PFC/JA-92-9
A Proton Activation Diagnostic to Measure
D-'He Reaction Yields
J. R. Lierzer, K. W. Wenzel, R. D. Petrasso,
D. H. Lo, J. W. Coleman, C. K. Li,
E. Hsieht and T. Bernatt
March 1992
Submitted for publication in: Review of Scientific Instruments
Plasma Fusion Center
Massachusetts Institute of Technology
Cambridge, MA 02139
t Lawrence Livermore National Laboratory
Livermore, CA 94550
This work was supported in part by U.S. DOE Grant No. DE-FG02-91ER54109
and LLNL Subcontract No. B116798
A Proton Activation Diagnostic to Measure
D- 3He Reaction Yields
J.R. Lierzer, K.W. Wenzelt, R.D. Petrasso, D.H. Lot,
J.W. Coleman, C.K. Li,
11IT Plasma Fusion Center, Cambridge, MA 02139
E. Hsieh and T.Bernat
Lawrence Livermore National Laboratory, Livermore, CA 94550
March 1992
Abstract
We are developing activation diagnostics for monitoring energetic chargedparticle fluxes in space and laboratory plasmas. More immediately, we plan
to use activation to measure the time-integrated proton flux from D- 3 He fusion reactions in Alcator C-MOD, providing a measure of the time averaged
D- 3 He fusion rate.
We demonstrated the technique's feasibility by inducing significant gamma
activity in a titanium sample exposed to D- 3 He protons created in our CockcroftWalton generator. The titanium target received a fluence of 5.5x109 protons
at 14.7 MeV (of order what a 3 cm 2 target should receive from one shot in
Alcator C-MOD) and became activated by the 48 Ti(pn) 48 V reaction.
The activity's spectrum from a high purity germanium (HPGe) detector showed the characteristic 0.984-MeV and 1.312-MeV lines of 4 8 V. The
measured activity agreed reasonably well with theory. An absence of activity at those energies before D- 3 He activation eliminated background or
D-D product-induced activity as the gamma source. We intend to repeat the
experiment with a chromium target to evaluate that material's diagnostic
potential.
1
Introduction
Activation analysis can be used to monitor energetic charged particles
from space and laboratory plasmas. We intend to initially apply activation
to the measure of time-averaged D-3 He proton yields and fusion rates in Alcator C-MOD. The application is particularly attractive for studying tokamak
plasma interiors since the high energy of the protons leaves them essentially
unconfined by the B field.
Chrien et al. were first to employ charged-particle activation as a tokamak
diagnostic in 1980.1 They qualitatively showed the presence of D-3 He protons
in PLT by activating titanium and iron targets through the "8Ti(p,n)4 sV and
"5Fe(p,n)"5 Co reactions.
We are extending this technique by using activation to quantitatively
measure time-averaged fluxes of 14.7-MeV protons striking activation targets
in tokamaks. We may then determine time-averaged D- 3 He reaction rates
from those fluxes (via orbit-unfolding computer codes). Additionally, we
shall use surface barrier diodes2 to measure the fluxes.
Using our Cockcroft-Walton generator to create D- 3 He reactions in a titanium target,3 we experimentally demonstrated the feasibility of measuring
the D- 3 He yield by way of the 48'Ti(pn) 4 V-induced gamma activity. We measured significant gamma activity in the target, exposed to 5.5x109 protons (a
fluence on order of that expected in a 3 cm 2 sample exposed to a single shot
in Alcator C-MOD). We shall activate chromium in our next experiment,
which should provide activity comparable to, or better than, titanium's.
Selecting Target Materials
A successful target material for 14.7-MeV proton activation must meet
several criteria. The activation spectrum should exhibit narrow, intense -y
lines, distinguishable from background (usually easy with our high purity
germanium's 1.8-keV resolution). The activation product must have a halflife long enough to allow sample retrieval and analysis before decaying away,
but short enough to produce measurably large activity. The reactive isotope
should also have a sufficiently large cross section near 14.7 MeV to produce
significant activity, and it should occur in large natural abundance to eliminate any need for expensive enrichment.
The isotope's reaction threshold should exceed 3 MeV, if possible, to
prevent D-D protons from contributing to activation (although these slower
2
protons could alternately be removed by a thin foil in front of the target).
After examining a number of potential target materials, including chromium,
iron, nickel, copper and silver, all of whose (p,n) cross sections were readily
available in the literature,4 - 7 we found that titanium and chromium best fit
the criteria, offering the most activity for a given proton fluence.
Titanium reacts with the 14.7-MeV protons through 48 Ti(pn)48 V. 4 The
48
V subsequently decays back to 4 'Ti by /+ and EC, emitting a 0.984-MeV
and a 1.312-MeV gamma with virtually each decay.",' The decay has a 16.2
day half-life, providing sufficient time for sample retrieval and analysis. 4 8 Ti
has a cross section reaching 500 mb at 11 MeV,4 and it occurs as 74% of
natural titanium, so D-He protons should produce significant activity in
a natural titanium sample. Fortunately, this reaction also has a 5-MeV
threshold.
