computerized 3-ghz multichannel soft x-ray diode spectrometer for high density plasma diagnosis

Nuclear Instruments and Methods in Physics Research B18 (1986) 101-110 101 North-Holland, Amsterdam

C O M P U T E R I Z E D 3-GHz M U L T I C H A N N E L S O F T X-RAY D I O D E S P E C T R O M E T E R F O R H I G H D E N S I T Y P L A S M A D I A G N O S I S *

G. P IEN and M.C. R I C H A R D S O N

Laboratory for Laser Energetics, Universily of Rochester. 250 E. Rit,er Road, Rochester, NY 14623-1299, USA

P.D. G O L D S T O N E , R.H. DAY, F. A M E D U R I and G. E D E N

Los A[amos National Laboratory, University of California, P.O. Box 1663, Los Alamos, NM 87545, USA

Received 1 April 1986 and in revised form June 24 1986

Investigations of high density plasma conditions, particularly those used in laser fusion and X-ray laser schemes often demand a quantitative characterization of the X-ray spectral emission in the 100-2000 eV range covering the peak of the thermal radiation energy distribution. We describe a 3 GHz bandwidth, 4-channel soft X-ray diode spectrometer coupled to a computerized digital oscillographic system. The characteristics of this system, and examples of its use in high density plasma experiments are described in detail.

1. Introduction

The analysis of transient high density plasmas such as those produced by high power lasers and pulsed power systems often demand a precise quantitative characterization of the emitted soft X-ray emission. In many cases, such a characterization provides a reliable estimate of the total radiated X-ray flux, and this has important implications on the radiation and electron thermal transport within the plasma, and affects the mechanisms responsible for energy absorption, wave propagation, and hydrodynamic instabilities in the plasma. In high density plasmas these processes occur on a nanosecond time scale. These considerations are behind the need for a calibrated, broadband soft X-ray (100-2000 eV) spectrometer having a signal bandwidth sufficient to resolve subnanosecond emission features.

There are, in principle, two architectural approaches to satisfying this requirement for subnanosecond resolution broadband soft X-ray spectrometry on high density experiments. In many respects they are complimentary. The first approach, implemented at several major high density plasma facilities, involves the use of a number of calibrated X-ray diodes [1] separately filtered for

* This work was supported at the Laboratory for Laser En- ergetics by the US Department of Energy Office of Inertial Fusion under agreement No. DE-FC08-85DP40200 and the Sponsors of the Laser Fusion Feasibility Project, and at Los Alamos National Laboratory by the US Department of Energy under contract W7405-EN6-36.

0168-583X/86/$03.50 �9 Elsevier Science Publishers B.V. (North-Holland Physics Publishing Division)

different bands of the soft X-ray region by the incorporation of either K-edge, L-edge, or M-edge filters [2-4] or specific bandpass X-ray mirrors coated with reflec- tive multilayer structures [5]. The stability of the simple metal cathodes used in the X-ray diode has been studied in detail [6]. With a few exceptions the calibration stability of most metal cathodes has been found to be constant over lengthy periods of time (several months) under the typical experimental conditions necessitating frequent vacuum cycling and occasional exposure to oxygen, air, and other elements. Unless special care is taken to minimize the X-ray response time, such as with the diodes developed by Kornblum [7] and Day [8], the majority of X-ray diodes have response times of nanoseconds or longer. A spectrometer incorporating diodes having subnanosecond time resolution signifi- cantly increases the complexity of the overall system. The diodes themselves must be impedance matched to maintain the signal integrity, and a GHz signal detection system must be used. The fastest readily available commercial oscilloscopes have detection bandwidths of - 1 GHz [9], though a special system has been built in France by Intertechnique and Hyperlec with a bandwidth - 7 GHz [10].

