Radiative properties of asymmetric and symmetric X

ARTICLE IN PRESS

Journal of Quantitative Spectroscopy & Radiative Transfer 99 (2006) 349–362 www.elsevier.com/locate/jqsrt

Radiative properties of asymmetric and symmetric X-pinches with two and four wires recently produced on the UNR 1 MA Zebra generator V. Kantsyreva,, A. Safronovaa, V. Ivanova, D. Fedina, R. Mancinia, A. Astanovitskya, B. LeGalloudeca, S. Batiea, D. Browna,c, V. Nalajalaa, I. Shresthaa, S. Pokalaa, N. Ouarta, F. Yilmaza, A. Clintona, M. Johnsona, T. Cowana, B. Jonesb, C.A. Coverdaleb, C. Deeneyb, P.D. LePellc, D. Jobec, D. Nielsonc a

Department of Physics, University of Nevada, Reno, NV 89557, USA b Sandia National Laboratories, Albuquerque, NM 87185, USA c Ktech Corp., Albuquerque, NM, USA Accepted 26 April 2005

Abstract Experimental results of studies of the 1 MA X-pinch X-ray source in a wide spectral region are overviewed. Implosion dynamics and radiative properties of various X-pinches were studied by spatially and time-resolved X-ray and optical diagnostics. In particular, dynamics of spatial and temporal developments of the structure of X-ray emitting regions (1–5 keV), temporal characteristics of X-ray pulses, X-ray radiation outputs and electron beam characteristics from symmetric and asymmetric Mo, Cu, and combined asymmetric Mo/W X-pinches with two or four wires were studied. The mechanisms of X-ray multiburst generation are discussed. The future applications of the high-current X-pinch as a 5–10 kJ subkeV–10 keV radiation driver are considered. r 2005 Elsevier Ltd. All rights reserved. Keywords: X-pinch; X-ray; Diagnostics

Corresponding author.

E-mail address: [email protected] (V. Kantsyrev). 0022-4073/$ - see front matter r 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.jqsrt.2005.05.028

ARTICLE IN PRESS 350

V. Kantsyrev et al. / Journal of Quantitative Spectroscopy & Radiative Transfer 99 (2006) 349–362

1. Introduction Wire array Z-pinches create substantial soft X-ray powers from hot dense plasmas. X-pinches create plasmas in the same regime employing the process of magnetically driving fine wires [1–5], which is the same process used in Z-pinches, and so X-pinch studies are applicable to wire array physics. Plasmas derived from 1 MA X-pinches can also serve as wideband pulsed X-ray sources of comparable brightness and energy yield to laser produced plasmas [4,6–9]. An X-pinch plasma is formed by touch-crossing two or more wires placed between the electrodes of an 0.1–1.0 MA pulsed power generator [1–5]. A fast rising current (50–100 ns) quickly vaporizes and ionizes the wire material. The X-pinch yields several short (0.1–10 ns [2–5]) thermal X-ray bursts (XRB) in the spectral range from sub-keV to several keV from bright plasma hot spots with dimensions from 1 to 100 mm surrounded by a cooler plasma seen in the softer wavelengths (with a size of several mm at the cross-wire region [3–8]). The number of hot spots is typically greater than 2. The bright hot spots can be seen even for lo223 A˚.The size of hot spots has been observed to be smaller when shorter wavelengths are observed. The total number of shorter wavelength sources is usually smaller than the number of sources found for longer wavelengths. Further, the X-ray radiation sources tend to be more isolated as the wavelength of the source becomes smaller. On the other hand, the source becomes larger with increasing wavelength. Previous studies of magnetically driven 1 MA X-pinches included the study of the X-pinch source structure, the time and energy scaling of X-ray radiation in the energy range from 0.01 up to 500 keV, and investigations of energetic electron beams (40.01–2 MeV) in the X-pinch plasma [2,4,6–8]. A magnetic field inside an X-pinch may reach 41000 T [9,10]. The hot spots generated during the X-pinch discharge are found to have different electron temperature Te and density Ne [6]. For example, for Mo and Ti X-pinches the Ne ranged between1019 and 2 1022 cm3, Te ranged from 0.85 to 2 keV, and a hot electrons beam fraction is up to 7% [6,11]. The probability of generating a single XRB during the X-pinch shot varied from 10% (for Fe) up to 30% for Ti [3,8]. The total yield of X-ray/EUV radiation reaches 410 kJ for 1 MA Mo X-pinches [3,8]. The mm-scale sources of softer (keV and sub-keV radiation) thermal X-rays are located typically near a cross-wire point. The mm-scale harder X-ray source (more than 10 keV radiation energy) is shifted to the anode side from the cross-wire point and is originated by electron beams of 2–3 mm in diameter. Three different types of the energetic electron beams, having energies more than or equal to several keV, were observed in X-pinches based on measurement of hard X-ray pulsed emission [12]. The first type of energetic electrons was generated in the vicinity of the original cross-wire point in 70–80% of discharges. They are observed to be temporally well correlated with the initiation of the first XRB and have a similar duration (1–2 ns) [10,12]. The mechanism of generation of this electron beam is connected with a hot spot formation [10]. The second type of electron beams (2.5–10 ns duration) occurs typically after the first XRB and typically is not correlated with the XRB [12]. The location of the region of production of these electron beams was near the cross-wire region in at least 60–70% of the X-pinch discharges. The third type of electron beams (10–30 ns duration) was observed several tens of ns after the current pulse maximum, and had no correlation with the thermal XRB [12]. In more than 60% of experiments these beams were produced in the anode region of the X-pinch. The appearance of this third type of electron beam may correlate with a late-time pinching of a major portion of the X-pinch plasma volume, as was observed in previous Z-pinch plasma experiments [13]. Recent applications

