Classical ghost imaging is an active imaging technique that produces the likeness of an object in a mathematical basis of its user ' s choice using light intensity distributions that did not directly interact with the object. Rather, single-pixel detector measurements of light reflected or transmitted by the object provide weighting coefficients for the mathematical basis functions. After conducting many measurements with differing intensity distributions, an expansion of the object ' s image function is produced. Recent studies demonstrated that nondiffracting, self-healing beam structures can be used as a mathematical basis with which to perform classical ghost imaging. These beam structures are particularly desirable for ghost imaging over long distances through optically turbulent environments. Indeed, these same studies demonstrated the ability to perform ghost imaging through optically turbulent mediums with unspecified turbulence strengths. This research seeks to experimentally quantify the relationship between ghost image quality and simulated atmospheric turbulence using nondiffracting, self-healing beam structures. Multiple spatial light modulators are used to produce the desired beam structures and simulate atmospheric turbulence. Results are compared with ghost images obtained using light structures that are prone to diffraction through comparable turbulence strengths. Ultimately, this research explores the possibility of using classical ghost imaging for practical applications over long distances.

We present the generation of MeV-energy ions, electrons, x-rays, and D-D fusion neutrons using milliJoule-energy laser pulses on a thin (< 1 mu m) liquid sheet. The system produces deuterium-deuterium (D-D) fusion at high (1 kHz) repetition rate, allowing for consistent neutron generation. High repetition-rate and a frequency-shifted pump-probe shadowgraphy scheme allow the evolution of the plasma expansion to be studied at the femtosecond level. Evidence of D-D fusion neutron production is verified by a measurement suite with three independent detection systems: an EJ-309 organic scintillator with pulse-shape discrimination, a He-3 proportional counter and a set of 36 bubble detectors. Additionally, high repetitionrate, large-scale data collection allows for the training of machine learning models, which we demonstrate can effectively control the number of MeV electrons ejected. This high-repetition-rate laser-driven mixed radiation source could provide a low-cost, on-demand test bed for radiation hardening and imaging applications.

Laser-Induced Breakdown Spectroscopy (LIBS) provides a promising alternative approach for rapid analysis of lithium isotopes. Broadening and self-reversal effects in LIBS have traditionally restricted the ability for experimental measurements to spectrally resolve fine structure emission characteristics. Using a TriVista-777 spectrometer, an ICCD detector, and a 532nm nanosecond pulsed laser, it is first shown that powdered lithium hydroxide (LiOH) deposited on stainless steel results in reduced spectral broadening and comparable peak intensity when compared to pressed pellet targets under atmospheric conditions. Next, by reducing the amount of LiOH powder deposited onto the stainless-steel substrate, self-reversal effects are shown to be eliminated entirely without significant decreases in the signal-to-noise ratio (SNR). Empirical results confirm sufficient suppression of broadening and self-absorption effects for clear spectral peak separation of lithium ' s 2s-2p D doublets using single-shot surface-enhanced LIBS (SE-LIBS). Resolution of lithium isotope fine structure emissions validate the experimental designs ability in limiting the combined effects of Stark, Doppler, and instrumental broadening to less than 15pm, while maintaining a high SNR for single-shot analysis. Operation of the Trivista-777 spectrometer in the three-stage configuration, using consecutive dispersion gratings of 1800g/mm, 1800g/mm, and 1200g/mm, provided a final resolution of approximately 8pm when combined with a Teledyne PI-MAX 4 intensified charge couple device (ICCD) camera.

With the rise of high repetition rate ultra-intense laser systems, there is a need for solid density targets to study relativistic laser-plasma interactions that can operate at the same repetition rate. Flowing liquid targets are attractive because they are self-replenished, debris free, cost effective and easy to use. Liquid targets have been used for high-repetition rate (up to kHz rate) generation of electrons, protons, x rays, and neutrons by our group and elsewhere. In this Letter, we demonstrate a kHz-rate generation of a variety of dynamically shaped complex-structured targets from the interaction of a 10(16) W/cm(2) focused laser pulse with a submicrometer liquid sheet. The repeatable structured target evolves over microseconds, forming different shapes, such as a hollow channel, a cone, a cone-wire, and a curved surface with a wire. Based on a particle-in-cell simulation, we show that with a cone-wire target, the kHz-rate laser-target interaction could enable the statistical study of warm-dense matter generation that may be useful as a high flux x-ray source. Each stage of the target evolution provides a unique target geometry for ultra-intense laser plasma interaction with numerous potential applications, such as pulsed secondary source generation for high-resolution imaging.