An alternative approach to establishing time-resolved broadband soft X-ray spectrometry with much greater ( - 10 ps) temporal resolution is through the incorporation of an X-ray streak camera [11[ with an X-ray energy filter system, such as a set of K-edge, L-edge, or M-edge filters [12] or an array of multilayer mirrors [13], or with a weak X-ray dispersive element

102 G. Pien et al. / Computerized spectrometer for high density plasma diagnosis

such as a free-standing grating [14,15]. The advantages of this approach are the greater tim~ resolution and the availability of greater spectral resolution ( ~ / A A - 100). However, systems incorporating streak cameras having soft X-ray photocathodes, usually utilizing thin layers of materials such as Al, Ag, or CsI, deposited on thin (< 1000 ,A,) plastic (polypropylene or Formvar) sub- strates, and transmissive dispersive optics are difficult to calibrate'absolutely [16,17]. Thus in many respects, these two approaches are complimentary. While a multichannel X-ray diode system provides the least calibration uncertainty, a streak camera-dispersive optics system [14] has greater spectral and temporal resolution.

In this paper we describe in detail a multichannel (4-channel) ultrafast (> 1 GHz) soft X-ray diode spectrometer and data recording system recently deployed

�9 at the OMEGA target irradiation facility [18]. This facility, primarily used for laser-fusion related experiments, is centered on a 24-beam, nanosecond duration, 351 rim, 2.5 ld laser system, which is focused onto targets 50-1000 gm in diameter by f/3.7, 600 mm focal length lenses in a vacuum chamber 172 cm in diameter. A broad array of X-ray diagnostics are now in use on this facility, including high resolution X-ray crystal and XUV grating spectroscopy, multichannel hard X-ray (1-300 keV) spectrometry, tin~e-resolved X-ray crystal spectroscopy, pinhole camera and microscopic X-ray imaging and time-resolved streak imaging.

The signals from the detectors in this spectrometer, which incorporate various thin film filters to dis- criminate between different X-ray spectral regions are recorded on a computerized multicbannel digital oscillographic system having an overall bandwidth in excess of 3 GHz. Section 2 describes the details of the detector and data acquisition system. In section 3 we give exam- pies of X-ray emission investigations in which this detector has been used. In section 4 we summarize the capabilities of the present system, and discuss the merits of making additional improvements.

signal recording, and a complete set of computer soft- w a r e to acquire, reduce, and analyze the diode signals has been developed. These are described in detail below.

2.1. The X-ray photodiodes

The spectrometer is built around four identical soft X-ray photodiodes. The principal components of each diode are shown in fig. 1. Each diode is composed of a 50 line/inch anode positioned some 0.050 in. in front of a micropolished (dry diamond machined) aluminum cathode with an active area of 0.235 in.2. Experience has demonstrated superior long term stability of micropolished aluminum photocathodes over rough finished aluminum photocathodes, with some loss of sensitivity for the micropolished cathodes. Similar aluminum photocathodes have been characterized to have nominal sensitivity from about 60 eV to over 2 keV [22]. They are operated with a cathode-to-anode potential of 1.8 kV. The photocathode is electrically and mechanically attached to a modified vacuum rf feedthrough connec-

�9 tor, in an evacuated body which also supports the anode. The electrical response characteristics of each diode have been measured using picosecond time domain reflectometry (using a Tektronix ,$-52 pulse gener- ator and a Tektronix S-6 sampling module in a Tektronix 7S12 Time Domain Reflectometer). Each diode dis- played a typical response bandwidth of 1.7 GHz. The overall diode sensitivity is dependent on the intrinsic photocathode response, the filter transmission function, and the attenuation of the gold mesh attenuators. The latter were measured using alpha particles and a surface barrier spectrometer and checked in situ by comparison with identically filtered unattenuated channels. The sensitivity of the A1 photocathodes was determined from detailed measurements made with dc X-ray sources and calibrated proportional counter detection [22]. The filter transmission functions were determined from tabulated absorption cross-sections and accurate mass

2. Multichannel GHz X-ray diode spectrometer

In this section we describe the details of a fast, four-channel soft X-ray spectrometer based on X-ray diode detectors filtered with different K-edge filters. This device was originally configured at Los Alamos National Laboratory with a 1 GHz bandwidth [19]. The system installed on the OMEGA facility is patterned on the 7-channel (3 GHz) MULTIFLEX [20] X-ray diode system developed at Los Alamos National Laboratory for laser fusion studies [21]. In deploying this system on the OMEGA facility substantial modifications have been made to the diode spectrometer to improve its detection bandwidth, a new computerized ultrafast (> 3 GHz) digital oscilloscopic system has been configured for

XRD

a n o d e ~ mesh

signal conneclor

AI photocathode (14-mm diam)

Fig. 1. Details of the soft X-ray diode used in the X-ray spectrometer.

