Status of modern concepts of high power 14 MeV neutron source
Different approaches to the problem of design and construction of high power 14 MeV neutron sources are described. It has been already well recognized that the problem of tests of existing structural materials so as problem of creation of new ones for future fusion power plants should be solved in t...
Збережено в:
| Дата: | 2000 |
|---|---|
| Автор: | |
| Формат: | Стаття |
| Мова: | English |
| Опубліковано: |
Національний науковий центр «Харківський фізико-технічний інститут» НАН України
2000
|
| Назва видання: | Вопросы атомной науки и техники |
| Теми: | |
| Онлайн доступ: | http://dspace.nbuv.gov.ua/handle/123456789/82369 |
| Теги: |
Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
|
| Назва журналу: | Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| Цитувати: | Status of modern concepts of high power 14 MeV neutron source / E. P.Kruglyakov // Вопросы атомной науки и техники. — 2000. — № 3. — С. 54-59. — Бібліогр.: 37 назв. — англ. |
Репозитарії
Digital Library of Periodicals of National Academy of Sciences of Ukraine| id |
irk-123456789-82369 |
|---|---|
| record_format |
dspace |
| spelling |
irk-123456789-823692015-05-30T03:01:34Z Status of modern concepts of high power 14 MeV neutron source Kruglyakov, E. P. Мagnetic Confinement Different approaches to the problem of design and construction of high power 14 MeV neutron sources are described. It has been already well recognized that the problem of tests of existing structural materials so as problem of creation of new ones for future fusion power plants should be solved in the nearest years. It is shown that among plasma-based NSs the neutron sources on the basis of mirror machines are able to solve the problems of materials tests with lowest capital and operating cost. At present, the most advanced candidate both: from experimental and theoretical point of view is the Gas Dynamic Trap (GDT). Recent experiment with oblique injection of fast deuterium atoms in warm target hydrogen plasma has demonstrated a good agreement with results of calculations as from the viewpoint of spatial distribution of the neutrons of D-D reaction, so from the viewpoint of absolute value of the neutron flux density. It should be noted that the GDT-based NS is the object of interest even with existing, at present, plasma parameters (more exactly the electron temperature of the target plasma should be increased two times in comparison with the present level). The increase of the temperature from 130 eV up to 250 eV makes it possible to produce a moderate neutron flux density only several times less than that in the full-scale projects. An obvious advantage of this moderate version of the NS consists in the fact that the plasma physics database for such a source has already existed. Thus, the NS with neutron flux density of order of 200-400 kW/m2 can be designed and constructed on the basis of the present day experience. As the next step of such approach significant increase of neutron flux density will be possible in result of increase of power of D-T neutral beam injection. 2000 Article Status of modern concepts of high power 14 MeV neutron source / E. P.Kruglyakov // Вопросы атомной науки и техники. — 2000. — № 3. — С. 54-59. — Бібліогр.: 37 назв. — англ. 1562-6016 http://dspace.nbuv.gov.ua/handle/123456789/82369 539.125.5; 539.164 en Вопросы атомной науки и техники Національний науковий центр «Харківський фізико-технічний інститут» НАН України |
| institution |
Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| collection |
DSpace DC |
| language |
English |
| topic |
Мagnetic Confinement Мagnetic Confinement |
| spellingShingle |
Мagnetic Confinement Мagnetic Confinement Kruglyakov, E. P. Status of modern concepts of high power 14 MeV neutron source Вопросы атомной науки и техники |
| description |
Different approaches to the problem of design and construction of high power 14 MeV neutron sources are described. It has been already well recognized that the problem of tests of existing structural materials so as problem of creation of new ones for future fusion power plants should be solved in the nearest years. It is shown that among plasma-based NSs the neutron sources on the basis of mirror machines are able to solve the problems of materials tests with lowest capital and operating cost. At present, the most advanced candidate both: from experimental and theoretical point of view is the Gas Dynamic Trap (GDT). Recent experiment with oblique injection of fast deuterium atoms in warm target hydrogen plasma has demonstrated a good agreement with results of calculations as from the viewpoint of spatial distribution of the neutrons of D-D reaction, so from the viewpoint of absolute value of the neutron flux density. It should be noted that the GDT-based NS is the object of interest even with existing, at present, plasma parameters (more exactly the electron temperature of the target plasma should be increased two times in comparison with the present level). The increase of the temperature from 130 eV up to 250 eV makes it possible to produce a moderate neutron flux density only several times less than that in the full-scale projects. An obvious advantage of this moderate version of the NS consists in the fact that the plasma physics database for such a source has already existed. Thus, the NS with neutron flux density of order of 200-400 kW/m2 can be designed and constructed on the basis of the present day experience. As the next step of such approach significant increase of neutron flux density will be possible in result of increase of power of D-T neutral beam injection. |