ARTICLE IN PRESS V. Kantsyrev et al. / Journal of Quantitative Spectroscopy & Radiative Transfer 99 (2006) 349–362

351

of X-pinch sources include X-ray spectroscopy [6,11], backlighting and microscopy [3,10], and spectropolarimetry [7,11]. As seen from the brief review above, a 1 MA X-pinch is a very complicated 3D plasma object with a sub-ns lifetime. At the present time existing MHD models can describe and predict X-pinch plasma behavior only partially [14,15]. Experiments will continue to provide more understanding about this Z-pinch-related plasma object. Some important problems of the 1 MA X-pinch research are: 1. Why does X-ray multiburst regime take place during an X-pinch discharge? 2. Why Fe X-pinches so poor as spectroscopic sources of line radiation compared with Al, Ti, and Mo X-pinches and is it typical for all mid-Z elements? 3. Is it possible to increase an X-ray/EUV flux (or energy) density at a specific distance from an X-pinch without additional optics, which can be important for X-pinch application as a X-ray radiation driver? 4. How does one decrease the size of an emitting region? We now consider some ideas leading to solution of these problems.

2. Experimental setup Experiments were performed with the X-pinch as the load on the Nevada Terawatt Facility ‘‘Zebra’’ generator [16] with a peak current of about 1.2 MA, a rise time of 100 ns, a maximum stored energy of 200 kJ, and a 1.9 O pulse-forming line impedance. The diagnostic complement includes X-ray diagnostics and, for the first time in the NTF experiments , multiframe laser probing (Fig. 1). The X-ray diagnostic complement covers the spectral region from 0.25 to X8 keV. This complement includes temporally, spectrally and

4

5

3

+ 1

2A&B

7 8

6

Fig. 1. 1. X-pinch load. 2. Axially resolved (A) and radially resolved (B) X-ray spectrometers (KAP, a-quartz crystals). 3. Hard X-ray spectrometer (LiF crystal). 4. High-resolution X-ray pinhole camera. 5. Time- gated X-ray pinhole camera. 6. PCD and XRD assembly. 7. Two-frame laser interferometry. 8. Two (or four)- frame laser shadowgraphy.