2.2. The X-ray diode detector assembly and thicknesss measurements of the individual foils. The convolution of these three functions then provides an overall sensitivity function for each diode. Typical response functions for a set of four X-ray diode channels used in subkilovolt X-ray emission measurements of laser produced plasmas are shown in fig. 2.

10"1

Four X-ray diodes are assembled ,into an array to form the detector assembly. ' The X-ray diodes are mounted in a KeI-F [23] plate, which serves as the rear cover for the evacuated stainless steel detector housing

s- o 10 .2 o J= o.

c

~ 10 -3 tU

c @

10 .4

10-5 I I I , , , , I , , , , , , 1 , , , , , ,

Filter Stack: 1.997 mg/cm 2 Be

' ' ' ' " ' 1 ' ' ' " ' " 1 ' ' ' ' ' "

Channe l No. 1

10-5

10-2

10-3

10-4

, , ~ , I ~

Filter Stack: 191 ~g/crd 2 CloHsO 4 (mylar) -- 12 ~g/cm 2 AI

8.1% Transmission Au Mesh

Channe l No. 2

G. Pien et a L / Computerized spectrometer for high density plasma diagnosis 103

l l l l , ] l i i i l l

r .

2 o

e -

o

U J

n -

10-2

10-3

10-4

10-5 10"2

Filter Stack: 60 ~,g/cm 2 CH 2 - 210 pg/cm 2 AI


, , , , 1 1 , I i , , , 1 1 1 , I , i , , , r ,

Channe l No. 3

A , , , / , & , l . . . . I

10 -1 100

Energy (keY)

10-2

10-3

10 .4

. . . . . . lo-S 101 10 .2

Filter Stack: 49 pg/cm 2 CH20 �9 20 pg/cm 2 AI


i i I l l I H I ' ' ' ' ' ' ' ' l ' ' ' ' ' ' ' ~ ' ~

Channe l No. 4

i i i i i i J

10 -1 10 0 101

Energy (keV)

Fig. 2. Detailed response curves for the four diodes used in the X-ray spectrometer. These are a convolution of the AI photocathode sensitivity, the filter transmission function, and the attenuation of the anode and the attenuator.


and provides electrical isolation. The detector head con- sists of four individual channels each comprising an X-ray attenuator mesh, a filter sta~k, sweeper magnet and collimator tube, and the X-ray diode, as shown in fig. 3. Sweeper magnets are installed to deflect the electrons out of the X-ray beam. In principle, the long time of flight of the charged particles compared to that of the X-rays should separate any background due to pa~:ticles. The detector assembly incorporates the four diodes into a single, small diameter head which attaches to the OMEGA target chamber, via a 1.5 m flight tube and a gate valve. The length of the flight tube is adjusted to ensure optimum soft X-ray radiation density on the surface of the diode photocathodes. The flight tube and detector head are evacuated to target chamber vacuum ( - 4 • 10 -6 Torr). The gate valve and vacuum interconnects allow inter-shot modification of filters of individual channels.

2.3. Signal recovery

The signal acquisition system used with this X-ray diode spectrometer is based on four Lockheed modified Tektronix LM7912A transient digitizers [24]. These digital oscilloscopes have been modified to exploit the inherent bandwidth of the Type 7910 Cathode Ray

Tube. (This modification resulted in the designation change from Tektronix R7912 to Lockheed/Tektronix LM7912A.) The LM7912A transient digitizer has a bandwidth of 3.5 GHz at the 3 dB point, with less than 5% undershoot and overshoot, and a 12 bit output (2 mV/bit), operating under CAMAC control via a Tektronix 7912DPO bus interface. The signal from each X-ray diode is fed via 3 m of RG-214 coaxial cable to a high speed biasing capacitor which is connected through 1 m of 50 ~ semi-rigid coaxial cable to each of the digital oscilloscopes. High bandwidth signal attenuators are optionally inserted into the semi-rigid leg of the system, as the LM7912As have fixed vertical amplification. The LM7912A digitizers and their CAMAC crate and controller, as well as timing control hardware and an onboard calibration pulser, are positioned some 3 m from target center inside an EMI-shielded rack. A schematic of the circuit from the X-ray diodes to the transient digitizer is shown in fig. 4.