| format |
Article |
| author |
Kruglyakov, E. P. |
| author_facet |
Kruglyakov, E. P. |
| author_sort |
Kruglyakov, E. P. |
| title |
Status of modern concepts of high power 14 MeV neutron source |
| title_short |
Status of modern concepts of high power 14 MeV neutron source |
| title_full |
Status of modern concepts of high power 14 MeV neutron source |
| title_fullStr |
Status of modern concepts of high power 14 MeV neutron source |
| title_full_unstemmed |
Status of modern concepts of high power 14 MeV neutron source |
| title_sort |
status of modern concepts of high power 14 mev neutron source |
| publisher |
Національний науковий центр «Харківський фізико-технічний інститут» НАН України |
| publishDate |
2000 |
| topic_facet |
Мagnetic Confinement |
| url |
http://dspace.nbuv.gov.ua/handle/123456789/82369 |
| citation_txt |
Status of modern concepts of high power 14 MeV neutron source / E. P.Kruglyakov // Вопросы атомной науки и техники. — 2000. — № 3. — С. 54-59. — Бібліогр.: 37 назв. — англ. |
| series |
Вопросы атомной науки и техники |
| work_keys_str_mv |
AT kruglyakovep statusofmodernconceptsofhighpower14mevneutronsource |
| first_indexed |
2025-07-06T08:52:32Z |
| last_indexed |
2025-07-06T08:52:32Z |
| _version_ |
1836886998991765504 |
| fulltext |
UDC 539.125.5; 539.164
Problems of Atomic Science and Technology. 2000. N 3. Series: Plasma Physics (5). p. 54-59 54
STATUS OF MODERN CONCEPTS OF HIGH POWER 14 MeV NEUTRON SOURCES.
Edward P.Kruglyakov
Budker Institute of Nuclear Physics, 630090, Novosibirsk, Russia
Different approaches to the problem of design and construction of high power 14 MeV neutron sources are
described. It has been already well recognized that the problem of tests of existing structural materials so as
problem of creation of new ones for future fusion power plants should be solved in the nearest years. It is shown
that among plasma-based NSs the neutron sources on the basis of mirror machines are able to solve the problems
of materials tests with lowest capital and operating cost. At present, the most advanced candidate both: from
experimental and theoretical point of view is the Gas Dynamic Trap (GDT). Recent experiment with oblique
injection of fast deuterium atoms in warm target hydrogen plasma has demonstrated a good agreement with results
of calculations as from the viewpoint of spatial distribution of the neutrons of D-D reaction, so from the viewpoint
of absolute value of the neutron flux density. It should be noted that the GDT-based NS is the object of interest
even with existing, at present, plasma parameters (more exactly the electron temperature of the target plasma
should be increased two times in comparison with the present level). The increase of the temperature from 130 eV
up to 250 eV makes it possible to produce a moderate neutron flux density only several times less than that in the
full-scale projects. An obvious advantage of this moderate version of the NS consists in the fact that the plasma
physics database for such a source has already existed. Thus, the NS with neutron flux density of order of 200-400
kW/m2 can be designed and constructed on the basis of the present day experience. As the next step of such
approach significant increase of neutron flux density will be possible in result of increase of power of D-T neutral
beam injection.
If to speak seriously about the next steps of future
fusion program (ITER, DEMO, first commercial fusion
power plant), one can see that this program can not be
realized without high power 14 MeV neutron source
for material tests of the main structural materials of
future fusion reactor. In fact, testing of these materials
should be finished before the end of the ITER program.
Typical neutron flux density of future fusion reactor
incident upon the first wall is equal to 2÷3 MW/m2.
The total time of influence of such a flux upon the
reactor components is estimated as 10÷20 years and
corresponds to a fluence of 3÷4.5⋅1022 neutron/cm2 or
6÷9⋅1022 neutron/cm2 respectively.
Thus, a neutron source with neutron flux density of
2 MW/m2 (1014 neutron/cm2⋅s) should operate within
10÷15 years (or even 20÷30 years) to accumulate the
fluence of 3÷4.5⋅1022 neutrons/cm2 (or 6÷9⋅1022
neutron/cm2). That is why the Fusion Program
Evaluation Board headed by Prof. U.Colombo [1] has
prepared special report for the Commission of the
European Communities. In particular, the following
statement was written in the report: «...the problem of
the need for a powerful source for high energy
neutrons for materials testing should be addressed with
the utmost urgency. Such a source should be made an
integral part of the ITER programme».
