of nanoclusters immersed in HeII was studied using the technique of nitrogen-helium jet injection into the HeII, while rotating the beaker at speeds of 3, 4, and 7.5 rad/s [35]. The primary finding was a direct correlation between the beaker's rotational speed and the intensity of luminescence from nitrogen atoms within the nanoclusters. Specifically, as the rotation speed in-creased, the emission from the nitrogen atom α-group ( 2 D → 4 S) grew substantially. Additionally, the study found that increasing the concentration of molecular nitrogen in the initial gas mixture also resulted in an increase in luminescence intensity at each rotational speed.
This phenomenon was explained through the behavior of quantum vortices in superfluid helium. According to the model presented in a paper, rotating the beaker increases the density of these quantum vortices, which are aligned with the beaker's rotation axis. Nanoclusters injected into the HeII become trapped within the cores of these vortices. This confinement increases the rate of collisions between nanoclusters, initiating the recombination of nitrogen atoms that are stabilized on their surfaces. The energy released from this recombination excites other nitrogen atoms, which promotes luminescence. It was concluded that this technique not only provides a method for initiating chemical reactions on nanocluster surfaces but could also be used to visualize vortex cores and study quantum turbulence. The β-group of oxygen atoms was always present in the nanocluster luminescence initiated by vortices.
As mentioned in the previous section, the method of condensing atoms and molecules from the RF discharge into bulk superfluid helium is highly effective for achieving a record high concentration of stabilized atoms [36]. Most of these atoms become stabilized in nanoclusters, which form in the cold helium gas above the surface of HeII [37].
When these nanoclusters enter bulk HeII, they create a stable, porous structure similar to aerogel, which remains stable at low temperatures [38,39]. This unique structure enables various studies, including X-ray, ultrasound, optical, and electron spin resonance (ESR) examinations of the collection of nanoclusters immersed in HeII.
In their 2002 paper, Kiselev et al. [39] report on the results obtained using X-ray and ultrasound methods to determine that these solids are composed of very small clusters of the impurities.
For nitrogen and neon impurities, these nanoclusters were found to have a characteristic size of 5 ± 2 nm [40,41]. These nanoclusters, with an estimated density of approximately 10 20 cm -3 impurities/cm³, form a highly porous, aerogel-like structure.
The porous nature of these Im-He solids was investigated using two distinct methods: ultrasound measurements and small-angle x-ray scattering. Ultrasound experiments revealed a wide distribution of pore sizes, ranging from approximately 8 nm to 860 nm [38,42,43,44]. This was determined by analyzing the sound attenuation changes with temperature, which is linked to the interaction between the pore walls and the normal component of the liquid helium. Small-angle
x-ray scattering experiments corroborated these findings, independently identifying a distribution of pore sizes ranging from 8 nm to over 40 nm [39].
A key finding of the study is that the structure of these Im-He solids is highly sensitive to temperature. Both experimental techniques detected significant and irreversible structural changes when the samples were warmed above the superfluid transition temperature (T λ =2.17 K). X-ray diffraction patterns taken after warming showed that the diffraction peaks became narrower and more intense, indicating that the impurity clusters were growing in size through aggregation and that their internal lattice defects were annealing. Similarly, ultrasound measurements showed that after a sample was warmed above T λ and cooled back down, its sound attenuation was irreversibly increased, confirming a permanent change in its porous structure.
These irreversible changes were attributed to thermally activated diffusion of the impurity clusters. The proposed mechanism centers on the unique thermal properties of liquid helium. Below T λ , superfluid helium is an excellent thermal conductor, and it efficiently dissipates the heat released during the slow aggregation of clusters. However, above T λ , normal liquid helium is a poor thermal conductor. The heat released by aggregation can no longer be carried away effectively, leading to the formation of localized "hot spots" that dramatically accelerate the diffusion process and cause the rapid formation of larger impurity aggregates. A major conclusion is that to preserve the fine nanostructure of these materials and prevent unwanted changes, they must be maintained at temperatures below the lambda point of helium-4.
When a gas mixture containing two heavy impurities is injected into cold helium gas, impurity nanoclusters with a shell structure were formed [45,46,47,48].
Mao et al. [45] conducted a studies of impurity-helium condensates, where record-high concen-trations of stabilized nitrogen atoms were achieved by adding krypton to the nitrogen-helium gas mixture used for sample preparation. In these studies, electron spin resonance (ESR) techniques were applied. The presence of krypton atoms significantly enhances the stabilization efficiency of nitrogen atoms. This method enabled the achievement of an average N atom concentration of 5.3×10 19 cm -3 and an extremely high local concentration of 2×10 21 cm -3 , more than doubling the highest values previously achieved without addition of krypton.
By analyzing the complex shapes of the N atom ESR spectra, it was found that the nitrogen atoms are stabilized in three distinct environments within the nanoclusters. This suggests a shelllike structure of nanoclusters. The three locations identified were: 1) inside the krypton core of the nanocluster, 2) within a layer of molecular nitrogen (N 2 ) that covers the krypton core, and 3) on the outer surface of the N 2 layer. The analysis revealed that the vast majority of the stabilized nitrogen atoms, between 75% and 85%, reside on the surface of the nanoclusters.
The extremely high concentrations of nitrogen atoms enabled the observation of phenomena related to strong magnetic dipole interactions. The broad wings of the ESR signal are a direct result of these interactions, from which the high local atom concentrations were calculated. Furthermore, the proximity of the atoms allowed for the formation of nitrogen spin-pair radicals. This led to the detection of a forbidden, ∆M=2, transition in the ESR spectrum, which occurs when two neighboring spins flip simultaneously by absorbing a single microwave photon. This signal, appearing at half of the magnetic field for the allowed, ∆M=1, atomic ESR transition, is a clear indicator of the high density and close proximity of the stabilized nitrogen atoms.
The thermal stability of these high-density samples was also investigated. Upon warming, the nitrogen atoms residing on the nanocluster surfaces become mobile and undergo rapid, explosive recombination, which is accompanied by bright flashes of light and the sublimation of a significant portion of the sample. For the most concentrated samples, this explosive destruction occurred at temperatures between 3.5 K and 5 K. In conclusion, the addition of krypton is a highly effective method for creating materials with exceptionally high concentrations of stabilized nitrogen atoms, allowing the storage of a high density of chemical energy. These achievements represent an important step toward the potential observation of collective magnetic ordering effects at low temperatures.
ESR studies have also provided insights into the internal structure of the nanoclusters, identified the trapping sites of the atoms within them, and estimated the average concentration of stabilized atoms to be around 10 19 cm -3 , with local concentrations reaching approximately 10 21 cm -3 [45,46,49,50,51,52].
ESR investigations of H-H 2 -Kr and N-N 2 -Kr nanoclusters immersed in HeII provided evidence that the heavier Kr atoms constitute the core of the nanoclusters, while the lighter H 2 and N 2 molecules form the outer layers. These experiments also demonstrated that most of the stabilized free radicals (H and N) reside on the surfaces of the nanoclusters [46,47].
It has been proposed that one or two layers of solidified helium form on the surfaces of impurity nanoclusters [53]. This structure provides stability to the atoms within the porous formations and accounts for the high concentration of stabilized atoms on the surfaces of the nanoclusters.
Subsequent spectroscopic studies of alkali and alkaline-earth atoms and ions embedded in HeII and solid helium have been conducted [54,55,56,57]. The absorption and emission spectra observed for atoms embedded in condensed helium were explained using the "bubble" model [54,57].
The development of impurity-doping techniques in helium nanodroplets has opened new avenues for studying HeII systems with impurity atoms [55,58,59,60,61,62,63]. This method facilitates the investigation of the properties of liquid helium, as well as the optical and magnetic characteristics of impurity atoms and molecules captured in liquid helium nanodroplets.
Helium nanodroplets have been utilized to study atomic and molecular aggregation because they provide a high degree of control over the aggregation process, efficient cooling down to 0.37 K, and enable measurements with minimal matrix effects compared to other noble gas environments [55,64,65].
In studies of atoms and molecules in liquid helium, a question arises regarding the structure of helium surrounding impurities. Depending on the nature of the interaction between the impurity and the surrounding liquid helium, the solvation structures can be classified into two distinct cases. The "bubble" structure occurs with impurities that exhibit repulsive interactions with helium, whereas the "snowball" structure forms around impurities that exhibit strongly attractive potentials with helium.
This thesis will specifically focus on impurities that possess attractive potentials toward the surrounding helium. The formation of a solvated layer around an impurity with a strongly bound potential to helium has been studied theoretically using various approaches [66,67,68,69,70,71].
Notably, research predicted a pronounced shell structure of helium atoms surrounding a single xenon (Xe) atom [66]. Later studies, utilizing density functional theory, demonstrated that in bulk liquid helium, all rare gas atoms are surrounded by strongly bound helium shells, where the helium densities are equal to or even exceed the solid helium density [68].
Furthermore, the potential for a quantum gel formed from impurities that are surrounded by solidified helium has been discussed in the literature [69]. Fluorine and neon atoms have been suggested as promising candidates for producing impurity-based quantum gels [70,71].
Theoretical challenges related to the collapse of helium shells around impurity atoms and the subsequent formation of impurity clusters in liquid helium have also been addressed. Recently, the possibility of forming Ne 2 van der Waals complexes within superfluid helium has been explored, leading to conclusions about the possibility for the formation of a "quantum gel" or "quantum foam," characterized by a layer of solidified helium that separates neon atoms [72].
The dynamics of argon atom clustering in superfluid 4 He nanodroplets were studied in real time using a density functional approach applied to liquid helium. Depending on the initial kinematic conditions, two different structures of argon can be formed: either a compact argon cluster or a loosely bound metastable cluster with helium density caged inside [73]. Theoretical predictions suggest the existence of structures formed by helium "snowballs" surrounding impurity species.