In order to calibrate the bandwidth response of the system, a Tektronix S-52 70 ps rise time pulser was

�9 substituted in place of the X-ray diode in each acquisition arm. Traces were then acquired of this fast rise pulse using the acquisition system in standard calibration mode (i.e., normal acquisition sequence, with a generated trigger and signal). As shown in fig. 5, the

XRD

Anode Mesh

AI Photo / Cathode

(14 mm diam

Signal Connector

Collimator ~ ":~ ' Tube j . ~ ~ , ~\~'~

Sweeper ~ Magnets ~;.

Au Mesh (attenuator)

Bandpass Filter

Fig. 3. Configuration of the four detector array in the soft X-ray spectrometer housing�9

G. Pien et al. / Computerized spectrometer for high density plasma diagnosis 105

target

x-ray attenuator

tilter

I sweeper I j / : magnet j . . . ~ " I

anode e- ~ I mesh I hv

I I

sweeper I magnet ~ : .

AI photocathode

L high voltage I - 1.8 kV

f LMT912A transient digitizer

I ~-~-~ attenuati~ ~ ~'~

Fig. 4. Schematic of the signal acquisition details of each soft X-ray diode.

measured 10-90% rise time of this 200 mV 70 ps rise time input pulse is observed as a 130 ps rise time 200 mv pulse. Thus we determined that each acquisition arm performs at a nominal 3.1 GHz bandwidth response at the 3 dB signal degradation point.

2.4. Data recovery

The four LM7912A transient digitizers are all controlled over a Tektronix CP bus, which connects via a Tektronix 7912DPO controller to a fiber-optically lin- ked CAMAC serial highway. The CAMAC crate and CP bus hardware are housed along with the LM7912A digitizers in the EMI shielded rack, while the fiber-optic link leads down through the concrete target bay floor to the OMEGA target diagnostics processor, a DEC PDP 11/23 + running the RSX-11M operating system [25]. This system is outlined in detail in fig. 6, which shows a

Input Pulse: 70-ps Rise Time 200 mV from Tektronix S-52 pulser

f I I I I I

J I I I I ] I

;hannel No. 3 250 mV/dlv 305 ps/dlv

10-90% Rise Time "- t30 ps

Bandwidth 2.66 GHz @ 3-dB point

Shot Configuration

Calibration Configuration

Fig. 5. Measurement of the bandwidth response of the signal acquisition system. The X-ray diode was replaced with a Tektronix S-52 70 ps, 200 mV pulser. The recorded signal has a risetime of -130 ps, while, deconvoluting the input signal,

implies a signal bandwidth - 3.1 GHz at the 3 db point.

schematic of the overall architecture of the transient digitizer acquisition system. The OMEGA target diagnostics processor interconnects into the OMEGA laser system computer network via DECNET [25]. Since many of the OMEGA target diagnostic systems operate under CAMAC, a modular, parameter driven software sched- uling system is used under which individual FORTRAN routines are separately controlled. Modules have been written which initialize and arm the LM7912A transient digitizers, acquire baseline data just prior to the shot, re-arm for the event, acquire and archive the shot data, and display and perform preliminary data reduction. These operations are done automatically. Alteration of the vertical amplification of the LM7912A digitizer is facilitated by an attenuation selection. The system may be disabled when desired via a menu driven, parameter configured operator interface. This interface is common to all of the OMEGA computer-controlled target diagnostic systems.