Chromium also satisfies the criteria well. The isotope 52 Cr, having 84%
abundance,8 undergoes a (p,n) reaction with 14.7-MeV protons through
5 2Cr(p,n) 2 Mn." The 52Mn decays back to 52 Cr by P+ and EC, emitting a
1.434-MeV gamma with each decay, plus a 0.744-MeV and a 0.935-MeV
gamma (each in about 80% of the decays).' Although 32 Cr has a smaller
(p,n) cross section than 48Ti, peaking at only 210 mb at 14 MeV,6 its shorter
half-life of 5.7 days8 compensates for this and produces more activity for a
given proton fluence than titanium. The shorter half-life, though, still allows
plenty of time for sample retrieval and analysis. Finally, 12 Cr has a reaction
threshold around 5 or 6 MeV.'
Activation as a Tokamak Diagnostic
As an Alcator C-MOD diagnostic, we propose mounting titanium and
chromium targets, along with surface barrier diodes, outside the vacuum
vessel in a reentrant port near the plasma edge. 2 Protons from D- 3 He reactions should escape the plasma, then pass through a collimator and a very
thin (-50 Am) beryllium window (losing only 300 keV) to strike the activation targets. The activated targets will be removed after one or several
tokamak shots, and their activities counted for several hours with our high
purity germanium (HPGe) detector.
Assuming targets thick enough to totally stop the protons (800 pm of
titanium or 500 pm of chromium), we calculated that one out of every 800
D-He protons entering a titanium target creates one 4 8V nucleus, and one
3
out of every 2000 protons entering a chromium target creates one 52 Mn nucleus. Anticipating a D- 3 He proton flux in Alcator C-MOD of order l0
protons/cm 2 s for each 1/2-second shot, we expect a 3-cm
2
target of titanium
to obtain an activity on the order of 3000 decays per hour from a single shot.
A similar chromium target incurs an activity on the order of 4000 decays per
hour under the same conditions.
To measure the activity, we place each exposed target directly against
the face of our HPGe detector, giving nearly a 2 7r solid acceptance angle, to
maximize the count rate. In the case of titanium, we estimate the absolute
efficiency of our HPGe detector to be on the order of 3% at 0.984 MeV and
on the order of 2% at 1.312 MeV. Thus we should measure on the order of
100 counts and 70 counts per hour at the two titanium lines, respectively,
integrated over the width of each peak after a one-shot exposure.
By comparison, background radiation, which we subtract off, generates
about 330 counts per hour at 0.984 MeV and 190 counts per hour at 1.312
MeV, integrated over the same energy ranges as the emission peaks. If we
accumulate counts for, say, ten hours, the statistical errors in the final activity
measurements should be on the orders of 9% at 0.984 MeV, 10% at 1.312
MeV, and 6% for the two lines combined. Of course, the errors can be reduced
further by counting longer or by accumulating activity over several shots.
For activated chromium under the same conditions we estimate that the
HPGe detector has absolute efficiencies on the order of 4%, 3% and 2% at
the respective 0.743-MeV, 0.935-MeV and 1.434-MeV lines. Thus we should
expect to measure on the order of 130, 100 and 80 gammas per hour from
the three lines, respectively, integrated over each peak width after a oneshot exposure. Background radiation will provide about 410, 370 and 120
counts per hour at the three lines, respectively, integrated over the same
energy ranges as the emission peaks. The activity measurements should
then contain statistical errors on the order of 7% at 0.743 MeV, 9% at 0.935
MeV, 7% at 1.434 MeV, and 5% for the three lines combined after a ten hour
count.
Experimental Proof of Principle
To demonstrate the. diagnostic's feasibility, we activated an 890 pIm thick
titanium sample using D-3 He protons created in our Cockcroft-Walton generator. We sought to expose the sample to roughly the same proton fluence
4
expected from one shot in Alcator C-MOD, about 109 protons. Our generator can produce, on average, 500,000 D- 3 He reactions per second, so we
needed to maximize the sample's solid angle of exposure to the protons to
complete the experiment in a reasonable time.
We chose to generate the fusion reactions directly in the titanium sample
itself. We first implanted or "loaded" a titanium disk target with deuterium
by bombarding it with a 150-keV, 50-1A deuteron beam in our CockcroftWalton for a period of 10 hours. Since titanium has one of the largest affinities
for hydrogen of any metal, 0 much of the deuterium remained trapped in the
target. We next bombarded the target with a 3 He beam at 150 keV and 60
[A for an additional 6 hours. A small fraction (of order 1 out of every 10')
of the 3 He nuclei fused with a deuterium nucleus trapped in the target to
release a 14.7-MeV proton and a 3.5-MeV alpha.
Since the fusion reactions took place in a thin layer at the surface of the
target, and since the distribution of reaction products was nearly isotropic,
about half of all the fusion protons escaped from the target's surface while
the other half traveled into the target to react with titanium nuclei. A
surface barrier diode (SBD) mounted 120* from the beam axis monitored
the escaping proton flux and, thus, the fusion reaction rate. We stopped the
activation when the titanium accumulated a 5.5x10 9 proton fluence, then
counted the induced gamma activity for ten hours with the activated target
placed directly against the HPGe face.