Recently, several experimental evidences have emerged supporting the presence of structures formed by solvated helium layers surrounding impurity atoms in helium nanodroplets. One exam-ple is the observation of a foam structure created by doping helium nanodroplets with magnesium (Mg) atoms [74,75,76,77]. In this foam structure, a collection of Mg atoms are separated by a layers of helium atoms, with the distance between the Mg atoms being approximately 1 nanometer. The electronic and optical properties of the Mg atoms within the foam are similar to those of a single atom, as demonstrated by resonance two-photon ionization. Interestingly, the metastable network of Mg atoms undergoes spontaneous collapse upon exposure to laser radiation, resulting in the formation of Mg n clusters during the relaxation process [74,75]. Conversely, recent studies of infrared spectra in magnesium-doped superfluid helium nanodroplets with varying sizes of He nanodroplets and the numbers of dopants indicate that only compact magnesium clusters are formed [78].
Another piece of experimental evidence for the foam structure comes from electron diffraction studies of xenon (Xe)-doped helium nanodroplets, which contain between 10 5 and 10 6 helium atoms and xenon structures with up to a hundred atoms [79,80]. For certain recorded diffraction patterns, reliable fitting indicates that, in addition to Xe-Xe and Xe-He distances, the longer distances between Xe atoms, such as Xe-He-Xe and Xe-He-He-Xe, need to be considered. This supports the observation of the Xe foam structure.
To form macroscopic quantities of quantum gel (foam structure), the experimental approach should be similar to that used in previous work involving the injection of impurity helium gas mixtures into bulk superfluid helium [22,23]. An additional condition should be implemented: a substantial reduction in the concentration of impurities in the helium gas jet to prevent the formation of impurity nanoclusters before entering bulk HeII. Previous experiments involving impurity-helium gas mixtures contained 1% or more impurities, leading to the exclusive formation of impurity nanoclusters in bulk HeII [40,53]. Even under these conditions, the impurity nanoclusters are covered by at least one layer of solidified helium.
The method of condensing atoms and molecules from the discharge zone into HeII enables studies of nanoclusters as they cool in cold helium vapor above the HeII surface [25,37]. This approach allows for the examination of nanocluster formation processes in cold helium gas and their interactions with atoms and molecules from the gas jet.
Recently, a notable phenomenon was observed: an increase in the β-group emission of oxygen atoms in N 2 nanoclusters during the condensation of nitrogen-helium gas mixtures. This occurred after the gas passed through the discharge zone and entered into cold helium gas at temperatures around 20-35 K [81]. The presence of O atoms in the condensed gas mixture was attributed to a small unavoidable impurity of approximately 1 ppm in the helium gas.
This effect can be explained by the enhanced emission of O atoms that are stabilized within the N 2 nanoclusters. This enhancement resulted from an increased recombination rate of nitrogen atoms on the surfaces of the N 2 nanoclusters at the specified temperatures. Excitations from metastable N 2 molecules formed on the surfaces of the nanoclusters were effectively transferred through a chain of N 2 molecules within the nanoclusters to the stabilized O atoms [7,82,83].
It is of great interest to study at low temperatures the processes of interaction of the atoms from the gas jet with the mixed nanoclusters characterized by a shell structure. In this work, the luminescence from nanoclusters during the condensation of nitrogen-helium, nitrogen-neon-helium, nitrogen-argon-helium, and nitrogen-krypton-helium gas mixtures after passing the discharge zone is studied. In these experiments, the enhanced emission of the β-group of oxygen atoms and metastable N 2 molecules (V-K-bands), for a shell-structured nitrogen-krypton-helium nanoclusters was observed. The mechanisms of these emissions are discussed, and conclusions about the shell structure of impurity nanoclusters were made.
Another topic in this thesis is the first experimental evidence of the influence of a solidified helium layer on the spectra of nitrogen atoms stabilized on the nitrogen-neon nanocluster surface.
High dilution of impurities by helium gas in the condensed gas mixture reduces the size of impurity clusters injected into HeII and increases the quantity of solidified helium in HeII. Up to date, there were no direct observations of the influence of solidified helium on the characteristics of impurities collected in HeII. Searching for an evidence of a solid helium layer surrounding impurities or clusters of impurities in superfluid helium is a challenging task. Recently, the luminescence of nitrogen-neon nanoclusters immersed in bulk HeII was studied. The ratios N/Ne in nitrogen-neon nanoclusters were 10% and 20%. The α-group spectra of N atoms (transition 2 D→ 4 S) were fitted by seven components. It was found that the narrow component at λ = 519.9 nm prevails in the late stages of luminescence decay [84]. Preliminarily, this spectral component was assigned to nitrogen atoms stabilized on the surfaces of neon nanoclusters surrounded by a layer of solid helium. The characteristic of these nitrogen atoms might be influenced by interaction with the solidified He layer surrounding the neon nanoclusters.
It was found that the shape of α-group spectra was similar for all studied nitrogen-neon nanoclusters with a small content of N atoms. The component at λ = 519.9 nm is the largest in the α-group spectra. The lifetime of this component is ≈ 280 s. The position of this component is close to that of the N atom in the gas phase, which has a much longer decay time (≈44 hours).
ESR spectra registration has also been made of N atoms stabilized in a collection of neon nanoclusters, and it was found that one of the N atom triplets in the ESR spectra corresponds to a hyperfine splitting constant of 3.95 G. This value is close to that of free nitrogen atoms. The above facts allow us to suggest that, for the first time, N atoms in bulk HeII, whose characteristics are influenced by solidified helium atoms, were observed.
Research, presented in this thesis, is motivated by the versatile nature of nanoclusters and their ability to stabilize chemically active atoms for an extended period of time. Stabilization of chemically active atoms on the surfaces of nanoclusters enables the study of atoms that do not naturally exist in atomic states for an extended period of time: N, O, F, etc. Phenomena of chemical reaction initiation by quantum vortices allow for initiating reactions with atoms stabilized on the nanocluster surfaces, and control reaction rate by the rotational speed of the beaker, which changes the density of vortices [33,34]. High concentrations of stabilized atoms allow the creation of High Energy Density Matter (HDEM) with energy densities up to 1.5 kJ/g [53]. In nanoclusters made with light atoms (H, D, T) and molecules, chemical tunneling reactions have been observed [85,86]. Research on these reactions and the nanocluster size impact on them can give insights into micro-scale quantum systems. Recently, the possibility of using deuterium nanoclusters to achieve a high flux of ultracold neutrons was discussed [87]. Neutrons from the source, when traveling through the volume of a specially made collection of deuterium nanoclusters immersed in HeII, will get trapped in the nanoclusters and can only escape the system when their de Broglie wave becomes larger than the nanocluster size, allowing for the realization of the high flux of ultracold neutrons [87]. We prepare the nitrogen-neon-helium gas mixtures using research-grade helium gas from Linde
Electronics & Specialty, which has a purity of 99.9999%. Additionally, we use research-purity nitrogen, neon, argon, and krypton gases (99.9999%) from Matheson. The oxygen content in the gas mixtures is approximately 1 ppm, which results from contamination in the helium gas. We vary the helium concentration in our gas mixtures to enhance the efficiency of dissociating impurity molecules during the discharge process. In the current work, nitrogen-helium, nitrogen-neonhelium, neon-helium, nitrogen-argon-helium, and nitrogen-krypton-helium gas mixtures have been used. The majority of gas mixtures consist of helium gas, with all other gases considered impuri- ties. Generally, gas mixtures can contain up to 1% of impurities; however, for impurities, such as neon, the content can be as high as 5%. Gas handling system schematic is shown in Fig. 2.3
The process of sample accumulation starts by sending the gas mixture into the helium Dewar through a quartz capillary. The gas mixture passed through the discharge zone (3), and the discharge products entered the helium Dewar through a 0.75 mm orifice in a quartz capillary. The quartz capillary and the RF cavity are surrounded by the larger diameter quartz tube, filled with LN 2 , which allows for gas and discharge products to cool down before entering the volume of the helium Dewar and provides cooling for the RF cavity. The RF signal for discharge is generated by a Hewlett-Packard Signal generator Model 8656B and sent to an E&I 3100L RF amplifier connected to an LN 2 cooled RF cavity. This is a custom-built cavity with additional shielding to minimize its influence on external electronics [46]. The resonance frequency of the RF cavity was varied in different experiments from 76.5 to 82.5 MHz. The Brooks Model 5850E Mass Flow Controller regulated gas flow and kept it at the value 5x10 19 particles/s. Inside the helium Dewar, the discharge products form a well-formed jet (14) in the cold, dense helium gas. The jet is directed to the center of the beaker filled with HeII, which is installed in such a way that the distance from the quartz capillary orifice to the superfluid helium surface in a beaker was 2 cm (see Fig. 2.1 zoomed section).
All presented optical spectra were recorded simultaneously by three spectrometers. Andor and Ocean Optics spectrometers are connected to the bifurcating optical fiber, the other end of which is channeled directly into the helium Dewar and aimed at the bottom of the beaker ( 16).
An Avantes spectrometer was used to record near-infrared (NIR) spectra with a resolution of 5 nm. The spectrometer signal was captured from outside the helium and nitrogen Dewars, which were coated with silver for thermal radiation insulation. Small strips were left clear for optical and visual observations.
Convex lenses, along with an optical fiber, were installed on a motorized lab jack. By adjusting the height of the lab jack, it is possible to select the region of the beaker from which the NIR signal registration light is collected.
Using both Ocean Optics and Avantes spectrometers simultaneously enabled us to obtain data over a wide wavelength range, from 200 to 1650 nm. Additionally, an Andor spectrometer was used to improve optical spectral resolution in the selected wavelength region. The luminescence of the jet and the impurity-helium sample was recorded using a digital camera.
Two thermometers were placed inside the beaker. The first thermometer is secured at the center of the bottom of the beaker, and the second is placed vertically on the side of the beaker. The first thermometer is shown in Fig. 2.1 and Fig. 2.2(2). Thermometers are LakeShore germanium resistors (GR-200A series). They were connected to the LR-400 resistance bridge. Temperature differences between two thermometers up to 1 K were recorded (after HeII evaporation). This difference is a result of different thermometers' locations.