Initial synchronization of the data acquisition cycle to the timing sequence of the OMEGA shot diagnostic acquisition system is done through the central computer system, while final synchronization to the shot is accomplished with a hard wired timing signal derived from a fast photodiode situated at the front end of the OMEGA laser chain. This photodiode signal is transmitted on a dedicated HELIAX [26] cable to the EMI shielded acquisition system and is then run through several delay stages before being used to trigger the separate LM7912A digitizers in synchronism. On the fastest time base setting used, the acquisition window of each LM7912A digitizer is only 2 ns wide. Digitizer triggering synchronization is accomplished by substitut- ing a signal from a photodiode mounted on the OMEGA target chamber for each of the X-ray diode signals. The substitute diode samples a lower power laser pulse propagated from the oscillator through the OMEGA laser system, which is used as a timing mark to synchro- nize the triggering of transient digitizers to the laser target interaction. Once each digitizer channel is synchronized to this signal, an offset delay is added to each

106 G. Pien et al. / Computerized spectrometer for high density plasma diagnasis

I l I I,

e

i I

I I I I I I I I t__'

-I

~ l - - - 1 f - - -

I I I L _ _ I I . . . .

h- I

~ F - ] I - - - I uI i l

L _ J ~ . . . . 1

~r-I F--' I i i 1 L _ J 1 . . . . ~ i ~

i t I ] E

o

E

w

@

8 0

0

o

E

8

e-,

o

G. Pien et al. / Computerized spectrometer/or high density plasma diagnosis 107

digitizer trigger input line to account for the additional flight time of the X-rays from the target to the X-ray diode detector array. This system therefore ensures minimal trigger jit ter in signal acquisition of the four digitizers. Observation of numerous shots indicates a triggering jitter of < 100 ps, but to date no common timing mark has been added to the input signals to facilitate more accurate unfolding of the time histories from the multiple detectors.

During each shot cycle, after being automatically armed for a single sweep, the LM7912A digitizers are triggered by the photodiode monitoring the laser pulse. The writing beam of the digitizer encodes the trace on a 512 by. 512 diode array in real time. The diode array is then recharged by a reading beam scanning the array over the next 85 ms. This data is stored in on-board memory, and each digitizer awaits recovery of the data by the controlling software routines. Two files are being written for each OMEGA laser target shot, one for baseline data and one for event data. Each file contains information identifying the shot number and recovery system parameters of each channel, as well as data sets for each channel shot. Preliminary reduction is performed under full automatic control. Scaling and cor- rection of the data is made for each channel using previously determined calibration data for the channel. The signal duration (fwhm), the peak voltage, and integrated charge are calculated from the digitized signal and baseline record. The baseline and event curves are then presented as a corrected plot with channel parameters and reduced data on a video display terminal, and a hardcopy of the data is created. The data files for that shot are then transmitted over the OMEGA data network for archiving along with other data files for that shot.

3. Soft X-ray measurements on laser produced plasmas

The soft X-ray diode spectrometer has been used in several experimental investigations on the OMEGA laser system. In this section we discuss interpretation of the time and energy resolved data acquired during two such experimental programs. Both of these experiments used the soft X-ray diode spectrometer in the configuration and mode of operation described in section 2, with the only variation in operation being different electrical attenuation values in front of the digitizers to accom- odate the particular X-ray fluences produced on each experiment.

3.1. X-ray conversion studies

The first [27] of these experiments was an investigation of X-ray conversion in plasmas produced from solid high-Z (Au) coated targets at intensities in the

range of 1.0 x 1013-4.0 x 10 is W / c m 2. The principle purpose of this investigation was to compare the total X-ray emission spectrum and X-ray conversion ef- ficiency with the results of previous planar target experiments, and with detailed hydrodynamic computer codes, in order to gain a better understanding of the radiation and plasma dynamics of the interaction.

As indicated by fig. 2, the energy resolution on this spectrometer is obtained by using different filter values for the four channels. The CHzO/AI filter used in channel 4 provides information across the 150 eV to 1.5 keV range, CH2/AI in channel 3 emphasizes the 700 eV region near the Au N-band emission, Mylar /Al in channel 2 observes 1 keV and the Au O-band emission at 250 eV, and the Be filter in Channel 1 accepts emission above 800 eV. In addition, the four channels are used to provide absolute calibration of a higher-resolution, film based soft X-ray grating spectrograph [28].

The signals from the four channels may then be iteratively deconvolved with the spectral response of each detector to determine the spectral characteristics of the emission. Alternatively, these signals may be used to define the absolute calibration of spectra obtained from higher resolution instruments.