spatially resolved imaging systems, spectrometers, and fast X-ray detectors. Two X-ray pinhole cameras, with magnification of 0.71 and spatial resolution of 170 mm, were used in the experiments for plasma imaging. A side-on time-integrated pinhole camera generated two plasma images using two different filters with the lower cutoff wavelengths, designated by the 10% transmission, of l1=10 o3:5 A˚ and l1=10 o10 A˚.The direction of view was 451 to the X-pinch plane. In the pinhole camera the images have been recorded using packages of 2–3 layers of DEF-5 Kodak X-ray film which extended a registration range of the wavelengths from 10.3 to 2.7 A˚ due to the additional filtering by the film layers. A side-on time-gated pinhole camera based on a microchannel plate detector (MCP) has 4–6 time-frames with an adjustable duration from 4–6 ns with an inter-frame time of 10 ns.The output image was registered on the Kodak Pan Film 2484. The direction of view was 67.51 to the X-pinch plane. The pinhole camera, with a spatial resolution of 230 mm, was composed of two identical parallel rows of Pt/Ir membranes with hole diameters of 70 mm and a 25 mm Be filter to protect the membranes from debris. The number of membranes in each row was 6 where one row formed images for the lower cutoff wavelength l1=10 o10 A˚ and one for l1=10 o3:5 A˚. A. In order to obtain more precise information about Te and Ne spectroscopic diagnostics were employed. These devices are three time-integrated X-ray convex crystal spectrometers. Two of them had spatial 1D resolution up to 0.3 mm. One spectrometer was axially resolved (direction of view was 451 to X-pinch plane) and another was radially resolved (direction of view was 67.51 to the X-pinch plane). A KAP (2d ¼ 26:62 A˚, radius of curvature 51 mm) and a-quartz (2d ¼ 6:687 A˚, radius of curvature 102 mm) convex crystals were employed in these experiments. A 12.5 mm Be was used as a filter. In the third spectrometer a LiF crystal (2d ¼ 4:027 A˚, radius of curvature 25.4 mm) is used with a 70 mm thick Be filter (direction of view was 22.51to the X-pinch plane). These spectrometers provide information on plasma condition in X-pinch experiments during the near-stagnation stages where the maximum plasma densities and temperatures occur. A study of time history and absolute measurements of X-ray yields were made using calibrated filtered photoconducting detectors (PCD) and X-ray diodes (XRD) with a 0.5 ns resolution. Be filters, kimfoil and kapton films were used for filtering. The XRD had a solid carbon photocathode and a Ni mesh anode. The PCD and XRD signals were recorded by a 2 Gs/s Tektronix TDS 640L oscilloscope. In order to obtain absolute yields of X-ray radiation measured by these devices, the oscilloscope waveforms were integrated over the time and corrected for the geometry of the experiment and the sensitivity of the detectors. Optical laser-based diagnostics, using a commercial Nd:YAG laser produces 0.2-ns 100-mJ pulses with a wavelength of l ¼ 532 nm, included two-frame shadow/schlieren with two-frame interferometry, or four-frame shadowgraphy. The angle between two probing channels is 22.51. Two optical delay lines deliver two couples of laser pulses with orthogonal polarization into diagnostic channels. The first optical delay line provides a 25-ns delay between p-polarized and s-polarized laser pulses. In the two-frame version both the shadowgraphy and interferometry recorded the same time interval. The four-frame shadow diagnostic used two combined, twoframe channels. The first two pulses are shifted from the other two by the second 12-ns optical delay line. Lenses relay the image of the plasma onto a four 12-bit CCD cameras. Polarizers separated two pulses with different polarization and pulses with a different delay time are transmitted to appropriate CCD cameras. Narrowband interference filters on the CCD cameras block broadband plasma radiation. The spatial resolution of the optical diagnostics was 10–35 mm


353

12 mm

Anode

20 mm

Cathode 1.

2.

3.

Fig. 2. 1. X-pinch load scheme. 2. Planar-loop 2-wire asymmetric X-pinch with a distance between a cross-wire point and an anode of 5 mm (potential high X-ray flux at the anode window). 3. Planar-loop 4-wire (901) X-pinch with a probably smaller emitting region and better time characteristics. X-pinches are shown with supporting rods that are removed after the X-pinch frame installed between Zebra electrodes.