Two automatic liquid nitrogen (LN 2 ) refill systems regulated the liquid nitrogen levels in both the LN 2 Dewar and the quartz tube, which houses the RF discharge cavity and the quartz capillary. Although the two refill systems are nearly identical, they differ in the locations of their thermocouples.
The first system includes a pair of thermocouples, an HP 3478A Multimeter, a microcontroller, a gas solenoid valve, a pressurized nitrogen gas supply, and an LN 2 refill Dewar. When the LN 2 level in the main (glass) Dewar falls below the thermocouple's designated location, the thermocouple begins to warm up. This heating causes a change in the potential difference detected by the multimeter. The microcontroller monitors these changes in the multimeter's readings and subsequently triggers the gas solenoid valve to open, increasing the pressure in the refill LN 2 Dewar.
The increased pressure allows liquid nitrogen to flow from the refill Dewar into the main Dewar (see Fig. 2.1).
The refill system for the outer quartz tube operates similarly, with the key distinction being that it contains two pairs of thermocouples located at the top and bottom of the tube, as well as a reference Dewar filled with LN 2 . This allows not only to refill the quartz tube, but also to know the temperature inside. In the earlier experiments, the quartz tube was filled manually, which led to several issues, including waste, inconsistent cryogen levels, and the potential for water to condense inside the tube. When moisture from the air condenses inside, it can create a dangerous condition known as an ice plug, trapping evaporating nitrogen vapor within the tube.
This pressurized gas can forcefully eject the tube, leading to its destruction along with the discharge insert. To address these problems, an automatic filling system is used to control the LN 2 level in the main Dewar and the quartz tube. When the helium Dewar is filled with liquid helium and pumped to become superfluid, the helium vapors in the Dewar start to cool the bottom of the quartz tube.
This allows the temperature of liquid nitrogen in the quartz tube to drop below the LN 2 evaporation temperature. If the LN 2 temperature reaches the solidification point, the nitrogen in the tube can freeze and potentially shatter the tube, leading to a failed experiment and costly, time-consuming repairs. Additionally, starting the discharge can cause the solid nitrogen to sublimate, generating a high pressure of vapor that can also damage the tube. To prevent this situation, a heater is located at the bottom of the quartz tube and operates within a closed-loop control system with a thermocouple, maintaining the LN 2 temperature above 63 K.
The automatic filling system works by pressurizing a nitrogen Dewar with nitrogen gas supplied from an external tank. The gas flows from the tank through a pressure regulator, which reduces the pressure to less than 5 PSI. It then travels through a rubber hose to a 3-way normally closed solenoid valve. This valve exhausts to the atmosphere when de-energized and connects the inlet when energized, allowing the storage Dewar to continuously vent, thereby maintaining atmospheric pressure until filling is required. Once energized, nitrogen gas flows through the valve into the Dewar, pressurizing it. A separate overpressure valve limits the pressure inside the Dewar, automatically opening when a predetermined pressure is exceeded.
Additionally, the pressure inside the helium Dewar was monitored using an MKS gauge and recorded by the PC data acquisition (DAQ) system.
In order to collect the maximum amount of information during spectroscopic study of the nanoclusters in various conditions, data from the following available sensors and spectroscopes were collected:
• Andor Shamrock SR-500i-D2 spectrograph
• Ocean Optics USB2000+ spectrometer • Avantes NIR512-1.7TEC spectrometer • The Brooks Model 5850E Mass Flow Controller • LakeShore GR-200A series germanium resistors (for temperature measurements inside the beaker)
• DV-4 and DV-6 vacuum gauge tubes
The Ocean Optics spectrometer is capable of registering light with wavelengths ranging from 200 to 900 nm, with a resolution of 1.5 nm. Utilization of it allows for an extensive range of data acquisition. However, unsatisfactory time resolution does not allow this detector to be used for spectroscopic recording during fast processes, such as sample destruction. Low spectroscopic resolution of the Ocean Optics detector is limiting its capabilities in resolving the fine structure of nitrogen α-group and oxygen β-group. In most of the experiments, the synchronization loop (which allows synchronization of data from all the sensors and spectrometers) has been run at a 5Hz cycle (200 ms), and the acquisition window for the Ocean Optics spectrometer was set to 150 ms (to allow data transfer to the PC).
Data streams from all sources must be synchronized, which is achieved through a synchronization circuit for the spectrometers integrated into a loop with a PC DAQ system. PC DAQ system also provides synchronisation and recording capabilities for all other sensors and controls via LabVIEW software.
The Andor spectrometer has a 150 lines/mm grating with a 500 nm blaze (maximum sensitivity region, as shown in Fig. 2.4) and a 200-900 nm spectral acquisition range. The integration time for the Avantes and Ocean Optics spectrometers was set at 150 ms, with an external synchronization trigger of 5 Hz. The Andor spectrometer had an integration time of 50 ms and a 20 Hz internal trigger.
The digital camera was used for experiment recordings. The camera is a Nikon model 7200D, equipped with a 25 mm sensor size. The camera was mounted with a 50 mm lens. During video recording, the shutter speed was set to 1/100 s, the f-stop was set to f/4.0, the sensitivity was set to ISO 4000, and the video was recorded at 720p resolution and 60 frames per second.
An experiment is a multi-day process. On the first day, at least half of the day is spent precooling the dewar and preparing for the experiment. After all equipment is prepared, the experiment can be conducted. For each gas mixture, multiple types of experiments can be conducted.
Accumulation: beaker is filled with HeII, discharge is on, fountain pump is on, discharge products form nanoclusters in cold helium gas, and they are collected in the beaker (see Fig. 2.
Sample luminescence decay: the sample is accumulated in the beaker (after 15-30 minutes of running discharge) and submerged in bulk HeII, the fountain pump is on, and discharge is turned off. One should start the registration luminescence of the sample while the discharge is still on, and only turn it off after 15-30 seconds after the start of the registration.
The sample is accumulated in the beaker and submerged in the bulk HeII. The discharge and fountain pump are turned off. Without a constant supply of HeII in the beaker, the level of HeII gradually decreases due to liquid helium evaporation and dripping from the beaker over the course of 15 to 20 minutes. Once all the HeII is removed, the temperature in the beaker begins to rise. This increase in temperature initiates the recombination of stabilized atoms on the surface of the nanoclusters, resulting in the release of more heat and triggering a chain reaction of atom recombination that leads to the rapid destruction of the accumulated sample, accompanied by bright flashes of light. This process can also be accelerated if the jet is left on, as shown in Fig. 2.5.
Enhanced β-group luminescence Beaker is filled with HeII, discharge is on, fountain pump is off. The discharge jet is evaporating HeII from the beaker, and when all helium is evaporated, a rapid change in the luminescence spectra takes place, nitrogen α-group is suppressed, while oxygen β-group is enhanced. As soon as HeII is evaporated, the temperature rises to 12-14K, and an enhanced β-group luminescence regime is sustained until the temperature in the beaker reaches 36 K. Synchronization of spectrometers, temperature sensors, and pressure gauges is ex- tremely important during this type of experiment. If a large amount of sample accumulates during HeII evaporation from the beaker, sample explosion during destruction might raise the temperature above the optimal for this regime, preventing the occurrence of the enhanced luminescence regime. Photos taken during the experiment are presented in Fig. 2.6. After the experiment is finished and the data is collected, we need to prepare the data for the analysis. The first step is to convert data from the proprietary spectrometer format to plain text that the analysis code can read. The Ocean Optics spectrometer can record data in plain text directly, and Avantes has a software option to convert all files in the folder. The Andor spectrometer also has this option; however, if the "Step and Glue" option was used during the experiment, data from separate files need to be stitched together. The Andor spectrometer has a rotating turret, equipped with three gratings. First grating has an aqueous range of 340 nm, and 0.3 nm resolution, while the third grating has a 38 nm range and 0.03 nm resolution. The "Step and Glue" option allows us to sweep through the entire spectrometer range (200 to 900 nm), using the third grating, and therefore obtain high-resolution spectra; however, it takes a long time (5-10 minutes) per sweep.
After conversion, the spectral data must be adjusted for wavelength calibration. Ocean Optics and Avantes spectrometers do not have any moving parts, and therefore, a single calibration spectrum can be collected during each day of the experiment. The Andor spectrometer must be recalibrated after every turret adjustment. Calibration spectra are obtained by connecting a bifurcating cable to the Ocean Optics Hg-Ar calibration source. with line positions from the source manual, and by performing a fit, with a 4th-order polynomial, a calibration function can be obtained, example of the fit shown in Fig. 2.8.
When the polynomial function is obtained, the spectra can be calibrated. To improve the signalto-noise ratio, the background signal is also measured for all detectors. Measurement performed between experiments, so there is no light pollution in the background spectra. The background proved stable across all spectrometers, so a single background measurement is sufficient for all experiments conducted in a single day. Examples of backgrounds collected for each spectrometer are shown in Fig. 2.9. Backgrounds are subtracted from the obtained experimental spectra during data analysis.
Solid molecular nitrogen nanoclusters are formed by injecting a nitrogen-helium gas mixture after the discharge zone into dense cold helium gas above the surface of HeII. After entering bulk HeII, nanoclusters form a porous gel-like structure inside HeII. Each nanocluster is covered with a few layers of solid helium, which impedes the recombination of atomic nitrogen, mostly stabilized on the nanocluster's surface. [45].
