close

Вход

Забыли?

вход по аккаунту

?

j.nima.2018.07.080

код для вставкиСкачать
Nuclear Inst. and Methods in Physics Research, A 906 (2018) 30–36
Contents lists available at ScienceDirect
Nuclear Inst. and Methods in Physics Research, A
journal homepage: www.elsevier.com/locate/nima
Study of MPPC damage induced by neutrons
S. Mianowski a ,∗, J. Baszak b , Y.M. Gledenov c , Y.N. Kopatch c , Z. Mianowska a , M. Moszynski a ,
P. Sibczynski a , T. Szczesniak a
a
National Centre for Nuclear Research, 05-400 Otwock, Soltana 7, Poland
Hamamatsu Photonics Deutschland GmbH, Arzbergerstr. 10, D-82211 Herrsching am Ammersee, Germany
c
Joint Institute for Nuclear Research, Frank Laboratory of Neutron Physics, Dubna, Joliot-Curie 6, Moscow region, Russian Federation
b
ARTICLE
INFO
ABSTRACT
Keywords:
MPPC
Neutron irradiation
Neutron fluence
Radiation damage
Energy resolution
GAGG:Ce scintillator
This paper presents the results of neutron irradiation of two types of Multi-Pixel Photon Counter (MPPC). The
measurements were taken at the National Centre for Nuclear Research (NCBJ), Poland and the Joint Institute
for Nuclear Research (JINR), Russia. Two types of neutron source were used: PuBe with a continuous energy
spectrum up to 11 MeV, and mono-energetic 4.8 MeV neutrons produced in a (d,d) reaction. For both sources,
fluence in the range of 1010 n/mm2 was achieved. A series of MPPC tests were performed after each irradiation.
The changes in MPPC properties, such as current–voltage (I–V) characteristics, breakdown voltage and energy
resolution of 662 keV from the 137 Cs gamma line for a Gd3 Al2.6 Ga2.4 O12 :Ce (1%) scintillator as a function of
neutron fluence are presented.
1. Introduction
respectively), which each have a different active area size: 3 × 3 mm2
and 6 × 6 mm2 , respectively, and a 50 μm pixel pitch size.
To minimize the gain — temperature dependence of the MPPCs, a
Hamamatsu evaluation board C12332-01 [4] was used. This electronic
circuit, equipped with a precision power supply and a temperature
sensor, allows a temperature compensation factor to be set for each type
of MPPC. Using this board and a 6 × 6 mm2 MPPC coupled to a GAGG
scintillator, a gain stability test was performed in a climate chamber.
The temperature changes were programmed to be in the range of 14◦ C–
26◦ C. In this controlled environment, the position of the 662 keV gamma
line from 137 Cs was checked. The centroid of the full energy peak was
determined by a Gaussian fit. Fig. 2 shows the obtained results. As can
be seen, the 662 keV peak position mirrors the temperature trend (black
line) and the calculated difference between the two opposite centroids
stays below 1%. This illustrates good stability of our experimental setup. It is worth mentioning that the temperature changes monitored
during the two experiments described later in this paper were about
two times lower than those tested in the climate chamber.
To determine the I–V characteristics of each MPPC, a Keithley 2400
Series SourceMeter was used. This instrument can register currents
starting from single picoampers. For safety reasons, the upper current
limit was set to 100 μA.
To observe energy resolution degradation caused by irradiation
of the MPPCs, a GAGG scintillator was used. This 5 × 5 × 5 mm3
A Multi-Pixel Photon Counter (MPPC), also known as a Silicon
Photomultiplier (SiPM), is a silicon photon counting device that uses
multiple avalanche photodiode pixels operating in Geiger mode. MPPCs
are widely used in science fields such as nuclear and high-energy
physics [1], astrophysics, and medical imaging [2,3]. Due to their very
good photon counting capability, they can also be used in various
applications for detecting extremely weak light at the level of single
photons. MPPC features include a high gain (105 to 107 ) comparable
with standard photomultiplier tubes, fast response time, compact size,
low bias voltage (below 100 V) and insensitivity to magnetic fields.
In our work we investigate the current–voltage (I–V) characteristics
and breakdown voltage changes that occur with the increase of neutron
fluence in two different types of MPPC. We determine the energy
resolution of the irradiated photodetectors coupled to a non-irradiated
Gd3 Al2.6 Ga2.4 O12 :Ce (1%) scintillator (later referred to as GAGG) and
show that the degradation of energy resolution should be taken into
account in gamma spectroscopy physics if a high neutron background is
present. Finally, we calculate the noise contribution to energy resolution
as a function of neutron fluence.
2. Experimental set-up
In our set-up (Fig. 1) we used two types of MPPC from Hamamatsu:
S13360–3050CS and S13360–6050CS [4] (called: 3050CS and 6050CS,
∗ Corresponding author.
E-mail address: slawomir.mianowski@ncbj.gov.pl (S. Mianowski).
https://doi.org/10.1016/j.nima.2018.07.080
Received 25 May 2018; Received in revised form 24 July 2018; Accepted 24 July 2018
Available online xxxx
0168-9002/© 2018 Elsevier B.V. All rights reserved.
S. Mianowski et al.
Nuclear Inst. and Methods in Physics Research, A 906 (2018) 30–36
Fig. 1. MPPC photodetectors from Hamamatsu: S13360–3050CS and S13360–6050CS,
GAGG scintillator and Hamamatsu evaluation board C12332-01.
crystal has a high light output (44600 ± 4400 photons/MeV) and,
most importantly, a well determined intrinsic resolution component
((2.7 ± 0.3)% for 662 keV) [5] measured with an APD. This parameter
was employed in our calculations. As a readout system, a CAEN digitizer
DT5720 was used.
Fig. 2. Test of position stability of a 662 keV gamma line from 137 Cs for a temperature
range of 14 ◦ C–26 ◦ C, performed in a climate chamber. Top — temperature change
programmed in climate chamber. Bottom — position of the full energy peak of 137 Cs
determined by a Gaussian fit. Each point corresponds to an 18 min measurement time.
2.1. Experiment timeline
(3050CS or 6050CS). To obtain the neutron fluence, the flux was
integrated over the irradiation time.
The first experiment was performed at the National Centre for Nuclear Research (NCBJ), Poland, with neutrons from a PuBe source. This
source is characterized by a continuous neutron energy spectrum with
average energy of about 4.6 MeV and internal activity of 8.0 × 105 n/s
in a 4 angle. Neutron flux and the calculated fluence chosen for
each measurement session were determined by MPPC — PuBe source
Before irradiation of the MPPCs, initial parameters such as I–V
characteristics, dark count spectra and energy resolution dependencies
for different operating voltages were collected for each type of MPPC.
Next, two independent experiments with different neutron energy
profiles were performed. In both experiments, during the irradiation
process, the geometry between the MPPC and neutron source was
chosen in order to keep the same neutron flux for each type of MPPC
Fig. 3. Changes in I–V characteristics with the increase in neutron fluence from a PuBe source for a 3 × 3 mm2 (upper part) and 6 × 6 mm2 (lower part) type of MPPC. Dashed vertical