depending on the magnification of image relying optics. Reference images were made for every diagnostic channel before the Z-pinch experiment. The reference images, firstly, show the initial position of wire and, secondly, can be used to provide the background of images in plot diagrams. The pulse of the diagnostic laser and the pulse of current on the load of the accelerator were synchronized to 75 ns.The diagnostic laser is synchronized with the X-ray time-resolved devices with accuracy 1 ns. Main configurations of loads are presented in Fig. 2. Two typical configurations of X-pinches were applied. The first is the planar-loop configuration (Cu, Mo, and Mo/W X-pinches with two and four wires) where the anode and cathode side-wire loops touched each other only in one central point. The second was a wire-twisted configuration (Ti X-pinch), where the wires were twisted together by 1801. For symmetric X-pinches (Ti, Cu and Mo), cross-wire points were at 10 mm from the anode and cathode, whereas for asymmetric X-pinches (Mo and combination Mo/W) cross-wire points were at 5 mm from the anode and at 15 mm from the cathode. The angle between wires was 621 for all experiments. In experiments anodes both with and without a central window were used. The following loads were studied: two-wire Mo (diameter F ¼ 50 mm) symmetric and asymmetric X-pinches with total mass of 0.94 mg; four-wire Mo (diameter F ¼ 30 mm) symmetric X-pinches with a total mass of 0.67 mg; two-wire Cu (diameter F ¼ 76:2 mm) symmetric X-pinches with a total mass of 1.9 mg; four-wire Cu (diameter F ¼ 63:5 mm) symmetric X-pinches with a total mass of 2.6 mg; and two-wire combined asymmetric X-pinches with an anode loop from a Mo wire (diameter F ¼ 65 mm) and a cathode loop from a W wire (diameter F ¼ 45:7 mm) with a total mass of 1.5 mg.

3. Experimental results We focused on the observation of new effects and on the study of the difference in radiation characteristics of two and four wires, and symmetric and asymmetric 1 MA X-pinches. A detailed



spectroscopic analysis of X-ray line spectra from these X-pinches were performed and discussed in [17]. 3.1. X-ray yield, dimensions of a central emitting region and an X-ray pulse timing for two- and four-wire X-pinches The main parameters of X-pinch radiation sources are X-ray yield, dimensions of a central (near a cross-wire point) emitting region and the XRB timing in different spectral bands. Studies of this type for symmetric 0.8–1.0 MA Al, Ti, Fe, Mo and W X-pinches have been conducted previously [2,4,6,8,9]. A typical set of new data obtained is shown in Fig. 3. Results of new recent

Fig. 3. X-ray time gated (A) and time-integrated (B) and laser-probing images (D), X-pinch current, X-ray image frames and PCD (with an 8 mm Be filter) signals response versus time (C) of two 76.2 mm wires Cu X-pinch implosions. Shot # 372. Such a set of data (parts A, B, C, and D) was collected in all shots, but a current, positions of X-ray images frames and PCD (with 8 mm Be filter) signals response versus time are shown only in this figure. The arrows indicate positions of hot spots. Shadowgram (D) corresponds to frame 3.


355

Table 1 X-ray yield and size of the X-pinch emitting region

Mo 2 50 mm wires Mo 2 50 mm wires asymmetric Mo 4 30 mm wires Cu 2 76.2 mm wires Cu 4 63.5 mm wires Mo 2 65 mm and 2xW 45.7 mm wires asymmetric

ER

F(lo10:3A˚) mm

F(lo3:1A˚) mm

1 1 0.6 0.6 0.4 1.2

3 4.5 3 5.4 2.7 4.4 5.4 3.7 4.3 4 33

2 2.5 1.3 3.2 1.1 2.4 2.6 2.9 1.5 2.2 1.8 2.4

Relative X-ray yield ER normalized to the total emitted energy E ¼ 7:5 kJ (XRD, 5 mm kimfoil) from a Mo 2 50 mm wire X-pinch, and sizes in mm of a central emitting region (lo10:3A˚ & lo3:1A˚, time-integrated images, perpendicular to the central X-pinch axis and along the central axis).