If the supply of HeII to the beaker is turned off, the discharge product jet evaporates all the liquid helium from the beaker. The temperature in the beaker is increased. High temperature (14-36 K) in the beaker stimulates atoms on the surface of nanoclusters to recombine. Energy generated in this process needs an efficient channel to be released. Oxygen atoms, present in the nanoclusters in small quantities (less than 5 per nanocluster), can not efficiently recombine into O 2 , and their fast decay time from the excited to the ground state allows to facilitate energy release from the nanoclusters. In this chapter, this mechanism will be explained in great detail, along with the insights into nanocluster structure one can obtain by studying this process in nanoclusters with different compositions. using a mixture of N 2 :He = 1:200 and having O 2 impurities in He gas of around 1 ppm, the oxygen atom β-group intensity is an order of magnitude higher than the nitrogen atom α-group intensity, as evident from the Figs. 3.1a and 3.1b. The dynamics of the luminescence spectra, recorded by the Andor spectrometer, are presented in Fig. 3.2 in logarithmic scale. Table 3.1: Wavelengths of the transitions of molecular nitrogen, which were identified in the obtained spectra.
nm 0 0 1051 3 4 1113 7 5 716.4 0 0 337.1 1 1240 4 0 618.6 8 4 595.9 1 357.6 2 1498 1 678.8 5 646.8 2 380.4 1 0 891.2 2 750.4 9 5 590.6 3 405.9 1 1022 4 943.6 6 639.4 4 434.3 2 1193 5 1078 10 5 544.2 1 2 353.6 3 1430 5 1 612.7 6 585.4 3 375.5 2 0 775.3 2 670.4 7 632.2 4 399.8 1 872.3 3 738.7 11 6 540.6 5 426.9 2 994.2 6 1044 7 580.4 2 3 350 3 1151 6 2 606.9 8 625.3 4 371 4 1364 3 662.3 12 7 537.2 5 394.3 3 0 687.5 4 727.3 8 575.5 6 420 1 762.6 7 1013 3 6 389.4 2 854.2 7 3 601.3 7 414.2 3 968.2 4 654.5 4 8 409.4
We will be focusing our attention on luminescence in a range from 550 to 570 nm, which belongs to the β-group ( 1 S → 1 D transition in oxygen atom), and in a range from 520 to 525 nm, which corresponds to α-group ( 2 D → 4 S transition in nitrogen atom). The dynamics of the integrated intensities of those bands is shown in Fig. 3.3, alongside the temperature. Noise in the spectrum integral (See Fig. 3.3) can be assigned to the slight instability in the discharge.
From Fig. 3.4 one can see that while the helium level in the beaker was relatively high the αgroup emission was stronger than the β-group emission (See Fig. 3.4 a and b), but when more and more helium was evaporated by the jet (helium supply to the beaker was terminated) the β-group emission started to be more prominent (around 300 s mark, see Fig. 3.3). At point c in Fig. 3.3, all helium was evaporated from the beaker, which led to the explosion of the accumulated sample (See Fig. 3.4 c and d).
Table 3.2: Wavelengths of the transitions of atomic nitrogen, atomic oxygen, and helium, which were identified in the obtained spectra.
Atomic lines and groups Element Energy levels λ, nm He The explosive destruction period is less than one second. Intense β-group emission starts almost instantly after the liquid helium is evaporated and, therefore, synchronized with the sample explosion. If the sample accumulation time was short, we don't have enough sample in a beaker for its explosion to be distinctive on top of the intense β-group emission regime. After the explosion, the nature of the emission, and correspondingly the spectrum, changed rapidly and drastically. The α-group almost completely disappears, and β-group luminescence increases tenfold In frame 1, one can see the usual regime after some time after the explosion. In frame 2, the bottom of the beaker is heated above 36 K (registered by the thermometer, one can see), and the β-group emission is weak. There is no green glow in the region directly under the jet at the bottom of the beaker. In image 3, the fountain pump is turned on and starts to supply HeII into the beaker. In image 4, the bottom of the beaker is cooled down, and the intensity of β emission is higher than in image 1. The gas mixture N 2 :He = 1:400 was used in this experiment.
We It has been observed that a longer accumulation time has a negative impact on the luminescence intensity of the β-group but leads to higher-intensity explosions due to the larger amount of accumulated nanoclusters.
In one of the experiments, bulk HeII was evaporated by the pure helium jet. When the beaker was empty, and the temperature in the beaker reached 25 K, the gas supply was switched from pure helium to the N 2 :He = 1:200 mixture, and intense β-group emission appeared in the jet and at the bottom of the beaker.
We studied the influence of nitrogen atom concentration in the gas jet on the observed enhanced β-group emission of O atoms in N 2 nanoclusters. We used N 2 -He gas mixtures with different N 2 content from 1/800 to 1/100 for these experiments. It is known that the concentration of nitrogen atoms in the discharge products of N 2 -He gas mixture is proportional to the content of N 2 molecules [50,52].
Observation of nanoclusters, formed in cold helium gas, with atoms from the discharge-jet experiment, is conducted in the same way for a nitrogen-neon-helium gas mixture. The experiment starts when the sample accumulation beaker is filled with superfluid helium, and the fountain pump is on. The gas mixture is supplied through the capillary, the discharge is turned on, and the sample accumulation is continued. As a next step, the fountain pump is turned off, and while HeII is evaporating from the beaker, we record the accumulating sample and gas jet luminescence spectra. The luminescence intensity dynamics of the nitrogen α-group and the oxygen β-group luminescence, as well as the temperature dynamics in this period, are shown in Fig. 3.8. Once all the HeII is evaporated from the beaker, we observe sample explosion, accompanied by the short, bright flash, see letter (b) in Fig. 3.8. Right after the explosion, the enhanced oxygen βgroup emission can be observed. It starts when the beaker temperature increases to 15 K, see letter (c) in Fig. 3.8, and continues until the temperature of ≈ 36 K is reached, see letter (d) in Fig. 3.8. The slightly enhanced oxygen β-group luminescence was observed in experiments with a nitrogenneon-helium mixture in the 15-36 K range, similar to the result obtained during this phenomenon's study with nitrogen-helium mixtures [81]. However, in contrast to the previous studies with N 2 :He mixtures, where a 10 times difference in intensity was observed, the β-group intensity during the enhanced emission was only two times bigger than that during the accumulation phase for N 2 -Ne-He mixture. During the experiments with N 2 :Ne:He = 1:5:1000, we also saw insignificant β-group luminescence enhancement, as well as the appearance of V-K bands. V-K bands serve as a clear indicator for the obstruction of the energy transfer mechanism from excited N 2 molecules to the stabilized oxygen. Comparison of the integrated spectra, recorded during experiments with N 2 :Ne: He = 1:5:1000 and N 2 :He = 1:200 is shown in Fig 3.9.
This fact confirms the findings, regardless of the mechanism underlying the enhanced β-group luminescence and the shell structure of nanoclusters. While in he experiment with nitrogen-helium mixture, entire nanoclusters have been formed from nitrogen atoms and molecules, in nitrogenneon-helium mixtures, nanoclusters with shell structure are formed. Since nitrogen-nitrogen van der Waals interactions are stronger than neon-neon and nitrogen-neon interactions (see Table 3.3 [88,89]), molecular nitrogen will be first to solidify, forming the core of the nanocluster, while neon atoms form the outer shell.
Nitrogen atoms from the jet are recombining on the surface of nanoclusters, and therefore, the nitrogen molecule chain, which facilitates energy transfer from excited nitrogen molecules, is interrupted by the layer of solid neon. Therefore, there is no efficent energy transfer mechanism for energy from excited nitrogen molecules formed on the surface of a nanocluster to the oxygen atoms stabilized in the interior of the nanocluster. In order to verify those findings regarding shell structure and energy transfer mechanism, experiments with nitrogen-krypton-helium and nitrogen- The regimes for obtaining integrated spectra are similar to those described in Fig. 3.1. In the nitrogen-neon-helium mixture, the content of nitrogen is reduced by 5 times compared to that of the nitrogen-helium mixture. This reduction led to a substantial decrease in all emissions.
argon-helium have been conducted.
Experiments of enhanced β-group luminescence have been conducted for various nitrogenkrypton-helium mixtures as well, and integrated spectra obtained during one of the experiments are shown in Fig. 3.10. While the beaker was filled with superfluid helium, the discharge products in the jet were rapidly cooled as they came into contact with the cold, dense helium gas above the surface of the HeII, forming nitrogen-krypton nanoclusters. After entering bulk HeII N 2 :Kr nanoclusters form a porous aerogel-like structure inside HeII. Due to larger van der Waals interaction, krypton atoms form the cores of the nanoclusters, which are later covered by layers of N 2 molecules and later by layers of He atoms. Nitrogen atoms stabilized mainly on the surface of outer N 2 layers [45,46].
During sample accumulation, intense luminescence from the gas jet and condensate was observed. The temperature rises as helium evaporates from the beaker, initiating a rapid chainrecombination reaction of stabilized atoms in the accumulated sample. Similar to the experiment with the nitrogen-helium mixture, after sample destruction, we observe a change in the luminescence spectra, with a significant decrease in nitrogen α-group ( 2 D → 4 S transition) luminescence and an increase in oxygen β-group ( 1 S → 1 D transition) luminescence compared to the spectra obtained during the accumulation phase. Fig. 3.10 displays the time-integrated luminescence spectra of the entire observation period, comprised of accumulation, explosion, and intense β-group luminescence phases, for the experiment conducted with a gas mixture of N 2 :Kr:He = 1:2:600. All identified lines are marked in Fig.3.10 and listed in Table 3.4 and Table 3.5.
The bands of first (B 3 Π g → A 3 Σ + u ) and second (C 3 Π u → B 3 Π g ) positive systems of molecular nitrogen, as well as atomic lines of helium, oxygen, nitrogen, and krypton, are present in the spectra. Nitrogen molecular bands and atomic helium and krypton lines correspond to the gas phase emission from the jet. Atomic αand δ-group of the nitrogen and β-group of oxygen, as Table 3.4: Wavelengths of the transitions for molecular nitrogen, which were identified in the spectra presented in Fig. 3.10.
λ, nm 0 0 1055 4 We will be focusing our attention on luminescence in the range from 550 to 600 nm, which belongs to the β-group ( 1 S → 1 D transition in oxygen atom), and in the range from 520 to 525 nm, which corresponds to α-group ( 2 D → 4 S transition in nitrogen atom) as well as V-K bands of N 2 molecules. The dynamics of the integrated intensities of those bands for the experiment with condensation of N 2 :Kr:He = 1:2:600 gas mixture are shown in Fig. 3.12, alongside the temperature. Fig. 3.13 shows luminescence spectra during different phases of the experiment. From Fig. 3.13,
one can see that while the helium level in the beaker was relatively high the α-group emission was stronger than the β-group emission (See Fig. 3.13 a and b), but when more and more helium was evaporated by the jet (helium supply to the beaker was terminated) the β-group emission started to be more prominent (around 400 s mark, see Fig. 3.12).
