Measurement of (γ, p) reactions with ∆E − E telescope at Max-lab facility
The paper considers the results of experiments on the reactions ¹²C(γ, p)¹¹B and d(γ, p)n in the energy range of tagged photons 35...80 MeV. Demonstrated the possibility identification of protons by ΔE - E using CsI/SSD telescope. Using the spectra of the missing energy defined the values of differe...
Збережено в:
| Дата: | 2015 |
|---|---|
| Автори: | , , , , , , , , , |
| Формат: | Стаття |
| Мова: | English |
| Опубліковано: |
Національний науковий центр «Харківський фізико-технічний інститут» НАН України
2015
|
| Назва видання: | Вопросы атомной науки и техники |
| Теми: | |
| Онлайн доступ: | http://dspace.nbuv.gov.ua/handle/123456789/112106 |
| 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: | Measurement of (γ, p) reactions with ∆E − E telescope at Max-lab facility / D. Burdeinyi, J. Brudvik, V. Ganenko, K. Hansen, K. Fissum, L. Isaksson, K. Livingston, M. Lundin, B. Nilsson, B. Schr¨oder // Вопросы атомной науки и техники. — 2015. — № 3. — С. 49-64. — Бібліогр.: 12 назв. — англ. |
Репозитарії
Digital Library of Periodicals of National Academy of Sciences of Ukraine| id |
irk-123456789-112106 |
|---|---|
| record_format |
dspace |
| spelling |
irk-123456789-1121062017-01-18T03:03:16Z Measurement of (γ, p) reactions with ∆E − E telescope at Max-lab facility Burdeinyi, D. Brudvik, J. Ganenko, V. Hansen, K. Fissum, K. Isaksson, L. Livingston, K. Lundin, M. Nilsson, B. Schroder, B. Ядерно-физические методы и обработка данных The paper considers the results of experiments on the reactions ¹²C(γ, p)¹¹B and d(γ, p)n in the energy range of tagged photons 35...80 MeV. Demonstrated the possibility identification of protons by ΔE - E using CsI/SSD telescope. Using the spectra of the missing energy defined the values of differential cross sections of these reactions in the range of photon energies. The good agreement of the experimental results with the available data in the literature. Розглянуто результати експериментів з вивчення реакцій ¹²C(γ, p)¹¹B і d(γ, p)n в області енергій мічених фотонів 35…80 МеВ. Показана можливість ідентифікації протонів методом ΔE-E за допомогою CsI/SSD-телескопа. Використовуючи спектри недостатніх енергій, визначені диференційні перерізи розглянутих реакцій в зазначених діапазонах енергій фотонів. Отримані експериментальні результати добре узгоджуються з літературними даними. Рассматриваются результаты экспериментов по изучению реакций ¹²C(γ, p)¹¹B и d(γ, p)n в области энергий меченых фотонов 35…80 МэВ. Показана возможность идентификации протонов методом ΔE-E с помощью CsI/SSD-телескопа. Используя спектры недостающих энергий, определены значения дифференциальных сечений рассматриваемых реакций в указанных диапазонах энергий фотонов. Показано хорошее согласие результатов экспериментов с имеющимися литературными данными. 2015 Article Measurement of (γ, p) reactions with ∆E − E telescope at Max-lab facility / D. Burdeinyi, J. Brudvik, V. Ganenko, K. Hansen, K. Fissum, L. Isaksson, K. Livingston, M. Lundin, B. Nilsson, B. Schr¨oder // Вопросы атомной науки и техники. — 2015. — № 3. — С. 49-64. — Бібліогр.: 12 назв. — англ. 1562-6016 PACS: 03.65.Pm, 03.65.Ge, 61.80.Mk http://dspace.nbuv.gov.ua/handle/123456789/112106 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 |
Ядерно-физические методы и обработка данных Ядерно-физические методы и обработка данных |
| spellingShingle |
Ядерно-физические методы и обработка данных Ядерно-физические методы и обработка данных Burdeinyi, D. Brudvik, J. Ganenko, V. Hansen, K. Fissum, K. Isaksson, L. Livingston, K. Lundin, M. Nilsson, B. Schroder, B. Measurement of (γ, p) reactions with ∆E − E telescope at Max-lab facility Вопросы атомной науки и техники |
| description |