lines show the calculated breakdown voltage parameter (
) for the non-irradiated (initial) MPPC. Right hand graphs present the current increase for 
compared to linear fit. (For
interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
31
S. Mianowski et al.
Nuclear Inst. and Methods in Physics Research, A 906 (2018) 30–36
Fig. 4. Changes in I–V characteristics with the increase in neutron fluence from a (d,d) reaction for 3 × 3 mm2 (upper part) and 6 × 6 mm2 (lower part) type of MPPC. Dashed vertical


lines show the calculated breakdown voltage parameter (
) for the non-irradiated (initial) MPPC. Right hand graphs present the current increase for 
compared to linear fit. (For
interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
der Graaf accelerator. The intensity of neutron flux was monitored by
two independent systems: a 3 He gas detector and a stilbene scintillator
coupled to a photomultiplier tube. The neutron fluence calculated from
these two acquisition systems are self-consistent and the average value
was used as a reference point. The presented results were acquired for
fluence values: 2.2 × 108 mm−2 , 4.9 × 109 mm−2 and 9.8 × 109 mm−2 .
The last value corresponds to over 14 h of neutron beam lifetime.
Both experiments were carried out as follows: MPPC units were
irradiated by neutrons at specified time intervals: 1 × 15 min and 3 × 8 h
for neutrons from a (d,d) reaction and interval time calculated before
irradiation for neutrons from a PuBe source. After each irradiation
session, the properties of the MPPC were investigated (about 2 h per
MPPC). To determine energy resolution dependencies, the MPPCs were
coupled to a non-irradiated GAGG scintillator and measurements with
gamma sources (22 Na, 54 Mn and 137 Cs) were taken. Finally, after each
measurement session, MPPCs waited approximately 14 h for the next
beam time for neutrons from a (d,d) reaction, or were immediately
exposed to neutron flux from a PuBe source.
Fig. 5. Example of dark count spectra measured for MPPC 3050CS for 54.6 V at room
temperature as a function of neutron fluence from a PuBe source. For better comparison,
all spectra were normalized to unit area. Dashed vertical line shows the position of a single
photoelectron peak (1 spe). Higher orders of photoelectron peaks are also visible only for
low neutron fluence. (For interpretation of the references to color in this figure legend,
the reader is referred to the web version of this article.)
3. Experimental results
distance. The presented results were obtained for nine fluence values:
7.5 × 105 mm−2 , 3.6 × 106 mm−2 , 1.5 × 107 mm−2 , 7.8 × 107 mm−2 ,
1.5 × 108 mm−2 , 4.8 × 108 mm−2 , 8.6 × 108 mm−2 , 1.9 × 109 mm−2
and 3.3 × 109 mm−2 .
The second experiment was performed at the Joint Institute for Nuclear Research (JINR), Russia, where a mono-energetic 4.8 MeV neutron
beam was used to irradiate the MPPCs. These neutrons were produced
in a (d,d) reaction, where incoming deuterons were accelerated by a Van
3.1. Changes in I–V characteristics with the increase of neutron fluence
In the first step, I–V characteristics were measured. As can be seen in
Fig. 3, neutrons from the PuBe source cause a very fast current increase
for both photodetectors, which is in agreement with data presented
in [6].
32
S. Mianowski et al.
Nuclear Inst. and Methods in Physics Research, A 906 (2018) 30–36
Table 1
Rate of change of breakdown voltage as a function of neutron fluence in units of [V⋅mm2 ],
where the minus indicates the decrease of breakdown voltage.
Neutron
source
MPPC type
3050CS
6050CS
(d,d)
PuBe
(−6.2 ± 0.5) × 10−12
(−8.0 ± 0.8) × 10−12
(−7.3 ± 1.2) × 10−12
(−3.5 ± 1.7) × 10−11
Assuming the same value of operating voltage, equal to the initial
breakdown voltage (before irradiation) (see Fig. 7 below), the relationship between current and neutron fluence deviates from the linear fit.
Similar behavior is observed for MPPCs irradiated by neutrons from a
(d,d) reaction (Fig. 4). In general, for all cases, the dark current increases
over a factor of 103 for the highest neutron fluence with the increase of
neutron fluence.
3.2. Changes in the breakdown voltage parameter with the increase of
neutron fluence
In this part, we investigate the breakdown voltage parameter ( ).
In literature [7,8] we can find two ways of determining breakdown
voltage. In the first case, dark current spectra and a single photoelectron peak measured for various temperature ranges and for different
operating voltages are used. The main disadvantages of this procedure
are its time consumption, and more importantly, the dark count rate
increases drastically with the irradiation process and the photoelectron
peaks become indistinguishable [6], eliminating the usefulness of this
