THIN FILM MASER EMITTER AND THIN PANEL PHASED ARRAY OF EMITTERS

Abstract

A MASER (Microwave Amplified Stimulated Emission of Radiation) emitter is fabricated of thin film components, including a thin film of nitrogen-implanted, epitaxial crystal diamond. The MASER elements can also include a controllable Q-switching layer and be arranged in a thin panel, phased array to generate a single beam of coherent, mode-locked, continuous wave MASER radiation.

Claims

1. A MASER emitter capable of outputting coherent microwave energy, comprising: a first dielectric polymer resonator layer; a second dielectric polymer resonator layer; a gain medium comprising a thin film epitaxial layer of nitrogen-implanted crystal diamond, a Q-switch CCD controlled layer of nematic molecules located adjacent the gain medium, wherein the gain medium and the Q-switch are located between the first and second dielectric polymer resonator layers, said Q-switch being controllable between a high Q factor setting and a low Q factor setting; an LED layer that outputs light that passes through at least the first dielectric polymer resonator layer to pump photons into the gain medium at a level sufficient to cause coherent microwave emission from the gain medium in continuous wave mode when the Q-switch is set to the high Q factor setting.

2. The MASER emitter recited in claim 1 further comprising a thermo-electric Peltier slab layer oriented adjacent the LED layer.

3. The MASER emitter recited in claim 2 wherein the thermo-electric Peltier slab layer is formed on an underlying substrate, the LED layer is formed on the thermo-electric Peltier slab layer, the first dielectric polymer resonator layer is formed on the LED layer, the Q-switch CCD controlled layer of nematic molecules is formed on the first dielectric polymer resonator layer, the thin film epitaxial layer of nitrogen-implanted crystal diamond is formed on the Q-switch CCD controlled layer of nematic molecules, and the second dielectric polymer resonator layer is formed on the thin film epitaxial layer of nitrogen-implanted crystal diamond.

4. A phased array comprising a plurality of MASER emitters as recited in claim 1, each arranged in a two-dimensional array and each capable of emitting continuous wave, coherent microwave radiation, wherein the respective Q-switch layers are controlled to appropriately delay MASER emission from the respective MASER emitters in order to form a combined emission from the plurality of MASER emitters in the phased array that is a single mode-locked, beam of coherent, continuous wave MASER radiation.

5. The phased array comprising a plurality of MASER emitters recited in claim 4 wherein a first group of emitters are located in a first plane, and a second group of emitters are located in a second parallel plane and offset spatially from the first plane.

6. The phased array comprising a plurality of MASER emitters recited in claim 4 wherein the emitters are arranged in column and rows.

7. The phased array comprising a plurality of MASER emitters recited in claim 4 wherein the emitters are arranged in a circular pattern or a star pattern.

8. The phased array comprising a plurality of MASER emitters recited in claim 4 wherein the distribution of energy in the coherent, mode-locked continuous wave MASER radiation is orthogonally Gaussian.

9. The phased array comprising a plurality of MASER emitters recited in claim 8 wherein Gaussian orthogonality is achieved by adjusting the LED pump energy and coordinating the operation of the Q-switches.

10. The phased array comprising a plurality of MASER emitters recited in claim 4 wherein the thickness of the epitaxial gain medium is no less than 5 micron and no more than 40 microns.

11. A MASER apparatus capable of outputting continuous wave, coherent microwave radiation, comprising: a gain medium comprising a layer of nitrogen-implanted, epitaxial diamond; a first resonator layer located on a first side of the layer of nitrogen-implanted, epitaxial diamond; a second resonator layer located on a second side of the layer of nitrogen-implanted, epitaxial diamond; and an LED that outputs light to pump photons into the gain medium at a level sufficient to cause continuous wave, coherent microwave emission.

12. The MASER apparatus recited in claim 11 wherein the layer of nitrogen-implanted, epitaxial diamond has a thickness of no greater than 40 microns.

13. The MASER apparatus recited in claim 12 wherein the nitrogen-implanted, epitaxial diamond has a thickness of no greater than 30 microns.

14. The MASER apparatus recited in claim 11 wherein the first resonator layer is a Spin Hall, spin torque oscillator.

15. The MASER apparatus recited in claim 11 wherein the second resonator is a lithographed chemical vapor deposited copper layer.

16. The MASER apparatus recited in claim 11 further comprising a heat sink thermally connected to the LED.

17. The MASER apparatus recited in claim 11 wherein the MASER apparatus operates between 29 C. and 43 C.

18. The MASER apparatus recited in claim 11 wherein the layer of nitrogen-implanted, epitaxial diamond is exposed to radiation by a beam of swift heavy ions during the fabrication process.

19. The MASER apparatus recited in claim 18 wherein the layer of nitrogen-implanted, epitaxial diamond is also exposed to radiation by a beam of low energy electrons during the fabrication process.

20. The MASER apparatus recited in claim 4 wherein the Q-switches are controlled to adjust phase and frequency of the beam such that the beam includes a first characteristic frequency for a coupling component of the beam and a second characteristic frequency for a modulation component of the beam.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0058] FIG. 1 is a diagram illustrating wavelength and frequency along the electromagnetic spectrum.

[0059] FIGS. 2A and 2B are Jablonski diagrams illustrating the relaxation of a molecule or atom to a lower energy state through emission of photons (FIG. 2A) and relaxation of a molecule or atom to a lower energy state through loss of heat and photon emission (FIG. 2B).

[0060] FIG. 3 is a diagram illustrating stimulated emission.

[0061] FIG. 4 is an illustration showing a MASER element constructed in accordance with an exemplary embodiment of the invention.

[0062] FIG. 5 is a diagram showing a thin panel, phased-emitter array constructed in accordance with an exemplary embodiment of the invention.