The paper considers the results of experiments on the reactions ¹²C(γ, p)¹¹B and d(γ, p)n in the energy range of tagged photons 35...80 MeV. Demonstrated the possibility identification of protons by ΔE - E using CsI/SSD telescope. Using the spectra of the missing energy defined the values of differential cross sections of these reactions in the range of photon energies. The good agreement of the experimental results with the available data in the literature. |
| format |
Article |
| author |
Burdeinyi, D. Brudvik, J. Ganenko, V. Hansen, K. Fissum, K. Isaksson, L. Livingston, K. Lundin, M. Nilsson, B. Schroder, B. |
| author_facet |
Burdeinyi, D. Brudvik, J. Ganenko, V. Hansen, K. Fissum, K. Isaksson, L. Livingston, K. Lundin, M. Nilsson, B. Schroder, B. |
| author_sort |
Burdeinyi, D. |
| title |
Measurement of (γ, p) reactions with ∆E − E telescope at Max-lab facility |
| title_short |
Measurement of (γ, p) reactions with ∆E − E telescope at Max-lab facility |
| title_full |
Measurement of (γ, p) reactions with ∆E − E telescope at Max-lab facility |
| title_fullStr |
Measurement of (γ, p) reactions with ∆E − E telescope at Max-lab facility |
| title_full_unstemmed |
Measurement of (γ, p) reactions with ∆E − E telescope at Max-lab facility |
| title_sort |
measurement of (γ, p) reactions with ∆e − e telescope at max-lab facility |
| publisher |
Національний науковий центр «Харківський фізико-технічний інститут» НАН України |
| publishDate |
2015 |
| topic_facet |
Ядерно-физические методы и обработка данных |
| url |
http://dspace.nbuv.gov.ua/handle/123456789/112106 |
| citation_txt |
Measurement of (γ, p) reactions with ∆E − E telescope at Max-lab facility / D. Burdeinyi, J. Brudvik, V. Ganenko, K. Hansen, K. Fissum, L. Isaksson, K. Livingston, M. Lundin, B. Nilsson, B. Schr¨oder // Вопросы атомной науки и техники. — 2015. — № 3. — С. 49-64. — Бібліогр.: 12 назв. — англ. |
| series |
Вопросы атомной науки и техники |
| work_keys_str_mv |
AT burdeinyid measurementofgpreactionswitheetelescopeatmaxlabfacility AT brudvikj measurementofgpreactionswitheetelescopeatmaxlabfacility AT ganenkov measurementofgpreactionswitheetelescopeatmaxlabfacility AT hansenk measurementofgpreactionswitheetelescopeatmaxlabfacility AT fissumk measurementofgpreactionswitheetelescopeatmaxlabfacility AT isakssonl measurementofgpreactionswitheetelescopeatmaxlabfacility AT livingstonk measurementofgpreactionswitheetelescopeatmaxlabfacility AT lundinm measurementofgpreactionswitheetelescopeatmaxlabfacility AT nilssonb measurementofgpreactionswitheetelescopeatmaxlabfacility AT schroderb measurementofgpreactionswitheetelescopeatmaxlabfacility |
| first_indexed |
2025-07-08T03:24:18Z |
| last_indexed |
2025-07-08T03:24:18Z |
| _version_ |
1837047544286281728 |
| fulltext |
NUCLEAR-PHYSICAL METHODS AND PROCESSING OF DATA
MEASUREMENT OF (γ, p) REACTIONS WITH ∆E − E
TELESCOPE AT MAX-LAB FACILITY
D.Burdeinyi1, J.Brudvik2, V.Ganenko1, K.Hansen3, K.Fissum3,
L. Isaksson3, K.Livingston4, M.Lundin2, B.Nilsson2, B.Schröder2,3
1National Science Center ”Kharkov Institute of Physics and Technology”, Kharkov, Ukraine;
2MAX-lab, Lund University, SE-221 00 Lund, Sweden;
3Department of Physics, Lund University, SE-221 00 Lund, Sweden;
4Department of Physics and Astronomy, University of Glasgow, Glasgow G12 8QQ, Scotland, UK
(Received April 2, 2015)
The paper considers the results of experiments on the reactions 12C(γ, p)11B and d(γ, p)n in the energy range
of tagged photons 35...80MeV. Demonstrated the possibility identification of protons by ∆E − E using CsI/SSD
telescope. Using the spectra of the missing energy defined the values of differential cross sections of these reactions
in the range of photon energies. The good agreement of the experimental results with the available data in the
literature.