At present, only two main approaches exist: neutron
source on the basis of accelerators and plasma-based
neutron sources.
There exist four ways to obtain high energy neutrons
with the did of accelerators. The first source is known
as spallation source [2]. It uses bombardment of a heavy
target (W, Pb, U) by protons (or deuterons) with energy
of 1÷1.5 GeV. About 30 neutrons can be produced per
one proton. In this case, very wide spectrum of
neutrons is obtained. Such a spectrum is hardly
appropriate for tests of fusion structural materials.
Similar accelerator
can produce negative µ-mesons in result of
interaction of accelerated protons, deuterons or tritons
with target [3]. The physical cost of the formation of
one µ-meson is about 20 GeV for proton beam case and
8 GeV in the case of deuteron beam [4]. In result of
interaction of the mesons with a dense gas target from
D2 and T2 molecules each meson produces for its
lifetime over a hundred DT mesomolecules and these
molecules will emit 14 MeV neutrons. In spite of very
simple general idea of such a source it will be hardly
constructed as one for fusion material tests. The walls
of the vessel with D-T mixture should be thick enough
because of high pressure (1000 atmospheres) and high
temperature of the mixture (1000°C). Under these
conditions 14 MeV neutrons passing through the thick
wall of the vessel will lose their monochromaticity.
In this sense the projects on the basis of low energy
accelerators look more realistic from technical point
of view. In the range of energies 20-40 MeV two
stripping reactions can be used:
7Li + D (35÷40 MeV) ⇒ 8Be + n,
H + T (21 MeV) ⇒ 3He + n.
In the first case (D-Li reaction) the maximum neutron
yield is obtained at energy of neutrons equal to 14
MeV. However, in this case, rather wide spectrum of
neutrones (±7 MeV) generates [5].
Second reaction also has a wide neutron spectrum [6]
but for definite conditions the spectrum is cut off
sharply at neutron energies En > 14 MeV (14.6 MeV).
At present, experts of International Energy Agency
(IEA) have selected the source on the basis of D-Li
reaction as candidate number one among the
accelerator-based schemes. The conceptual design of
the International Fusion Materials Irradiation Facility
(IFMIF) project is widely developed [7,8]. Final
55
designed parameters of D-Li source are as follows: the
deuteron beam energy is 35-40 MeV, the beam current is
2x125 mA, the testing volume with high (2 MW/m2)
neutron flux density is 0.5 liter. The most evident
disadvantages of the project are as follows: too small
irradiated volume and area. But, perhaps, even more
significant disadvantage of such a source is the existence
of high energy neutron tail (En > 14 MeV) which does
not exist in the neutrons of fusion D-T plasma. As one
can see in Fig. 1, a plasma-based neutron source GDT NS
(see below) has the same spectrum of secondary
neutrons as that in the ITER case. At the same time, the
spectrum of D-Li source (the IFMIF project) differs
significantly from that in the plasma-based neutron
source case. A lot of neutrons are obtained with energies
larger than 14 MeV. This circumstance can lead to errors
during the material tests. Besides, the volume of testing
zone in any accelerator-based source is too small. Thus,
accelerator-based sources cannot solve many problems
of materials tests. Therefore, a plasma-based neutron
source is required by all means.
A special issue of the journal «Nuclear Instruments and
Methods» devoted to different proposals of high power
14 MeV neutron sources has appeared in 1977 [10].
However, plasma parameters in those proposals were far
enough from required ones. The first realistic projects of
plasma-based NSs were proposed seven years later (see
references of Ref. [11]). These projects were based on
mirror machines. Among them the most promising were
the NSs based on the Gas Dynamic Trap concept [12] and
on the 2XIIB [13]. The tokamak-based NS projects have
appeared only at the beginning of 90s. Most of these
proposals dealt with large scale tokamaks. In this case,
the level of plasma parameters is close enough to
required ones. However, besides physical requirements
there exist several economic limitations.
One of the significant factors which determine the
operational cost is the power consumption. As it follows
from data presented in Ref. [14], the capital cost of large
tokamak-based NS can achieve 0.45÷0.48 of the ITER
direct cost. Besides, the required power consumption is
within 500÷1000 MW. For the typical energy cost (10
cents per kilowatt-hour) the annual energy expenses
would amount to $M 438÷876.