At point c in Fig. 3.12, all helium was evaporated from the beaker, which led to the explosion of the accumulated sample (See Fig. 3.13 c and d). The explosive destruction period was less than one second. Intense β-group emission began immediately after the liquid helium was evaporated and synchronized with the sample explosion, and temperature increase to ≈20K. If the sample accumulation time was short, there was not enough sample in a beaker for its explosion to be distinctive on top of the intense β-group emission regime. After the explosion, the nature of the emission and observed spectrum changed rapidly and drastically. The α-group intensity dropped, and β-group luminescence increased tenfold for 1:2:600 mixture (See Fig. 3.13 e and f). The enhanced β-group emission continued for 160 s and decayed with time (See Fig. 3.13 g and h).
We also studied the influence of N 2 -Kr nanoclusters composition on the oxygen atom β-group, nitrogen atom α-group, and molecular nitrogen Vegard-Kaplan bands emissions. The results of these investigations are shown in Fig. 3.14. In these experiments, gas mixtures with N 2 /Kr ratios from 1/2 to 1/50 were used, and the impurities gas to helium ratio was either 1/200 or 1/400. It is known from ESR studies [49,50,52] that the concentration of nitrogen atoms in the nitrogenhelium jet after passing the discharge zone is proportional to the content of N 2 molecules in the gas jet [52]. A similar tendency was observed for N 2 -Kr-He gas jet for the ratios of N 2 /Kr used in these experiments [45]. We observed enhanced emission of O atom β-group and N 2 molecule V-K bands for N 2 -Kr gas mixtures with N 2 /Kr ratios larger than 1/20. From Fig. 3.14, one can see that decreasing N 2 content in the N 2 :Kr:He gas mixture led to a decrease of the O atom β-group and N 2 V-K band emissions in N 2 -Kr nanoclusters. The effect of enhanced emission was not observed for the gas mixture with N 2 /Kr = 1/50 ratio.
Positions of the V-K bands during enhanced emission for different N 2 -Kr-He gas mixtures are shown in Table 3.6 [90,91].
Table 3.6: Positions of Vegard-Kaplan bands observed during enhanced emission phase upon different N 2 -Kr-He gas mixtures condensation. The position of V-K bands in solid N 2 and solid krypton are also listed for comparison. Wavelengths are given in nm. The detailed spectra of O atom β emission during enhanced luminescence period for different gas mixtures are shown in Fig. 3.15. The decrease of N 2 /Kr ratio in the N 2 :Kr:He mixture decreased β-group intensity and shifted β-group spectrum to the longer wavelength region. Fig. 3.16(1) show luminescence spectra during the explosion of the samples accumulated in HeII using different N 2 :Kr:He gas mixtures. N atoms α-group, O atoms β-group, and N 2 molecule V-K bands are present in the spectra. All these emissions occur from the solid phase. The strong line at λ = 587 nm is the emission of the He atom from the gas phase. An explosion occurs in the presence of gas jet luminescence. (1) (2) Figure 3.16: Spectra of luminescence during sample explosive destruction (1) and enhanced βgroup emission (2) for different gas mixtures taken by Ocean spectrometer: a -N 2 :Kr:He = 1:2:600, b -N 2 :Kr:He = 1:2:1200, c -N 2 :Kr:He = 1:5:2400, d -N 2 :Kr:He = 1:20:4000, e -N 2 :Kr:He = 1:50:10000. The integration period is 3s.
Enhanced β-group experiments have been conducted for multiple gas mixtures: N 2 :Ar:He = 1:5:600, 1:20:2000, 1:50:5000, 1:100:10000. Similar to the experiments with nitrogen-krptonhelium gas mixture, enhanced emission was observed for the gas mixtures with N 2 /Ar ratios up to 1/50, and in all mixtures α-group luminescence diminishes as soon as all helium is evaporated from the beaker. Plots of α-group luminescence integral and β-group luminescence integral during condensation of N 2 :Ar:He = 1:5:600 gas mixture, as well as the temperature are shown in Fig. 3.17. In contrast to the experiments with nitrogen-krypton-helium mixtures, enhanced emission for nitrogen-argon-helium mixtures continues during higher temperatures, up to 42 K. In addition, there is no enhancement of Vegard-Kaplan bands. Dynamics of the luminescence spectra of nitrogen-argon nanoclusters is shown in Fig. 3.18. Observation of enhanced β-group emission confirms the previously assumed mechanism of this enhancement, as well as the shell structure of nanoclusters. Argon, similarly to krypton, forms the core of the nanocluster, while nitrogen molecules form the outer shell. When nitrogen atoms from the gas jet recombine on the surfaces of nanoclusters, they can transfer their excitation through the molecular nitrogen matrix in the outer shell to the stabilized oxygen atoms.
Active solid nitrogen containing stabilized nitrogen atoms exhibits a strong optical emission at low temperatures [7]. Previous studies of N-N 2 systems include condensation of products from gaseous discharge on cold surfaces as well as γ-rays and electron bombardment of solid nitrogen.
Intense "green" emissions were observed in these experiments. Most intense bands in these emissions were assigned to the forbidden transition 2 D → 4 S of nitrogen atoms and 1 S → 1 D transition of oxygen atoms [7]. These transitions are strongly forbidden in a gas phase. The lifetime of N( 2 D → 4 S) transition in gas phase is ≈ 44.5 hours [89] and of O( 1 S → 1 D) transition is 0.75 s [92]. The N( 2 D → 4 S) transition is also forbidden by the angular momentum selection rule and the parity rule for dipole transition at centrosymmetric atomic sites in the solid N 2 lattice. However, it occurs as induced dipole radiation due to the weak dynamical perturbing fields arising from lattice vibrations [13]. The lifetime of N atom α-emission in solid N 2 is ≈30 s [8,13,14]. The influence atoms.
In our experimental method, we injected the nitrogen-helium gas mixture discharge products into the bulk superfluid helium. On the way from the entering helium Dewar orifice to the surface of HeII, the nanoclusters of solid N 2 , containing a high concentration of N atoms, are formed [49,50,51,52]. Most of the N atoms are stabilized on the surface of nanoclusters [45,46,50]. The nanoclusters form an aerogel-like structure inside HeII [42,44,94]. The abovementioned mechanism also explains the green luminescence of the collection of nanoclusters during accumulation.
During sample accumulation, the integral of α-group emission is much larger than that of β-group emission.
The investigation of the emission spectra during the explosive destruction of the collection of the nanoclusters shows that in these fast flashes, the emission intensity of β-group is much larger than that of α-group. The relative increase of β-group intensity was explained by reduced concentration of stabilized nitrogen atoms due to their recombination and the much smaller lifetime of the oxygen atom transition, which provides an efficient channel for releasing the energy created in the sample due to the recombination of stabilized nitrogen atoms.
In this thesis, we described the regime of enhanced β-group emission of oxygen atoms in solid molecular nitrogen nanoclusters. The regime was realized during condensation of activated (passing RF discharge) nitrogen-helium gas jet into dense cold helium gas at temperatures 16-36 K.
The phenomenon of the enhanced oxygen atom β-group emission during nitrogen-helium gas jet condensation in a cold helium vapor can be explained by the following mechanism. The nanoclusters of molecular nitrogen are formed upon cooling down of the discharge products in the cold helium gas. As an example, the N 2 :He = 1:200 gas mixture contains 0.5% of N 2 molecules and ≈ 10 -3 % of O 2 molecules. The O 2 molecules appear as an impurity in the helium gas. The flux of the helium gas is ≈ 5 • 10 19 s -1 , so the flux of N 2 molecules is dN 2 /dt ≈ 2.5 • 10 17 s -1 , and the flux of O 2 molecules is ≈ 5 • 10 14 s -1 . In a cold helium gas, the nanoclusters of N 2 molecules with a size of ≈ 5-10 nm (≈ 2500 N 2 molecules) are formed [40,41,95]). The rate of cluster formation is
Therefore, the N 2 nanoclusters formation rate is of the same order as the flux of O 2 molecules. At such a small content of O 2 molecules in the He gas, all O 2 molecules are dissociated in the discharge zone [96]. Therefore, after passing the discharge, only the atomic form of oxygen should be present in the gas jet. These O atoms might be at the center of nanoclusters.
The We suggested that the observed difference in β-group luminescence enhancement is due to the difference in the nanocluster structure. In the case of the pure N 2 nanoclusters, the nanoclusters are composed of nitrogen molecules, allowing easy energy transfer from the excited nitrogen molecules, produced during nitrogen atom recombination on the surfaces of the nanoclusters, to the oxygen atoms, stabilized in the interior of the nanoclusters. In the case of the N 2 -Kr nanoclusters, composed of a krypton core and nitrogen molecules in the outer layer, energy can also be easily transferred through the surface molecular nitrogen layer to the stabilized oxygen atom. However, according to the ESR study [46], the N 2 :Ne nanoclusters consist of nitrogen molecules core with an outer neon layer and N atoms stabilized in both.
In all conducted experiments, starting from N 2 :Ne ratio of 1:10, all of the nitrogen atoms stabilized in the outer layer will be surrounded by Ne atoms, so in the event of their recombination, the excited N 2 molecules will not have an efficient way to transfer their excitation to the oxygen atoms, stabilized in the inner N 2 core of nanoclusters. This mechanism is supported by the experiment with the N 2 :Ne:He = 1:20:200 mixture (see Fig. 3.8), where oxygen β-group emission enhancement in N 2 -Ne nanoclusters was 10 times smaller than in pure N 2 nanoclusters and 5 times smaller than in N 2 -Kr nanoclusters. The presence of Ne atoms changes the structure of nanoclusters. The neon layer has covered the N 2 nanoclusters core [46]. In this case, the N atoms from the jet recombine on the surface of nanoclusters with the formation of excited N 2 molecules. Ne atom layer suppresses the transfer of excitation from the N 2 molecules formed on the surfaces to the interior of N 2 core of nanoclusters since the lowest excited state of Ne atom (≈16.6 eV) is much higher than the energy released in N( 4 S) atom recombination (≈9.8 eV) [25]. Therefore, the emission of β-group of O atoms inside the core of N 2 nanoclusters is substantially suppressed, as observed in experiments with the gas mixture containing neon.