method. This effect is also clearly visible in our data. An example of dark
count spectra for different neutron fluences is presented in Fig. 5.
The second method is based on I–V curve analysis. Different analyses
are proposed in this case [8]. Usually, the breakdown voltage parameter
is extracted from ln(I)–V dependency. In our case, due to the relatively
small voltage probing (0.2 V), we decided to fit the function defined by
the following equation:
{
(
)
1 ⋅ exp ( − 1 ) ⋅ 1 ,
 <  ,
ln(I) =
(1)
[
(
)]
2 ⋅ 1 − exp −( − 2 ) ⋅ 2 ,  >  ,
Fig. 6. Top — Example of ln(I)–V characteristics in the range of the breakdown voltage
parameter with a fitted function. For better visualization, all point uncertainties were not
drawn on the graph. Bottom — the second derivative of the fitted curve.
we used three additional gamma lines from 22 Na (511 keV, 1274 keV)
and 54 Mn (835 keV) as calibration points. This scanning procedure was
performed after each irradiation session. Examples of obtained results
before irradiation of the MPPCs and for the highest neutron fluence
measured from a PuBe source are presented in Fig. 9.
Fig. 10 presents the degradation of energy resolution with the
increase of neutron fluence from a PuBe source. The difference in initial
energy resolution (5% for 6050CS and 7% for 3050CS) is due to the
better light collection efficiency of the 6 × 6 mm2 MPPC. As can be seen,
for both photodetectors, the strongest change in degradation is observed
for relatively low neutron fluence. For higher values we observe the
saturation effect in the range of E = 10.5%. To describe the observed
dependency, the logistic function (red curve) was chosen. For better
data visualization, the log-lin scale was chosen. Initial values (before
irradiation) correspond to fluence equal to 1 mm2 for all presented
fluence dependencies.
For neutrons from a (d,d) reaction, the energy resolution was determined for the maximum available neutron fluence of 9.8 × 109 mm−2 —
one irradiation session fewer than for the presented I–V characteristics.
This was because the dark current exceeded the current limit for the
Hamamatsu evaluation board C12332-01. In this case, the saturation
effect was observed for about 12% for 3050CS and 9.5% for 6050CS.
Due to the low number of experimental points, the red curve in Fig. 11
is drawn only to guide the eye. Fig. 12 shows the energy spectra of
the 662 keV gamma line from 137 Cs before and after irradiation. For
better comparison the amplitudes of full energy peaks were normalized
to unity. Due to the small scintillator size, the K–X escape component
from gadolinium was also observed. In general, we observed energy
resolution degradation by a factor of 2.
where  was chosen as a maximum of the second derivative of
the experimental ln(I)–V dependency. As can be seen in Fig. 6, the
intersection point of the fitted functions for two regions (A and B) is
(as expected) in excellent agreement with the maximum of the second
derivative of the fitted function. This point defines  . The changes
in the breakdown voltage parameter for different MPPCs and different
neutron sources are presented in Fig. 7. All results are presented in a
log-lin scale for better visualization of the low neutron fluence region.
The fitted linear function determines the rate of change of  with the
increase of neutron fluence.
As can be seen in Table 1, the breakdown voltage changes are in
the same order for both neutron sources. The high uncertainty obtained
for MPPC 6050CS and the PuBe source does not allow this result to be
interpreted as significantly different from the others. Fig. 8 shows the
breakdown current determined by the fitted breakdown voltage from
Fig. 7 and measured I–V characteristics. The obtained dependencies of
breakdown current are well described by a linear function, which is in
contrast to current characteristics presented in Fig. 3 and Fig. 4, where
a constant (initial) breakdown voltage was assumed.
3.3. Energy resolution degradation for an irradiated MPPC
The best operating voltage of an MPPC is defined by the value
for which the best energy resolution ( ) during spectroscopy measurements is obtained [9]. In our case, a 662 keV line from 137 Cs
was chosen as a reference point. We scanned the value of  for a
different operating voltage in 0.3 V steps to find the optimal voltage
for each MPPC. To obtain a true value of energy resolution, correction
for the non-linearity effect of the MPPC was included. For this reason
3.4. Noise component for irradiated MPPC
The energy resolution of the full energy peak measured with a
scintillator coupled to a MPPC photodetector can be expressed in general
as [10–12]:
2
2
2
2

