вход по аккаунту



код для вставкиСкачать
Nuclear Inst. and Methods in Physics Research, A 906 (2018) 83–87
Contents lists available at ScienceDirect
Nuclear Inst. and Methods in Physics Research, A
journal homepage:
Physical–chemical characterization of a GEM side-on 10 B-based thermal
neutron detector and analysis of its neutron diffraction performances
A. Santoni a , G. Celentano a , G. Claps a , A. Fedrigo b , C. Höglund c,d , F. Murtas e,f , F. Rondino a ,
A. Rufoloni a , A. Scherillo b , S. Schmidt d,g , A. Vannozzi a , A. Pietropaolo a ,∗
ENEA, Department of Fusion and technologies for Nuclear Safety and Security, via E. Fermi 45 I-00044 Frascati, Roma, Italy
Science and Technology Facilities Council, ISIS Facility, Chilton Didcot Oxfordshire OQ11, UK
European Spallation Source ERIC, P.O. Box 176, 22 100 Lund, Sweden
d Thin Film Physics Division, Department of Physics, Chemistry and Biology (IFM), Linköping University, 581 83 Linköping, Sweden
e Istituto Nazionale di Fisica Nucleare, Laboratori Nazionali di Frascati, via E. Fermi 44, I-00044 Frascati, Roma, Italy
f CERN, 1211 Geneva 23, Switzerland
IHI Ionbond AG, Industriestraße 211, 4600 Olten, Switzerland
Boron coating
Gas Electron Multiplier
The synergic interplay between nuclear physics, detector technology and solid state and surface sciences is a
fundamental aspect of the development of new neutron detection devices. The synthesis technique and the
physical–chemical properties of the B4 C films used as a neutron-to-charged particle converter are described
in relation to the GEM side-on thermal neutron detector. Neutron detection is performed allowing scattered
neutrons to be converted into charged particles by means of a series of sheets covered by 10 B-enriched boron
carbide (B4 C) layers placed along their flight path inside the detector. The extremely interesting performance
shown by the detector in neutron diffraction measurements at the ISIS spallation neutron sources are discussed
and related to the chemical–physical properties of the converting layers.
1. Introduction
The lack of 3 He experienced in the late 2000s stimulated the advent
of different approaches to thermal neutron detectors [1]. Among all
the different approaches summarized in Ref. [1], in this experimental
work a special focus is devoted to Gas Electron Multiplier (GEM)-based
devices [2] in the so-called GEM side-on configuration [3–5]. The development of these novel thermal neutron detection concepts in neutron
scattering applications relies on a synergic and interdisciplinary link
between neutron science and materials science. In a recent experimental
work [6], the capability of this device to measure neutron diffraction
patterns on a reference sample with good efficiency and excellent time
and d-spacing resolution was assessed.
In the present paper, a thorough and systematic discussion of the
device’s main physical, chemical and engineering features is presented.
2. Experimental
2.1. Detector layout
Fig. 1 shows a schematic drawing of the GEM side-on detector and
a picture of the sheets distribution glued onto the cathode.
Three different regions are individuated: (1) the drift region located
in between the cathode and the first GEM foil (12 mm deep), (2) the
inter-GEM region (inter-GEM gaps are 1 mm, 2 mm and 1 mm) and (3)
the induction region between the last GEM foil and the pads readout. In
the drift region, two adjacent series of 15 silicon sheets (400 μm thick,
1 × 5 cm2 area) are glued onto the cathode. Each sheet is coated on
both sides with a B4 C layer (enriched in 10 B at 96%, 1.2 μm thickness)
with the inter-sheets distance set at 6 mm as found to be the optimized
choice in this case [3]. The GEM foils were 10 × 10 cm2 , the same area
is featured by the cathode and the anode. The latter was divided into
128 rectangular pads 3 × 24 mm2 .
The internal part of the cathode in contact with the gas features a
series of 15 straight grooves with a pitch of 6 mm. The drift region
thus contains 30 thermal neutron converting layers. Neutrons enter the
detector through a 1 mm thick 1 × 9 cm2 area glass window located
on one detector side in front of the silicon sheets (see Fig. 1). The
flushing gas used is a Ar/CO2 (70/30) mixture. The GEM foils were
biased at 870 V total high voltage, a value that guarantees a good
charged particle collection and high -ray background rejection, while
the drift electric field applied on the drift gap was 3 kV cm−1 . For the
other gaps, i.e. inter-GEM and induction region, the applied electric
fields were 3 kV cm−1 , 3 kV cm−1 and 5 kV cm−1 , respectively. Fig. 2
shows a schematic of the distribution of the readout pads into four
∗ Corresponding author.