Figure 1 shows the gamma spectrum of the titanium sample immediately after bombardment by 'He. The spectrum contains sharp gamma lines
at 0.984 MeV and 1.312 MeV, demonstrating that even one shot in Alcator C-MOD should provide enough titanium activation for detection by our
equipment.
We also took spectra of the sample before 'He bombardment, both before
and after deuterium loading, and those spectra showed no activity beyond
background at 0.984 or 1.312 MeV. Thus the detected activity could only be
due to titanium activation by D-IHe protons.
The activation line intensities agree reasonably well with our calculations (to within 50%). We suspect that the discrepancy is due to our overestimating the absolute efficiencies of our HPGe in the given geometry, i.e.
with the source placed directly against the detector face. We intend to measure the absolute efficiency for this geometry more accurately before our next
set of experiments.
5
Future Work
In the future, this experiment will be repeated with a chromium target
to also demonstrate that material's feasibility as a D-3 He activation diagnostic. Unlike titanium, though, chromium has almost no ability to absorb
hydrogen10 ; we must evaporate a very thin layer of titanium (about 3 pim)
onto the chromium surface to retain the deuterium during loading. Assuming
the success of these steps, the chromium activation experiment is essentially
the same as for titanium activation.
Conclusion
We have successfully demonstrated the feasibility of using proton activation of titanium to monitor the time-integrated proton flux from tokamaks. We exposed a titanium sample to a D-3 He proton flux (i.e. 14.7-MeV
protons) with a total fluence of 5.5x10' protons, comparable to the fluence
expected in a 3 cm 2 titanium target from a single Alcator C-MOD shot.
The protons reacted with the titanium by means of the 8Ti(pn)'4 V
reaction. We measured significant activity, which agreed reasonably well
with theory (to about 50%), from 41V decay at 0.984 MeV and 1.312 MeV.
Even though our loading process produced copious numbers of D-D fusion
neutrons (of order 10" total), no activation lines were seen at 0.984 MeV
or 1.312 MeV after the loading process and before activation. Hence the
observed activity comes entirely from D- 3 He proton activation. Also natural
chromium should reach activation levels from D-3 He protons comparable to,
or better than, that achieved by titanium.
Acknowldgements
We thank Dr. Fredrick Seguin and Ms. Cristina
ful reading of this manuscript. This work is supported
DE-FG02-91ER54109, by LLNL subcontract B116798,
Energy Postdoctoral Research Program,t and by U.S.
ship Program.*
Borris for the careby U.S. DOE Grant
by U.S. DOE Fusion
DOE MFET Fellow-
1. R.E. Chrien, P.L. Colestock, H.P. Eubank, J.C. Hosea, D.Q. Hwang,
J.D. Strachan, and H.R. Thompson, Jr., Phys. Rev. Lett. 46 535
(1981)
6
2. D.H. Lo, K.W. Wenzel, R.D. Petrasso, J.W. Coleman, C.K. Li, J.R.
Lierzer, E. Hsieh, and T. Bernat, submitted to Rev. Sci. Instrum.
(1992)
3. K.W. Wenzel, D.H. Lo, R.D. Petrasso, J.W. Coleman, C.K. Li, J.R.
Lierzer, C. Borras, T. Wei, E. Hsieh, and T. Bernat, submitted to Rev.
Sci. Instrum. (1992)
4. S. Tanaka and M. Furukawa, J. Phys. Soc. Japan 14 1269 (1959)
5. J. Wing and J.R. Huizenga, Phys. Rev. 128 280 (1962)
6. D.B. Smith, ChargedParticle Cross Sections: Neon to Chromium, LA2424, Los Alamos National Laboratory (1960)
7. F.K. McGowan, W.T. Milner and H.J. Kim, Nuclear Cross Sections
for Charged-ParticleInduced Reactions - Ni, Cu, ORNL CPX-2, Oak
Ridge National Laboratory (1964)
8. C. Lederer, J. Hollander, I. Perlman, Table of Isotopes, 6th Ed., John
Wiley and Sons, Inc., New York (1967)
9. R.L. Heath, Scintillation Spectrometry Gamma-Ray Spectrum Catalogue, 2nd Ed., Vol. 2, IDO-16880-2, Phillips Petroleum Co. (1964)
10. N.A. Galaktionowa, Hydrogen-Metal Systems Databook, Ordentlich Publishers, Holon, Israel (1981)
7
Activated Titanium Spectrum
Counts After 10 Hours
600
984 keV
Line
0
U
1312 keV
Line
300
900
1000
1300
Photon Energy (keV)
Figure 1. The gamma spectrum of the titanium target immediately after
'He bombardment. Note the two lines at 984 keV and 1312 keV, which
appear only after 'He bombardment. These lines result from D-3 He 14.7
MeV proton activation.
8