PACS: 03.65.Pm, 03.65.Ge, 61.80.Mk
1. INTRODUCTION
The (γ, p) reactions are of the most studied pho-
tonuclear processes, which have been widely investi-
gated in the energy range between the Giant Dipole
Resonance and the pion production threshold, see
[1] and reference in their. The purpose of these re-
searches was, on the one hand, to study nuclear struc-
ture, and on the other hand, to determine the mech-
anisms of photon absorption by nuclei in this energy
range, in particular, to study relative role of the direct
knock-out and quasi-deuteron mechanisms. Produc-
tion of linearly polarized photon beam at MAX-lab
[2] opened a new possibility for investigations.
One of a simple experimental technique, available
in the MAX-lab at present time and which could
be applied for the (γ, p) reactions investigations is
a ∆E-E CsI/SSD telescope [3]. The telescope con-
sists of two single-sided silicon strip detectors and
CsI counter which function as (∆E) and (E) detec-
tors, respectively. In order to study the telescope
characteristics, and its experimental possibility of the
(γ, p) reactions identification, the measurements of a
deuteron and a carbon photodisintegration have been
performed. In this paper results of the data process-
ing are presented, and methods of the (γ, p) reactions
selection are analyzed.
2. EXPERIMENTAL APPARATUS AND
TECHNIQUE
The measurements have been produced at the
MAX-lab nuclear physics facility, described in [4] in
detail. The facility has advanced infrastructure for
precision photonuclear experiments in energy range
from Giant Dipole Resonance and to some ten MeV
above the pion threshold: (i) The electron beam
with maximal energy E0 ≈ 200MeV, duty cycle
df ≈ 50...70% and current up to 20 nA; (ii) Two tag-
ging systems which cover energy interval from 10 to
180MeV with energy resolution 0.5...1MeV; (iii) Sys-
tems of the beam diagnostic and control.
2.1. Beam and beam line
The electron beam was extracted from the MAX-
I storage ring which worked in a stretcher mode.
Injection of electrons into the ring was performed
by a double-section linear accelerator at a frequency
of 10Hz and duration of the injected electron pulse
about 200 ns. The electron energy was E0 =
192.7MeV. The electrons were slowly extracted from
the MAX-I ring during 100ms and by a beam trans-
portation system delivered into experimental hall.
A schematic picture of the beam line and experi-
mental set up is shown in Fig.1. A dipole magnet (1)
directed the electron beam towards 50µm Al pho-
ton radiator fixed in a target holder of a goniometer
(5). The goniometer was placed in a vacuum cham-
ber between magnets of the end-point tagger (ET)
(4) and the main tagger (MT) (6). The electron cur-
rent on photon radiators was ∼ 5...10 nA. The beam
size on the radiators was no more than 2mm. A non-
interacting part of the electron beam was deflected
to the beam dump (8) by the MT magnet, where it
was absorbed by a Faraday cup. At the MT magnet
setting, used in the experiment, the electrons passed
a section of air (∼ 135 cm) and two steal foils 25µm
thick on its way to the beam dump [4].
ISSN 1562-6016. PROBLEMS OF ATOMIC SCIENCE AND TECHNOLOGY, 2015, N3(97).
Series: Nuclear Physics Investigations (64), p.49-64.
49
A photon collimator was placed on the dis-
tance 2140mm from the photon radiators, before
the shielding wall. The collimator total length
is 40 cm. It consists of heavy metal main colli-
mator 108.5mm long with the variable entrance
openings followed by a scrubber magnet ∼ 100mm
long, and a scrubber collimator 200mm long. The
12mm opening was used in the measurements.
Fig.1. Scheme of the MAX-lab beam line for
photonuclear researches. 1-bending magnet (500);
2-vertical and horizontal steering magnets; 3-photon
radiator for end-point tagger; 4-end-point tagger
(ET); 5-goniometer; 6-main tagger (MT);
7-additional dump magnet; 8-Faraday cup;
9-shielding; 10-photon collimator; 11-gamma
monitor (IBM); 12-SSD/CsI telescope; 13-photon
beam dump
It provided the collimation angle θc ≈ 1.2θγ for a
point-like electron beam (θγ = mec
2/E0 is the char-
acteristic angle of bremsstrahlung, E0 and me are
the electron energy and mass). The intensity of the
photon beam was controlled by a scintillation gamma
beam monitor (IBM) (11) [4].
