The magnetic geometry of the Prototype Material Plasma Exposure eXperiment (Proto-MPEX) was recently modified to enable more effective utilization of 28 GHz microwave auxiliary power, specifically: (1) to heat plasma electrons in the radial core of the device and (2) to deliver the heated plasma to the target plate of the device. To achieve this goal, the microwave launcher geometry and placement were significantly re-engineered, guided by previous experimental results and computational modeling. The core electron temperature in the launcher region is observed to increase from 3 eV to 11 eV with 30 kW of auxiliary power after the improvements, and an increase from 3 eV to 6 eV is concurrently measured in the target region (∼1 m from the launcher) at electron density above O-mode cutoff. Radially resolved measurements in the launcher region exhibit a strong dependence on the magnetic geometry. The results of a magnetic field scan reinforce the effectiveness of the intended O-X-B mode conversion scenario that is currently planned for microwave heating of the Material Plasma Exposure eXperiment (MPEX).

Previous results from the Prototype Material Plasma Exposure eXperiment (Proto-MPEX) have demonstrated the efficacy of using 28 GHz microwaves to heat over-dense plasma.1 The electron temperature was observed to increase off-axis in the launcher region by a factor of 4, from ∼5 eV to ∼20 eV, when ∼20 kW of 28 GHz auxiliary microwave heating was applied to deuterium plasma sourced by a ∼100 kW rf helicon antenna operating at 13.56 MHz. The radial location of the electron heating was consistent with predictions from modeling, which indicated that launched O-mode microwaves first convert to X-mode microwaves and then subsequently to electron Bernstein waves (EBW), so-called O-X-B mode conversion in over-dense plasmas.2 While the observed electron heating was substantial in the launcher region of those experiments, the electron heating measured at the target plate 1.5 m away was more modest (from ∼2 to ∼3 eV). A significant “magnetic hill” existed in those experiments between the launcher and the target, influencing the kinetic transport of energetic plasma particles to the target. (This is a hypothesis supported by preliminary, on-going Monte Carlo modeling that will be discussed in a subsequent publication.) The mission of Proto-MPEX is to verify the plasma source concept (including auxiliary heating) for the MPEX device; hence, the ability to deliver heat and particle flux onto target is a key metric.3 Both in the launch region and at the target, the predominant electron temperature rise was radially off-axis and asymmetric, consistent with the physical geometry of the launcher. The understanding afforded by those experiments motivated a significant re-engineering of the Proto-MPEX vacuum and magnetic geometry “downstream” from the helicon plasma source, including the microwave launcher region.

This paper will briefly describe the modifications of the Proto-MPEX device,4 which enabled the re-engineered microwave launcher to be placed in a magnetic geometry that was more favorable for delivering auxiliary heated plasma to the target region. Results from GENRAY-C modeling will be shown, which detail the O-X-B mode conversion process that is consistent with observed 2D heat flux patterns on the target plate for the new microwave launcher geometry. Radial profiles from double Langmuir probes (DLP) will be shown, demonstrating the effect on the plasma electron density and temperatures at both the launcher and the target. Lastly, measurements of core electron temperature at the target during a scan of the magnetic field in the launcher region will be presented to investigate the O-X-B mode conversion process and identify core microwave heating scenarios that are applicable to MPEX.

The linear machine geometry of Proto-MPEX is fairly flexible, including modular magnetic field coils which can be used to achieve on-axis fields approaching 2 T. As such, a variety of magnetic configurations have been explored.4–10 During the 3 month period spanning June to August of 2018, the portion of Proto-MPEX downstream of the helicon source region was reconfigured, as shown in Fig. 1. The helicon plasma source remains unchanged, allowing the auxiliary heating research program to progress without re-addressing the established helicon source physics, including specifically the differential pressure requirements for high-density, light ion plasma production.10–12 The diagnostic suite is similar although the number of diagnostic ports at the target region has been increased from 4 to 16.

FIG. 1.

Proto-MPEX cross-sectional view with key features labeled. The colored overlay sections from coils 7–13 were re-engineered for the experiments reported here. Magnetic field value along the axis for a variety of magnetic field configurations (coil 7 and 8 B-field scan reported in this paper). The locations of double Langmuir probe profiles are indicated, along with the plane of focus for the 28 GHz launcher at z ∼ 3.16 m.

FIG. 1.