The second component of the operational cost is
determined by value of tritium consumption. Typical
surface area of a vacuum chamber of large scale tokamak
exceeds 100-200 m2. Correspondingly, for desirable
neutron flux density of 2 MW/m2, the annual tritium
consumption should be of the order of 14-28 kg (420-
840 million USD). It is important to note that besides
too high operational cost there exists very serious
limitation of tokamak scale: the annual world production
of tritium is of the order of 5 kg. Thus, as it follows from
these comments, there exist two possibilities for the
large scale tokamak-based NSs. At first, one can
construct a neutron source with lithium blanket for a
reproduction of tritium. Indeed, the projects of tokamak-
based NS with a tritium production system using lithium
containing blanket appeared [15]. Of course, such
approach should significantly increase the capital cost of
the NS. It is hardly possible to wait that such a source
will be constructed. Really, in order to solve the
problems of material tests, a neutron source is required
to be operated very reliably over ten years. If one
considers the concept of large scale tokamak with
lithium blanket from this point of view, one should come
to conclusion that such a source is similar to simplified
fusion reactor. The neutron flux density (2 MW/m2) will
be practically the same as in the fusion reactor, but the
lifetime of the first wall of the NS is unknown. If one
adds the high capital and operating cost and the tritium
problems, it becomes clear that there is no solution in
this way. Really, in recent years the large scale tokamak-
based projects of NS have disappeared.
However, one can try to decrease significantly the sizes
of the tokamak-based NS and the area of the first wall.
Indeed, appearance of compact spherical tokamaks with
low aspect ratio initiated design works in this direction
[16,17].
The decrease of the aspect ratio allows one to increase
the parameter β (the ratio of plasma pressure to magnetic
field pressure). Not long ago at START facility β=0.48
was achieved [18]. On the basis of this result two
projects of compact volumetric NSs were proposed:
Material Test Facility (MTF) in the United Kingdom [16]
and ST VNS in the USA [17]. One should note, that very
high electron and ion temperatures (Te = Ti = 20 keV)
are assumed in these projects, so as high plasma density
(ne = 1014cm-3). At present, real plasma parameters in the
compact spherical tokamaks are very far from mentioned
above. Besides, it should be mentioned that low aspect
ratio can be obtained only for spherical tori (the internal
torus diameter should be very small). This implies the
impossibility of using a neutron shield and forced the
authors of the projects to abandon not only from
superconducting windings but even from warm multiturn
windings of a toroidal magnetic field (the insulation does
not withstand neutron irradiation).
Area of the vacuum chamber wall is planned to be
30 m2. Thus, the annual tritium consumption should be a
little more than 4 kg and even this compact system
should have a lithium blanket.
As to the mirror-based NSs, at the moment, the most
Figure 1. First wall neutron spectra in ITER,
plasma-based neutron source (GDT type, see later)
and IFMIF [9].
56
advanced concept is based on the so called Gas
Dynamic Trap (GDT).
The GDT is one of the simplest systems for magnetic
plasma confinement. As a matter of fact, it is an
axisymmetric mirror machine of the Budker-Post type,
but with a very high mirror ratio (R>10) and with a
mirror to mirror length L exceeding an effective mean
free path λ/R for the ion scattering into loss cone [19].
Thus, due to frequent collisions the plasma confined in
the trap is very close to isotropic Maxwellian and
therefore many instabilities can not excite and plasma
behaviour is similar to classical one.
If the total number of particles in the trap is equal to
LSno (here no is a plasma density and S is the plasma
cross section at the central part of the trap) and if the
number of particles leaving the trap through mirrors per
second is no VTi
Sm (here VTi
is ion thermal velocity and
Sm is the plasma cross section in the end mirrors), then
the confinement time can be determined as
τ ≈ LSno/SmnoVTi
= RL/VTi
.
Using an oblique injection of fast deuterium and
tritium atoms into warm target plasma one can obtain a
population of unisotropic fast D-T ions which oscillate
back and forth between the turning points near the end
mirrors. As calculations show, there should be an
intensive radiation of 14 MeV neutrons in the vicinities
of these turning points where the fast ions density has
strong peaks. It is necessary to note that the GDT NS is
the only one plasma-based source where the neutron
flux density is strongly inhomogeneous. Due to that
this source has the lowest tritium consumption (of the
order of 100-200 gram per a year) among all the types
of plasma-based NSs. At the same time the GDT NS
can provide the required by material scientists neutron
flux density of 2 MW/m2 or even more within the
testing zone area of the order of 1 m2. Another
advantage of the concept considered is the fact that the
plasma diameter is substantially (by an order of
magnitude) smaller than that of vacuum chamber.