We propose a novel method for controlling quantum states of light through the manipulation of vacuum fields, enabling the deterministic generation of strongly correlated and entangled photons. This approach provides new capabilities for quantum information processing and long-distance quantum communication. We consider a Λ -type three-level atom embedded in a QED cavity, where one atomic transition is driven by a classical field while the other interacts with a cavity-modified vacuum field with chirped frequency modulation. The coherent interplay between the classical and vacuum fields allows the generation of quantum optical states with tunable correlations and spectral properties. These results suggest promising applications in quantum sensing, single-photon spectroscopy, and quantum teleportation.

In this paper, we study interaction of quantum fields and a single three-level atom placed in the Mach–Zender interferometer. We demonstrate that the phases acquired by quantum fields depend on the number of photons in the quantum states. Notably, the phases differ between single- and two-photon states, enabling the separation of multiphoton states. This finding highlights new applications related to the dispersion of three-level atoms, which are important in advancing quantum information processing and enhancing quantum communication technologies. The results are crucial for long-distance quantum communication and hold potential for developing quantum field-based linear devices such as beam splitters, lenses, and quantum prisms capable of separating different components of quantum fields. The findings can have interesting applications for manipulating and assembling of multiphoton entanglement states.

Long-range quantum communication requires sending qubits, usually photonic qubits, over long distances through free space. Photons propagating through free space encounter optical turbulence, causing scattering and entanglement degradation. In recent decades, adaptive optics has been used to correct imaging errors introduced by turbulence. An atmospheric turbulence simulator has been used at AFIT to measure the effect of optical turbulence on single photon propagation, and an adaptive optics system is in development to be incorporated. The turbulence simulator uses two rotating phase plates to emulate the effects of random dispersive media and two afocal systems to scale long propagation distances down to lengths realizable in a laboratory. The adaptive optics system consists of a Shack-Hartman wavefront sensor used to measure wavefront error, and a deformable mirror used to correct this error. The use of adaptive optics allows us to correct wavefront errors introduced by turbulence. Adding an adaptive optics system into this simulator facilitates more robust characterization of the strength of the simulated turbulence, as well as correction of some of the negative effects the turbulence introduces. This paper discusses the effects of turbulence on single photon propagation and progress towards using an adaptive optics system to correct for these effects in order to compare the results with previous measurements and quantify the improvement adaptive optics introduces.

The emergence of quantum sensing as an active subject of interest heralded a potential watershed moment in defense information collection capabilities. However, for these concepts to leave the laboratory, they must first demonstrate their usefulness under less-than-ideal conditions. One of these concepts in particular, quantum illumination, merits increased interest if it can overcome the effects of atmospheric turbulence. Quantum illumination theoretically promises exponentially higher resolution of object images. This research characterizes the effects of turbulence on the ability to successfully resolve an object image using photon pairs within the optical regime. It is inspired by an experiment conducted in the Canary Islands in 2007 that demonstrated free- space propagation of entangled photons over a distance of 144 km, while it fundamentally imitates the experimental design of "ghost imaging" conducted at the University of Maryland, Baltimore County in 2009. However, it differs through its inclusion and expansion of research results obtained at the Air Force Institute of Technology in 2023, which established a relationship between the second-order correlation function, aperture diameter, and Fried coherence diameter. Ultimately, this research better quantifies the fundamental scientific challenges that must be overcome prior to the development of an operational quantum illumination system and makes recommendations regarding potential solutions.