2.2. Photon beam tagging system
The bremsstrahlung spectrum, generated due to
interaction of high energy electrons with a photon ra-
diator, contains energies from zero to maximal elec-
tron energy E0. The photon energy value is ob-
tained from the relation Eγ = E0 − E′
e, where E′
e
is the energy of a post-bremsstrahlung electron. The
post-bremsstrahlung electrons are detected by a focal
plane (FP) hodoscope, placed along the MT magnet
focal plane. Their energy is determined by the MT
magnet field and position of the counters of the FP
hodoscope along the focal plane. In our case the focal
plane position was different from the standard loca-
tion due to upstream shift of the position of the pho-
ton radiators, fixed in the goniometer target holder.
The new focal plane position was calculated by trac-
ing electrons in the magnetic field of MT and pre-
sented in [4].
The FP hodoscope consists of two rows of scintil-
lators 25mm wide and 3.2mm thick. The first row,
facing the exit window of the MT, contains 31 and
the second row, behind the first, contains 32 scintil-
lators. The overlap between scintillators of the rows
was 50%. The coincidence requirement to overlap-
ping scintillators resulted in 62 channels for detecting
the post-bremsstrahlung electrons.
The momentum acceptance of the MT spectrom-
eter is ±40% of PC , where PC is the central momen-
tum of the MT tagging spectrometer. Its value is
determined in [4]. In our beam run, in order to ex-
tend the tagging interval to lower photon energies,
the hodoscope was shifted above the position corre-
sponding to momentum acceptance +40%, so that
the energy interval above the +40% mark was in-
cluded. It covered the tagger channels 50-62, corre-
sponding to energies Eγ = 21.9...33.8MeV. Thus, the
total tagged energy range of the photon beam tagging
was Eγ = 21.9...78.8MeV.
Energy resolution of the tagger system in the case
of coincidence requirement between the scintillator
rows depends on the dispersion of the MT magnet
(presented in [4]), the effective width of the tagging
channel along the focal plane, which depends on over-
lap of the scintillators, and an angle between the di-
rection of the post-bremsstrahlung electron trajec-
tories and the MT focal plane. The energy resolu-
tion of the tagged photon channels was calculated
for 50% overlap and normal incident of the scat-
tered electrons to the MT focal plane. The resolu-
tion smoothly varies from ∆Eγ ≈ 0.8MeV, for the
high end of the tagged range, to ∆Eγ ≈ 1MeV at
Eγ ≈ 38MeV, and then it is almost constant in the
interval Eγ = 22...38MeV. However, the real an-
gle between the direction of the post-bremsstrahlung
electron trajectories and the new MT focal plane is
52.60. This results in a difference between the mo-
mentum acceptance of odd and even FP channels,
when coincidence requirement is applied to the over-
lapping scintillators. This brings to different rate of
the post-bremsstrahlung electrons detected by the FP
channels. To exclude this effect, summation of the
yields over neighboring FP channels was produces.
2.3. Targets
The measurements have been carried out using
a CD2 and a CH2 targets. The CH2 target was
applied both for carbon disintegration measurement
and for removing the background contribution from
this process to the deuteron disintegration. The CD2
target had a disc shape, 75mm in diameter and
1mm thick. The CH2 target had a square form,
150.2 × 150.2mm2 and 1.1mm thick. The density
of the targets was determined by their weighing and
calculation their volume. The calculated density of
50
the CD2 target was ρ = 1.026 g/cm3, and the CH2
target ρ = 0.937 g/cm3.
The targets were positioned on the distance ≈ 2m
from the photon collimator under angle θm = 600
to the photon beam direction. The effective number
of deuteron per area was ND = 1.541 × 1022 cm−2,
and the effective number of 12C nuclei per area was
NC = 0.771 × 1022 cm−2 for the CD2 target and
NC = 0.885× 1022 cm−2 for the CH2 target.