measurements of the relative X-ray yield ER normalized to previously obtained Mo 2 50 mm wire X-Pinch data [6,8,9] and sizes of a central emitting region are shown in Table 1. For softer X-ray yields no difference was observed between symmetric and asymmetric Mo two-wire X-pinches. The application of asymmetric geometry to an X-pinch (Fig. 2) provides a simple way to increase the flux density of X-ray radiation near an anode window without implementing an expensive focusing X-ray/EUV optics. In recent experiments the flux density was 4 times larger than that obtained with a symmetric X-pinch. In addition, replacing one cathode Mo wire with a W wire in an asymmetric X-pinch increases the X-ray yield (Table 1). Meanwhile, when changing from two wires to four wires a decrease in source size and relative softer X-ray yield was observed for both Cu and Mo X-pinches. Here we present new data on softer X-ray XRBs from Mo X-pinches (XRD with 5 mm kimfoil filter, spectral regions 0.2–0.3 keV and 40.8 keV) considered important for X-pinch radiation driver applications. The minimum number of softer X-ray XRBs during the Mo X-pinch discharges were 2–4 and their duration was about 15–20 ns. A total duration of an XRB cluster was about 50–90 ns starting 30–50 ns after the beginning of current. 3.2. Spatial structure and timing of X-ray pulses of two- and four-wire X-pinches Initial experiments with 1 MA two-wire Cu X-pinches demonstrated a significant difference from 1 MA two-wire Ti and Mo X-pinches. Time-gated and time-integrated X-ray images show the appearance of hot spots, mainly in a softer spectral region l44.4 A˚, not only near a cross-wire point but also along both anode and cathode parts of wires (Fig. 3). Similar structures were observed previously for 1 MA Fe and W X-pinches [8], but low resolution and clarity of images in Ref. [8] made a correct interpretation difficult. Some of the recently observed hot spots along the wires sufficiently stable that their positions did not change for420 ns (Fig. 3, frames 3–5). At times the hot spots had the shape of jets directed toward the X-pinch axis and perpendicular to the wire. The separation of the hot spot was 1 mm and a number of hot spots on the anode side was 2–3 times bigger than on the cathode side. The correspondence in positions of hot spots from ‘‘left’’ and ‘‘right’’ sides of an X-pinch was observed with probability of 60%. Note that the number and probably the positions of strong constrictions on the shadowgram (Fig. 3) were close



to the number and the positions of hot spots seen in softer X-ray images. Measurements of positions and parameters of hot spots along wires can be used for diagnostics of the X-pinch local magnetic field. For two-wire Cu X-pinches, Te and Ne of the hot spots in a cross-wire region was determined by spectroscopic modeling to be 2.2570.25 keV and 1021 cm3, respectively [17]. Meanwhile, hot spots along the wires have a lower T e 2002500 eV with a lower density N e 1018 21019 cm3 . Another interesting object was a strong plasma jet that seen only in softer X-ray images (for lo10:3 A˚ and lo4:4 A˚) and directed perpendicular to the longitudinal X-pinch axis. The maximum dimension of a jet was about 15–20 mm and as such was much larger than the hot spot size which are all smaller than several hundred mm. New results were also found for a two-wire asymmetric Mo/W X-pinch, where the Mo wire forms a short anode-side loop, and the W wire forms a cathode-side loop. For such X-pinch hot spots appeared not only near a cross-wire point and along the wires but also formed arcs between the wires (Fig. 4). In addition, these hot spots were clearly seen in a much shorter spectral region lo3:1 A˚, in distinction to the Cu X-pinches. In time-integrated and time-gated images at least two such hot spot arcs were present. Some of the arcs in recent images did not change position between the wires for at least 10 ns (Fig. 4, frames 4 and 5). Note that X-ray images of the arcs,

λ < 3.5A° anode

cathode ° λ < 3.7A

λ < 3.1A° anode

cathode ° λ < 10A

1

2

3

4

5

X-ray time-gated ∆fr = 4 ns

(A)

° λ < 10.3A

° λ < 4.4A

X-ray time-integrated anode

Field of view 15 mm

cathode (B)

Shadowgram. Frame 4 (after x-ray frame 4)

Fig. 4. X-ray (A) and laser-probing images (B) of combined two-wire Mo (65 mm) and W (45.7 mm) asymmetric X-pinch implosions. Shot # 379. The arrows indicate positions of hot spots.