[0063] FIG. 6 is a schematic cross-sectional view of the thin panel, phased-emitter array shown in FIG. 5.

[0064] FIG. 7 is a schematic diagram illustrating irradiation conditions of nitrogen-implanted epitaxial diamond by swift heavy ion (SHI) radiation and low-energy electron radiation.

[0065] FIG. 8 is a hyperspectral image of photoluminescence (PL) intensity from NV.sup. centers (635-642 nm) in epitaxial diamond under various experimental conditions of irradiation.

[0066] FIG. 9 is a plot depicting PL intensity under the various experimental conditions illustrated in FIG. 8, before thermal annealing. The inset is a plot of Raman intensity for experimental conditions 2 and 3 in FIG. 8.

[0067] FIG. 10 is a plot depicting PL intensity under the various experimental conditions illustrated in FIG. 8, after thermal annealing.

[0068] FIG. 11 is a plot illustrating relative NV.sup. PL intensities from exposure to SHI and to low energy electrons for a series of N+ implantation energies and fluences without thermal annealing. For SHI alone and both treatments, the damage signal from SHI irradiation of pristine diamond areas was subtracted. The PL intensity before background subtraction is indicated by the top of the error bars. For the electron treatment, the background signal measured from pristine diamond was subtracted. The data is normalized to an NV.sup. signal from a 10.sup.13/cm2 10 keV N+ implant followed by exposure to 10 keV electrons.

DETAILED DESCRIPTION

[0069] In 2018, Breeze et al., successfully demonstrated a room-temperature MASER operating in continuous wave mode (Breeze at, 2018). The most critical factor in their success was the use of a nitrogen-doped, bulk diamond. The device used a laser to pump photons into the nitrogen-doped, bulk diamond, which was housed in a copper resonator cavity. At 0.4 W of optical power, the masing threshold was reached. The device generated MASER emission in continuous wave mode, and was stable for all experiments, the longest being more than 10 hours.

[0070] In contrast to the Breeze et al. device, the present invention uses layers of thin film to fabricate MASER emitters, and phased arrays of MASER emitters. FIG. 4 illustrates the layer components of a single MASER emitter 10 constructed in accordance with an exemplary embodiment of the invention. The gain medium 12 is a nitrogen-implanted layer of epitaxial crystal diamond, e.g., about 10 in thickness. The layered components 12, 14, 16, 18, 20, and 22 are layered for example on a standard printed circuit board substrate. A thermo-electric Peltier slab 22 is the base layer and is adjacent the thin film, high-output LED (Light Emitting Diode) layer 18. The LED layer is desirably made of an alloy of Ga, Al, and In as cations, and N as the anion and is controlled to receive the required forward voltage to generate light waves at 530 nm. The current level is adjusted to control the lumens of light output from the LED and ensure that the masing threshold is reached. The Peltier slab 22 is controlled to reduce heat build-up in the LED layer 18 and otherwise ensures that the temperature of the LED 18 is within the range for suitable effectiveness. The thin film, high-output LED layer is the photon pump for the MASER emitter 10.

[0071] Alternating layers of dielectric polymers 14, 16 are made of materials chosen to reflect microwaves 24 emitted in the active layer 12 (i.e. the gain medium) and provide the sides of the resonator cavity for masing in the active layer 12. MASER wavelengths are orders of magnitude greater than the dimensions of a single emitter 10. The alternative approach would be to house the entire thin panel array in a physical resonant cavity. This would severely limit its practical application, and being a rigid physical structure, also limit the potential for tunability. In FIG. 4, reference number 24 refers to the microwaves being generated by the gain medium 12 and reflected by the layers of dielectric material 14, 16. Reference number 26 refers to the continuous emission of coherent microwave emission once the masing threshold is met.

[0072] A layer of Q-switching material 20 made of nematic molecules deposited onto the active layer 12. The Q-switching material 20 is capable of being CCD (charge coupled device) controlled and provides an interface for addressable control of the emission and entrainment of coherent emissions across a phased array of emitters 10. Accordingly, in accordance with another aspect of the invention, a phased array 28 (FIG. 5) of thin panel MASER emitters is controlled to output a single beam of mode-locked, coherent MASER radiation.

[0073] Still referring to FIG. 4, the layers of the emitter 10 are desirably formed on a substrate using lithographic and/or thin-film epitaxial techniques. The thickness of the layers in the exemplary embodiment is 50 nm120 nm and 220 nm.

[0074] Referring to FIGS. 5 and 6, use of a phased array 28 of thin panel MASER emitters 10a, 10b is capable of generating MASER emissions that are stable and have long length coherence. A representation of a portion of a multi-element phased array 28 is shown in FIGS. 5 and 6. A plurality of emitters 10a, 10b are fabricated on or embedded in a thin panel substrate 30. Reference number 10a designates emitters fabricated on the surface of the thin panel substrate 30, and reference number 10b designates emitters embedded in the thin panel substrate 30, with a vertical spacing offset. The offset can be selected, e.g., to be compatible with a desired resolution of the interferogram if the phased array 28 is used in an imaging system as described in co-pending application Ser. No. ______, entitled Phased-Array MASER Detector for Synthetic Aperture Interferometric Imaging, by James Joseph Cohen and Emad N. Eskandar, Attorney Dkt. No. 6756-00005. filed on even date herewith and incorporated by reference herein.

[0075] The assembly 28 is desirably manufactured using an epitaxial technology or similar means to generate an implantable, zero dangling-bond gain medium, that can be ion milled and implanted to achieve transitional vacancies at the desired Masing frequencies. Further, the stacked emitter 10a, 10b forms an addressable synthetic-aperture emitter, transparent to the optical pump except for the quantum lattice transition idealized by the ion implantation admixture.