Proto-MPEX cross-sectional view with key features labeled. The colored overlay sections from coils 7–13 were re-engineered for the experiments reported here. Magnetic field value along the axis for a variety of magnetic field configurations (coil 7 and 8 B-field scan reported in this paper). The locations of double Langmuir probe profiles are indicated, along with the plane of focus for the 28 GHz launcher at z ∼ 3.16 m.

Close modal

A key feature of this device reconfigure is that the microwave launch location has been moved from the “central chamber” (between coils 6 and 7) to the new “launch region” (between coils 8 and 9). The launcher itself (described below) has been significantly re-engineered compared to what was described in Ref. 1. In general, the physical spacing between magnetic coils 7–12 has been reduced to enable higher on-axis magnetic field values for equivalent coil currents and to reduce potential energetic particle trapping due to magnetic field ripple between the launch and the target regions. An additional coil, #13, was added behind the target, with the effect that the overall device length has not meaningfully changed. The magnet power supplies were configured such that coils 9–13 are in series and can be used to establish a high (∼1 T) magnetic field under the (now external to vacuum) ion cyclotron resonant heating (ICRH) antenna. For these experiments, the auxiliary ICRH system was not used. Coils 7 and 8 are in series on an independent power supply, allowing the magnetic field in the electron cyclotron resonant heating (ECRH) launch region to be increased up to 2 T. For the 28 GHz gyrotron used in these experiments, the fundamental ECH resonance occurs at 1 T, and the second harmonic ECH resonance occurs at 0.5 T. Figure 1(b) shows a variety of axial magnetic field profiles for a range of currents in coils 7 and 8. The value of the 1 T fundamental and 0.5 T 2nd harmonic fields is indicated. In this way, the O-X-B mode conversion physics for the launched 28 GHz microwaves can be investigated, as will be reported below.

Since the coil spacing in the launch region is reduced from previous experiments,1 the microwave launcher has been substantially redesigned. The previous design had the microwaves injected vertically from above in O-mode polarization. The results published in Ref. 1 had a single, large bore launcher. A subsequent design with defocusing subreflector and focusing reflector was also employed in an effort to match the microwave beam profile more closely to the plasma diameter and to make the wave fronts conformal with the plasma surface. Similar results were observed in each case: the over-dense plasma electrons in the launch region were observed to be heated off-axis (vertically above) up to a factor of 4, consistent with the launcher geometry. However, the magnetic field at the launcher in those experiments was relatively low compared to the neighboring regions. This “magnetic well” geometry tends to kinetically trap particles with high perpendicular energy, based on simple arguments related to conservation of magnetic moment13 and deflection time.14 Experimentally, the heat flux to the target was not substantially increased with the application of microwave power.

In the experiments described in this paper, the large bore launcher was again employed, with a horizontal launch as shown in Fig. 2. The launched microwaves impinge on a shaped mirror which directs and focuses them, so that the beam diameter roughly matches the plasma diameter in the launch region in O-mode polarization. It is recognized that for oblique launch in the Proto-MPEX geometry the wave polarization will consist of mixed O- and X-mode components. Due to the diameter of the launch beam relative to the plasma and the local curvature of the Proto-MPEX magnet field, the wave polarization will consist of primarily O- and some X-mode components. Unfortunately, no measurements of the actual polarization fractions are available in Proto-MPEX. A rough estimate, supported by the launcher modeling, suggests that ∼75% of the wave remains O-mode polarized at first-pass interaction with the plasma column. When facing the target plate, the microwaves strike the plasma column at roughly the 7 to 8 o'clock position, which is consistent with the observed heat flux pattern on the target (discussed in Sec. IV), and notably different from previous experiments where the microwaves were injected from the 12 o'clock position. Importantly, the magnetic geometry of this launch location [see Fig. 1(b)] is now much more kinetically favorable for energetic particles to reach the target from the launch location. The unabsorbed microwave power is measured at several axial locations in Proto-MPEX using microwave-sensitive diodes attached to a wave scattering cavity, also commonly called “sniffer probes.” However, this effort was time constrained, and the diodes were uncalibrated, providing only qualitative measurements. Experiments with and without the presence of a back-reflector were not completed. Moreover, in these experiments, the launcher angle was not changed, and only the optimal angle predicted by modeling was explored.

FIG. 2.