2.4. CsI/SSD telescope
The emitted protons were detected by a CsI/SSD
telescope [3], schematically shown in Fig.2. The tele-
scope consists of two identical single-sided silicon
strip detectors (SSDs), and a CsI(Tl) counter. The
SSD detectors (∆E) are of octagonal shape, have an
active area 3300mm2 and a thickness 0.5mm. They
have 64 strips, each with a width of 1mm. The strips
are paralleled in groups of two for the read-out, thus
yielding an effective strip width of 2mm. The ac-
tive aria of the detectors was wrapped Al foil 15µm
thick. The CsI(Tl) detector (E) is of a cylinder shape
12.5 cm in diameter and 10 cm long, placed in Al con-
tainer with the standard Al-front-foil.
Fig.2. Scheme of the CsI/SSD telescope, see text
The telescope was placed under angle θp = 900 to
the beam axis. The distance between the first SSD
and the center of the photon beam sport on the tar-
get was 98.5mm, the distance between the SSDs was
15mm, and the distance between the second SSD and
CsI was 10mm.
3. DATA ANALYSIS
The measurements have been carried out in the
course of three short (∼ 1 hour) beam runs for both
targets. The total duration of the data taking was
∼ 3.52 hour for the CD2 and ∼ 2.92 hour for the CH2
targets.
3.1. Proton identification
Proton identification has been performed by stan-
dard ∆E − E method, based on relationship be-
tween the energy losses in a ∆E detector and full
energy E of the particles with different masses. Typ-
ical two-dimensional plot pairs of ADC signals from
the SSD (∆E) and CsI (E) detectors is shown in
Fig.3, where one can see separation of the protons
from other particles (electrons, deuterons, etc.) into
clear band. For further analysis the background par-
ticles were removed by special soft cut of the ∆E−E
plot, shown in Fig.3. The selected proton yield re-
sults from various reactions of the carbon disintegra-
tion, and the deuteron disintegration, as well, if the
CD2 target is used. At that, for photon energies
Eγ < 52MeV, due to threshold of the proton de-
tection by the telescope, only two-body reactions of
the carbon disintegration, 12C(γ, p)11B, give contri-
bution to the proton yield, in which the residual nu-
cleus is in the ground state or one of the low-lying ex-
cited states with excitation energies Eex =2.13, 5.02,
6.74, 6.79 and 7.29MeV. At higher photon energies,
there is some contribution from the processes with
higher exited states of 11B and two-nucleon emission.
0 200 400 600 800 1000 1200 1400 1600 1800 20000
200
400
600
800
1000
1200
1400
C
h
an
n
el
, S
S
D
1
Channel, CsI
Proton
Deuteron
Fig.3. Two-dimensional plot pairs of corresponding
signals from the SSD and CsI detectors. Lines
demonstrate separation of the proton band by special
soft cut
3.2. Time coincidence focal plane hodoscope
detectors with CsI/SSD telescope
The stretched electron beam at MAX-lab has a
rather complicated time structure which complicates
the search for coincidences between the FP detectors
and the CsI/SSD telescope signals. The CsI/SSD
telescope (the coincidences between the SSDs and the
CsI detector) produces trigger signals, which are gen-
erated by protons, electrons and other background
particles, and electronic noise. Their number is iden-
tified by a selfcoincidence in the CsI/SSD trigger
TDC. The trigger signals start all time TDCs of the
FP array which are stopped by the pulses from the
51
FP detectors. Thus, the first step in the analysis
consists of selecting events triggered by the CsI/SSD
telescope, corresponding to the proton band, Fig.3.
After such selection the random coincidences contri-
bution in the FPtdc spectrum was strongly decreased.
The trigger start timing was determined by the
OR of the SSDs signals. These SSD signals came
from leading edge discriminators, and thus they had
a significant time walk (∼ 20 ns). As a result, the
time coincidence FPtdc spectra were spread and there
was no clear peak of the time coincidence. In order to
cancel this time walk, the CsI detector was used as a
time reference, i.e., the CsI TDC was subtracted from
all other TDCs, taking the different TDC conversion
gains into account. In this case only events above
the CsI discriminator threshold were used. Having
performed the steps outlined above, coincidence peak
appears in the individual tagger TDCs, as shown in
Fig.4, where one can see a strong prompt peak of
the time coincidence of the FP and CsI/SSD tele-
scope signals on top of random background. The
time resolution of the FPtdc coincidence is ≈ 2...3 ns.