E-mail address: (A. Pietropaolo).
Received 27 June 2018; Received in revised form 13 July 2018; Accepted 17 July 2018
Available online xxxx
0168-9002/© 2018 Elsevier B.V. All rights reserved.
A. Santoni et al.
Nuclear Inst. and Methods in Physics Research, A 906 (2018) 83–87
Fig. 1. (Color online) (left): schematic drawing of the triple-GEM side-on thermal neutron detector; (right) picture of the
B4 C covered silicon sheets inside the detector before final
Fig. 2. (Color online) Schematic of the distribution of readout sectors of the GEM side-on
detector. The last two sectors were not available during the measurements.
Fig. 3. Sample mounting in the CC800/9 deposition system during deposition of
layers onto Si-sheets, using 2-axis planetary rotation.
different sections, namely the upper first sector (S1U ), the first lower
sector (S1D ) etc. Each sector collects signal coming from the above-lying
eight borated sheets.
10 B
VG SSI monochromatized AlK and a PSP Vacuum Technology nonmonochromatized MgK. The spectrometer was a VG CLAM2 calibrated
on the Ag3 d5∕2 and Ag M4NN lines according to [10] and by setting
at 284.6 eV the binding energy (BE) of the adventitious carbon C1s
core-level measured on the as-inserted sample. Least squares fits for
data analysis were done by using a Voigt profile on a Shirley-type
background [10,11]. The estimated BE maximum error is ±0.1 eV.
For quantification, data were normalized to the photoelectron cross
section, analyzer transmission function and electron mean free path.
For sample etching a PSP Vacuum Technology ISIS 3000 system was
used as the Ar+ ion source. The sample was mildly bombarded with a
defocused beam of 500 eV Ar+ ions at about 12 μA total sample current.
XPS measurements were performed on the nominal B4 C-type sample at
cumulative Ar+ ion sputtering times of 0 min (i.e. as-inserted), 12 min,
42 min and 72 min.
2.2. Growth of B C
The coatings were deposited onto 2-side polished Si-sheets in an industrial CC800/9 deposition system manufactured by CemeCon AG (see
Fig. 3) situated at the ESS Detector Coatings Workshop in Linköping,
Sweden. The depositions were done using direct current magnetron
sputtering and the conditions were close to the those used in Refs. [7–9].
The samples were mounted in a 2-fold planetary rotation setup to allow
for 2-side deposition, see Fig. 3. The deposition system was evacuated
and the samples were heated and degassed for 3 h at a temperature of
about 350 ◦ C, to a base pressure of 0.2 mPa. Thereafter, Ar working gas
was let in and the deposition pressure was kept constant at 0.2 Pa, while
a power of 4 kW was applied to two magnetrons equipped with 10 B4 C
The film thickness on Si-wafer reference samples was measured to
be 1.2 μm, using a Bruker DektakXT profilometer. A uniform thickness
along the 5 cm length of the Si-sheets is expected due to the mounting
near the center of the sputtering targets, as discussed in Ref. [9].
SEM analysis shows B4 C films with a uniform morphology consisting
of mainly equal-axed grains with an average size of about 15 nm,
clustered in larger aggregates of 60–130 nm in size (Fig. 4). The film
shows a constant thickness of about 1.2 μm (inset of Fig. 4).
XPS data after the first sputtering cycle, showed the C1s photoemission peak changing shape and its binding energy position moved from
284.6 eV (i.e. C–C bonds or graphitic C) to 282.6 eV. On the other hand,
the B1s after the first sputtering did not change its original as-inserted
position and remained stable at 188.3 eV. These BE positions did not
change by further sputtering. The measured B1s at 188.3 eV BE and the
C1s at 282.6 eV BE indicate the presence of a B4 C bond structure and
are in agreement with Refs. [12,13]. Cumulative sputtering revealed the
presence of O, Fe (about 0.5 at.%) and traces of Cr and Ni on the etched
2.3. Morphology and chemical–physical analysis
A sample of the B4 C films was analyzed using Scanning Electron
Microscopy (SEM) and also by means of X-ray Photoelectron Spectroscopy (XPS). A LEO 1525 field-emission, SEM was used to study
the morphology of B4 C films. The thickness of B4 C has been evaluated
by cross-sectional SEM by peeling off the film from the substrate. XPS
data were acquired by means of two different soft X-ray sources: a
A. Santoni et al.
Nuclear Inst. and Methods in Physics Research, A 906 (2018) 83–87
Fig. 4. SEM picture of B4 C film grown on Si substrate (scale bar: 100 nm). The inset
shows a picture of the cross section of B4 C film peeled off from the substrate (scale bar:
1 μm).