357

i.e., between W wires of the cathode side, correlated with some of jets-like structures on shadowgrams. Modification from a two-wire to a four-wire X-pinch dramatically changed the structure of a radiation source as shown in Fig. 5. For example, only a central hot spot, or a cluster of hot spots, was seen in both softer and harder spectral range in time-integrated images of four-wire Cu X-pinches. Time-integrated X-ray pictures contained a bright central region and low-intensity images of jets directed toward the X-pinch axis perpendicular to the wire, as in Refs. [4,8]. For 4-wire Cu X-pinches, Te and Ne of the hot spots in a cross-wire region were determined by spectroscopic modeling [17] to be approximately the same as those for two-wire X-pinch, i.e., 2–2.5 keV and 1021 cm3, respectively. For the two-wire Cu X-pinch, a strong plasma jet perpendicular to longitudinal X-pinch axis was observed, in regions lo10:3 A˚ and lo4:4 A˚, at the cross-wire point. The maximum dimension of jet was about 15–20 mm which is much larger than any hot spots size as mentioned above. A plasma jet similar to the jet produced from twowire Cu X-pinches is seen on shadowgrams in Fig. 5. Note, that four-wire Cu X-pinch has a smaller emitting region size for the shorter wavelengths compared with two-wire scheme. Two-wire Mo X-pinches have been extensively studied [6–9]. For this reason only new results for four-wire Mo X-pinches are shown. The structure of four-wire Mo X-pinches, in Fig. 6, had bright central hot spot clusters and relatively weak plasma images at the former positions of wires. For 4-wire Mo X-pinches, the Te and Ne of the hot spots in a cross-wire region was determined by ° λ < 3.5A

anode

cathode ° λ < 3.7A

anode

° λ < 3.1A

cathode ° λ < 10A

1

2

3

4

5

X-ray time-gated ∆fr = 4 ns

(A)

° λ < 10.3A

° λ < 4.4A

X-ray time-integrated anode Field of view 9 mm cathode

Shadowgram. Frame 3 (x-ray frame 4) (B)

Fig. 5. X-ray (A) and laser-probing images (B) of four 63.5 mm wire Cu X-pinch implosions. Shot # 374.


V. Kantsyrev et al. / Journal of Quantitative Spectroscopy & Radiative Transfer 99 (2006) 349–362 ° λ < 3.5A

anode

cathode anode

cathode °

λ < 10A (A)

1

2

3

4

5

° λ < 3.7A

° λ < 3.1A

° λ < 10.3A

° λ < 4.4A

X-ray time-integrated

X-ray time-gated ∆fr = 4 ns cathode

Field of view 15 mm

anode (B)

Shadowgram. Frame 4 (end of x-ray frame 4)

Fig. 6. X-ray (A) and laser-probing images (B) of four 30 mm Mo X-pinch implosions. Shot # 375. The arrows indicate positions of hot spots. Note that hot spots shift in axial and radial directions of 2–3 mm during X-pinch discharge. Xray bursts were during the end of X-ray frame #1, frame #2, the end of frame #4, and the beginning of frame #5.

spectroscopic modeling to be 0.8 keV and 1022 cm3, respectively, with hot electron fractions up to 5%, which are similar to those found for the two-wire Mo X-pinches [6]. Time-gated imaging has shown that hot spots near a cross-wire point shift axially and radially 2–3 mm, covering up to 25% of the distance from a cross-wire point to the anode at a speed of 1.5 107 cm/s, during the discharge from their initial positions near a cross-wire point as can be seen in Fig. 6. The wire structure, seen at the anode side due to excitation of ‘‘cold’’ X-rays in a remaining wire material by an electron beam [9], has been seen at the same position even after several plasma implosions near a cross-wire point. A multicomponent structure of a Mo X-pinch during an implosion, see the end of the X-ray frame 4 of Fig. 6, is shown on the shadowgram in frame 4. A gap between the anode and cathode parts of plasma was 0.6 mm. From the image it was clear that the density of a gap plasma was significantly lower than the plasma on the cathode or anode sides because a gap was transparent for laser probing radiation with l ¼ 532 nm. The interferometric measurements of Ne in such a gap during intervals between XRB were performed recently for a two-wire (diameter F ¼ 25 mm) Ti X-pinch and is shown in Fig. 7, where an electron density of 1018–1019 cm3 was measured in a gap with a width of 0.3–0.6 mm. The structure of an asymmetric Mo X-pinch had a larger central emitting region compared with a symmetric two-wire or four-wire X-pinches (Table 1). Also, as with the four-wire symmetric Mo


359

° λ < 3.5A

anode

cathode anode

° λ < 3.7A

° λ < 3.1A

° λ < 10.3A

° λ < 4.4A

cathode ° λ < 10A

(A)

1

2

3

4

5

X-ray time-gated ∆fr= 4 ns

X-ray time-integrated cathode Field of view 9 mm anode

(B)

Shadowgram. Frame 3 (End of x-ray frame 2)

Fig. 7. X-ray (A) and laser-probing images (B) of asymmetric two 50 mm wire Mo X-pinch implosions. Shot # 377.