Studies of luminescence spectra during the injection of N 2 -Kr-He gas jets after passing the discharge zone, into cold helium gas allowed us to verify suggestions regarding the luminescence enhancement mechanism.
At the early stage of cooling in the helium vapors, the Kr atoms form a core of clusters due to strong van der Waals interaction between Kr atoms. After that, the N 2 molecules, N and O atoms condense on the Kr core, forming the outer layers [45]. Upon condensation, the impurity-helium gas mixture with impurity concentration ≈ 1% the clusters with characteristic sizes 5±2 nm are formed. By studying the luminescence spectra, we have observed the effect of emission enhancement of O atoms β-group and V-K bands of N 2 molecules in N 2 -Kr nanoclusters. This effect was observed in the temperature range 20-36 K. Fig. 3.11 illustrates the luminescence dynamics of the gas jet and N 2 -Kr nanoclusters during condensation N 2 :Kr:He = 1:2:600 gas mixture at the time of appearance of enhanced emission. The increase in emission intensity of oxygen atoms and nitrogen molecules is clearly demonstrated in Fig. 3.12 and 3.13 for this gas mixture.
Before elaborating on the mechanism for enhanced emission in N 2 -Kr solid clusters, it is essential to make an assumption for the structure of these clusters. During condensation of nitrogen- krypton gas mixture N 2 :Kr:He = 1:2:600, the average nanocluster should be composed of ≈33% nitrogen with ≈66% Kr. Utilizing the established crystal structure of solid krypton, which adopts a face-centered-cubic configuration with a lattice constant a 0 = 5.59 Å (with the nearest neighbor distance d N N = 3.96 Å), we can calculate the characteristics of a hexagonal icosahedral (hico)
nanocluster [40,41,95,98]. This cluster, containing seven icosahedral layers, would exhibit a diameter of about 5.56 nm and comprise a total of 1975 sites available for occupation by the constituent molecules and atoms. Assuming that krypton atoms occupy 66% of these sites, we estimate an average of 1317 Kr atoms per nanocluster, allowing approximately 658 sites for nitrogen molecules and nitrogen and oxygen atoms.
of the jet. This recombination process occurred in the outer layers of clusters primarily composed of N 2 molecules. The formation of metastable N 2 in the A 3 Σ + u state on the surfaces of these nanoclusters led to the emission of the V-K bands. Moreover, the rapid exchange of energy through N 2 molecules in the outer layer facilitated energy transfer from exited N 2 molecules to stabilized oxygen atoms, and the short lifetime of the O ( 1 S → 1 D) transition, measured at just 0.2 ms, enabled an effective channel for emissions of O atom β-group.
The N atoms from the gas jet interact with N atoms in outer N 2 layers of N 2 -Kr clusters. The mechanism of enhanced emissions of N 2 -Kr clusters was shown to be similar to that of pure N 2 clusters. However, the proximity of Kr atoms could alter the spectra of emission of atoms and molecules residing in the N 2 surface layers. The enhanced emissions of O atom β-group and V-K bands of N 2 molecules were observed during condensing gas mixture with N 2 /Kr ratios from 1/2 to 1/20. However, for gas mixture with N 2 /Kr ratio 1/50, there was no enhanced emission. The absence of enhanced β-group emission for gas mixture N 2 :Kr:He = 1:50:10000 can be explained by small quantities of N atom and N 2 molecules in the clusters. Following the cluster model with 1975 sites, we can estimate that only 1975/51 = 38 sites on the cluster's surface are occupied by N atoms and N 2 molecules. For this gas mixture, the N/N 2 ratio equals to 6.28×10 20 cm -3 /2.28×10 22 cm -3 = 0.0275 [45]. Therefore, on average, only 38×0.0275 ≈ 1 N atom is present in each cluster. In this cluster, nitrogen atoms and molecules, as well as oxygen atoms, are surrounded by Kr atoms.
Even if the recombination reaction of N atoms takes place on the cluster's surface, the energy transfer of the energy to the O atom is suppressed by the Kr atoms.
Adding Kr atoms to the condensed N 2 -He gas mixture changes the nanocluster structure and The most intense features in the spectra of luminescence during the destruction of N 2 -Kr clusters are the O atom β-group and N 2 molecule V-K bands (see Fig. 3.16(q)). This is in contrast with the spectra recorded during the destruction of N 2 -Kr nanoclusters with high concentrations of N atoms, initially collected in bulk HeII [26,27,28,30]. The last spectra are characterized by intense emission of O atom β-group and M-bands of NO molecule. The difference can be explained by different conditions of destruction observations. In our experiments, destruction was initiated not only by increasing temperature but also by the clusters interaction with gas jets containing free radicals and excited atoms and molecules.
The spectra during enhanced emission of N 2 -Kr nanoclusters observed at temperatures ≈20-36 K (see Fig. 3.16( 2)) are similar to that recorded during explosive destruction of N 2 -Kr nanoclusters (see Fig. 3.16( 1)). The similarity of the spectra provides evidence for the resemblance of the emission mechanism for these two cases. The positions of N 2 molecule V-K bands during the enhanced emission period for all gas mixtures studied in this work are shown in Table 3.6.
The positions of VK bands were close to that recorded for N 2 molecules in solid N 2 . In contrast, the spectra of N atom α-group and O atom β-group exhibit substantial modification due to the influence of Kr atoms residing in the cores of the nanoclusters.
Experiments with N 2 -Ar nanocluster showed very similar behavior of enhanced β-group lumiscence to that in N 2 -Kr nanoclusters. This is explained by their similar shell structure.
We observed enhanced β-group emission of oxygen atoms in the collection of molecular nitrogen nanoclusters at temperatures 16-36 K. The energy for excitation of O( 3 P) atoms was transferred through the N 2 matrix from the surface of nanoclusters. This energy results of the recombination of N atoms from the gas jet on the surfaces of N 2 nanoclusters, leading to the formation of the metastable N 2 (A 3 Σ + u ) molecules. The fast process of excitation transfer through the N 2 matrix and the relatively small lifetime of O( 1 S → 1 D) transition provide an efficient mechanism of enhanced luminescence of oxygen atoms in solid molecular nitrogen nanoclusters. To confirm this mechanism, experiments with nitrogen-neon, nitrogen-argon, and nitrogen-krypton nanoclusters have been conducted.
In experiments with N 2 -Ne nanoclusters, we observed insignificant enhancement of β-group emission of oxygen atoms at temperatures of 15-36 K. The β-group emission of oxygen atoms in the collection of N 2 -Ne nanoclusters was weaker than that observed in the collection of pure N 2 [81] and mixed N 2 -Kr [97] nanoclusters in the same temperature range due to solidified neon shell surrounding N 2 core of N 2 -Ne nanocluster, preventing efficient energy transfer from exited nitrogen molecules on the surface, to stabilized oxygen atom in the core.
We observed oxygen atoms β-group and nitrogen molecules V-K bands emissions enhancement in the collection of nitrogen-krypton nanoclusters at temperatures ≈16-36 K. The addition of Kr atoms changes the spectra of enhanced emission compared to that observed for pure N After N atom recombination on the surface of N 2 -Kr or N 2 -Ar nanoclusters, the excitation energy transfers to O atoms through the N 2 molecules in the outer surface layer. Accordingly, the enhanced atoms stabilized in Ne nanoclusters immersed in bulk HeII were studied for nanoclusters with different ratios N/Ne ranging from 0.016% to 5%. We found that the shape of α-group spectra was similar for all studied nitrogen-neon nanoclusters with a small content of N atoms. The component at λ = 519.9 nm is the largest in the α-group spectra. The lifetime of this component is ≈ 280 s. The position of this component is close to that of the N atom in the gas phase, which has a much longer decay time (≈44 hours). We have also made registration of ESR spectra of N atoms stabilized in the collection of neon nanoclusters and found that one of the N atom triplets in the ESR spectra corresponds to a hyperfine splitting constant equal to 3.95G. This value is close to that of free nitrogen atoms. The above facts allow us to suggest that for the first time we observed N atoms in bulk HeII, whose characteristics are influenced by solidified helium atoms. varying the environment of the N( 2 D) atom in the solid matrix [90,99].
In this work, we studied the luminescence of nitrogen-neon nanoclusters with different compositions. The obtained spectra of nitrogen atom α-group emission and measured characteristic decay times of the spectra components provide additional information on the structure of nitrogen-neon nanoclusters. the spectra decay at the same rate. However, the narrow line at the wavelength 519.9 nm decays slower. The initial intensity of this line is smaller than the others, but due to its longer decay time, it is still clearly visible at t = 220 s. Fitting spectra for all the time windows involves multiple steps. First, we fit each spectrum with amplitude, width, and function center as free parameters for each of the five Gaussian functions.