Fig. 6. (Color online) ToF-ERDA depth profile from a 1.2 μm10 B4 C thin film deposited
onto Si.
confirmed to be around 97%. The C-content is 15.8 at. % and the total
amounts of the common impurities H, O, and N are below 1 at.%. Like
in the XPS measurements, the presence of Fe was observed and some
traces of the sputtering gas Ar were found. Also, the increased amount
of O in the surface layers was confirmed.
3. Neutron diffraction measurements
As thoroughly described in Ref. [8] the detector was used for a
neutron diffraction measurement on a reference Cu sample using the
INES beamline [18,19]. The layout of the experiment is schematically
drawn in Fig. 7. The GEM side-on was placed at a scattering angle 2=
90◦ , almost in the same angular position of one of the 144 squashed 3 Hetubes of the INES detectors array namely the one at 2 = 89.6◦ . During
the measurements only the first two sectors, namely S1=S1U +S1D and
S2=S2U +S2D , where operative. The neutron diffraction experiment was
performed using a standard Cu powder sample and the following set
of measurements were performed: (1) A measurement without sample
(empty instrument) for background estimation (Ip = 1200 μAh); (2) a
measurement of a 11 mm diameter vanadium rod for detector efficiency
correction (Ip = 1800 μAh) and (3) a measurement of a Cu standard
powder in a vanadium can of 6 mm diameter (Ip = 1145 μAh), where Ip
is the total beam charge delivered for a given measurement (the average
ISIS proton beam current on Target Station 1, where INES operates, is
about 165 μA).
Fig. 8 shows the neutron ToF diffraction pattern measured by
means of the GEM side-on detector. The upper panel shows the ToF
spectra recorded by S1U , S1D , S2U and S2D separately, while the lower
panel those obtained by summing signals coming from (S1U +S1D ) and
(S2U +S2D ). It can be noted that the diffraction peaks for the S1 and S2
are different both in intensity and in ToF position. The former effect is
due to absorption of the neutron beam in passing through the borated
sheets array, while the latter to the time needed by a (thermal) neutron
to cover the distance between sector 1 and 2. As a matter of fact, it is
more simple to explain that the axial symmetry with respect to the 90◦
position allows the equivalence between the upper and lower sectors.
The same applies to S2U and S2D .
Referring to Fig. 8, a preliminary analysis on the main peak parameters was performed, Table 1 summarizing the main results for the single
GEM sectors and compared to 3 He tube.
In order to evaluate the detection performance of the GEM side-on
detector, the ToF spectra in the lower panel of Fig. 9 were converted in
-spacing using the Bragg’s law and the De Broglie’s equation obtaining
Fig. 5. (Color online) 72’ sputtered B1s least-squares core-level fitting. Data: black dots;
red continuous line: best fit; background: green short-dashed line. The components of the
fit are shown with blue lines and triangles.
surface. Oxygen was observed to decrease from about 4.8 at.% after 12
min sputtering to 2.0 at.% after 72 min sputtering. Fig. 5 shows the B1s
core-level obtained from a surface that was sputter-cleaned for 72 min.
The B1s lineshape can be fitted with a main peak at 188.3 eV, assigned
to B–C bonds in B4 C, and a smaller but wider component at about 1eV
toward higher binding energy. The latter component, featuring a nearly
constant area of about 20% of the total peak area, was observed to
move from about 189.8 eV in the 12’ etched sample to 189.4 eV in the
42’ etched surface up to 189.2 eV after 72 min sputtering. Components
around these energies have been associated to B–C–O bonds [14,15] and
B oxides or suboxides [16]. The smaller component may be assigned
to B bonded to O and C in a disordered chemical environment. This
is corroborated by the fact that the component shifts by 0.6 eV toward
lower BE throughout the course of the sputter cleaning by which surface
O is removed.
The film composition was investigated using isotope-specific time-offlight elastic recoil detection analysis (ToF-ERDA), with a 36 MeV 127 I8+
beam at 66◦ incidence and 45◦ recoil scattering angle. The recoil energy
of each element was converted to relative elemental depth profiles using
the CONTES code [17]. The ToF-ERDA depth profile from the 10 B4 C
on Si is plotted in Fig. 6 and shows a uniform elemental distribution
throughout the measurement depths of a few hundred of nanometers.