Launch region configuration for the 28 GHz power launcher. (a) Schematic of the vacuum spool piece between coils 8 and 9, including the horizontal, large bore launcher and shaped mirror. (b) COMSOL calculation of the focused microwave beam spot size on the plasma column. (c) COMSOL calculation of the ray trajectories of the launched O-mode polarized microwaves as they are reflected by the shaped mirror and strike the plasma column “from below.”

FIG. 2.

Launch region configuration for the 28 GHz power launcher. (a) Schematic of the vacuum spool piece between coils 8 and 9, including the horizontal, large bore launcher and shaped mirror. (b) COMSOL calculation of the focused microwave beam spot size on the plasma column. (c) COMSOL calculation of the ray trajectories of the launched O-mode polarized microwaves as they are reflected by the shaped mirror and strike the plasma column “from below.”

Close modal

Microwave heating schemes rely on a variety of wave-physics effects to transfer energy from the launched electromagnetic wave to plasma electrons (Refs. 15, 16, and references therein). One such scheme is O-X-B mode conversion, which aims to mode-convert the original microwave energy into an electrostatic Bernstein wave propagating in the plasma medium, whose energy is absorbed by the plasma. Resonant and nonresonant heating of plasmas in toroidal geometries by O-X-B mode conversion has been shown to efficiently raise the electron temperature to hundreds of electron volts in a stellarator.17 The detailed wave physics associated with O-X-B in Proto-MPEX is given in Ref. 2 and references therein. The magnetic field strength in Proto-MPEX has azimuthal symmetry, with weak radial dependence due to the large bore (relative to the plasma diameter) magnetic coils, but strong axial variation as shown in Fig. 1(b). Measurements of the plasma electron density in Proto-MPEX consistently show strong radial peaking.5 Hence, the mode conversion physics must contend with 3D geometry effects even in a nominally “linear” device. The GENRAY-C code has been used to model the 3D cylindrical geometry of Proto-MPEX to examine O-X-B mode conversion. (GENRAY-C is a cylindrical version of the extensively used GENRAY code.18) An example result is shown in Fig. 3. An O-mode ray is launched, mode-converts to X-mode at the O-mode cutoff, and then refracts and mode-converts to an EBW at the upper-hybrid (UH) resonance layer. The power in the ray is approximately lossless at the O-mode cutoff, based on an analytical calculation of O-X mode conversion transmission efficiency.2 For an optimized oblique launch angle (∼30° between the launch angle and the magnetic field, as utilized in both the experiment and in the modeling), very strong single-pass absorption of the Bernstein wave is observed. In the absence of collisions, the wave energy is absorbed at Doppler shifted cyclotron resonances (associated with plasma flowing in regions of strong axial B-field variation), which can be located near the radial core of the plasma column. Collisional damping on neutrals or ions can result in absorption of the wave energy radially off-axis. In Sec. IV, experimental results will be shown that are consistent with expectations from this modeling. Examination of the 2D infrared heat flux images suggests strong single-pass radial absorption, following X-B conversion at the UH layer, propagating inward. These images are consistent with collisional damping of the ray energy predicted by modeling in the presence of neutrals. The (wall measured) ambient neutral pressure is below the threshold identified for strong collisional damping,1,2 but may nevertheless be playing a role. Conversely, as will be shown in Fig. 6, the power absorption on-axis peaks as the magnetic field is swept through the 2nd harmonic resonance, suggesting that resonant absorption is the dominant process; however, it is conceivable that more than one power absorption mechanism (collisional damping at the edge and resonant absorption in the center) is present.

FIG. 3.

GENRAY-C calculations of the wave absorption of the 30° oblique launched microwaves in Proto-MPEX via O-X-B mode conversion and collisional damping.

FIG. 3.

GENRAY-C calculations of the wave absorption of the 30° oblique launched microwaves in Proto-MPEX via O-X-B mode conversion and collisional damping.