[0076] In a typical MASER emitter or laser, the gain media is contiguous, and all the active moieties are essentially in one unit (such as a doped crystal) and potentially participate in the process of amplification. Coherent emission is achieved by stimulated emission through a population inversion. In essence, the gain media, composed of a great plurality of coupled re-radiative components are stimulated with a pump of energy. Thereafter, a preferred step of energy conversion is selected by stimulating the transition in a uniform manner. This causes a cascade of emission from the gain media in an energy signature and vector consistent with the overall resonation of the gain media. Other radiative transitions are minimized and occupy a fractional component of the energy conversion. The overall effect is that the light emission appears to be radiating from a single radiative element. The photon wave front is coherent and is synthesized from the overall emission topology.

[0077] A different approach is used to generate coherent emissions from the array 28 of discrete emitters 10a and 10b shown in FIGS. 5 and 6. As described above, Q-switching is a technique for disrupting transmission in a resonant cavity to allow for storage and emission of energetic pulses. In one embodiment of the invention, however, the goal is not to generate pulses but rather to drive continuous wave in-phase emissions from the individual emitters 10a, 10b. In this embodiment of the invention, the Q-switch layer 20 is mated to the active diamond layer 12, both of which are within the resonant cavity 14, 16 of each emitter 10. The Q-switch layer 20 is transparent to photons (in the visible range) pumped from the LED layer 18, but selectively scatters photons in the microwave range of interest. Depending on its state (low Q or high Q), the Q-switching layer 20 selectively interferes with microwave transmission, favoring coherent emission across the entire array 28, and dispersing the rest. It is this effect by which a quasi 2-dimensional emitter/gain media can produce a diffraction limited beam of consequence.

[0078] Provided that the wave propagating emitter 10 is stable and controllable, the Q-switch layer 20 is used to generate a coherent beam through the combined emissions of the individual emitters 10a, 10b. Before use, the thin panel array 28 of microwave emitters 10a, 10b is calibrated. Once the array 28 is powered and stable, the Q-switch layer 20 of each element 10a, 10b is individually flipped from low-Q (non-permissive) to high-Q (permissive) to determine the time needed to reach the masing threshold. This is a stable quantity reflecting features peculiar of each element. Coherent emissions always start at the lowest point of the waveform.

[0079] Q-switching is achieved by putting a variable attenuator inside the microwave resonator cavity. When the attenuator is functioning, microwave radiation leaving the gain medium does not return, and masing cannot begin. This attenuation inside the cavity corresponds to a decrease in the Q factor or quality factor of the maser resonator. A high Q factor corresponds to low resonator losses per roundtrip, and vice versa. Initially, the gain medium is pumped with LED light energy (530 nm) while the Q-switch is set to prevent feedback of microwave radiation into the gain medium (producing a microwave resonator with low Q). This produces a population inversion, but MASER emission cannot yet occur since there is no feedback from the resonator. Since the rate of stimulated emission is dependent on the amount of light entering the gain medium, the amount of energy stored in the gain medium increases as the gain medium is pumped. Due to losses from spontaneous emission and other processes, after a certain time the stored energy will reach a maximum level; the medium is said to be gain saturated. At this point, the Q-switch device is quickly changed from low to high Q, allowing feedback and the process of microwave amplification by stimulated emission begins.

[0080] Once the timing of all the emitters 10a, 10b is ascertained, the Q-Switch layer 20 of each element 10 is programmed with a small delay specific to that element. The slowest element has zero delay while faster elements have proportionately longer delays. Subsequently, coherent emission is initiated by first turning all the switches to low-Q (non-permissive). At the desired time, the individual emitters are flipped to high-Q (no disruption) with the programmed delays so that the faster elements begin emission at the same time as the slowest element. In this fashion, coherent emission from all the arrays starts at the same time. Since MASER emission always starts at the lowest point of the waveform, they are in-phase and the resultant beam, or wavefront, is coherent and mode-locked. Once masing begins, the Q-switches can be controlled to adjust phase and frequency of the beam, and can be controlled to include two or more characteristic frequencies, e.g. a coupling component and a modulation component of the beam.

[0081] ExampleNitrogen-Implanted, Epitaxial Diamond MASER. This example describes experiments showing that MASER emission is possible from a singular isotropic MASER emitter having an epitaxial layer as a gain medium. In this example, a 0.6 film of single crystal, epitaxial diamond was formed atop a bare Luxeon Rebel LED operating at 530 nm. Reported operational parameters of the LED are listed in Table 2:

TABLE-US-00002 TABLE 2 LED Color 530 nm Lumens @ 350 mA 102 lm Lumens @ 700 mA 161 lm Efficacy @ 350 mA 100 Lm/W Efficacy @ 700 mA 68 Lm/W Typical Wavelength (1) 530 nm Wavelength Range (1) 520 to 540 nm Beam Angle (2) 125 Rec. Operating Current 700 mA Max. Rated Drive Current 1000 mA Typical Forward Voltage 2.9 Vf Max. Forward Voltage 3.51 Vf Thermal Resistance 16 C. /W Max Rec. Junction Temp 150 C. Operating Temperature 40 to 135 C. Overall Dimensions 20 20 5 mm Active LED .2 mm .2 mm 60

[0082] The LED was supplied on a ceramic substrate and included conductive landing pads. The LED itself was supplied bare without an optical dielectric. The active LED had dimensions of about 0.20.20.060 mm (along with a thermal portion having a thickness of about 0.5 mm). The LED was prepared for chemical vapor deposition (CVD) of a Spin Hall spin-torque oscillator (STO) with microwave stimulated molecular beam epitaxy of a layer of cadmium telluride. The cadmium telluride is transparent at 530 nm and provided a smooth surface without defects over the area of major activity. Metrologic characterization of the CVD STO was done with an optical gonioreflectometer incremented in 200 nm steps.