0 500 1000 1500 2000 2500 3000 35000
100
200
300
400
500
600
700
800
C
ou
nt
s
Channel
Prompt
Random
600 650 700 750 800 850 900
0
100
200
300
400
500
600
700
800
σ−3 σ+3
Fig.4. The time coincidence spectra (FPtdc)
between the SSD/CsI telescope and the FP channel,
corresponding to photon energy
Eγ = 50.4± 0.5MeV, after selecting events from the
proton band, and cancellation the SSD signals time
walk. Blue color shows the prompt peak range,
brown color shows range of the random coincidence
which is used for the background spectrum
construction. Curves in insertion are Gauss fits of
the background and the prompt peak, see text
Due to slightly varying delays for the tagger sig-
nals, the prompt peaks will not be in exactly the same
position for each TDC channel but can be differed up
to several tens of the TDC channels. In order to en-
able summing the FPtdc spectra, the prompt peaks
for all FP channels were shifted to identical position.
In order to produce the shift, the coordinates of the
peak’s positions were determined for all FP channels.
For that the time structure of the random background
under the prompt peak of the FPtdc spectrum was
described by approximating the distribution with a
Gaussian, Fig.4. Then the whole spectrum was fit-
ted with a sum of two Gaussians: a broad Gaussian
approximating the background under the peak, and
a narrow Gaussian describing the coincidence peak.
From the fit the position µ and the standard deviation
σ of the peak were determined. The range µ±3σ was
used to define the prompt region, shown blue shaded
in Fig.4. Having the prompt peaks coordinates for
all FP channels, the prompt peaks in all spectra were
shifted to the same 750 channel of the FPtdc.
In the prompt region, µ± 3σ, there are two types
of events: true coincidence events (from a deuteron
and various channels of carbon disintegration, an
electromagnetic background), and random events.
The region to the right of the peak, shown brown
shaded in Fig.4, where only random events are pre-
sented, are subsequently labeled as the random re-
gion. Using the prompt and random region so de-
fined, two different missing energy spectra can be
generated: the spectrum for events in the prompt
region, and one for events in the random region.
3.3. Tagging efficiency
A large portion of the tagged-photon beam was
collimated away, thus not every electron registered in
the focal plane detectors corresponded to a photon in-
cident on the target. This loss is given by the photon
beam tagging efficiency, which relates the number of
photons, impinging on the target, e.g., corresponding
to the tagger channel (i), Nγ,ch(i), to the number of
post-bremsstrahlung electrons, detected by the focal-
plane array channel (i), N ′
e(i),
εtag(i) =
Nγ,ch(i)
N ′
e(i)
. (1)
Measurements of the tagging efficiency are usually
performed at strongly reduced the electron beam cur-
rent by a special detector, placing in the photon beam
path, which detects all photons that passed through
the collimator. In our case the tagging efficiency was
measured with a NaI detector 25 × 25 × 25 cm3 [5].
The number of photonsNγ,ch(i) was determined from
ADC spectrum of the NaI signals, taking into account
only the part, corresponding to the tagged photon
interval. The numbers of electrons, recorded in the
focal plane detectors NFP (i), were determined using
the TDCs spectra, in which ”self-coincidence” peaks
were generated by the electrons, registered by the FP
hodoscope counters. The peaks include also back-
ground contributions,
NFP (i) = N ′
e(i) +Nbg,beam(i) +Nbg,const(i), (2)
Nbg,beam(i) is the background yield generated by the
beam and detected by the i-th FP counter. This
background may come from various sources, includ-
ing Meller electrons from the radiator and the beam
halo striking a vacuum chamber. The N ′
e(i) and
Nbg,beam(i) rates are proportional to the beam inten-
sity. Nbg,const(i) is the background resulting from a
constant room background (cosmic radiation, activa-
tion), which does not depend on the beam intensity.