The 10 B content is found to be above 80 at. % and the 10 B enrichment is
A. Santoni et al.
Nuclear Inst. and Methods in Physics Research, A 906 (2018) 83–87
Fig. 7. Schematic layout of the neutron diffraction measurement at INES with the side-on GEM detector.
Table 1
Comparison of the main parameters for a representative selection of diffraction peaks in figure 8, for the different sectors of the GEM detector, as
compared to the corresponding ones for the 3 He tube: p is the ToF position of the peak, R is the peak to background ratio and FWHM is the Full
Width at Half Maximum.
GEM (S1U +S1D )
p [μs]
FWHM [μs]
GEM (S2U +S2D )
p [μs]
FWHM [μs]
He tube
p [μs]
FWHM [μs]
Table 2
Value of -spacing for different crystallographic planes of the Cu sample, identified by the Miller indexes [hkl] as determined from
the diffraction patterns registered by two adjacent 3 He tubes with angular positions close to the GEM detector and by the two
sectors of the GEM Side-on detector (see text for details). The -spacing uncertainty can be estimated as 0.001 Å.




the following formula:  = 2 ⋅⋅sin
, where ℎ is the Plank constant, 

is the neutron mass,  is the total flight path of the neutron, and  is half
of the scattering angle. L=L1 +L2 , where 1 is the moderator to sample
distance and 2 is the sample to detector distance. As  and  are known
for the 3 He detector and sector 1 and 2 of the GEM side-on, ToF spectra
can be converted into d-spacing diffraction patterns. Fig. 9 shows the
neutron diffraction pattern in the -spacing domain for the GEM side-on
detector and the 3 He tube: the pattern in (a) is obtained summing S1U
and S1D while (b) summing over S2U and S2D (see Fig. 2). In Table 2,
the -values obtained from the diffraction patterns in Fig. 9 are shown
for a selection of representative crystallographic planes defined by the
Miller’s indexes [hkl].
show that the detector has good performances (although not optimized
because it was only 50% operating during the measurements).
Nevertheless, it can be highlighted how the combination of the good
chemical composition of the B4 C coating, the choice of its thickness providing the best trade-off between neutron absorption and the escaping
charged particles and the design of the readout electronics provided an
excellent ToF resolution, good efficiency and high count rate capability
thanks to the detector’s intrinsic fast detector response.
By means of the GEM side-on detector, neutron diffraction spectra on
a reference sample were successfully recorded at the INES diffractometer
operating at the ISIS spallation neutron source in UK. The performance
of the detector, compared with one of the standard 3 He tube of the instrument, can be summarized as follows: (1) a quite good ToF resolution,
in turn improvable by a finer knowledge of the readout pad position with
respect to the sample (i.e. a well calibrated secondary neutron flight
path); (2) although only half of the detector was operating the overall
efficiency is found to be around 15%. Although a clear evaluation is
only possible by means of experimental measurements (foreseen in a
next experimental campaign on INES), nevertheless relying on previous
Monte Carlo simulations described in Ref. [3], a rough extrapolation can
be made that provided a projected efficiency for the whole operating detector with the present B4 C sheets arrangement and thickness, of 30%–
35%. This estimation can be improved by considering the possibility
4. Conclusions
SEM and XPS data show that films consist of aggregated nano-sized
grains of B4 C with a low oxygen content slowly decreasing after the first
sputtering cycle. The analysis of the B1s core level lineshape showed the
presence of two B components. The more intense one is indicative of the
B–C bond in B4 C and the other may be assigned to B bonded to O and C
in a disordered chemical environment. The neutron diffraction recorded
by the GEM side-on and compared to that obtained by the 3 He tube
A. Santoni et al.
Nuclear Inst. and Methods in Physics Research, A 906 (2018) 83–87
Fig. 9. (a) Diffraction pattern in the -spacing domain for the first sector (S1U +S1D ) of
the GEM side-on; (b) the same as (a) but for the second sector (S2U +S2D ); (c) diffraction
pattern in the d-spacing domain for the 3 He tube placed opposite to the GEM detector (see
Fig. 7) at 2 = 89.6◦ .
[2] F. Sauli, Nucl. Instrum. Methods A 386 (1997) 531.
[3] A. Pietropaolo, F. Murtas, G. Claps, L. Quintieri, D. Raspino, G. Celentano, A.
Vannozzi, O. Frasciello, Nucl. Instrum. Methods A 729 (2013) 117.
[4] G. Claps, F. Murtas, A. Pietropaolo, G. Celentano, A. Vannozzi, A. Santoni, L.
Quintieri, R.A. Riedel, Europhys. Lett. 105 (2014) 22002.