Although the soft X-ray diode spectrometer provides only modest spectral resolution, the high temporal resolution, absolute calibration, and stability of each detector provides for reliable, shot-to-shot comparisons of a variety of target parameters. In addition the ease with which this system can be synchronized to the irradiating laser pulse permits accurate temporal comparison of the spectra1 characteristics of the X-ray emission during the interaction with the predictions of the hydrodynamic codes. At present, no other diagnostic system can em- body all these particular characteristics.

Fig. 7 gives an example of the four channel output from one of the target shots in this series. This particular target was coated with a thin (0.035 Fm) layer of gold. The sharp termination in the rise of the signal observed in channels 2-4 results from the penetration of the ablation region of the plasma through the Au coating into the CH target. Note also that in order to calculate the pulse duration (fwhm) and integrated charge we have also plotted the baseline response of the system acquired just prior to the event.

3.2. Time resolved X-ray spectrometry of imploding laser fusion targets

The soft X-ray diode spectrometer has also been used to acquire data in fusion experiments with DT filled glass microbaUoon targets. These experiments [29] were performed to examine the degree to which sym- metric implosions of spherical shell targets could be driven by the 24, 351 nm beams of the OMEGA facility. One such series of target shots utilized large targets


Shot No. 11547

I r rad ia t i on Cond i t i ons :

24 Beams UV

Energy on ta rge t : 1.7 kJ

Energy ba lance on target : 8.2% In tens i ty : 4.49 • 1014 W / c m 2

Pulse w id th : 600 ps

, I I I I I I I

2",._ I I l i I 1 I I

! 1 I I i l f i

l l l l l l l l

Channel No. 1 1.997-mg/cm 2 Be, 100% transmission 22.7 V/dtv 241 ps/dlv Bias: - 1800 V Integral Charge: 0.106 • 10 -8 C Peak Voltage: 70.9 V FWHM: 720 ps

Channel No. 3 60-~g CH 2 + 210-pg AI, 13.8% transmission 13.8 V/dlv 305 ps/dlv Bias: - 1800 V Integral Charge: 0.595 • 10 .9 C Peak Voltage: 30.1 V FWHM: 971 ps

Ta rge t

R o = 221 ~m

t l f l l l l l l

3so-A Au

3 l l l | l l l l

I I I I I l I I I

l [ l l l | l l l

Channel No. 2 191-,ug mylar + 12-,ug AI, 8.1% transmission 13.3 V/div 362 ps/div Bias: - 1800 V Integral Charge: 0.425 ~. 10 -9 C Peak Voltage: 17.1 V FWHM: 1.29 ns

Channel No. 4 49-,ug CH20 + 20-~g AI, 6.2% transmission 12.5 V/div 351 ps/div Bias: - 1800 V Integral Charge: 0.719 '~ 10 "9 C Peak Voltage: 29.9 V FWHM: 1.31 ns

Fig. 7. Typical spectrometer data acquired in experiments on high-Z coated targets.

S h o t No. 11633

I r r a d i a t i o n C o n d i t i o n s :

24 b e a m s UV

E n e r g y on ta rge t : 2005 J

B e a m b a l a n c e on ta rge t : 7%

Pu lse w i d t h : 600 ps

' ' ' ' ' ' ~ ' Channel No. 1 1.997-mg/cm 2 Be 100% transmission 22.7 V/div 241 ps/div Bias: - 1800 V Integral Charge: 0.111 / 10 .8 C Peak Voltage: 44.1 V FWHM: 1.35 ns

Channel No. 3 60-jug CH 2 + 210-~g AI 13.8% transmission 13.8 V/div 305 ps/div Bias: - 1800V Integral Charge: 0.47 ~ 10 .9 C Peak Voltage: 18.3 V FWHM: 1.46 ns

l ~ i I I I I I I

i i i I | I I I !