Close modal

Once the machine was reassembled, Proto-MPEX embarked on an experimental campaign (from September to November 2018) to investigate O-X-B mode conversion and heating of plasmas using 28 GHz microwaves. Some of the results are reported here. The two primary diagnostics in this paper used to determine the effectiveness of microwave coupling are an infrared (IR) camera viewing the target plate and double Langmuir probes (DLPs) at various locations. An infrared camera monitors at a 100 Hz frame rate the plasma-facing surface of the Proto-MPEX target plate via a periscope.19,20 DLPs are used to make point measurements at 200 Hz sweep rate of the plasma electron density and temperature, when inserted into the plasma column at a given radius and axial location.21 Profiles are achieved by shot-to-shot radial movement of the DLP. Figure 4 shows the IR camera measured heat flux on the target plate for a ∼100 kW rf helicon-only plasma discharge and for a similar discharge with ∼30 kW of auxiliary 28 GHz microwaves. The field of view of the IR camera through the periscope does not include the entire plasma footprint on the target plate, but it has been aligned to observe the lower region. The heat flux on the target during microwave application rises significantly off-axis in the region expected for the absorbed microwave beam. The power arriving on the target plate indicates that there is strong single-pass absorption of the microwaves, consistent with the GENRAY-C modeling in the presence of collisions. The heat flux on the target plate rises strongly off-axis from ∼0.5 MW/m2 to ∼4 MW/m2. The greatest heat flux rise occurs radially between the upper-hybrid layer (mapped to target from DLP measurements of the electron density profile in the launch region) and the left-hand X-mode cutoff layer. The location of the X-mode cutoff layer should not be a factor in the O-X-B mode conversion scenario since the radially inward propagating Bernstein waves do not experience a density cutoff. But the layer is shown for reference, especially since significant IR measured heat flux is located radially inside this layer. The heat flux on the target plate from the core of the plasma column increases from ∼0.4 MW/m2 to ∼1 MW/m2 during 28 GHz microwave application. It is worth restating that the heat flux on target in Proto-MPEX (e.g., as reported here) is a key metric for assessing the plasma source performance expected for the MPEX device.

FIG. 4.

IR camera measured heat flux on the plasma facing side of the Proto-MPEX target, shown during a plasma discharge produced by 13.56 MHz rf helicon waves and compared to a discharge where auxiliary 28 GHz microwave power is added. The heat flux rises strongly in the azimuthal region consistent with the direction of the launched microwaves, and in the radial region consistent with O-X-B mode conversion. The path of the Launcher and Target DLPs is indicated, for the profile measurements shown in Fig. 5.

FIG. 4.

IR camera measured heat flux on the plasma facing side of the Proto-MPEX target, shown during a plasma discharge produced by 13.56 MHz rf helicon waves and compared to a discharge where auxiliary 28 GHz microwave power is added. The heat flux rises strongly in the azimuthal region consistent with the direction of the launched microwaves, and in the radial region consistent with O-X-B mode conversion. The path of the Launcher and Target DLPs is indicated, for the profile measurements shown in Fig. 5.

Close modal

The heat flux on target depends on a number of plasma parameters, including the electron temperature and density, the ion temperature and density, magnetic boundary sheath conditions, etc.20 Moreover, heat flux is not sufficient to illustrate the O-X-B mode conversion plasma heating process that is discussed by this paper. To examine this point, radial profile measurements with DLPs were made of the plasma electron temperature and density in both the launch region and near the target. Figure 5 shows the example profiles of the effect of adding auxiliary 28 GHz microwave power to rf helicon Proto-MPEX plasmas. The DLP in the launch region has a horizontal range of motion and, hence, does not sample the hottest part of the discharge indicated in the IR heat flux profile. For the discharge shown with launched 28 GHz microwaves, the O-mode cutoff density occurs at 0.97 × 1019 m−3, and the discharge is clearly over-dense to the propagation of O-mode microwaves. For the oblique launch of Proto-MPEX, GENRAY-C calculations show that the inward moving O-mode waves convert to radially outward propagating X-mode polarized waves. When the X-mode waves encounter the upper-hybrid (UH) resonance layer for 28 GHz, they are further mode converted into inwardly propagating Bernstein waves, which then experience strong single-pass absorption. This process is observed to modify the measured electron density profile (understanding the modification of the density profile is a work in progress and outside the scope of this paper) as shown, but it is important to note that the discharge remains over-dense to O-mode propagation, so that the circumstances for further mode conversion (in time) continue throughout the applied microwave pulse (∼50 ms in duration). Moreover, examination of the electron temperature profiles clearly shows an overall doubling (or more) of the electron temperature during the period of microwave application in both the launcher and subsequently the target region. Off-axis, the concurrent rise of electron temperature and density results in the strong increase in edge heat flux on target, which is consistent with IR camera measurements. On-axis, the electron temperature also rises, but there is a concurrent drop in electron density, with the net effect being that a more modest (but important) rise in core heat flux occurs during microwave application. Inside the radius of the left-hand X-mode cutoff, the only waves which can be carrying the energy necessary to affect this rise are electrostatic Bernstein waves.