[0083] A diamond epitaxial layer was formed by microwave stimulated molecular beam epitaxy through NdYag ablation of a ricochet carbon target set at 90 degrees and about 1000 microns from the CVD STO layer. Ultra-high vacuum was maintained by an ion pump. In this manner, a diamond film of 635 nm was developed over a deposition time of about 175 seconds. Deposition was halted due to thermal limits of the substrate as measured by a solid-state temperature probe beneath the LED.

[0084] The diamond film was then ion implanted with free disassociated nitrogen at an energy of 64 KEV with a fluence of 1017 ions/sec/cm2. Implantation time was approximately 3 seconds. It was estimated that implantation efficiency was about 3%. A measurable and uniform color change was recorded.

[0085] Atop the diamond layer, a copper microstrip resonator was formed by lithographed CVD Cu. The microstrip geometry was controlled to achieve optimal resonance at 36.8 Ghz. This corresponds to the third resonant octave of the expected hyperfine transition sought in the Nitrogen center formed in the diamond layer.

[0086] The hetero-structure was powered by two lithium ion battery cells (3.4 V output) connected in parallel with a simple resistive load inserted to ensure a stable supply of voltage. The battery leads were soldered to the conductive landing pads on the ceramic substrate. The LED was observed to emit brightly upon powering on. The operating temperature of the system was between 29 C. and 43 C. after about an hour of operation.

[0087] MASER experimentation was temperature controlled by means of a large (30 gm) copper heatsink (i.e. thermal bonded to the ceramic substrate). Operation of the system was at 34 C. within 0.2 C. during the course of the testing. The testing period was about 10 minutes. The experiment was replicated several times.

[0088] MASER emission was recorded by a Gunn diode detector in a small tuned resonator mounted on a micro stage. The detector was incrementally moved by manual rotation of the thimble in increments of 2 microns. The resultant waveform resolved to be 3.7471 mm with a typical sinusoidal power decrement. Auxiliary noise was also present, but the main line predominated by at least one order of magnitude. The experiments showed that coherent single frequency MASER activity can be produced using nitrogen implanted, epitaxial crystal diamond.

[0089] In the exemplary MASER emitters depicted in FIGS. 4 through 6, the epitaxial crystal diamond should be made thicker than the 635 nm thickness in the experiments. Increased thickness renders the layer mechanically and thermally robust for implantation and use, and also increases efficiency. As discussed above, the thin film crystal diamond should be formed via epitaxial growth.

[0090] A more detailed description of the fabrication and use of epitaxial thin diamond film, secondarily implanted with nitrogen ions, as gain media for generating coherent microwave emissions, in the experiments is now described.

[0091] The only previously described method for generating continuous wave microwave emissions at room temperature relies on a bulk diamond crystal, doped with nitrogen. The use of such doped diamonds, crystals grown with nitrogen as an impurity, is impractical for commercial fabrication into flat panel arrays because it is difficult to control the density and uniformity of nitrogen doping to a sufficient degree. Moreover, the self-annealing aspects of bulk diamond crystals generally produces a plentitude of dangling bonds or rotated crystal nucleation centers. Such deformations result in an inelastic coupling sapping the efficiency of the elevation of electrons into a metastable transition inversion. Such an inversion is necessary for coherent emission ie. Masing. It is therefore understood that unannealed epitaxial material will elastically accommodate Nv.sup.1 vacancies and preferentially permit energy to be stored in a metastable state creating a significant population inversion.

[0092] In contrast to bulk material used in the prior art, the invention as mentioned is based on epitaxial thin diamond films secondarily implanted with nitrogen Ions. This technique can be recapitulated for large scale manufacturing by adapting existing processes, and generates gain media with highly favorable features, along with being of sufficient quality and uniformity, to be used in commercial applications.

[0093] Epitaxial Fabrication of Stress-Tolerant Thin Diamond Film. Diamond, along with its carbon-cousins graphite and graphene, is the best thermal conductor around room temperature, having thermal conductivity of more than 2,000 watts per meter per Kelvin, which is five times higher than the best metals such as copper.

[0094] The absorption coefficient of the color centers in the nitrogen implanted diamond filled at the pump wavelength is relatively high. Depending on the atomic percent in the implanted regime, absorption will range between 30 and 90 percent of the incident fluence. However, the conversion efficiency is low, averaging about 3%. The balance of this energy is converted through kinetic sinks into intense heat islands. Spreading the generated heat uniformly and removing it from the system is a vital process to ensure stable operation. Bulk material is best for thermally sinking active material under high stress. Yet, for the purposes described herein, the controlled gradient uniformity is critical as is the defect free lattice. These characteristics are not available in bulk material. The implantation depth in the hetero-epitaxial diamond film will hover about 2 microns. A thermal sink of 5 mass units is a nominal design goal. Further, implantation of nitrogen into otherwise single crystal zero dangling bond films sets up great stress. Unlike other materials, diamond is not readily annealable. Thin films close to the implantation depth will yield under the stress without creating dangling bonds. It is understood and accounted for that notwithstanding the toughness of diamond significant stress, both structurally and thermally induced, will result in instabilities and may cause potential catastrophic failure in the lattice structure.

