HT-7 Multipoint Nd Laser Thomson Scattering Apparatus
J.S.Mao, J.Y.Zhao, Y.D.Li, A.G.Xie, Z.S.Fang,
V.Sannikov* A. Gorshkov*
*
ASIPP, Hefei, Anhui, 230031, P.R.China.
PRC Kurchatov Institute, Moscow 123182. RUSSIA
Abstract A compact, low cost, multipoint Thomson scattering diagnostic for HT-7
Superconducting Tokamak has been in operation since 1999. Its capability of measuring
electron temperatures is 200ev to 2kev at densities of a few times 1012 cm-3, a spatial
resolution of 2.4cm for 5 spatial points and a temporal resolution of 1ms-1s for 8 time
points. The major components of the diagnostic system consist of 20-25J Nd:glass laser
with 35ns width (8 pulses per burst), KDP doubling unit, spherical mirrors of multipass
input optical system, wide-angle collection objective, bandpass glass filter for reducing
the stray light to zero, f/2.5 polychromator, the fiberglass collimator, the photomultiplier’s
box with electronic preamplifier, high gain and high signal/noise ratio, CAMAC data
acquisition and so on. The multipass optical system has been successful for increasing the
quantity of scattered photons by the probing laser beam passes 10 times through the
investigated plasma. The HT-7 Thomson scattering diagnostic has provided successfully
the information on two dimensional electron temperature in the plasma of HT-7 tokamak
with LHCD and IBW.
1. INTRODUCTION
Thomson scattering diagnostic has been an important and standard method for measuring
temperature and density profiles on all modern tokamaks, such as the TV Thomson
Scattering system on TFTR[1], the LIDAR system on JET[2] and the Nd:YAG laser
Thomson scattering system on DIII-D[3]. It has the attractive characteristics of not
perturbing the plasma to be investigated and of giving the absolute values of electronic
temperature Te and density ne. From the knowledge of temperature and density and of their
profiles， it is possible to derive such fundamantal parameters as poloidal bata value, the
energy confinement time and the effective ion charge of plasma. This paper describes the
design and performance of the compact, low cost, multipoint Thomson scattering
diagnostic on HT-7 Superconducting Tokamak and also describes the multipass optical
system, which can be useful for measuring temperature and density profiles during low
density discharge on HT-7 tokamak.
2. DESIGN OF THOMSON SCATTERING APPARATUS
When an electromagnetic wave is incident on a free charged particle, the particle is
accelerated and hence radiates. This interaction is generally refered to as Thomson
scattering when the scattering process can be analyzed classically (i.e.  << mc2). A
diagram of the scattering geometry is presented in fig 1. The theory of Thomson
scattering from a collection of free electrons is well developed[1,2]. The time-averaged
radiated power per unit solid angle is
dPs
  
c q2
 ( )( 2 ) 2 E 02 [n  (n  )] 2
d
8 mc
where
(1)
Ps = scattering power
= speed of light in vacuum
c
=
time average
E0 = the plane electromagnetic wave with electric field
Because the radiated power depends inversely on the mass of the charged particle the
radiation of the ions is negligible compared to the radiation of the electrons. In a plasma,
collective behavior only arrises over distances greater than the Debeye length,  D ,
inconherent scattering occurs for  in   D . In HT-7 Thomson scattering experiment
discussed in  in /  D  .01 , so that only the theory of incoherent scattering need be
considered.
The frequency spectrum of the scattered power for a plasma has been calculated by
several authors[2,3]. The scattering intensity is shown to be propotional to the dynamic
form factor S(k,  ).
S(k, ) 
1
( 2 ) 2



dtn k ( t ) exp( it )
2
 
n k   dr[  (r  r j ( t ))  n ] exp(  jk  r )
v
rj = location of the jth electron
n = mean density
< >= statistical average over an ensemble of systems
Physically, the observed frequency spectrum is du to a double Doppler shift of the
incident electromagnetic wave. The first shift is caused by the motion of the electrons
relation to the incident radiation and the second shift is due to the motion of the radiating
electrons relative to the observer. Fig. 2 shows the scattered power spectrum for various
electron temperatures.
For a plasma in thermodynamic equilibrium the frequency spectrum of the scattered
power is given by
where
S ( ) 
1
1
C me  
2
1
1
 2 4  2k T e 2
)
sin (
 0 2 me
exp[ 
8  02 sin 2
I( s )d  ne . I0 . T ..S()d

2
2
]
(2)
kTe
(3)
where
 0 = incident wavelength
 s = observed wavelenth
   s   i
 = scattering angle (see fig.1), =900 in HT-7 Thomson scattering system
ne = electron density
I0 = incident laser power
 T = differential scattering cross section
 = length of scattering volume
C = speed of light in vacuum
 = observed solid angle
2
The differential scattering cross section is defined as
 q2     2
d T
 [n  (n   )]
 