= 
+ 
 +  ,
33
(2)
S. Mianowski et al.
Nuclear Inst. and Methods in Physics Research, A 906 (2018) 30–36
Fig. 7. Breakdown voltage dependency as a function of neutron fluence from a (d,d) reaction (top) and from a PuBe source (bottom). Red and blue curves represent the linear fit in the
log-lin scale . (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Fig. 8. Breakdown current dependencies determined by fitted breakdown voltage (Fig. 7) and measured I–V characteristics for neutrons from a PuBe source (left) and neutrons from a
(d,d) reaction (right). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Fig. 9. Energy resolution of the 662 keV line from 137 Cs for MPPC S13360–3050CS (left) and S13360–6050CS (right) coupled to a GAGG scintillator for different operating voltages.
Obtained results were corrected for the non-proportionality effect of the MPPC.
where  is the intrinsic resolution of the scintillator ((2.7 ± 0.3)%
 = (2440±150) for 3050CS and  = (3840±100) for 6050CS are the
in our case [5]),   is the statistical contribution of a single
numbers of photoelectrons, which were determined by the position of
the full energy peak of 137 Cs and the position of the single photoelectron
photoelectron and  is the noise contribution. The statistical noise
peak in the dark current spectrum. These two  parameters were set
contribution for the MPPC can be expressed as:
as constant in our calculations for both types of MPPC. Parameter  is
2