FIG. 5.

DLP measured plasma electron density and temperature profiles in the launcher region and the target region of Proto-MPEX, during helicon-only discharge and during the application of auxiliary 28 GHz power. The electron density value associated with the O-mode cutoff of the launched 28 GHz microwaves is indicated.

FIG. 5.

DLP measured plasma electron density and temperature profiles in the launcher region and the target region of Proto-MPEX, during helicon-only discharge and during the application of auxiliary 28 GHz power. The electron density value associated with the O-mode cutoff of the launched 28 GHz microwaves is indicated.

Close modal

This experiment was repeated as the magnetic field at the launcher was scanned over a wide range (as shown in Fig. 1), by varying the current in coils 7 and 8 while keeping all other source parameters (helicon power, 28 GHz power, gas fueling, etc.) nominally constant. In particular, the magnetic field at the launcher was varied from below the 2nd harmonic ECH resonance (0.5 T) through the fundamental ECH resonance (1 T). During this scan, key plasma parameters were measured, especially at the launch and target locations. Figure 6 shows the variation of some parameters during this scan. All of the key parameters shown indicate that there is significant coupling of microwave power to the core plasma when the magnetic field at the launcher is near 2nd harmonic for 28 GHz, i.e., 0.5 T. An explanation of the results in the context of O-X-B mode conversion at 2nd harmonic follows in Sec. V, including a discussion of why the coupling does not occur for fundamental ECH relevant magnetic field values.

FIG. 6.

The value of certain key plasma parameters is shown for helicon-only (black) and helicon plus 28 GHz (red) discharges, as a the value of the magnetic field in the microwave launch region is varied. The DLP was located on-axis (i.e., “core”) in the target region, but was removed for IR camera measurements of the target surface. For 28 GHz microwaves, the 2nd harmonic EC resonance occurs at 0.5 T and the fundamental resonance occurs at 1 T.

FIG. 6.

The value of certain key plasma parameters is shown for helicon-only (black) and helicon plus 28 GHz (red) discharges, as a the value of the magnetic field in the microwave launch region is varied. The DLP was located on-axis (i.e., “core”) in the target region, but was removed for IR camera measurements of the target surface. For 28 GHz microwaves, the 2nd harmonic EC resonance occurs at 0.5 T and the fundamental resonance occurs at 1 T.

Close modal

The O-X-B mode conversion process relies on a careful tailoring of the microwave launch geometry, the magnetic field, the neutral gas pressure, and the plasma electron density profile.1,2 An illustrative example of the radial O-X-B mode conversion (at a given axial slice) is shown in Fig. 7. A trivial electron density profile is assumed, which is simply peaked on-axis at 2 × 1019 m−3 and decreases linearly to zero at the plasma edge. The horizontal axis in Fig. 7 is the plasma radius normalized to the edge of the column. The vertical axis in Fig. 7 is frequency normalized to the 28 GHz drive frequency of the microwave source. In this illustration, the electron plasma frequency, ωpe, electron cyclotron frequency, Ωce, and upper-hybrid frequency, ωUH2 = ωpe2 + Ωce2 are plotted, normalized to the 28 GHz microwave frequency. In this example, the magnetic field at the microwave launcher is ∼0.55 T, the fundamental Ωce is the horizontal (solid green) line, and the 2nd harmonic is shown as a horizontal (dashed green) line, which is slightly greater than 28 GHz. The O-X-B mode conversion process occurs along the solid red horizontal line: electromagnetic O-mode microwaves arrive from the launcher and propagate into the plasma until they encounter the O-mode cutoff density surface where ωpe = 28 GHz. At this radius, the microwaves are mode converted into radially outward propagating electromagnetic X-mode waves until they encounter the UH resonance surface where ωUH = 28 GHz and they are converted into electrostatic Bernstein waves. Power is absorbed by plasma electrons from the Bernstein wave inwardly propagating in the plasma medium, particularly on nearby Doppler shifted nΩce resonances.

FIG. 7.

Illustration of the O-X-B mode conversion process.

FIG. 7.

Illustration of the O-X-B mode conversion process.