Thanks to that the neutron load at the chamber wall is
substantially lower compared with other schemes. The
inhomogeneity of sloshing ions distribution acts in the
same direction. Thus, the main part of the vacuum
chamber is irradiated to much weaker fluence
compared with the test zones.
End mirror coils placed far enough from the turning
points. Thus, there is no serious problem to shield
these coils from the neutron irradiation. Besides, in the
GDT case it is not necessary to heat all the plasma till
high temperature. Most of 14 MeV neutrons creates in
fast-fast collisions of D-T ions.
At present, the works under the problem of the GDT
NS are doing in several directions. Among them one
should point out the experimental studies on the acting
model of the Gas Dynamic Trap. Besides there should
be mentioned the works under mathematical model of
plasma, under codes describing plasma and sloshing
ions behavior in the GDT and GDT NS, the design
studies of elements of future neutron source, etc.
Very important experiments have been already done
on the Gas Dynamic Trap. In particular, it has been
demonstrated that even in axisymmetric geometry large
scale MHD instabilities can be suppressed [20].
At present, fast ions population with β=0.3 has
already been obtained. However, up to now
microinstabilities driven by the strongly anisotropic
distribution function of hot ions in the velocity space
have not been observed. Based on the results of
previous studies of the microinstabilities in mirrors
(see, for example, [21]) one could conclude that for the
plasma conditions in the GDT NS, the most dangerous
microinstabilities could be stabilized by the warm
plasma background.
The issue of vital importance for the whole project is
how to achieve the bulk electron temperature as high as
0.5 -1.2 keV whereas a maximum of 0.26 keV has been
obtained as yet for relevant plasma density [22].
Generally, the longitudinal electron heat conduction to
the end walls may be a critical issue for open-ended
systems. In accordance with theoretical predictions
[23] these heat losses can be strongly suppressed if the
magnetic field on the end wall is less than (m/M)1/2
compared to the end mirror field. Following to the
increasing area of the magnetic flux tube, plasma
density reduces that gives rise to ambipolar potential in
the expander. This potential forces dragged back the
central cell electrons. A depth of the potential well for
the electrons increases when the magnetic field at the
end wall reduces. The electrons emitted by the end wall
are prevented from entering the central part of the trap
by reflection back from the magnetic mirror. Therefore
there is no an intense energy exchange between the
different electron populations.
Experimental data obtain in the GDT device generally
well agree with the theory predications of Ref.[23].
Fig.2 presents the potential in the central cell of the
GDT for different positions of the segment which can
move in the expander from the end wall till the mirror.
The position of the movable segment is marked by the
expansion ratio 1/K=Bm/Bz on its surface. As it is seen
in Fig.2, for large values of the expansion ratio (≥50),
the central cell potential (that is Te) is not sensitive to
the position of the wall segment. When the expansion
ratio decreases further (1/K<40-50), the potential
drops down. Consequently, it was observed in this case
that
Figure 2. The potential in the central cell of the GDT
for different positions of the emissive plasma absorber
57
the electron temperature in the center cell decreases
thus indicating an increase of longitudinal heat losses
[24].
In the first versions of the GDT NS it was supposed to
use combined mirror coils: superconducting and warm
with total magnetic field strength up to 26 T [12, 25].
Two first basic versions have been analized. In the first
one the two component case (GDT-2) was examined
where 240 keV tritium beam should be injected into a
warm deuterium plasma [12]. Later the conceptual
design of three- component system with 80-100 keV
D-T neutral beam injection was studied [25]. The
weakest point of these two versions was low lifetime of
mirror coils (of the order of two weeks [26]). Besides,
one should add that 30 MW is required to obtain 26 T
field in the mirror coils. Therefore an attempt was
undertaken to avoid using the resistive coils in a
subsequent design modifications. In Ref. [27]
optimization of the SC mirror magnet with maximum
on-axis magnetic field strength equaled to 13 T is
presented. Corresponding version of neutron source
with fully superconducting magnetic system was
calculated by making use of a self-consistent numerical
model [28]. Comparison of parameters of this source
with ones of the GDT device is shown in the Table 1. It
is important to note that quite a moderate energy of
neutral beam injection is used in this version (65 keV).
One can conclude that the neutron source with
superconducting magnetic system has reasonable
parameters and becomes more realistic.