Ultra-intense laser and plasma interactions with their ability to accelerate particles reaching relativistic speed are exciting from a fundamental high-field physics perspective. Such relativistic laser-plasma interaction (RLPI) offers a plethora of critical applications for energy, space, and defense enterprise. At AFIT's Extreme Light Laboratory (ELL), we have demonstrated such RLPI employing a table-top ∼10mJ, 40 fs laser pulses at a kHz repetition rate that produce different types of secondary radiations via target normal sheath acceleration (TNSA). With our recent demonstration of laser-driven fusion, the secondary radiations generated are neutrons, x-ray emission, and MeV energy electrons and protons-all at a kHz rate. To achieve the high repetition rate, we developed the enabling kHz-repetition-rate-compatible liquid targets in the form of microjets, droplets, and submicron-thick sheets. These targets, combined with high repetition rate diagnostics, enable a unique, real-time feedback loop between the experimental inputs (laser and target parameters) and generated sources (x-rays, electrons, ions, etc.) to develop machine learning (ML)-based control of mixed radiation. The goal of this paper is to provide an overview of the capabilities of ELL, describe the diagnostics and characteristics of the secondary radiation, data analysis, and quasi-real-time ML functionality of this platform that have been developed over the last decade and a half.

Laser-induced breakdown self-reversal isotopic spectrometry (LIBRIS) is implemented to record the Li 670.8 nm peak self-reversal shift with varying 6Li atom percent compared to 7Li. The self-reversed peak center wavelength is shown to shift across a range of 13.813±1.21pm in LiOH⋅H2O6Li samples varying from 3 to 95 6Li atom percent. Supervised machine-learning regressions are trained on the self-reversal shift in order to quantify the isotopic concentration of the samples. A stacked ensemble model using multiple supervised regression base learners is found to yield the superlative characterization of isotopic content with an RMSE of 5.66 at.% and a detection limit of 18.8 at.%.

Quantum sensing offers extraordinary potential in a multitude of applications. "Ghost imaging" is one such technique that enables the imaging of objects using only reflected or transmitted power measurements. This is in contrast with other forms of light imaging in that no spatial information about the imaged object is contained in the measured return signal. Classical and quantum variations of ghost imaging exist, each with its own strengths and weaknesses. Both variations depend upon correlated measurements of light generated by the imager, with only a single-pixel "bucket" detector needed to produce a two-dimensional image of the target object. However, a comprehensive study of atmospheric turbulence on the quality of both classical and quantum ghost images is necessary to fully understand the potential of this technique outside the laboratory. In this research, both classical and quantum ghost imaging techniques are subject to the same atmospheric turbulence generated by a numerical simulation, thereby providing a common basis upon which to judge the relative performance of each. An array of turbulence values, as measured by the Fried coherence diameter, are considered to evaluate the merits of ghost imaging under less-than-ideal environmental circumstances and to identify potential conditions under which one type of ghost imaging is preferable to the other.

Ultra-intense laser–matter interactions are often difficult to predict from first principles because of the complexity of plasma processes and the many degrees of freedom relating to the laser and target parameters. An important approach to controlling and optimizing ultra-intense laser interactions involves gathering large datasets and using these data to train statistical and machine learning models. In this paper, we describe experimental efforts to accelerate electrons and protons to ∼MeV energies with this goal in mind. These experiments involve a 1 kHz repetition rate ultra-intense laser system with ∼10 mJ per shot, a peak intensity near 5 × 1018 W/cm2, and a “liquid leaf” target. Improvements to the data acquisition capabilities of this laser system greatly aided this investigation. Generally, we find that the trained models were very effective in controlling the numbers of MeV electrons ejected. The models were less successful at shifting the energy range of ejected electrons. Simultaneous control of the numbers of ∼MeV electrons and the energy range will be the subject of future experimentation using this platform.

Advances in ultra-intense laser technology have increased repetition rates and average power for chirped-pulse laser systems, which offer a promising solution for many applications including energetic proton sources. An important challenge is the need to optimize and control the proton source by varying some of the many degrees of freedom inherent to the laser-plasma interactions. Machine learning can play an important role in this task, as our work examines. Building on our earlier work in Desai et al. 2024, we generate a large similar to 1.5 million data point synthetic data set for proton acceleration using a physics-informed analytic model that we improved to include pre-pulse physics. Then, we train different machine learning methods on this data set to determine which methods perform efficiently and accurately. Generally, we find that quasi-real-time training of neural network models using single-shot data from a kHz repetition rate ultra-intense laser system should typically be feasible on a single GPU. We also find that a less sophisticated model, like a polynomial regression, can be trained even faster and that the accuracy of these models is still good enough to be useful. We provide our source code and example synthetic data for others to test new machine learning methods and approaches to automated learning in this regime.