 = 2.355 ×
(


)1∕2
an excess noise factor, which describes effects such as cross-talks and
.
(3)
after-pulses.
34
S. Mianowski et al.
Nuclear Inst. and Methods in Physics Research, A 906 (2018) 30–36
Fig. 10. Energy resolution of a 662 keV line from 137 Cs for MPPC 3050CS (left) and 6050CS (right) coupled to a GAGG scintillator as a function of neutron fluence from a PuBe source.
Inset graphs present a zoomed region of low neutron fluence.
Fig. 11. Energy resolution of a 662 keV line from 137 Cs for MPPC 3050CS (left) and 6050CS (right) coupled to a GAGG scintillator as a function of neutron fluence from a (d,d) reaction.
The red curve is drawn to guide the eye only.
Fig. 12. Example of full energy peak of 137 Cs measured with a GAGG scintillator coupled to MPPC 3050CS (left) and 6050CS (right) before neutron irradiation and for the highest
neutron fluence from a PuBe source. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Knowing the energy resolution and intrinsic resolution, we are able
to determine   . As F is not well specified (below 2 for MPPC [10]),
we decided to discuss two extreme cases:  = 1 (no fluctuation) and
2
2
2 −
 =  , where  satisfies the equation 
= 
− 
2
  > 0. Fig. 13 presents the results obtained for irradiation from a
PuBe source based on the fitted function presented in 10.  values
were determined as 1.70 for 3050CS and 1.22 for 6050CS.
For the (d,d) reaction, numerical values for two opposing fluences
are presented in Table 2.  values were determined in this case to
be 1.77 for 3050CS and 1.36 for 6050CS. For comparison, results for
neutrons from a PuBe source are also shown. As we can see,  is
dominant for high neutron fluence in both cases.
4. Conclusions
The experimental results show that the two types of MPPC are sensitive to neutron irradiation from a PuBe source and from a (d,d) reaction.
The dark current strongly increases with the increase of neutron fluence,
35
S. Mianowski et al.
Nuclear Inst. and Methods in Physics Research, A 906 (2018) 30–36
Fig. 13. MPPC noise contribution to energy resolution for MPPC 3050CS (left) and 6050CS (right) as a function of neutron fluence from a PuBe source. Two extreme values of  (red
region) were tested . (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Table 2
Contribution of the noise component to energy resolution measured for the GAGG
scintillator before and after irradiation of the MPPCs by neutron flux.
Source: Presented at [13].
Fluence (mm−2 )
Neutrons from (d,d) reaction
1a
9.8 × 109
Neutrons from PuBe source
1a
3.3 × 109
a
[2] N.J. Cherepy, J.D. Kuntz, J.J. Roberts, T.A. Hurst, O.B. Drury, R.D. Sanner, T.M.
Tillotson, S.A. Payne, Transparent ceramic scintillator fabrication, properties and
applications, in: Conference on Hard X-ray, Gamma-Ray, and Neutron Detector
Physics X, Spie-Int Soc Optical Engineering, San Diego, CA, 2008.
[3] K. Shimizu, H. Uchida, K. Sakai, M. Hirayanagi, S. Nakamura, T. Omura, Development of an MPPC Detector for PET, in: IEEE Nuclear Science Symposium/Medical Imaging Conference (NSS/MIC)/18th International Workshop on
Room-Temperature Semiconductor X-Ray and Gamma-Ray Detectors, IEEE, Valencia, Spain, 2011, pp. 2991–2994.
[4] Technical specification available at http://www.hamamatsu.com/.
[5] P. Sibczynski, J. Iwanowska-Hanke, M. Moszynski, L. Swiderski, M. Szawlowski, M.