TABLE 1
Design parameters GDT GDT NS
Injection energy, keV 15 65
Injection power, MW 4 60
Neutron flux density,
MW/m2
- 2
Injection angle, degree 45 30
Magnetic field strength in
end mirrors, T
15 13
Magnetic field strength in
the mid plane, T
0.2 1.3
Plasma density in the mid
plane, cm-3
1013-
1014
1.16⋅1014
Plasma radius in the mid
plane, cm
10 8
Electron temperature, keV 0.13 0.75
That’s why it was interesting to revise the results of
previous calculations, in particular, to extend ranges of
the injection energies, electron temperatures, neutron
flux density, etc.
Fig.3 demonstrates the dependence of neutron flux
density on the injection energy of deuterium and
tritium atoms into plasma. The electron temperature
Te=10-2Einj is assumed in these calculations (it is well
istablished that under this condition microinstabilities
are not excited in a mirror plasma [21]). At present,
there are no experimental data in the range Te>10-2 Einj
concerning microinstabilities excitation. The
efficiency of the neutron source and neutron flux
density strongly depends on the electron temperature
of plasma. Fig.4 demonstrates the dependence of the
neutron flux density on the electron temperature of
target plasma. These calculations were made for the
case Einj=65 keV.
Figure 3. Neutron flux density in testing zone of GDT
NS as a function of injection energy of deuterons and
tritons (Te is assumed to be 10-2 Einj) [29].
Thus, most part of the curve corresponds to the case
when Te ≤ 10-2 Einj. One can see that even in this case
the desired neutron flux density (≈2MW/m2) can be
obtained. If one supposes that microinstabilities do not
excite at Te ≅ 3⋅10-2 Einj then GDT NS will be able to
produce up to 5 MW/m2 neutron flux density at the
same injection energy 65 keV (see Ref. [29]).
As to the results of calculations obtained with the aid
of developed mathematical model of plasma [30], Fast
Ion Transport code (FIT) based on the Monte Carlo
method [31], and Fokker-Plank code FPM (Fast
Particle Model) [32], one can note that the degree of
reliability
of the simulations is very high. Several figures
illustrate this statement. The energy content of plasma
in the GDT device as a function of time is shown in
Fig.5. It is seen that the difference between
experimental data and ones of calculations are
insignificant. Besides, measured angular spread of fast
ions corresponds well with results obtained by FIT
code calculations and with analytic estimates [34].
Recently an experiment with injection of fast
deuterium atoms was done [35]. In Fig.6 a longitudinal
distribution of neutron flux density (D-D reaction) is
presented in the vicinity of turning point. The solid
line is the result of simulation, the discrete marks
are the results of measuring. Again the agreement
is rather
Figure 4. Neutron flux density vs electron
temperature. Calculations were made at power of
neutral beam injection W=60 MW and energy of D-
T injection Einj=65 keV.
58
TABLE 2
Plasma radius in
the mid plane, cm
8 8 8
Injection angle 30° 30° 30°
Magnetic field in
the end mirrors, T
13 13 13
Mirror ratio 15 15 15
Injection energy,
keV
65 65 65
Electron
temperature, eV
200 250 300
Electron density
in the mid plane,
cm-3
1.2⋅1014 1.1⋅1014 1.1⋅1014
Density of fast
ions in the mid
plane, cm-3
0.32⋅1014 0.37⋅101
4
0.49⋅101
4
Electron density
in the test zone,
cm-3
2.5⋅1014 2.8⋅1014 3.0⋅1014
Density of fast
ions in the test
zone, cm-3
1.87⋅1014 2.29⋅101
4
2.53⋅101
4
Power
consumption of
injectors, MW
60 60 60
Neutron flux
density in the test
zone / in the mid
plane, kW/m2
230/7 350/10 450/20
Thus, it follows from the obtained data that the degree
of confidence of the results of calculations for the
GDT NS should be high enough. There is only one
serious objection against this statement. The highest
electron temperature in the GDT experiments, at
present, is only 130 eV. In such situation the most
reasonable strategy could look as follows. Taking into
account that the experiments on suppression of
electron heat conductivity have been done on the level
of Te=100-130 eV and that the electron temperature of
260 eV has been obtained in the mirror experiments, it
is reasonable to estimate the parameters of the GDT-
based neutron source at the range of low electron
temperatures. The results of self-consistent
simulations for moderate versions of the GDT NS are
presented in the Table 2 [36].