2 
d
 mc 
This is a very small quantity and for most Thomson scattering experiments only one out
of every 1013 incident photons is scattering by the plasma electrons.
For electron temperature up to 1 keV, the relativistic effects can be neglected, but
when temperature of electron is higher it must be taken into account using the formula for
thetheoretical spectrum[4.5]
2
2


   


B
  
   

  exp  


 
1  3.5
S( ) 
 7.6
1

2

 0
 0  sin ( / 2)Te   0  
sin(  / 2) Te 

  0  
 

A and B are numerical constants. The shape of the spectrum is changed due to the
relativistic effect(see fig 2)
When designing a Thomson scattering system, the principal problems are getting
sufficient scattering signal; suppression of stray light and suppression of plasma light.
Increasing the laser energy clearly helps the first of these problems. Increasing the laser
power is for increase the ratio of scattering signal to plasma light. For these reasons the Q
switched solid body laser has been the most commonly used light source for incoherent
Thomson scattering measurements. For HT-7 tokamak the Nd:glass laser is designed.
An e
3. THOMSON SCATTERING APPARATUS ON HT-7 TOKAMAK
The capability measuring electron temperatures of HT-7 Thomson scattering system is
200ev to 2kev at densities of a few times 1012 cm-3, a spatial resolution of 2.4cm for 5
spatial points and a temporal resolution of 1ms-1s for 8 time points. The major
components of the diagnostic system consist of 20-25J Nd:glass laser with 35ns width (8
pulses per burst), KDP doubling unit, spherical mirrors of multipass input optical system,
wide-angle collection objective, bandpass glass filter for reducing the stray light to zero,
f/2.5 polychromator, the fiberglass collimator, the photomultiplier’s box with electronic
preamplifier, high gain and high signal/noise ratio, CAMAC data acquisition. The
Nd:glass laser for HT-7 tokamak is not directly useful, because good photocathodes
doesn’t yet exist in the spectral region near micron. The advantage of working at 5300Å is
the sensitive for photocathodes (15%) more than at 6943 Å (ruby laser 3%). Reduction of
stray light was achieved both by a complicated baffles system, by using a spectrograph
with a high contrast, and especially by using a laser line rejection filter, so the stray light
was suppressed completely to zero. To increase signal-to-noise ratio and reduce the
contribution of plasma radiation, two Q switched was used inside laser system, to make
the width of laser pulse 35ns. The multipass optical system has been successful for
increasing the quantity of scattered photons by the probing laser beam passes 10 times
through the investigated plasma. Layouts of HT-7 T.S. diagnostic components is shown
in Fig.3.
3.1 ND-GLASS LASER SYSTEM
In the HT-7 T.S. system, the solid state Nd:glass laser is used. Nd-glass laser is consist
of an oscillator and an amplifier. The oscillator consists of a ring resonator and produces
very stable radiation during the burst. A Dove prism with Brester’s angles on both sides is
located inside the resonator( see fig.4). Due to Dove prism the cross-section of the light
beam is rotated inside the resonator along the optical axis of the oscillator to reduce the
thermal stress which are accumulated during the operation of the master oscillator. The
Q-switch is done by an plate of LiF crystal with colored centers. Near diffraction-limited
performance was achieved with a high quality a thermal phosphate Nd-glass rod (GLS-22).
The oscillator is constrained to operate in a TEMoo mode and produces about 50mJ. The
radiation line width is typically less than 0.01Å and pulse duration is about of 40 nsec. The
oscillator is capable to work with repetition rate 1Hz and in a single shot model. The
model with repetition rate 1Hz is very convenient for alignment or checking operation
system. The laser light amplifier is designed with telescopic resonator with 7 passes of the
beam and can operate in regimes with 1 or 8 pulses in the burst (see fig.5). This thermally
stable and compact design provides the multipulse regime with high output power. A
LiF-2 plate with colored is located inside the telescopic for Q switch and also for
protecting the master oscillator from back-emission. The Laser pulse duration is about of
30-35 nsec. The high power levers of Nd-glass laser make him well suited for harmonic
generation by passing the beam through nonlinear KDP crystal, which halves the
wavelength to 5300Å and produce the 4-5 joules of green light. The increasing laser power
density becomes more efficient by KDP’s nonlinear process.
3.2 PHOTODETECTOR’S SUBSYSTEM
The wide-angle collection objective (F=200mm) passes the scattered light to the image
fiberglass dissector. The five vertical fibers work as the input slits of the spectrometer. The
five sets are used for spatial resolution, each set corresponds to an individual spatial point
along the small radius of plasmacolumn. The polychromator is designed as high
luminosity system in Clocktype configuration[6] having a lage value of slit’s height and
f=1/2.5.( see fig. 6) Fiberglass couples the polychromater output to the photomulti-tubes.