T a r g e t ~.~,~,

,o,o

1 I i I I I 1 I i

Glass she l l , 1 . 75 -~m t h i c k

Ro -- 3 6 6 / L m

I I I I I I I I I

I I I I i i 1 |

Channel No. 2 191-~g mylar + 12-#g AI 8.1% transmission 13.3 V/div 362 ps/div Bias: 1800 V Integral Charge: 0 .332 . 10"9C Peak Voltage: 10.5 V FWHM: 1.68 ns

i , i i , , ' ' ' Channel No. 4 49-,ug CH20 + 20-#g AI 6.2% transmission 12.5 V/div 341 ps/div Bias: - 1800 V Integral Charge: 0.695 ~. 10 .9 C Peak Voltage: 22.3 V FWHM: 1.53 ns

L 1

Fig. 8. Typical spectrometer data recorded on laser implosion experiments with large initial aspect ratio targets imploded with modest ( < 2 • 10 z4 W / c m 2) irradiance.

G. Pien et aL / Computerized spectrometer for high density plasma diagnosis 109

having high initial aspect ratios ( R / z a R - 200). These 700-800 gm diameter targets were irradiated at modest intensities (2 x 1014 W/cm2). The soft X-ray emission generated in these experiments has two principal sources. This is clearly evident from the set of 4 channel X-ray diode data for a typical shot in this series in fig. 8. In the initial stages of the interaction the laser light heats the solid glass shell of the target, resulting in X-ray emission from the hot dense coronal plasma surround- ing the target. As the shell accelerates inward, it decom- presses and the coronal plasma becomes somewhat more tenuous, resulting in a progressive reduction in the X-ray emission. Then, as the shell stagnates close to the center .of the target, the temperature of the shell is heated producing a second burst of emission. This is evident in all four of the spectral channels in fig. 8. Deconvolution of the signal data from each of these detectors can thus provide valuable information on the coronal conditions during the implosion, on the X-ray emission at peak compression and on the implosion time. For example, in the data shown in fig. 8, the implosion time of the target can be deduced from the temporal separation ( - 7 0 0 ps) of the two peaks in X-ray emission.

4. Summary

We have described in detail the four channel soft X-ray diode spectrometer and data recovery system activated on the OMEGA irradiation facility for laser fusion and high density plasma physics experiments. This instrument is now providing calibrated spectral and temporal data of soft X-ray emission from a number of different target experiments.

The present instrument provides modest ( - 160 ps) time-resolved data of X-ray emission produced by nanosecond laser 13earns. However, the current system bandwidth is somewhat limited by the signal bandwidth of the X-ray diodes. Anticipated future improvements include the implementation of diode assemblies having a bandwidth greater than the signal bandwidth of the recording digitizers. This would permit the detection of X-ray emission with a time resolution of - 100 ps.

The four channel diode system described here uses foil filters for spectral resolution, permitting a coarse interpretation of the X-ray spectrum. Future improvements include the addition of more diode channels and the use of multilayer X-ray optical components to provide higher resolution spectral discrimination.

This work was supported by the US Department of Energy Office of Inertial Fusion at the University of Rochester under agreement No. DE-FC08-85DP40200 and at the Los Alamos National Laboratory under contract W7405-ENG-36; and by the Laser Fusion

Feasibility Project at the Laboratory for Laser Energet- ics which has the following sponsors: Empire State Electric Energy Research Corporation, General Electric Company, New York State Energy Research and Devel- opment Authority, Ontario Hydro, Southern California Edison Company, and the University of Rochester. Such support does not imply endorsement of the con- tent by any of the above parties.

References

[1] For a comprehensive review of the characteristics of soft X-ray detectors see: R.H. Day, Proc NATO Advanced Study Institute on High Density Plasma Diagnostics, Pascali, Italy (1983) to be published.

[2] J.L. Gaines, H.N. Kornbhim, V.M. Slivinsky, Lawrence Livermore Laboratory Rep. UCRL. 75987A (1974) unpublished.

[3] E.J.T. Burns Adv. X-Ray Anal. 18 (1974) 117. [4] R.H. Day, Bull. Am. Phys. Soc. 22 (1977) 1196. [5] A recent detailed review of the properties of X-ray mirrors

employing multilayer coatings is given in: T. Barbee, Proc SPIE 563 (1985) 2.

[6] B.L. Henke, J.A. Smith, and D.T. Attwood, J. Appl. Phys. 48 (1977) 1852.