[0095] Basic understanding of the fundamental processes in crystal growth as well as the structural quality of diamond synthesized by epitaxial deposition on iridium surfaces has reached a high level for samples with (001) orientation. Diamond deposition on the alternative (111) surface, as used in the preferred embodiment of the invention, is generally more challenging. Heteroepitaxy of diamond on Ir/YSZ/Si (111) with different off-axis angles and directions is the desired target. During the growth of the first microns, strong and complex intrinsic stress states were rapidly formed. They restricted the range of suitable temperatures in this study to values between 830 C. and 970 C. At low-stress conditions, the maximum growth rates were about 1 /h. They facilitated long-time processes which yielded pronounced structural improvements with minimum values of 0.08 for the azimuthal mosaic spread, 4107 cm2 for the dislocation density and 1.8 cm1 for the Raman line width. This refinement is even faster than on (001) growth surfaces. It indicates substantial differences between the two crystal directions in terms of merging of mosaic blocks and annihilation of dislocations. Crystals with a thickness of up to 330 were grown but were not used for testing. The correlation of photoluminescence and -Raman tomograms with topography data also revealed fundamental differences in the off-axis growth between (001) and (111) orientation. The analysis of the microscopic structures at the growth surface provided the basis for explaining the intriguing reversal of stress tensor anisotropy caused by a simple inversion in sign of the off-axis angle.

[0096] Optimization of Nitrogen Doping for Coherent Emission. Nitrogen doping is used to achieve field emission from diamond, and hence it is important to be able to dope diamond with nitrogen in a controlled way, such as that offered by ion implantation. The procedure developed for optimizing p-type doping of diamond by B ion-implantation (cold implantation and in situ rapid annealing (CIRA) followed by high-temperature annealing) has been implemented here for N implantations. However, the preferred conditions would avoid the consequential annealing phase. Thereby building in stronger stressed structures and a refined energy coupling by Rabi processes. In the experiments performed, such annealing could not be avoided. Care was taken to minimize the annealing components. The implanted and annealed layer has been studied by EPR and by cathodoluminescence and photoluminescence measurements. These indicate that a high fraction of N can be incorporated into substitutional sites by the disclosed implantation procedure and that the so-treated N-implanted diamond much resembles N-containing type Ib diamond. In this manner, we achieved a gradient implantation of N into 111 oriented diamond film.

[0097] Local formation of nitrogen-vacancy centers in diamond by swift heavy ions. After nitrogen-implanting diamond films it was decided to expose the films to beams of swift heavy ions (1 GeV, 4 MeV/u) and find that these irradiations lead directly to the formation of nitrogen vacancy (NV) centers, without thermal annealing. Analysis required comparing the photoluminescence intensities of swift heavy ion activated NV.sup. centers to those formed by irradiation with low-energy electrons and by thermal annealing. NV.sup. yields from irradiations with swift heavy ions are 0.1 of yields from low energy electrons and 0.02 of yields from thermal annealing. Possible mechanisms of NV center formation by swift heavy ions such as electronic excitations and thermal spikes are expected. While forming NV centers with low efficiency, swift heavy ions could enable the formation of three-dimensional NV.sup. assemblies over relatively large distances of tens of micrometers. Results show that NV center formation is a local probe of (partial) lattice damage relaxation induced by electronic excitations from swift heavy ions in diamond.

[0098] The following description is directed to the mechanism of formation of NV.sup. centers, why these centers are essential for the masing action and how SHI and low energy electron (e) irradiation lead to the production and activation of NV.sup. centers. The primary goal in this experiment was to raise the conversion efficiency for stable masing. In diamonds that were doped with nitrogen during the growth process NV centers can be formed by creating vacancies using irradiations with energetic photons, neutrons, electrons or ions followed by thermal annealing above 600C. It was observed that vacancies become mobile and substitutional nitrogen atoms (Ns or P1 centers) can capture single vacancies, forming NV centers. In pure diamonds, nitrogen can be added by implantation of nitrogen ions. Ion implantation offers control over the local nitrogen and NV center concentrations with high spatial resolution. When implanting a diamond with nitrogen ions, e. g. of keV to MeV energies, ions transfer kinetic energy to carbon atoms in elastic collisions, leading to collision cascades that generate vacancies and carbon interstitials. Implanted nitrogen ions come to rest mostly on interstitial lattice sites, though replacement collisions are also possible. The formation of NV centers in nitrogen implanted diamond has been assumed to follow a two-step process. During annealing above 600 C., implanted nitrogen atoms can first be incorporated on substitutional lattice sites, forming P1 centers. Vacancies can diffuse through the diamond lattice during thermal annealing and the second step in NV formation is the trapping of a vacancy at the site of a substitutional nitrogen atom. The charge state of the resulting NV centers is affected by the local nitrogen concentration, the defect environment and by surface conditions. Only NV.sup.0 and NV.sup. centers are observed in photoluminescence (PL) or cathodoluminescence (CL) measurements. Reported efficiencies for the conversion of implanted nitrogen atoms to NV centers depend on the nitrogen implant energy and processing conditions (including implantation with elevated target temperatures) and range from below 1% to about 25% for low energy nitrogen implants (<100 keV).

[0099] Hydrogen, present from growth or introduced into diamonds during annealing can interact with NV centers and quench their photoluminescence. Care was taken to avoid hydrogen contamination. However, the star-ion pump saturation threshold was reached quickly and the resultant pollution of the lattice by incorporated hydrogen impurities was unavoidable. The atomic percent was quite small on the order of 0.01 log atomic percent and was ignored as insubstantial.

[0100] Recent simulations of NV center formation in diamonds with high concentrations of substitutional nitrogen suggest that NV centers form during exposure to vacancy-producing irradiation and not during thermal annealing. The reason for this is the apparently higher likelihood for di-vacancy vs. NV formation during annealing.

[0101] NV centers also form in nitrogen-implanted, electronic grade diamond during exposure to low-energy electrons (1-30 keV) without any thermal annealing. This result showed that the commonly accepted two-step model of NV formation in electronic grade, nitrogen-implanted diamond is incomplete.