As it is seen from the Table the neutron flux density
of 350 kW/m2 can be obtained for achieved in previous
experiments with mirror machines electron
temperature (250 eV) on the basis of present day
plasma data base and technology. Even insignificant
excess of Te above this level till 300 eV leads to the
value of this flux of 450 kW/m2. At Te=400 eV this
value increase up to
710 kW/m2.
Some of material scientists believe that even with
moderate parameters of neutron flux they would be
able to make important conclusions concerning the
quality of main structural materials NS. Anyhow, the
construction of powerful neutron source should begin
from the moderate versions presented in the Table 2. It
should be noted that the level of the neutron flux
densities shown in the Table is minimum one. For 60
MW power consumption of injectors (roughly of 30
MW in the neutral beams) the electron temperature
should be significantly higher than it is shown in the
Table (the Te values presented there is specially limited
by adding of cold plasma). Thus, if the mechanism of
electron heat conductivity wi ll act on the level of Te≈1
keV, in this case, neutron flux density will be
significantly higher than one can see in the Table 2.
CONCLUSIONS
The urgency of design and construction of the
dedicated neutron source, which is absolutely needed
to perform material development for a future fusion
power reactor, is obvious.
D-Li source can solve only a small part of the
problems.
Among plasma-based sources only compact spherical
tokamak and mirror machines can compete.
The axisymmetric GDT NS looks like the simplest
and cheapest one. It requires minimum power and
tritium consumption among plasma-based NSs but
makes it possible a creation of testing zone area of the
order of
1m2.
Figure 6. Distribution of the maximum neutron
flux density (D-D reaction) along the GDT device
length. Discrete marks are the experimental data,
solid curve is the result of simulation.
Figure 5. The energy content of target plasma and
fast ions vs time [33].
59
A moderate GDT-based neutron source with the
neutron flux density not less than 350-450 kW/m2 can
be designed and constructed on the present day
technology level as a first step of the program of
material tests for fusion reactors.
Recent progress of superconducting technique (21 T
was obtained in the coils of appropriate size [37])
allows one to revise the former GDT NS projects with
high magnetic field in the end mirrors.
REFERENCES
1. Colombo U., Iaumotte A., Kennedy E., et al, Report
of the Fusion Programme Evaluation Board
prepared for the Commission of the European
Communities, Bruxelles, (1990).
2. Carpenter J.H. Nuclear Instruments and Methods,
v.145, p.91, (1977).
3. Petitjean C., Atchinson F., Heidenreich G., et al,
Fusion Technology, v.25, p.437, (1994).
4. KuzminovV.V., Petrov Yu.V. and Shabelsky Y.M.,
Hyperfine Interactions, 82, p.423, (1994).
5. Ikeda Y., Konno C., Oishi K., et al, Activation
Cross Section Measurements for Fusion Reactor
Structural Materials at Neutron Energy from 13.3
to 15 MeV Using FNS Facility, JAERI, 1312,
Japan, March 1998.
6. Kondo T., Doran D.G., Ehrlich K., Wiffen F.W.,
Journ. Nucl. Materials, v.191, p.100, (1992).
7. Shannon T.E., Rennich M.J., Kondo T., et al, Proc.
16th Fusion Energy Conference IAEA (Montreal,
Canada, Oct. 7-11, 1966). IAEA, Vienna, 3, p.437,
(1997).
8. IFMIF CDA Final Report. Edited by M.Martone,
ENEA Frascati, Roma, 1997.
9. Fisher U., Möslang A., Ivanov A.A., Fusion
Technology Transactions, v.35, ¹1T, p.160,
(1999).
10. Nuclear Instruments and Methods, v.145, ¹1,
p.p.1-128, (1977). Special issue of the journal
devoted to the high power neutron sources.
11. Kruglyakov E.P., Fusion Technology
Transactions, v.35, ¹1T, p.20, (1999).
12. Mirnov V.V., Nagornyi V.P. and Ryutov D.D.,
Preprint ¹40, Budker Institute of Nuclear Physics,
Novosibirsk, (1984);
Kotelnikov I.A., Mirnov V.V., Nagornyi V.P.,
Ryutov D.D., in: «Plasma Physics and Controlled
Fusion Research», IAEA, Vienna, 2, p.309, (1985).
13. Coensgen F.H., Casper T.A., Correl D.L., et al,
Journ. Fusion Energ., 8, p.237, (1989).
14. Kruglyakov E.P., Intern. Conf. on Plasma Physics,
ICPP 1994, Foz do Iguacu, Brazil, Oct.-Nov. 1994.