[7] H.N. Kornblum and V.W. Slivinsky, Rev. Sci. Instr. 49 (1978) 1204.

[8] R.H. Day, P. Lee, D. VanHuylsteyn and T.E. Elsberry, presented at APS Div of Plasma Phys. Ann. Mtg. (1977).

[9] For example Tektronix 7104, oscilloscope 1 GHz. [10] Hypeflec has built a CRT possessing a 7 GHz which has

been incorporated into an oscillographic system by Inter- technique as model IN 7000.

[11] For reviews of current developments on X-ray streak cameras see Proc. the 12-16th Congresses on High Speed Photography published by SPIE, Bellingham, Washington.

[12] R.S. Marjoribanks, M.C. Richardson, J. Delettrez, S. Letzring, W. Seka, and D.M. Villeneuve, Opt. Commun. 44 (1982) 113.

[13] G.L. Stradling, T.W. Barbee, R.L. Henke, E.M. Campbell, W.C. Mead, Proc. Top. Conf. on Low Energy X-Ray Diagnostics, (AIP Conf. Proc. 75 (1981) 292.

[14] A.M. Hawryluk, N.M. Ceglio, R.H. Price, J. Melngaillis and H.I. Smith, Proc. Top. Conf. Low Energy X-Ray Diagnostics eds., D.T. Attwood and R..L. Henke, AIP Conf. Proc. 75 (1981) 286.

[15] M.C. Richardson, R.S. Marjoribanks, S.A. Letzring, J.M. Forsyth, and D.M. Villeneuve, IEEE J. Quant. Electr. QE-19 (1983) 1861.

[16] P.A. Jaanimagi and B.L. Henke, Rev. Sci. Instr. 56 (1985) 1637.

[17] G. Stradling and J. Chmielewski, Proc. Ann. Mtg. Opt. Soc. Am., Optics News (Sept. 1985) p. 104.

[18] J.M. Soures, R.J. Hutchison, S.D. Jacobs, L.D. Lund, R.L. McCrory and M.C. Richardson, Proc. 10th Ann. Mtg. of Nucl. Soc., Philadelphia PA (1983) p. 1392.

[19] The original 4-channel (MINIFLEX) soft X-ray diode spectrometer was developed and built at Los Alamos National Laboratory by EG & G Energy Measurements Group.

110 (7,. Pien et al. / Computerized spectrometer for high density plasma diagnosis

[20]' R.H. Day, R. Hackaday, F.P. Ameduri and E.W. Bennett, Bull. Am. Phys. So(:. 25 (1980) 962.,

[21] P.D. Rockett, W.C. Preidhorsky and D.V. Giovanielli, Phys. Fluids 25 (1982) 1286.

[22] R.H. Day, P. Lee, E.B. Saloman, and D.J. Nagel, J. Appl. Phys. 52 (1981) 6965.

[23] Ke!-F, proprietary product of Minnesota Mining and Manufacturing Co.

[24] This modification was implemented by Lockheed Palo Alto Research Laboratories, Palo Alto, California.

[25] DEC, PDP 11/23 + DECNET, and RSX-11M are registered tr/ldemarks of the Digital Equipment Corporation, Maynard, MA, USA.

[26] HELIAX is a registered trademark of the Andrew Corpo- ration, Orland Park, IL, for a semi-rigid, air-filled coaxial cable.

[27] P.D. Goldstone, S.R. Goldman, J.A. Cobble, A. Hauer, G. Stradling, W.C. Mead, M.C. Richardson, R.S. Marjori- banks, G. Pien, O. Barnouin, and B. Yaakobi, Bull. Am. Phys. Soc. 30 (1985) 1364.

[28] J. Cobble, P.D. Goldstone, A. Hauer, S.R. Goldman, W.C. Mead, G. Stradling, M.C. Richardson, E. Kowaluk, Bull. Am. Phys. Soc. 30 (1985) 1364.

[29] M.C. Richardson et al., Laser Interaction and Related Plasma Phenomena, eds., H. Hora and G. Miley vol. 7 (Plenum, New York, 1986) in press.

computerized 3-ghz multichannel soft x-ray diode spectrometer for high density plasma diagnosis

Documents