[0102] Electronic excitations induced by swift heavy ions (SHI) also lead to the local formation of NV centers without thermal annealing. SHI such as gold or uranium ions with kinetic energies of 1 GeV, near the Bragg peak in electronic energy loss, deposit kinetic energy mostly through inelastic collisions with target electrons at a rate of about 50 keV/nm in diamond. Energetic electrons quickly thermalize in cascades, forming many thousands of electron-hole pairs per SHI. Lattice heating through electron phonon coupling and from bond reorientation can lead to thermal spikes effects in the material surrounding the ion trajectory. Relative yields of NV.sup. centers formed by SHI were compared to exposure to low energy electrons and compared to thermal annealing and alternative possible mechanisms of NV.sup. formation by SHI were also noted. Insights into the interplay of electronic excitations and thermal annealing might lead to improved processing strategies for reliable formation of NV.sup. centers with high placement resolution, spectral stability, and favorable spin properties.

[0103] An experiment was performed to determine the result of nitrogen implantation and SHI and electron radiation. After cleaning in a mix of sulfuric, nitric and perchloric acid (1:1:1) two electronic grade diamond samples (<5 ppb nitrogen content, size of 4 mm.sup.21 thick were implanted with .sup.14N.sup.+ ions. The implant spots were 1 mm in diameter, and we aligned them so that they did not overlap. Each diamond was implanted with 10 and 30 keV nitrogen ions and for both implant energies two fluences, 10.sup.12 and 10.sup.13 N cm.sup.2 were applied for a total of four implant spots on each diamond. The implantation depths for these implant conditions were less than 130 nm, including channeling tails. Irradiations with SHI were performed at the GSI Helmholtz Center using 1.14-GeV U ions of fluence 510.sup.11 cm.sup.2 and 1.16-GeV Au ions (10.sup.11 cm.sup.2). The projected range of these ions in diamond is in excess of the full thickness of the film and extended about 30 into the substrate, far beyond the 100 nm thick nitrogen-implanted layer. The higher fluence, uranium-implanted sample showed a visible greenish discoloration presumably from vacancy related color centers, such as GR1 centers, formed by displacement damage preferentially at the end of the ion range. The focus of this work was on NV.sup. and did not track details of GR1 center formation and annealing.

[0104] During SHI irradiations the samples were masked with a honeycomb patterned metal grid with a thickness much larger than the range of the SHI. Analysis was concentrated on results from the uranium irradiated sample because signal levels from the sample irradiated with the lower fluence of gold ions were low.

[0105] In order to compare results from SHI irradiation to our previous results from low-energy electron bombardment, three different sample areas (pristine diamond, nitrogen implanted, nitrogen implanted and SHI irradiated) were exposed to a high fluence of 10 keV electrons in a scanning electron microscope (SEM). The applied electron fluence was 90 C/cm.sup.2 corresponding to saturation of the observed NV.sup. yield reported earlier for diamonds that had been implanted with the same fluences of nitrogen ions. A schematic of the implant conditions is illustrated in FIG. 7. The masked SHI irradiations resulted in sharp contrast in hyperspectral imaging between the neighboring irradiated and masked regions as seen in FIG. 8.

[0106] For optical characterization hyperspectral imaging at room temperature using a grating spectrometer Raman Microscope with 0.8 NA air objective and a laser of wavelength 532 nm was used. All different exposure combinations were analyzed and compared by integrating the peak area around the NV.sup. zero-phonon line (ZPL) before and after annealing at 850 C. FIG. 8 shows a hyperspectral image obtained by integration over the wavelength range of 635-642 nm around the ZPL of the NV.sup. centers. The large circular zone (lighter shade, indicated by the white line) corresponds to the spot of the nitrogen implantation. Regions covered with the honeycomb shaped metal mask during SHI irradiation appear as dark lines. The dark area (1) outside the nitrogen implant spot thus corresponds to pristine diamond that was neither implanted with nitrogen nor exposed to SHI. Area (2) outside the implantation spot was exposed only to 1.14 GeV uranium ions (510.sup.11 cm.sup.2) and shows a PL signal from single vacancies (GR1 center) caused by lattice displacements from the SHI impacts, but no NV.sup. signal. Region (3) was implanted with nitrogen as well as exposed to SHI and shows NV.sup. centers with a relatively low intensity. Areas that were implanted with nitrogen but then masked from SHI impacts (4) show NV.sup. only after irradiation with low-energy electrons. We also irradiated N-implanted and SHI-irradiated areas in micron scale spots with a 10-keV electron beam (5) to explore possible additive effects in NV formation from consecutive SHI and electron exposures.

[0107] Photoluminescence (PL) spectra for each of the six different areas in FIG. 8 before (FIG. 9) as well as after thermal annealing at 850 C. for 1 h (FIG. 10) was collected.

[0108] The intensity of NV.sup. ZPL signals from exposure to SHI was relatively low. The NV.sup. ZPL signal also overlaps with the tail of the GR1 center signal and this led to uncertainty in the assessment of the relative NV.sup. formation efficiency from SHI. For background correction in N-implanted regions that were masked from SHI exposures (4), we used data from pristine diamond areas (1), which were found to be spectrally equivalent to N-implanted areas prior to thermal annealing. With a given sensitivity and signal to noise, there was no observed NV.sup. centers in pristine diamond areas or nitrogen implanted areas that had not been exposed to either SHI or electrons prior to thermal annealing.

[0109] The passage of SHI through nitrogen implanted layers in diamond leads to the formation of NV.sup. centers (FIG. 9, N and SHI) without thermal annealing. The yield of NV.sup. centers formed directly by a fluence of 510.sup.11 U-ions/cm.sup.2 (1.14 GeV) is about 0.1 of yields from exposure of nitrogen implanted diamonds to 10 keV energy electrons (90 C/cm.sup.2) and about 0.02 of NV.sup. yields from thermal annealing. The highest NV.sup. intensities when nitrogen implanted areas were first exposed to SHI and then also to low energy electrons (FIG. 8, N, SHI and electrons). But this spectrum includes a contribution from the tail of the GR1 center line and after subtracting it from the NV.sup.ZPL area, the NV.sup. yields from electron exposures and electron plus SHI are about equal (FIG. 11).