[5] F. Murtas, A. Pietropaolo, G. Claps, L. Quintieri, O. Frasciello, D. Raspino, G.
Celentano, IEEE Nuclear Science Symposium Conference Record, 2012, pp. 4301–
4303, Art. no. 6551980.
[6] A. Pietropaolo, G. Claps, A. Fedrigo, F. Grazzi, C. Höglund, F. Murtas, A. Scherillo,
S. Schmidt, E.M. Schooneveld, Europhys. Lett. 121 (2018) 62001.
[7] C. Höglund, J. Birch, K. Andersen, T. Bigault, J.C. Buffet, J. Correa, P. Van Esch,
B. Guerard, R. Hall-Wilton, J. Jensen, A. Khaplanov, F. Piscitelli, C. Vettier, W.
Vollenberg, L. Hultman, J. Appl. Phys. 111 (2012) 104908.
[8] C. Höglund, K. Zeitelhack, P. Kudejova, J. Jensen, G. Greczynski, J. Lu, L. Hultman,
J. Birch, R. Hall-Wilton, Rad. Phys. Chem. 113 (2015) 14.
[9] S. Schmidt, C. Höglund, J. Jensen, L. Hultman, J. Birch, R. Hall-Wilton, J. Mater.
Sci. 51 (2016) 10418.
[10] M.T. Anthony, Spectrometer calibration, in: D. Briggs, M.P. Seah (Eds.), Practical
Surface Analysis By Auger and X-Ray Photoelectron Spectroscopy, in: App., vol. 1,
Wiley, New York, 1983, pp. 429–435.
[11] D.A. Shirley, Phys. Rev. B 5 (1972) 4709.
[12] R. Bao, D.B. Chrisey, Thin Solid Films 519 (2010) 164.
[13] W.J. Pan, J. Sun, H. Ling, N. Xu, Z.F. Ying, J.D. Wu, Appl. Surf. Sci. 218 (2003)
[14] S. Jacques, A. Guette, X. Bourrat, F. Langlais, C. Guimon, C. Labrugere, Carbon 34
(1996) 1135.
[15] L.G. Jacobsohn, R.K. Schulze, M.E.H. Maia da Costa, M. Nastasi, Surf. Sci. 572
(2004) 418.
[16] G. Celentano, A. Vannozzi, A. Mancini, A. Santoni, A. Pietropaolo, G. Claps, E.
Bemporad, M. Renzelli, F. Murtas, L. Quintieri, Surf. Coat. Technol. 265 (2015)
[17] M.S. Janson, CONTES conversion of time-energy spectra—A program for ERDA
data analysis, Internal Report Uppsala University, 2004.
[18] F. Grazzi, M. Celli, S. Siano, M. Zoppi, Nuovo Cimento C 30 (2007) 59.
[19] S. Imberti, W. Kockelmann, M. Celli, F. Grazzi, M. Zoppi, A. Botti, A. Sodo, M.
Leo Imperiale, M. de Vries-Melein, D. Visser, H. Postma, Meas. Sci. Technol. 19
(2008) 034003.
Fig. 8. (Color online) (upper panel) ToF neutron spectra recorded by the separate sectors
S1U , S1D , S2U and S2D ; (lower panel) comparison between sector (S1U +S1D ) and sector
(S2U +S2D ) of the GEM side-dn. The shift in the ToF peaks is due to the different sampleto-detector distance of sector 1 and sector 2 (1 m and 1.024 m respectively). The lower
intensity of sector S2 with respect to S1 is due to beam attenuation in passing through the
B4 C coated sheets.
to make B4 C converting layers featuring about 2 μm thickness and/or
using pure 10 B layers. These options may provide an enhancement of
the efficiency so to reach about 50% efficiency at 25 meV neutron
energy (very close to that of the 3 He tubes used in the experiment).
Experimental tests are currently carried out about deposition of pure 10 B
layers on different substrate materials. This experimental work shows
the importance of the interplay of nuclear physics, detector technology
and solid state and surfaces science in achieving new developments in
neutron detection concepts.
This work was partially funded under the CNR (Italy)-STFC (UK)
06/20018 Cooperation Agreement. Consiglio Nazionale delle Ricerche
is hereby acknowledged.
[1] N. Colonna, A. Pietropaolo, F. Sacchetti (Eds.), Focus Point on 3 He replacement in
neutron detection: Current status and perspectives, Eur. Phys. J. Plus 130 (2015)
Без категории
Размер файла
2 419 Кб
niman, 2018, 045
Пожаловаться на содержимое документа