Close modal

Raising the magnetic field at the microwave launcher (while keeping other parameters constant) effectively raises the horizontal level of the Ωce resonances in this illustration. In this way, the constant 28 GHz microwaves can be scanned across the 2nd harmonic Ωce resonance (including through the fundamental resonance) to investigate the O-X-B process and optimize the on-axis power delivery to the Proto-MPEX target as shown in Fig. 6. However, since the UH resonance also depends on the magnetic field value, the blue curve in this illustration is modified as the magnetic field is raised. The effect of this is that the X-B mode conversion location, which occurs at the UH resonance layer, is pushed further out radially as the magnetic field is raised. When the magnetic field approaches the condition for fundamental ECH in this linear magnetic geometry, the UH resonance leaves the plasma, and consequently the O-X-B process cannot occur (since Bernstein waves are electrostatic and cannot propagate in vacuum). This is supported by the observation in Fig. 6 that for magnetic fields relevant to fundamental ECH, there is no apparent coupling of power to the plasma. For a fixed magnetic field value, the illustration in Fig. 7 can also be used to understand the effect of lowering the electron density below the over-dense condition for 28 GHz microwaves. Lowering the peak value of density lowers the black (and blue) curves in the illustration, which moves the location of the O-X and X-B conversion layers radially inward. When the core value of density drops below 0.97 × 1019 m−3, ωpe < 28 GHz, and there is no surface where the O-X mode conversion can take place, again terminating the O-X-B process.

These results lead to the conclusion that 2nd harmonic O-X-B mode conversion in over-dense plasmas is an effective heating scenario for delivering significant heat flux plasma to target in Proto-MPEX. The DLP data from Fig. 6 can be replotted in the context of the density and magnetic field space of Proto-MPEX, as shown in Fig. 8. The data points are shaded according to the measured value of electron temperature on-axis during the period of auxiliary 28 GHz power application. As the magnetic field at the launcher is scanned, it is clear that significant core electron heating at the target is only observed as the launcher magnetic field goes through the 2nd harmonic O-X-B relevant region of parameter space. Figure 9 shows an expanded region of (ne, B) parameter space that is relevant to the MPEX operational space. In particular, the 2nd harmonic O-X-B heating ranges associated with higher frequency gyrotrons are indicated. At this time, the gyrotron frequency for MPEX has not been chosen. The Proto-MPEX helicon plasma source without auxiliary heating has produced discharges with an electron density of 8 × 1019 m−3 at a launcher magnetic field of 1.5 T. This suggests that a 70 GHz gyrotron installed on MPEX (operating with a similar helicon source) should be able to utilize 2nd harmonic O-X-B electron heating scenarios to achieve both high heat flux and high particle flux plasma delivery onto material targets.

FIG. 8.

Regions in (ne, B) space relevant to 28 GHz microwave heating in Proto-MPEX. The plotted points are measurements from on-axis DLP at the target region, corresponding to a magnet field scan at the launcher shown in Fig. 6. The data points are shaded according to the temperature scale shown.

FIG. 8.

Regions in (ne, B) space relevant to 28 GHz microwave heating in Proto-MPEX. The plotted points are measurements from on-axis DLP at the target region, corresponding to a magnet field scan at the launcher shown in Fig. 6. The data points are shaded according to the temperature scale shown.

Close modal
FIG. 9.

Regions in (ne, B) space relevant to O-X-B microwave heating in MPEX for a variety of gyrotron source frequencies. The black * are Proto-MPEX achieved data points, measured by an on-axis DLP in the microwave launch region.

FIG. 9.

Regions in (ne, B) space relevant to O-X-B microwave heating in MPEX for a variety of gyrotron source frequencies. The black * are Proto-MPEX achieved data points, measured by an on-axis DLP in the microwave launch region.

Close modal

The mission of Proto-MPEX is to perform research and development (in a pulsed device) which validates the plasma source concept for MPEX (a steady state device), including auxiliary heating power systems. The heat flux on target is a key metric for assessing the performance of the Proto-MPEX device. The results presented here demonstrate that the recent reconfiguring of the magnetic geometry of the auxiliary heating sections of Proto-MPEX has been successful at producing plasma pulses that deliver high heat flux onto target. The on target heat flux is increased due to measured heating of plasma electrons by the application of 28 GHz microwaves through an O-X-B mode conversion process. While the O-X-B scenario predominantly produces off-axis heat flux improvements (nearly 10×), the on-axis heat flux to target is nevertheless increased by a factor of 2. While keeping the helicon source plasma and applied auxiliary microwave power constant, a magnetic field scan in the launcher region was performed. The core electron temperature rise during 28 GHz power application was found to be strongly peaked in a region around 0.5 T at the launcher, corresponding to 2nd harmonic EBW heating via O-X-B mode conversion. Projecting these results to a steady state MPEX device indicates that 70 GHz microwave heating via the same O-X-B mode conversion process should lead to the capability of simultaneous high heat and high particle source plasma for fusion plasma relevant material exposure conditions.