X-pinch, a wire structure at the anode side was seen in X-rays even after several plasma implosions near a cross-wire point (Fig. 7). The plasma electron temperatures in a cross-wire region of an asymmetric Mo X-pinch were the highest of the Mo X-pinch experiments, with Te up 22 3 to1.3 keV at Ne near 10 cm and no hot electrons. The presence of images of hot spots along the wires in a spectral region 1–1.5 keV in 1 MA twowire Cu and W-containing X-pinches and their absence in two-wire Ti [4,8], Mo and four-wire Mo X-pinches may suggest the existence of the critical current Ic, in kA per wire, for generation of such hot spots. The value of Ic was estimated to be between 250 and 500 kA for Cu and W (see Figs. 3 and 4) wires, and may be more than 500 kA for Ti and Mo wires.

4. Multiburst regime of X-ray generation in 1 MA X-pinches The scenario of the X-pinch first implosion and XRB generation is described, for example, in Ref. [10]. It includes appearance of m ¼ 0 instabilities near a cross-wire point, that produces a small neck with a hot spot in a plasma that emits an XRB and can generate the first type of an electron beam (Fig. 9A). Before this occurs a precursor plasma from the wires forms an outer corona, and central and side plasma jets. After an XRB a narrow and short neck (200–400 mm) in a gap between anode and cathode parts of plasma explosively disassembles as in Fig. 9B [10]. The measurements presented show that the density of the remaining plasma in the gap dropped from


1.0

0.5

0.0 2.E - 07

3.E - 07

3.E - 07

4.E - 07

2.5E - 07

3.0E - 07

3.5E - 07

1.0

18

Ne = 1.2.10 cm

-3

19

Ne = 1.3.10 cm

-3

360

0.5

0.0 2.0E - 07

Fig. 8. Interferograms of a Ti X-pinch plasma in a cross-wire region (Ne is plasma electron density in cm3 in the gap between anode and cathode parts of plasma). In the picture the initial position of wires is shown.

1021–1022 to 1017–1018 cm3 at distances smaller than 0.5–0.6 mm (Fig. 8), and only a small part of the initial X-pinch mass was involved in radiation processes. The majority of material remained at the initial position. As this occurs it becomes more difficult to support a current through a gap and some portion of the current could reconnect to a coronal plasma that surrounded the central part of an X-pinch. The remaining part of a continuous X-pinch current could, in some cases, form the second type of an electron beam in a plasma gap, that could lead to an additional heating of the near part of the remaining plasma at the initial position material of wires (Fig. 9C). Due to a strong pressure gradient plasma filled the gap again and a magnetic-field-pinched plasma and a new XRB occurred (Fig. 9D). Such pulsation process continued until all material of wires finally collapsed on a central axis of the X-pinch under the magnetic forces of the current (Fig. 9E). These described processes can be affected by local plasma instability and reconnection of the current in plasmas. As a result, different time intervals between XRB will be observed, and axial and radial shifts of hot spots, i.e., the m41 instability growth, may occur during an X-pinch discharge.

5. High-current X-pinch as a possible sub-keV–10 keV radiation driver Significant X-ray/EUV output from 1.0 MA high-Z X-pinches in a 1–10 kJ range, together with a ns-scale radiation pulse places such a university-scale source at the same level as much larger facilities such as the Omega laser (Table 2). With value of flux density, Q4108 21010 W=cm2 and energy density, E d 41000 J=cm2 the 1 MA X-pinch source can be used in material science study, pulsed power research, atomic physics, and plasma physics.