The measured tagging efficiency is no usually cor-
rected for the beam-related background. On the one
52
hand, it is no possible to determine correctly the real
background value, because there is no monitor of the
beam intensity when the radiator is removed from
the low-intensity electron beam, while the intensity
may vary significantly over a run. Furthermore, re-
moving the radiator one may change the background
situation, e.g., with the radiator-in measurements,
the residual electrons from the low-energy part of
the bremsstrahlung spectrum may hit the beam pipe
on their way to the dump, and this background is
not present, as well as the Meller electrons, when
the radiator is removed. On the second hand, be-
cause this background presents in the same propor-
tion in the low and high intensity runs, one may re-
late the number of photons with sum value N ′
e(i) +
Nbg,beam(i). Besides, the high intensity radiator-
off measurements demonstrated that this background
contribution was usually no more than 1...2%.
0 10 20 30 40 50 60 70 800
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
T
ag
g
in
g
e
ff
ic
ie
n
cy
, MeVγE
Fig.5. Tagging efficiency, measured (full circles)
and simulated (empty circles), as a function of
photon energy, corresponding to the focal-plane
channels. See text for details
Thus, correction of the tagging efficiency only on
the constant room background contribution was pro-
duced. The average value of this background was
determined in the beam-off measurements and was
nb ∼ 0.1Hz for all focal-plane channels, so at high
beam intensity this background count rate was negli-
gible on comparison with the count rates of the resid-
ual electrons and the background generated by the
beam. However, at low intensity, being at the tagging
efficiency measurement, the room background contri-
bution can reach ∼ 2...3%, and should be taken into
account. Thus, the tagging efficiency εtag is given by
(see [6] for details)
εtag(i) =
Nγ,ch(i)
NFP (i) + tlivnb
, (3)
where tliv is the live time during the tagging efficiency
measurement. The results of the tagging efficiency
measurements are presented in [5] and shown in Fig.5.
The tagging efficiency is weakly decreased at photon
energy decreasing in the interval Eγ ≈ 40...80MeV,
then it steeply decreases for energies Eγ < 40MeV.
Such behavior may be due to increasing the back-
ground contribution to the FP rate, small at the high
energies and strongly increasing at Eγ < 40MeV.
Value of the measured tagging efficiency, averaged
over Eγ ≈ 50...80MeV is εtag = 0.350± 0.002.
3.4. Energy calibration and solid angle of the
telescope
Because of large elliptic beam sport on the target,
small distance between the target and the telescope,
various materials on the proton way, there no sim-
ple analytical expressions for determination the pro-
ton energy losses, angular capture, and effective solid
angle of the telescope. Thus, a Monte Carlo simula-
tion of the setup was performed using the GEANT-4
software package [7]. Three associated blocks of the
modeling were realized: (i) Simulation of interaction
of an electron beam with a photon radiator, and the
emitted photon passing through photon collimator;
(ii) Simulation of passing of the protons, produced
in the target, through the CsI/SSD telescope, and
calculation their energy losses; (iii) Determination of
the telescope angular capture and the solid angle.
3.4.1. Simulation of electron beam
interaction with a radiator
Simulation began with modeling interaction of the
electron beam with energy of E0 = 192.7MeV with
an aluminium photon radiator of 50µm thick. The
interaction points of the electrons with the radiator
were randomly distributed within the electron beam
spot on the radiator, of 0.9× 2.2mm2 in vertical and
horizontal directions, respectively. Momenta of the
bremsstrahlung photon and the post-bremsstrahlung
electron were obtained from the simulation of the
bremsstrahlung process with the GEANT-4 package
code. Then it was checked, firstly, if the photon
passed the photon collimator, described above, hav-
ing an aperture of 12mm in diameter, and, secondly,
if the post-bremsstrahlung electron exited from the
MT magnet. It can be under condition that vertical
displacement of the post-bremsstrahlung electron on
its way in the magnet chamber was less the vertical
size of the magnet chamber (±75mm from the middle
plane).
The numbers and energies of the simulated ini-
tial electrons (Ne) and secondary particles - the post-
bremsstrahlung electrons (N ′
e,mc), which passed the
MT magnet chamber, the emitted bremsstrahlung
photons (Nγ,tot) and the photons, which passed the
photon collimator (Nγ,mc), were fixed. Then the sec-
ondary particles were sorted on the energy bins, cor-
responding to the FP channels energy and width.