[0110] After thermal annealing, the NV.sup. signal is higher in areas that had been irradiated with either SHI or keV electrons compared to areas that had been implanted with nitrogen but not irradiated further. Combining both irradiations and thermal annealing produces the highest NV.sup. yields (FIG. 8, N plus SHI plus electron, and FIG. 11), higher than thermal annealing alone.

[0111] To compare the PL intensities more quantitatively the SHI induced GR1-center contribution were subtracted from the NV.sup. ZPL area (635-642 nm) and normalized all measurements to PL signals from 10 keV N-implants (10.sup.13 cm.sup.2) followed by exposure to 10 keV electrons (>90 C/cm.sup.2). These data were extracted from spectra similar to those in FIG. 3. The N to NV.sup. conversion efficiency from electronic activation using low energy electrons is 25% of that from standard thermal annealing (850 C., 1 h). Thermal annealing itself has a formation efficiency for NV.sup. centers from 10 keV nitrogen implants of only a few percent. The formation of NV.sup. centers by SHI (U ions, 1.14 GeV, 510.sup.11 cm.sup.2) yields approximately 10% of the PL intensity of a 10 keV e-beam (90 C/cm.sup.2). The estimate for the absolute conversion efficiency from implanted nitrogen to NV.sup. centers, NV.sup./N, from the uranium ion irradiations is 510.sup.4 to 10.sup.3. In FIG. 11 relative NV.sup. yields before global annealing, from spectra as in FIGS. 9 and 10, are compared. The data in FIG. 11 was recorded before thermal annealing and were normalized to the reference implant (10.sup.13/cm.sup.2, 10 keV N.sup.+) that had also been irradiated with a saturation dose of low energy electrons. The top of the error bar indicates the yields from raw data before subtraction of contributions from GR1 centers to the NV.sup.ZPL. Within the uncertainty of background subtraction, exposing a nitrogen-implanted diamond first to 510.sup.11 U-ions/cm.sup.2 and then to low energy electrons does not yield significantly more NV.sup. centers compared to the exposure to low energy electrons alone for the sample implanted with a fluence of 10.sup.13 N/cm.sup.2. Trends are similar for the 10 keV and 30 keV nitrogen implants and for the sample implanted with 10.sup.12 N/cm.sup.2. However, the latter uncertainties are increased due to lower signal levels and background contributions from GR1 centers.

[0112] The formation of NV.sup. centers by means of low-energy electron exposure does not exhibit a threshold in electron beam energy and does not depend on the e-beam current density, indicating that the formation process is not dominated by momentum transfer or target heating but rather by electronic excitations induced by the electron beam. Deposition of kinetic energy of SHI along their trajectory in diamond is also dominated by electronic excitations with an electronic stopping power of 50 keV/nm and only a small contribution from elastic collision processes (<0.1 keV/nm and 0.25 vacancies) in the 130 nm deep nitrogen-implanted layer. Collisions of 1.14 GeV U-ions with target electrons can produce electrons with up to 10 keV of kinetic energy. Subsequent electron-electron collisions lead to a cascade with a distribution of low-energy electrons. The cascade electrons thermalize within a few picoseconds, followed by radiative and non-radiative recombination or charge trapping at defect sites. The kinetics and relaxation mechanisms of the electron cascade are similar to those occurring during irradiation with keV electrons, though details of the energy, depth, and lateral distributions differ greatly. The main difference is the high excitation density of SHI with electronic energy loss rates of tens of keV/nm that can induce local thermal spike effects including localized melting and rapid re-crystallization. Whether SHI form NV centers directly via electronic excitations similar to processes during irradiation with keV electrons, or if thermal spike effects play a role is an unknown. In thermal spikes, the high electronic energy loss of projectiles can induce local melting of materials along the ion trajectory.

[0113] In diamond, no permanent tracks have been observed, indicating that re-crystallization leads to at least partial lattice reconstruction following passage of SHI. The observation of NV.sup. centers does indicate partial lattice reconstruction of the nitrogen implanted layer in diamond following passage of SHI, while the presence of substitutional nitrogen atoms (P1 centers) is a measure of more complete damage repair. NV.sup. can be detected by optical microscopy while the analysis of P1 centers and other non-luminescent defect complexes can be probed e.g. via spin resonance, dark state spectroscopy or structural probes such as transmission electron microscopy.

[0114] In order to elucidate possible mechanisms of NV.sup. formation by SHI consider the qualitatively electronic excitation processes from SHI and exposure to low energy electrons. For low energy electrons we found an increase of the NV.sup. intensity with increasing electron exposure up to a saturation fluence of 10 C/cm.sup.2. If one considers the electronic stopping of 50 keV/nm for 1.14 GeV U-ions, a fluence of 510.sup.11cm.sup.2 and an energy for the formation of an electron-hole pair in diamond of 13 eV, then it can be estimated that the SHI irradiation generates of the order of 210.sup.17 electron-hole pairs/cm.sup.2 in the 100 nm wide nitrogen implanted layer. For comparison to earlier measurements of the fluence dependent NV.sup. formation from low energy electrons, the equivalent low electron fluence from SHI is then about 30 mC/cm.sup.2. For low energy electrons, this fluence would corresponded to a relative NV.sup. intensity of 500 from a 100 nm thick, nitrogen implanted layer, about 4 times lower than a relative intensity of 2000 for a saturation fluence (>10 C/cm.sup.2) of low energy electrons. The experiment demonstrates NV.sup. are formed by SHI with a relative intensity ten times lower compared to the NV.sup. center intensity from a high fluence of low energy electrons. Within this very rough estimate, SHI appear to be 2.5 less efficient in forming NV.sup. centers than electrons per keV of electronic energy loss in the nitrogen implanted layer.