Grodzicka, T. Szczesniak, K. Kamada, A. Yoshikawa, Characterization of GAGG:Ce
scintillators with various Al-to-Ga ratio, Nucl. Instrum. Methods Phys. Res. A 772
(2015) 112–117.
[6] I. Nakamura, Radiation damage of pixelated photondetector by neutron irradiation,
Nucl. Instrum. Methods Phys. Res. A 610 (2009) 110–113.
[7] Z.W. Li, C.Z. Liu, Y.P. Xu, B. Yan, Y.G. Li, X.F. Lu, X.F. Li, S. Zhang, Z. Chang, J.C.
Li, H. Gao, Y.F. Zhang, J.L. Zhao, A novel analog power supply for gain control of
the Multi-Pixel Photon Counter (MPPC), Nucl. Instrum. Methods Phys. Res. A 850
(2017) 35–41.
[8] F. Nagy, G. Hegyesi, G. Kalinka, J. Molnar, A model based DC analysis of SiPM
breakdown voltages, Nucl. Instrum. Methods Phys. Res. A 849 (2017) 55–59.
[9] M. Grodzicka, M. Moszynski, T. Szczesniak, M. Szawlowski, D. Wolski, J. Baszak,
MPPC Array in the Readout of CsI:Tl, LSO:Ce:Ca, LaBr3:Ce, and BGO Scintillators,
IEEE Trans. Nucl. Sci. 59 (2012) 3294–3303.
[10] M. Grodzicka, M. Moszynski, T. Szczesniak, M. Szawlowski, J. Baszak, Characterization of 4x4 ch MPPC array in scintillation spectrometry, J. Instrum. 8 (2013)
23.
[11] L. Swiderski, M. Moszynski, W. Czarnacki, A. Syntfeld-Kazuch, M. Gierlik, Nonproportionality and energy resolution of NaI(Tl) at wide temperature range (-40
degrees C to +23 degrees C), IEEE Trans. Nucl. Sci. 54 (2007) 1372–1377.
[12] M. Moszynski, M. Szawlowski, M. Kapusta, M. Balcerzyk, Large area avalanche
photodiodes in scintillation and X-rays detection, Nucl. Instrum. Methods Phys.
Res. A 485 (2002) 504–521.
[13] S. Mianowski, J. Baszak, Y.M. Gledenov, Y.N. Kopatch, Z. Mianowska, M. Moszynski, P. Sibczynski, T. Szczesniak, Evolution of MPPC properties as a function of neutron fluence, in: IEEE Nuclear Science Symposium/Medical Imaging Conference,
NSS/MIC, Atlanta, USA, 2017.
 (%)
3050CS
6050CS
0.3–4.2
10.0–10.9
0.3–2.4
7.5–7.9
0.5–4.0
7.9–8.8
0.3–1.8
9.2–9.5
Indicates before neutron irradiation.
which makes the photoelectron peaks indistinguishable. The breakdown
voltage, determined by analyzing I–V characteristics, changes slowly
(factor of 10−12 V⋅mm−2 ) with the increase in neutron fluence. Finally,
there is strong energy resolution degradation of the 662 keV gamma line
by a factor of 2, to about 11%–12% for both tested MPPCs. The highest
change is observed for relatively low neutron fluence <0.5 × 109 mm−2 ,
with the saturation effect above this value. This observation should be
taken into account in gamma spectroscopy measurements, which are
performed in environments with a high neutron background.
Acknowledgments
S.M. would like to thank the operation crew of the 5-MV Van der
Graaf accelerator at JINR.
This work was supported in part by the Polish Program of Applied
Research, grant number: PBS2/B2/11/2014, RaM — scaN and by
Research Program For the Research Group at JINR and Research Centers
in Poland: 03-4-1104-2011/2016.
References
[1] M. Bazzi, C. Berucci, C. Curceanu, A. D’Uffizi, K. Piscicchia, M.P. Lener, A.R. Vidal,
E. Sbardella, A. Scordo, O.V. Doce, Characterization of a scintillating fibers read
by MPPC detectors trigger prototype for the AMADEUS experiment, J. Instrum. 8
(2013) 10.
36
Документ
Категория
Без категории
Просмотров
2
Размер файла
3 735 Кб
Теги
niman, 2018, 080
1/--страниц
Пожаловаться на содержимое документа