Recent developments in atomic spectroscopy techniques enable rapid quantitative analysis of nuclear material through the implementation of data science techniques. Atomic emission spectra of such materials are often convoluted, owing to their complex makeup and electronic structures. Consequently, performing a chemical analysis using such spectra requires the implementation of advanced analytical methods to understand the relationship between spectral emission features and the chemistry of the material. We present an implementation of spectral analysis methods that combine principal component analysis with supervised machine learning regressions, enabling the quantification of gallium in cerium matrices with superior precision and sensitivity. The proposed principal component-generalized spectrum-machine learning (PC-GS-ML) approach yields marked improvements in model accuracy compared to regressions trained using solely spectral features or PC reduced spectral features. Prediction errors as low as 0.08 wt% Ga are achieved by training boosted ensemble and Gaussian kernel regression models on PC-GS preprocessed laser-induced breakdown spectroscopy data, showing an order of magnitude improvement in the Ga quantification error over prior studies in literature.

We study the propagation of a quantum field composed of a few photons interacting with a three-level Λ-atom driven by a coherent classical field. The quantum field acquires a phase shift, which can be interpreted as a dispersion effect on the photon wave packet and described by the refractive index for quantum fields down to the single-photon level. In this paper, we demonstrate that the phases acquired by quantum fields depend on the number of photons in the quantum states. Notably, the phases differ between single- and two-photon states, enabling the separation of multiphoton states. This finding highlights new applications related to the dispersion of three-level atoms, which are important in advancing quantum information processing and enhancing quantum communication technologies. The results are crucial for long-distance quantum communication and hold potential for developing quantum field-based linear devices such as beam splitters, lenses, and quantum prisms capable of separating different components of quantum fields. The findings can have interesting applications for manipulating and assembling of multiphoton entanglement states.

We use the Fourier-based coherent atomic transfer (CAT) function for EIT storage in a Λ atomic system to reconstruct an arbitrary pulse shape given the retrieved output using numerical deconvolution algorithms.

In recent work, we have demonstrated the capability of directional beam forming in the microwave portion of the electromagnetic spectrum using linear superconducting ring arrays, triggered by an ultra-fast laser pulse. This paper extends these developments by presenting modeling of microwave radiation patterns covering the entire four pi steradian sphere, and outlining an experimental plan for detailed characterization of planar arrays. Experimental data is compared with the simulations conducted in this study, showing good agreement that builds confidence for further work. Planned research is presented to include array design optimizations for varying frequencies into the THz region. Additional research will include experimental and computational investigation of THz propagation using previously developed laser atmospheric propagation software. If successful, the numerical models used could prove beneficial in predicting real time absorption characteristics.

Lithium compounds such as lithium hydride (LiH) and lithium hydroxide (LiOH) have a wide range of industrial applications, but are highly reactive in environments with H2O and CO2. These reactions lead to the ingrowth of secondary lithium compounds, which can alter the homogeneity and affect the application of particular lithium chemicals. This study performed an exploratory analysis of different lithium compounds using laser-induced breakdown spectroscopy (LIBS) and Raman spectroscopy. Machine learning models are trained on the recorded spectral data to discriminate emission features that differ between LiH, LiOH, and Li2CO3 to perform high-fidelity classification. Support vector machine classifiers yield perfect prediction accuracy between the three compounds with optimal training time. Multivariate methods are then used to produce regression models quantifying the ingrowth of LiOH in LiH. Performing a mid-level data fusion of selected LIBS and Raman features with partial least-squares regression produces the superlative model with a root mean square error of 2.5 wt% and a detection limit of 6.3 wt%.

In this study, we consider three different machine-learning methods-a three-hidden-layer neural network, support vector regression, and Gaussian process regression-and compare how well they can learn from a synthetic data set for proton acceleration in the Target Normal Sheath Acceleration regime. The synthetic data set was generated from a previously published theoretical model by Fuchs et al. 2005 that we modified. Once trained, these machine-learning methods can assist with efforts to maximize the peak proton energy, or with the more general problem of configuring the laser system to produce a proton energy spectrum with desired characteristics. In our study, we focus on both the accuracy of the machine-learning methods and the performance on one GPU including memory consumption. Although it is arguably the least sophisticated machine-learning model we considered, support vector regression performed very well in our tests.