Then, we find optimal values for the function center and width of the Gaussian function, assuming that the shape of individual lines should not change over time. Then, the width and center values are fixed, and for each line, only the amplitude is left as a free parameter. Amplitudes of all fitted lines for all integration windows (10 s each) for N 2 :Ne:He = 1:5:120 gas mixture are presented in Fig. 4.4. From Figs. 4.2 and 4.3b, one can see that at the later stage of α-group decay, the narrow component of the spectra at the λ = 519.9 nm dominates. The observation of a narrow component in the α-group spectra of mixed nitrogen-neon nanoclusters with disordered structure is very surprising. The decaying exponent function fits amplitude dynamics to find the characteristic decay time of each component. This procedure was repeated for the luminescence spectra of nitrogen-neon nanoclusters formed by condensation of the N 2 :Ne:He = 1:5:240 and 1:10:220 gas mixtures. The wavelengths, widths, and decay times of nitrogen α-group spectra components are presented in
Table 4.1. We studied the emission of two different collections of nitrogen-neon nanoclusters formed by N 2 :Ne:He gas mixture with different N 2 /Ne ratios. We found that for gas mixture N 2 :Ne:He = 1:5:120, the spectra of α-group contained five main components. One of the components has a long decay time of 280 s, and the four other components' decay times are in the range from 25 to 49 s. The components originate from the sites of different neighboring environments of N( 2 D) atoms. The first component corresponds to a greater amount of Ne atoms in the N atom environment. The smaller inducing moment of a lighter, spherical neon atom compared to the nitrogen molecule leads to an increased decay time of the emitting N( 2 D) atom. When the nitrogen-neon-helium mixture was further diluted by helium (from N 2 :Ne:He = 1:5:120 to N 2 :Ne:He = 1:5:240), the influence of neon atoms on the α-group emission was reduced. The intensity of the long-lived component stayed the same, but the intensity of other components dropped. Characteristic decay time for the components does not change significantly, and for short-lived components remains closer to that group components are in the 267 to 367 s range, which is closer to that in pure Ne matrix (≈360 s) [90]. This case demonstrates the substantial influence of the Ne atom on the emission N atom α-group in nitrogen-neon nanoclusters. However, the narrow line did not become broader and retained its characteristic decay time. In the α-group spectra of all nitrogen-neon nanoclusters, we observed a narrow component at λ = 519.9 nm, which becomes dominant at the later stage of the decay. The position of this line is close to the positions of the lines in the gas phase.
Therefore, in this case, the environment of the emitting N( 2 D) atoms only decreases the lifetime of the transition compared to that in the gas phase. We suggest that this component might be assigned to the nitrogen atoms on the surfaces of the Ne layers of N 2 -Ne nanoclusters. This assignment is supported by the electron spin resonance studies of N atoms in N 2 -Ne nanoclusters [46]. Solidified helium atoms can also be in the closest shells of these nitrogen atoms. shows spectra for samples prepared from Ne-He gas mixtures, where nitrogen was present only as an impurity in the He gas. To estimate the quantity of nitrogen impurity in He gas, we performed a special experiment in which we compared the integrated spectra of the sample produced from a neon-helium (Ne:He = 1:50) mixture and from a nitrogen-neon-helium gas mixture (N 2 :Ne:He = 1:1000:50000). In the preparation of the last gas mixture, we used purified Ne gas after cleaning it in a liquid nitrogen trap and a known quantity of added nitrogen gas. The obtained integrated spectra for samples prepared from the above-mentioned two mixtures are shown in Fig. 4.6. The spectra were obtained under similar experimental conditions. They were almost identical after scaling the second spectra by a factor of 3.5. Both samples are prepared from the mixtures which contain 2% of impurities, and have been created under the same conditions. This, and the identical shapes of the spectra, allow us to conclude that the different nitrogen concentrations in the mixtures cause the difference in luminescence intensity. If we put the nitrogen concentration in the helium gas to be C He , then the concentration in the N 2 :Ne: He = 1:1000:50000 mixture would be 20ppm + C He . Fig. 4.7 shows the dynamics of the α-group intensity decay for the collection of nitrogen-neon nanoclusters prepared from three different gas mixtures. During the decay process, the shape of the α group spectra changes slightly, but the narrow line at λ = 519.9 nm was always strongest in the spectra. Spectra of α-group show complex structure in the wavelength range from 519 to 524 nm.
The complicated spectrum was fitted by the sum of seven components as shown in Fig. 4.8. The Voigt functions were used to model the spectra components. Fig. 4.8a shows the α-group spectrum at the beginning of decay and the corresponding seven fitting Voigt functions for the collection of nanoclusters prepared from the gas mixture N 2 :Ne:He = 1:50:1000. Fig. 4.8b shows the spectrum in the middle of the decay (120 s after beginning) and the seven corresponding fitting componenets. Fig. 4.8c shows the spectrum in the late stages of the decay (240 s after beginning) and seven fitting components. Table 4.2 and Table 4.3.
For modeling α-group spectra, we used the following approach. We assume that the positions and widths of each component of the spectra should not change during the decay process. There-fore, the fit function used for fitting the α-group spectra during the early stages of the spectra decay and the later stages of the decay should only differ by the amplitudes of the components, but not their shapes. As a first step of our fit procedure, we fitted the early stage of the decay spectra with Voigt functions using seven components for which the position, width, and amplitude were left as free parameters. The decay spectra were split into 10 to 12 s periods. Spectra accumulated during each period were fitted by the composite Voigt function, where the position and the width of the components were fixed according to the fit from step one. Only amplitudes were left as free parameters.
Amplitudes extracted from the fits, for each individual component, have been fitted with a decay exponent, as shown in Fig. 4.9. Table 4.2. in which the exponents are straight lines. Only the blue (fast decay) line and green line (slow decay, narrow line) can be fitted by a straight line, and therefore can be represented by a single exponent. For other lines, we have to use a sum of two decaying exponents (fast τ ≈ 30 s and slow τ ≈ 300 s). A similar analysis of decay was performed for all samples with small nitrogen content. The results of these analyses are presented in Table 4.2 and Table 4.3, where wavelengths, widths, and decay times of components for nitrogen α-group are listed. Most of the spectral components presented in Table 4.2 have a long decay time ranging from 224 to 332 s. Only the second components at λ ≈ 520.1 nm have relatively short decay times ranging from 28 s to 58 s. The decay times for second components were determined by the presence of N 2 molecules in the neon nanoclusters. For the samples with largest contents of nitrogen the spectra of α-group had stronger intensities, but during decay process the components which influenced by nitrogen molecules disappeared faster. Fig. 4.10 shows the dynamics of the ratio of narrow line intensity to the total intensity of α-group spectra. For the samples with larger content of nitrogen (2% -5%) the ratio at the beginning of decay is growing during ≈ 240 s and after reaching maximum is decreasing linearly. The initial growth of the ratio is due to the faster decay of α-group components, which were influenced by the presence of N 2 molecules in the neon nanoclusters [13].
The decay of components corresponding to N atoms surrounded by Ne atoms is much slower [90].
For samples with small (0.04% -0.1%) content of nitrogen in neon nanoclusters, the weight of the narrow line linearly decreased in the decay process. This might be a consequence of longer decay times for the other components of α-group spectra for the latter samples.
The shapes of the α-group spectra in neon nanoclusters with strong maxima at λ ≈ 519.9 nm were similar for all as-prepared samples with small nitrogen content. Warming of the collection of neon nanoclusters from 1.5 K to 13 K initiated the recombination of the stabilized nitrogen atoms and changes in the nanocluster structures. group acquired during the destruction of the sample is completely different from that obtained during luminescence decay of the as-prepared sample. Spectrum during destruction has a maximum at λ ≈ 521.9 nm. This spectrum corresponds to the α-group spectra of nitrogen atoms in N 2 matrix [102]. The intensity of this spectrum is 20 times smaller than that of the as-prepared sample during the decay process. The spectrum obtained during sample destruction is the result of a sequence of processes, including the recombination of stabilized nitrogen atoms in Ne nanoclusters, the formation of N 2 clusters and excited N 2 molecules with the following energy transfer from excited N 2 molecules to N atoms stabilized in N 2 clusters.
Recent experiments provide evidences for the formation of solvation layers of solidified helium around neutral impurity atoms injected into superfluid helium (HeII) [74,75,76,77,78,79,80].
These experiments described the formation of a foam structure by Mg [74,75,76,77] and Xe solvation layers of He atoms [69,71].
The first attempts to create solidified helium were made by injecting gas discharge products into the bulk HeII [22,36,102]. The high concentrations of stabilized nitrogen atoms inside superfluid helium obtained in these experiments were explained by the formation of metastable impurityhelium solids in which the impurity atoms and molecules are surrounded by solidified layers of He due to van der Waals forces between the helium atoms and heavy impurities [22,36,102,103].
Following X-ray and ultrasound investigations of these solid samples revealed only the presence of impurity nanoclusters which form a porous structure inside HeII [38,40,41,42,43,44,95].
The high concentrations of nitrogen atoms in HeII were explained by the efficient stabilization of these atoms on the surfaces of N 2 nanoclusters [46,52,53]. Also, it was suggested that layers of solidified helium are formed on the surfaces of impurity nanoclusters. Solidified helium layers provide stability for porous structures containing a high concentration of chemically active atoms.
However, up to now, there have been no direct experimental observations of the solidified He layers in the collection of nanoclusters.
The goal of this work is an observation of the He atoms influence on the characteristics of nitrogen atoms stabilized on the surfaces of neon nanoclusters surrounded by solidified helium.
In our recent experiment on luminescence of nitrogen-neon nanoclusters immersed into HeII, we observed a narrow line at λ = 519.9 nm in the N atom α-group spectra [84]. During the decay of luminescence, this line became the largest component in the spectra. The decay lifetime of this line is 280 s. The characteristics of this line might be explained by the specific environments of N atoms responsible for the emission of this line. It was suggested that this line might be assigned to N atoms on the surfaces of Ne nanoclusters and surrounded by a solidified layer of He atoms.
To provide additional evidences for this suggestion, we performed systematic studies of the luminescence of N atoms stabilized in neon nanoclusters with smaller nitrogen contents than those used in previous work. In our work, the nitrogen to neon (N/Ne) ratio in nanoclusters was varied in the range from 5% to 0.04%. In all spectra obtained in this work, the intensity of the narrow line at λ = 519.9 nm was maximal. Furthermore, the relative intensity of this line was growing relative to all other components α-group when the content of nitrogen in Ne nanoclusters was reduced (see Fig. 4.5). It was found earlier that upon condensation, nitrogen-neon-helium gas mixtures into HeII, most of the N atoms are stabilized on the surfaces of nanoclusters [46]. The enhancement of the narrow line and the increase in relative intensity upon reducing the nitrogen content support the assignment of the narrow line to N atoms on the surfaces of neon nanoclusters. For the small content of N atoms, the larger portion of them should be stabilized on the surfaces of Ne nanoclusters. The neon nanoclusters are surrounded by a layer or two of solidified helium [53]. It is natural to expect that characteristics of N atoms stabilized on surfaces of nanoclusters will be influenced by the close proximity of solidified He atoms.