[0115] Thermal spike effects could also play a role in NV.sup. formation by SHI. SHI can induce thermal spikes by their high density of electronic energy deposition. Along the projectile trajectory, a hot cylindrical zone is created with a heating and cooling time of a few ps. It is predicted that temperatures well above the (pressure dependent) stability limit of NV.sup. centers of about 1200 to 2000 C. can be reached within the cylinder core of 5 nm radius. This hot core is surrounded by progressively cooler regions. Within an area of about 200 nm.sup.2 the temperature can still be above 600 C. and induce vacancy mobility for a brief period. It seems plausible that SHI can stimulate NV.sup.formation in a region surrounding the ion trajectory that is heated, but not to a temperature above the NV stability limit. Here, NV.sup. centers can form by the two-step process of nitrogen incorporation into the lattice followed by capture of a mobile vacancy, and by a combination of electronic excitation and thermal activation for reconstruction of defect complexes that contain interstitial nitrogen and vacancies. At the given fluence for U ions of 510.sup.11 cm.sup.2 and an estimated heated area per ion of 200 nm.sup.2, all nitrogen ions (implanted with a with fluence of 10.sup.13 cm.sup.2) experience both a local thermal transient together with electronic excitations.

[0116] Following SHI irradiation and keV-electron exposures, a standard thermal annealing process was applied. After annealing, the NV.sup. yields are higher in areas that were exposed to electrons or first exposed to SHI and then to electrons, compared to thermal annealing alone. SHI irradiations form NV's directly, but with relatively low efficiency. Apparently, SHI irradiation also modifies the structure of the nitrogen implanted layer such that consecutive exposure to keV electrons produces slightly more NV centers than electron exposure alone, without SHI pre-treatment. For the 10 keV nitrogen implant with fluence of 10.sup.13 cm.sup.2, it is found that the combination of SHI, followed by electron beam and finally thermal annealing produces the highest NV.sup. yields, more than thermal annealing alone, and more than electron exposure plus thermal annealing. For the 30 keV nitrogen implant, NV.sup. formation by 10 keV electrons is less efficient, probably because electrons lose more energy before reaching the peak of the implanted nitrogen distribution. Trends are similar for nitrogen implants with 10.sup.12 cm.sup.2 but signal levels were lower.

[0117] Both low energy electrons and SHI can also affect defect complexes of NV centers with hydrogen. No characterization of the hydrogen content of the diamonds used here. This is likely a small effect present where hydrogen related NV.sup. PL quenching effects were observed following specific hydrogen plasma treatments.

[0118] Simple steps enable the formation of three-dimensional NV.sup.-N assemblies by means of SHI irradiation. First, electronic grade diamond is implanted with a selected depth profile of nitrogen ions, e. g. using multiple implantation energies. In a second step, the diamond is irradiated with SHI and NV.sup. centers are formed along the ion trajectories. Assemblies of NV.sup. can be tailored with a distribution of NV.sup. spacings resulting from the local nitrogen concentrations and N to NV.sup. conversion efficiencies.

[0119] Local NV.sup. formation without global thermal annealing also enables a try-and-repeat approach to NV.sup. array formation. After a single nitrogen ion is implanted and registered in a selected location, an NV.sup. center can be formed by low energy electron irradiation, by targeted SHI irradiation or by another form of local excitation. The location can be repeatedly probed for the presence of an NV.sup. center by PL or CL. Once an NV.sup. is detected, the process is repeated at the next location or more nitrogen and vacancies can be introduced by local ion implantation, followed by local excitation until an NV-center is observed. In this way, arrays of single NV centers can be formed. Coherence times of locally formed NV.sup. have to be quantified to evaluate their potential uses.

[0120] NV.sup. yields from exposure of nitrogen implanted diamonds to SHI and keV electrons followed by a standard thermal annealing step are 1.7 higher than yields from thermal annealing alone. But absolute conversion yields, NV.sup./N, are still low, of order a few percent. Results show that the simple model of NV.sup. formation in nitrogen implanted diamond in a two-step process of P1 center formation followed by vacancy trapping at substitutional nitrogen sites is incomplete. SHI can form NV.sup. centers along tracks like pearls on a string over distances of tens of m's. Local NV.sup. formation by SHI or low-energy electrons enables try-and-repeat approaches for the deterministic formation of NV center arrays in which nitrogen is first implanted and then locally activated. Further, findings support the confirmation of predictions that NV formation by the passage of SHI in pre-implanted and pre-damaged diamond shows that NV.sup. centers are probes of (partial) lattice damage recovery. Tracking of NV.sup. centers and other color centers may be useful in studies of lattice damage and recovery dynamics in the interplay of elastic and inelastic energy loss process during irradiations.

[0121] NV.sup. lattice defects for use in continuous wave MASER films is achievable in unannealed systems provided thermal runaway is controlled and preferential annealing is prevented. The conversion of NV to NV.sup. lattice vacancies is a critical threshold which must be observed. Localized bond shifts will occur at energies far below bulk annealing temperatures. Such shifts will prevent metastable inversion and broaden the hyperfine transitions. In such a case, masing efficiency will be greatly reduced and may prevent a population inversion.

[0122] In the foregoing description, certain terms have been used for brevity, clarity, and understanding. No unnecessary limitations are to be inferred therefrom beyond the requirement of the prior art because such terms are used for descriptive purposes and are intended to be broadly construed. The different configurations, systems, and method steps described herein may be used alone or in combination with other configurations, systems and method steps. It is to be expected that various equivalents, alternatives and modifications are possible within the scope of the appended claims. Each limitation in the appended claims is intended to invoke interpretation under 35 U.S.C. 112, sixth paragraph, only if the terms means for or step for are explicitly recited in the respective limitation.