AIP Conference Proceedings 345, Woodbury, New
York, p.247, (1995).
15. Tsuji-Jios, Tsutsui H., Kondoh J., et al, Proc. 16th
Fusion Energy Conf. IAEA (Montreal, Canada, Oct.
7-11, 1996), IAEA, Vienna, 3, p.685, (1997).
16. Buttery R., Counsell G., Cox M., et al, in: Plasma
Phys. and Controlled Nucl. Fusion Research
(Seville, Spain, Sept.26-Oct.1, 1994), IAEA,
Vienna, 2, p.633, (1995).
17. Cheng E.T., Intern. Workshop on ST VNS and
Applications, Del Mar, California, USA, Oct. 13-
15, (1997).
18. O’Brien M.R., Akers R., Bamford R.A., et al, Proc.
16th Fusion Energy Conf. IAEA (Montreal, Canada,
Oct. 7-11, 1996), IAEA, Vienna, 2, p.57, (1997).
19. Mirnov V.V., Ryutov D.D., Sov. JTP Letters, v.5,
p.279, (1979).
20. Anikeev A.V., Bagryansky P.A., Ivanov A.A., et al,
Plasma Phys. and Contr. Fusion, v.34, p.1185,
(1992).
21. Post R.F. Nuclear Fusion, v.27, p.1579, (1987).
22. Simonen T.S. (Ed.), «Summary of Results from
Tandem Mirror Experiment», UCRL 53120,
LLNL, (1981).
23. Konkashbaev I.K., Landman I.S., Ulinich F.R.,
Sov. JETP, v.74, ¹3, p.956, (1978).
24. Anikeev A.V., Bagryansky P.A., Deichuli P.P., et al,
in: Proc. 23rd EPS Conference on Plasma Physics
and Controlled Fusion, Kiev, 20C, pt.2, p.688,
(1996).
25. Ivanov A.A., Kotelnikov I.A., Kruglyakov E.P., et al,
Proc. 17th Symp. on Fusion Techn. (SOFT), Rome,
Italy, 2, p.1394, (1992).
26. Klyavin P.P., Levashov A.D., Rozhdestvensky B.V.,
et al, Pribory i Tekchnika Eksrepimenta (in
Russian), ¹5, p.232, (1976).
27. Ivanov A.A., Kruglyakov E.P., Tsidulko Yu.A., et al,
Proc. of 16th Symp. on Fusion Engineering, Urbana,
USA, Sept.30 - Oct.5, 1995 (SOFE’95).
IEET/NPSS, v.1, p.66, (1995).
28. Ivanov A.A., Kruglyakov E.P., Tsidulko Yu.A., et al,
Proc. of 16th Fusion Energy Conf. IAEA (Montreal,
Oct. 7-11, 1996). IAEA, Vienna, 3, p.667, (1997).
29. Ivanov A.A., Kruglyakov E.P., Tsidulko Yu.A., Proc.
17th Fusion Energy Conf. IAEA (Yokohama, Japan,
Oct. 19-24, 1998), FTP/31.
30. Kotelnikov I.A., Ryutov D.D., Tsidulko Yu.A., et al,
Preprint ¹105 of Budker INP, Novosibirsk, (1990).
31. Noack K., Otto G., Collatz, Fusion Technology
Transactions, v.35, ¹1T, p.218, (1999).
32. Ivanov A.A., Karpushov A.N., Preprint 96-2, Budker
INP, Novosibirsk, (1996).
33. Anikeev A.V., Bagryansky P.A., Bender E.D., et al,
in: Proc 22nd EPS Conf. on Controlled Fusion and
Plasma Physics, Bornemouth (UK), 3-7 July, 1995,
v.19C, pt.IV, p.193, (1995).
34. Anikeev A.V., Bagryansky P.A., Ivanov A.A., et al,
Nuclear Fusion, v.40, ¹4, p.753, (2000).
35. Anikeev A.V., Bagryansky P.A., Ivanov A.A., et al,
in: Proc. 26th EPS Conf. on Controlled Fusion and
Plasma Physics, Maastricht, v.23J, p.1781, (1999).
36. Ivanov A.A., Kruglyakov E.P., Tsidulko Yu.A., 27th
EPS Conf. on Controlled Fusion and Plasma
Physics, 12-16 June, 2000, Budapest.
37. Kiyoshi T., Kosuge M., Inone K., and Maeda N.,
IEEE Trans. on Magnetics, 32, p.2478, (1996).
|