361

Anode 1

9

3 4 ((

5

))

7

8

6

2 Cathode (A)

(B)

(C)

(D)

(E)

Fig. 9. Proposed stages (A–E) of X-ray multiburst and electron beam generation in X-pinches. Stages A–D repeat several times, generating multiple X-ray bursts, until the plasma finally collapses in a stage E. 1, 2. Wires initial positions with possible generation of hot spots along wires 3. Narrow plasma neck between anode and cathode parts of the X-pinch. 4. Hot spot in the neck. 5. The direction and location of the first type of the electron beam (shown alongside the neck for clarity). 6. Expanding plasma. 7. The direction and location of the second type of the electron beam. 8. Plasma column formed near the end of the current pulse. 9. The direction and location of the third type of the electron beam.

Table 2 Flux and energy densities of X-ray/EUV radiation R(cm)

Q(W/cm2)

Ed (J/cm2)

0.5 1 3

7 1010/2 1010 1.8 1010/5 109 2 109/5.5 108

1400/380 350/95 40/10

Flux density Q and energy density Ed of X-ray/EUV radiation from a 1 MA Mo X-pinch at a distance R from plasma. In each column, data are estimated without a filter (at the left) and with a 1–2 mm thick plastic filter with a 0.1 mm A1 coating (at the right).

A natural limitation for 1 MA X-pinch driver applications is plasma jets and electron beams that can destroy a test-object before the end of an X-ray/EUV pulse. Estimates show a minimum source to test-object distance of 0.5–1 cm would be needs and strong magnets could provide



protection from ion and electron beams. In addition, one of the more promising applications could be experimental modeling of processes in a hohlraum and verification of theoretical models. 6. Conclusion Experimental results of studies of the 1 MA X-pinch X-ray source in a 0.15–500 keV spectral region have been presented. Implosion dynamics and radiative properties of symmetric and asymmetric Mo, Cu, and combined asymmetric Mo/W X-pinches with two or four wires were studied by spatially and time-resolved X-ray and optical diagnostics. The mechanisms of generation of hot spots and the dissipation of the initial discharge X-pinch energy are discussed. Future applications of 1 MA X-pinch as a 5–10 kJ sub-keV–10 keV radiation driver are considered. Acknowledgements The authors wish to thank L.I.Rudakov for helpful discussions. Work was supported by the DOE NNSA/NV Cooperative Agreement DE-FC52-01NV14050, and by Sandia National Laboratories under DOE contract DE-AC04-94AL85000.

References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17]

Zakharov S, Ivanenkov G, Kolomenskii A, et al. Sov Tech Phys Lett 1982;8(9):456. Kalantar D, Hammer D, Mittal K, Qi N, et al. J Appl Phys 1993;73:8134. Sinars D. Ph.D. dissertation, Cornell University, Ithaca, NY, 2001. Kantsyrev V, Bauer B, Shlyaptseva A, et al. Rev Sci Instr 2001;72:663. Pikuz S, Sinars D, Shelkovenko T, et al. Phys Rev Lett 2002;89:035003. Shlyaptseva A, Hansen S, Kantsyrev V, et al. Phys Rev E 2003;67:026409. Shlyaptseva A, Hansen S, Kantsyrev V, et al. Rev Sci Instr 2001;72:1241. Fedin D. Ph.D. dissertation. University of Nevada, Reno, NV, 2004. Kantsyrev V, Fedin D, Shlyaptseva A, et al. Phys Plasmas 2003;10:2519. Shelkovenko T, Sinars D, Pikuz S, Hammer D. Phys Plasmas 2001;8:1305. Shlyaptseva A, Fedin D, Hamasha S, et al. Rev Sci Instr 2003;74:1947. Kantsyrev VL, Fedin DA, Shlyaptseva AS, et al. Rev Sci Instr 2004;75:3708. Robledo A, Mitchell IH, Aliaga-Rossel R, et al. Phys Plasmas 1997;4:490. Ivanenkov GV, Pikuz SA, Sinars DB, et al. Plas Phys Rep 2000;26:868. Ivanenkov GV, Stepniewski W. Plasma Phys Rep 2002;28:814. Scudder DW, Shlachter JS, Hammel JE, et al. SIF Proceedings, 1991, 519. Safronova A, Kantsyrev V, Ouart N, et al. Spectroscopic modeling of radiation from Cu and Mo X-pinches recently produced on the UNR 1 MA Zebra generator, SIF Proceedings, 1991, JQSRT.