This manuscript has been authored by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the U.S. Department of Energy. The United States Government retains and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a nonexclusive, paid-up, irrevocable, world-wide license to publish or reproduce the published form of this manuscript, or allow others to do so, for United States Government purposes. The Department of Energy will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan (http://energy.gov/downloads/doe-public-access-plan).

Grateful appreciation is due to the Proto-MPEX team of scientists, engineers, students, technicians, and support staff that made this work possible, especially C. Josh Beers, Jeff Bryan, Andy Fadnek, E. Wayne Garren, Seungsup Lee, Elizabeth Lindquist, Jason McDaniel, Pawel Piotrowicz, Holly Ray, Melissa Showers, and Mark Watson.

1.
T. M.
Biewer
,
T. S.
Bigelow
,
J. F.
Caneses
,
S. J.
Diem
,
D. L.
Green
,
N.
Kafle
, and
J.
Rapp
, and
the Proto-MPEX Team
.
Phys. Plasmas
25
,
024501
(
2018
).
2.
S. J.
Diem
,
D. L.
Green
,
R. W.
Harvey
, and
Y. V.
Petrov
,
Phys. Plasmas
25
,
072124
(
2018
).
3.
J.
Rapp
,
C. J.
Beers
,
T. M.
Biewer
,
T. S.
Bigelow
,
J. F.
Caneses
,
J. B. O.
Caughman
,
S. J.
Diem
,
R. H.
Goulding
,
R.
Isler
,
N.
Kafle
,
L. W.
Owen
, and
M.
Showers
,
Fusion Eng. Des.
136
,
712
716
(
2018
).
4.
J.
Rapp
,
A.
Lumsdaine
,
C. J.
Beers
,
T. M.
Biewer
,
T. S.
Bigelow
,
J. F.
Caneses
,
J. B. O.
Caughman
,
R. H.
Goulding
,
N.
Kafle
,
C. H.
Lau
,
E.
Lindquist
,
P. A.
Piotrowicz
,
H.
Ray
, and
M.
Showers
, “
Latest results from Proto-MPEX and the future plans for MPEX
,”
Fusion Sci. Technol.
(to be published).
5.
J.
Rapp
,
T. M.
Biewer
,
T. S.
Bigelow
,
J. F.
Caneses
,
J. B. O.
Caughman
,
S. J.
Diem
,
R. H.
Goulding
,
R. C.
Isler
,
A.
Lumsdaine
,
C. J.
Beers
,
T.
Bjorholm
,
C.
Bradley
,
J. M.
Canik
,
D.
Donovan
,
R. C.
Duckworth
,
R. J.
Ellis
,
V.
Graves
,
D.
Giuliano
,
D. L.
Green
,
D. L.
Hillis
,
R. H.
Howard
,
N.
Kafle
,
Y.
Katoh
,
A.
Lasa
,
T.
Lessard
,
E. H.
Martin
,
S. J.
Meitner
,
G.-N.
Luo
,
W. D.
McGinnis
,
L. W.
Owen
,
H. B.
Ray
,
G. C.
Shaw
,
M.
Showers
, and
V.
Varma
, and
the MPEX Team
,
Nucl. Fusion
57
,
116001
(
2017
).
6.
J.
Rapp
,
T. M.
Biewer
,
T. S.
Bigelow
,
J. B. O.
Caughman
,
R. C.
Duckworth
,
R. J.
Ellis
,
D. R.
Giuliano
,
R. H.
Goulding
,
D. L.
Hillis
,
R. H.
Howard
,
T. L.
Lessard
,
J. D.
Lore
,
A.
Lumsdaine
,
E. J.
Martin
,
W. D.
McGinnis
,
S. J.
Meitner
,
L. W.