We can compare spectra of the α-group in luminescence of nitrogen-neon nanoclusters immersed in HeII and in the luminescence of N atoms in solid N 2 /Ne films [11,90,101]. The In future experiments, the direct registration of solidified helium can be realized. For this purpose, the collections of neon nanaoclusters with smaller sizes shall be investigated. It is known that dilution of impurities by helium gas in condensed impurity-helium mixtures allows the formation in HeII the collections of nanoclusters with smaller sizes [40,41]. In the past, the gas mixtures with more than 1% of impurities had been used. As a result, the average size of nanaoclusters was of order 5 nm (≈ 2000 impurity atoms/molecules) [40,41]. The preparation of neon nanoclusters by using impurity-helium gas mixtures with Ne/He ratios from 0.1 % to 0.01 % should be realized. The ratio of the quantity of solidified He atoms to the quantity of Ne atoms in the samples is expected to be larger in a collection of nanoclusters with smaller sizes. The X-ray method might be used for the determination of nanocluster sizes and for the direct observation of the solidified helium on the surface of neon nanoclusters. Earlier experiments showed that the collection of neon nanoclusters can survive after being removed from liquid helium (dry samples) [40]. For dry samples, the background signal from liquid helium will be considerably reduced, providing a better chance for detection of a signal from the solidified He layers.
The luminescence of neon nanoclusters doped by nitrogen atoms was studied. Neon nanoclusters were collected inside bulk superfluid helium at temperatures below 1.5 K. We found that the spectra of N atoms α-group have a specific shape with an enhanced narrow line at λ = 519.9 nm.
The measured spectroscopic characteristics of this line, such as the position, the narrow linewidth, and the relatively large intensity, allow assignment of this line to the N atoms on the surfaces of neon nanoclusters surrounded by layers of solidified helium. This is the first observation of the solidified helium influence on the spectral characteristics of nitrogen atoms stabilized inside the bulk superfluid helium.
This thesis has presented a comprehensive investigation into the low-temperature luminescence properties of nanoclusters doped with nitrogen and oxygen atoms, formed by condensing radiofrequency discharge products in a cold helium gas atmosphere and bulk superfluid helium (HeII).
The research was centered on two primary objectives: first, to explore and understand the phenomenon of enhanced oxygen β-group emission observed during the interaction of nitrogen and nitrogen-rare gases nanoclusters with a gas jet, and second, to search for the first direct spectroscopic evidence of solidified helium layers forming on the surfaces of nanoclusters in bulk HeII.
Through a series of experiments employing luminescence spectroscopy, this work has successfully explained energy-transfer mechanisms occurring during enhanced emissions. It also provided compelling evidence for the influence of a solid helium layer on the surface of nanoclusters accumulated in bulk HeII, on the spectral characteristics of stabilized atoms. The findings not only confirm longstanding hypotheses about the structure of impurity-helium condensates, but also establish a new area of research: methodology for studying nancluster interaction with a gas jet, in a cold, dense, helium gas.
A central achievement of this research is the detailed characterization and explanation of the enhanced luminescence of oxygen atom β-group (O( 1 S→ 1 D) transition). This phenomenon was consistently observed when nanoclusters interacted with gas in cold helium gas at elevated temperatures (16-36 K), after all the bulk superfluid helium evaporated from the beaker. The underlying mechanism, as established in this work, involves the recombination of nitrogen atoms (N( 4 S)) from the gas jet on the surfaces of formed nanoclusters. The recombination of two nitrogen atoms in the ground state produces metastable nitrogen molecules, N 2 (A 3 Σ + u ). Excitation from the nitrogen molecule can be efficiently transferred through the matrix of solidified N 2 molecules. This process was found to be critically dependent on the nanoclusters' internal structure, providing a powerful of two. This critical finding supports an inverted shell-core structure for N 2 -Ne nanoclusters, where stronger N 2 -N 2 van der Waals forces, compared to the N 2 -Ne and Ne-Ne, cause molecular nitrogen to form the core, which is subsequently covered by a shell of neon atoms. In this configuration, the outer neon shell acts as an insulating barrier. The high excitation energy of neon (≈16.6 eV)
prevents it from participating in the energy transfer from the recombining N atoms (releasing ≈9.8
eV) on the surface to the oxygen atoms stabilized in the N 2 core.
In summary, the investigation of enhanced β-group emission, presented in this work, allowed for a model that directly links the luminescence behavior to nanocluster structure. The dramatic difference in emission enhancement between Kr/Ar-containing and Ne-containing nanoclusters provides a clear spectroscopic signature of their core-shell structure, validating the proposed energy-transfer mechanism.
The second major contribution of this thesis is the observation of the first direct spectroscopic evidence for the existence and influence of solidified helium layers on the surfaces of impurity nanoclusters immersed in HeII. While the existence of such layers has long been hypothesized to explain the stability of impurity-helium condensates, direct observational proof has remained elusive. This work successfully identified a unique spectral signature associated with this quantum solid environment by using the spectra of nitrogen atom α-group luminescence.
The Narrow Spectral Component at λ=519.9 nm: The key piece of evidence is a distinct, narrow emission line observed at λ=519.9 nm in the nitrogen atom α-group N( 2 D→ 4 S) spectrum. This feature consistently dominated the spectra for all samples prepared with a low nitrogen-to-neon ratio (ranging from 5% to 0.04%), where surface-residing atoms are expected to be the majority. The position of this line is remarkably close to the gas-phase transition of an unperturbed nitrogen atom (λ=519.8 nm), indicating that the emitting atoms experience a minimal matrix shift. Such a weak interaction is strong evidence of a helium environment, which is significantly less perturbing than a neon or nitrogen solid matrix.
Luminescence Decay Kinetics: Analysis of the luminescence decay provided further differentiation. The λ=519.9 nm component exhibited a long decay, which can be fitted by a singleexponential decay with a lifetime of approximately 280 seconds. In contrast, other components within the α-group showed more complex, multi-exponential decays, including faster initial decay times attributed to interactions with nearby N 2 molecules. The simple, slow decay of the narrow line is consistent with an isolated N atom in a highly inert, symmetric environment, as would be provided by a surrounding solidified helium layer.
Effect of Thermal Destruction: The most definitive evidence came from observing the spectra during thermal destruction. When a sample was warmed from 1.5 K to 13 K, the delicate structure stabilized in the HeII bath was destroyed. During this process, the sharp λ=519.9 nm line completely vanished. It was replaced by a broad spectrum with a maximum near λ=521.9 nm, which is characteristic of nitrogen atoms embedded within a solid nitrogen matrix. This transformation demonstrates that the environment responsible for the unique narrow line is fragile and exists only in the bulk superfluid HeII environment. The destruction of the solidified helium layer upon warming allows the surface N atoms to recombine, forming a completely different, more strongly interacting environment.
These three converging lines of evidence -the near-gas-phase line position and narrow width of the λ=521.9 nm line, the simple, long decay lifetime, and the disappearance of the line upon thermal destruction -collectively build an undeniable case. This work concludes that the narrow spectral line at λ=519.9 nm is the first observed spectroscopic signature of atoms on the surface of a nanocluster whose characteristics are predominantly influenced by a surrounding layer of solidified helium.
In conclusion, this thesis has made two significant advances in the study of impurity-helium condensates. First, it covers the discovery of the phenomenon of enhanced oxygen β-group luminescence and its mechanism. It also uses this mechanism as an optical probe to confirm the nanocluster core-shell structure. The developed mechanism, based on surface recombination and structure-dependent energy transfer, was successfully validated across pure N 2 , and mixed N 2 -Ne, N 2 -Ar, and N 2 -Kr systems. Second, and perhaps more fundamentally, this research has provided the first compelling spectroscopic identification of atoms whose properties are influenced by a solidified helium layer. The distinct spectral features of the N atom probe, its near-gas-phase emission wavelength, narrow linewidth, and unique decay kinetics offer a definitive signature of a solid helium environment.
These findings open up several promising avenues for future research:
Exploring Quantum Gels: The methodology developed here, particularly the use of highly dilute impurity-helium gas mixtures, should be pushed further to reduce nanocluster sizes. By using Ne/He ratios of 0.1% or lower, it may be possible to move beyond nanoclusters covered in helium toward the formation of a true macroscopic "quantum gel," where individual impurity atoms or very small clusters are suspended in a solid helium matrix.
Direct Structural Characterization: The samples prepared with high helium dilution should be investigated with structural methods, such as X-ray scattering. This would allow for a direct measurement of nanocluster sizes and could potentially provide a direct diffraction signal from the ordered solid helium layers, correlating the structural reality with the spectroscopic signatures identified in this work.
Generalizing the Probe Technique: The use of nitrogen atoms as a probe for the solidified helium environment should be extended to other systems. Investigating probe atoms on the surfaces of Ar or Kr nanoclusters could determine if the observed spectral characteristics are universal to the helium layer itself or are specific to the N-on-Ne system.
Investigating V-K Band Enhancement: The observation that strong V-K band enhancement occurs in N 2 -Kr nanoclusters but not in N 2 -Ar or pure N 2 clusters is intriguing. Further studies are needed to understand the specific role the krypton matrix plays in promoting or allowing this radiative decay pathway for the metastable N 2 (A 3 Σ + u ) molecules formed on the surface.
This work lays a solid foundation for future explorations of the fascinating physics and chemistry that occur at the interface of impurities and quantum fluids. The spectroscopic tools and physical insights developed herein will be invaluable for the continued pursuit of creating and understanding novel forms of quantum matter.