The transmission of the guides is about 0.7-0.8.
The stray light of second harmonic (5300Å), which has passed through the collection
optic , is reduced by bandpass glass filter. Two pieces of the filters are installed before the
input fiberglass guides. The shap boundary of spectral curve (see fig.7) provides the
strong decreasing of stray light (near zero) and high transmission in measured part of the
scattered spectrum(> 90%). The bandpass glass filter allows to measuure the wide region
of electron temperature without loss of the signals. Another special glass filter is used to
cut off of fundamental laser radiation(1.06μ) before input window of the vacuum
chamber. This filter has high damage resistant for high power laser.
The photomultiplier tube (PMT-84) with S-20 photocathode is used for the detection of
scattered light. The quantum efficiency (near λ=5300Å ) is 10%-12%. The boxes are
designed to protect PMT from outside light and electromagnetic radiation. To improve the
operation of PMT with very fast impulse, the last dynodes of PMT are protected by the
capacities. The two stagers of amplifiers are coupled with PMT. The gain of preamplifier
is about 50 and the frequency range is about 0.15-35MHz in order to suppress the
radiation from plasma bremsstrahlung and macroscopic instability, which are two main
sources of unwanted radiation that decrease the ratio of signal to noise. A low frequency
limit is about 400KHz in order to pass low frequency signal from tungsten ribbon lamp
with mechanical chopper for calibration unit. The signals from PMT are sent to analyzing
electronic units though the gate in Data Acquisition system.(CAMAC 2250)
3.3 THE CALIBRATION SYSTEM
In order to evaluate the measured signals quantitatively, the spectral sensitivity of 25
channels and polychromator has to be attained in a preceding calibration measurement
with relatively high wavelength resolution. The light of a stabilized tungsten ribbon lamp
is passed through the polychromator with appropriate resolution. (see fig.8) Each
individual channel is measured as a function of wavelength. The light is chopped with the
duration about 60-70μsec, which is produced by mechanical chopper-wheel. The light
has to pass through the polychromator just as the Thomson scattered light. The correct
calibration of the photodetectors and processing system is needed about 1000 accounts.
4. MULTIPASS OPTICAL SYSTEM
Due to low value of T.S cross-section   6  10 cm , so the quantity of scattered
photons are too poor( N   ). In order to increase the scattered signal, it is possible to
pass few times the probing beam through investigated plasma region. Multipass Optics
System (MOS) is used to increase the accuracy of the T.S measurements in tokamak[7,8].
It is the first time to be used on HT-7 tokamak.(see Fig.3). The MOS is consist of two
spherical mirrors, which is formed a cavity(the radius of curvature R=1.6m, the mirror
diameter D=5.2cm,R>>D) and the mirrors are installed on a distance 3.2m,allowing to
pass through plasma volume 10 times.
25
2
0
0
5. MEASUREMENT OF TEMPERATURE PROFILE
The HT-7 is a superconducting tokamak. It was reconstructed from the original Russian
T-7 tokamak in 1994. The feedback control system to simultancously control plasma
current,density and displacement was developed and put into daily operation since Spring
in 1998. [9.10]The HT-7 tokamak has a major radius of R=1.22m, minor radius of
a=0.26-0.28m defined by a full circular limiter. The layout of HT-7 superconducting
tokamak is shown in fig9. The major research fields on the HT-7 are steady-state operation,
high-performance discharge, the fuelling study, LHCD and ICRF heating. After good
alignment and calibration, the HT-7 Thomson scattering diagnostic has provided
successfully the information on two dimensional electron temperature profiles in the
plasma of HT-7 tokamak with Ohmic, LHCD and IBW heating. Fig.10 shows the electron
temperature profile Te(r) during Ohmic heating (Ip~130KA,Ne~1.1, Te
(0)~0.75kev-0.85kev). Fig .11 shows the electron temperature profile Te (t) by 8 peices of
laser pulse. Fig 12 shows Te(0,t) during LHCD. The electron temperature of center plasma
is lager than 1KeV.
REFERENCES
1. M.J.Forrest ,et al Culham Plasma Physics Laboratory, Report CLM 107 (1970)
2. J. Sheffield, Plasma Scattering of Electromagnetic Radiation Academic Press, New York(1975)
3. K.W. Gentle, “TEXT User Handbook”, University of Texas, Fusion Research Center (1984)
4. B.F.M.Potz, J.J.H. Coumans and D.C. Schram, Phys. Flluids.,24(3),517,(1981)
5. S.E.Segre, Thomson scattering from a plasma in Course on Plasma Diagnostic and Data Acquisition
System (Ed.A.Eubank and E.Sindoni, International School of plasma Physics, Varenna,(1975).
6. D.Johnson, N.Bretz et al , Rev. Sci. Instrum. 57, 1856(1986).
7. H. Salzmann et al., Rev. Sci. Instrum. 59,1451 (1988)
8. C.L. Hsieh, J. Haskovec et al, Sci. Instrum.61,2855(1990)
9. Xie,J.K. and the HT-7 Group, in Plasma Physics and Controlled Nuclear Fusion Reserch 1996( proc.
16th Int. Conf. Montreal, 19960, Vol.1, IAEA(1997)685.
10. Li,J., et al., Nucl. Fusion 41 (1999) 1349.