Owen
,
H. B.
Ray
, and
V. K.
Varma
,
IEEE Trans. Plasma Sci.
44
,
3456
(
2016
).
7.
J. B. O.
Caughman
,
R. H.
Goulding
,
T. M.
Biewer
,
T. S.
Bigelow
,
I. H.
Campbell
,
J.
Caneses
,
S. J.
Diem
,
A.
Fadnek
,
D. T.
Fehling
,
R. C.
Isler
,
E. H.
Martin
,
C. M.
Parish
,
J.
Rapp
,
K.
Wang
,
C. J.
Beers
,
D.
Donovan
,
N.
Kafle
,
H. B.
Ray
,
G. C.
Shaw
, and
M. A.
Showers
,
J. Vac. Sci. Technol. A
35
,
03E114
(
2017
).
8.
R. H.
Goulding
,
J. B.
Caughman
,
J.
Rapp
,
T. M.
Biewer
,
T. S.
Bigelow
,
I. H.
Campbell
,
J. F.
Caneses
,
D.
Donovan
,
N.
Kafle
,
E. H.
Martin
,
H. B.
Ray
,
G. C.
Shaw
, and
M. A.
Showers
,
Fusion Sci. Technol.
72
,
588
594
(
2017
).
9.
C. J.
Beers
,
E. G.
Lindquist
,
T. M.
Biewer
,
J. F.
Caneses
,
J. B. O.
Caughman
,
R. H.
Goulding
,
N.
Kafle
,
H.
Ray
,
M. A.
Showers
,
S. J.
Zinkle
, and
J.
Rapp
,
Fusion Eng. Des.
138
,
282
288
(
2019
).
10.
J. F.
Caneses
,
P. A.
Piotrowicz
,
T. M.
Biewer
,
J. B. O.
Caughman
,
R. H.
Goulding
,
N.
Kafle
, and
J.
Rapp
,
Phys. Plasmas
25
,
083518
(
2018
).
11.
J. F.
Caneses
,
P. A.
Piotrowicz
,
T. M.
Biewer
,
R. H.
Goulding
,
C.
Lau
,
M.
Showers
, and
J.
Rapp
, “
Ion fluxes and neutral gas ionization efficiency of the 100-kW light-ion helicon plasma source concept for the Materials Plasma Exposure eXperiment
,”
Fusion Sci. Technol.
(to be published).
12.
P. A.
Piotrowicz
,
J.
Caneses
,
M. A.
Showers
,
D. L.
Green
,
R. H.
Goulding
,
J. B. O.
Caughman
,
T. M.
Biewer
,
J.
Rapp
, and
D. N.
Ruzic
,
Phys. Plasmas
25
,
052101
(
2018
).
13.
T. D.
Rognlien
and
T. A.
Cutler
,
Nucl. Fusion
20
,
1003
(
1980
).
14.
L.
Spitzer
,
Physics of Fully Ionized Gases
, 2nd ed. (
Courier Corporation
,
2013
).
15.
16.
V. K.
Yadav
and
D.
Bora
,
Phys. Plasmas
11
,
4582
(
2004
).
17.
H. P.
Laqua
,
V.
Erckmann
,
J. J.
Harfuss
,
H.
Laqua
, and
the W7-AS Team and ECRH Group
,
Phys. Rev. Lett.
78
,
3467
(
1997
).
18.
A. P.
Smirnov
and
R. W.
Harvey
, “
The GENRAY ray tracing code
,”
CompX Report No. CompX-2000-01
(
2001
).
19.
M.
Showers
,
T. M.
Biewer
,
J. B. O.
Caughman
,
D. C.
Donovan
,
R. H.
Goulding
, and
J.
Rapp
,
Rev. Sci. Instrum.
87
,
11D412
(
2016
).
20.
M.
Showers
,
P. A.
Piotrowicz
,
C. J.
Beers
,
T. M.
Biewer
,
J.
Caneses
,
J.
Canik
,
J. B. O.
Caughman
,
D. C.
Donovan
,
R. H.
Goulding
,
A.
Lumsdaine
,
N.
Kafle
,
L. W.
Owen
,
J.
Rapp
, and
H.
Ray
,
Plasma Phys. Controlled Fusion
60
,
065001
(
2018
).
21.
J. F.
Caneses
and
B.
Blackwell
,
Plasma Sources Sci. Technol.
24
,
035024
(
2015
).