The accumulated data base allows one to calculate
the photon beam intensity distribution on the target
and calculate the tagging efficiency,
εtag,m(i) =
Nγ,hit(i)
N ′
e,m(i)
. (4)
The simulated tagging efficiency is shown in
Fig.5. It is practically constant in the range Eγ =
40...80MeV, the average value in this interval is
53
εtag,m = 0.374 ± 0.002 that is ∼ 6.8% more than
the experimental value. The difference may be due
to the beam dependent background contribution to
the FP detectors counting rate. The estimation of
this background contribution can be made by re-
lation fbg = 1 − εtag/εtag,m, which give the value
fbg = 0.06 in the range Eγ = 40...80MeV. This esti-
mation is close to the approximate experimental esti-
mation of the background, ∼ 1...3%, obtained usually
in the radiator-off measurements (see note above).
−5 −4 −3 −2 −1 0 1 2 3 4 5−5
−4
−3
−2
−1
0
1
2
3
4
5
0
50
100
150
200
250
Y
, c
m
X’, cm
Fig.6. Two-dimensional distribution of the photon
beam intensity (shown by color scheme) on the
target
Results of the measurements and the simulation
have shown that ∼ 35% photons passed the col-
limator and hit the target. The calculated beam
spot size on the target is shown in Fig.6. It is
∼ 13.8mm (FWHM) in vertical, and 28.1mm in
horizontal planes, respectively. Large size of the
beam spot increased the telescope angular capture
and worsened its angular resolution.
3.4.2. Proton energy losses
On the second stage, the simulation of the pro-
ton passing through the CsI/SSD telescope was pro-
duced, and the proton energy losses on this way were
calculated. Because the target was thin, one can con-
sider the horizontal and vertical coordinates (X,Y) of
the photon hit of the target, as coordinates of the
photon interaction point with the target nucleus. Dis-
tribution of these points within the beam spot is pro-
portional to the photon beam intensity distribution.
The third coordinate (Z) was played randomly along
the photon trajectory across the target. Array of all
these interaction points determined active volume of
the target.
Energy of protons emitted from the interaction
point was calculated using the reaction kinematics,
d(γ, p)n or 12C(γ, p0)
11B, and values of the photon
energy and polar angle of the proton emission. The
photon energy values were obtained on the previous
stage of the bremsstrahlung simulation, but on this
stage only energies being within the tagged interval,
Eγ = 21.9...78.8MeV, were considered. The polar
and azimuthal angles, defining direction of the proton
emission, were played isotropic. The other parame-
ters included in the simulation, describing the targets,
geometry of the CsI/SSD telescope, and the matter
on the protons way, have been partly indicated above
in the sections 1.3 and 1.4, and the others are pre-
sented below:
- the target matter in which the protons pass dis-
tance from the interaction point to exit from the tar-
get;
- the air interval from the target to the first SSD
detector, which depends on the exit point from the
target and the point of the proton hitting into the
SSD;
- two silicon strip detectors of hexagonal shape
and of 0.5mm thick;
- two Al foils, each of 15µm thick;
- two air intervals, of 15mm between the first and
the second SSDs, and of 10mm between the second
SSD and the CsI detector;
- the Al wall of the CsI box of 0.5mm thick.
From all generated protons (Np,tot), referred as
”events”, there were fixed those (Np,det) which passed
through the telescope within the active aria of the
SSD detectors and entered into the CsI. Such protons
were counted as detected ones. For these ”events”,
there were fixed: initial conditions (location, mo-
mentum) in the point of origin, the energy losses in
each materials, being on its way, energy deposited in
the SSDs and CsI detectors, coordinates of the pro-
ton hits the SSD detectors, and points of entering
into the CsI. The energies deposited by protons in
the detectors and their energy losses in the materi-
als were averaged over intervals of the emitted pro-
tons energies, corresponding to the energy width of
the photon tagger channels. The calculated energy
losses are shown in Fig.7, as a function of the initial
proton energy (in the point of origin in the target).
15 20 25 30 35 40 45 50 55 600
2
4
6
8
10
12
14
16
P
ro
to
n
e
n
er
g
y
lo
ss
, M
eV
Tp, MeV
Fig.7. Proton energy losses on its way from the
point of origin in the target to the CsI detector,
averaged over intervals of the initial proton
energies, corresponding to the photon energy
resolution of the FP hodoscope
The energy losses are ∆Ep ≈ 7MeV for pro-
tons with initial kinetic energy Tp = 36MeV and
∆Ep ≈ 13MeV for Tp = 20MeV. The accuracy of the
energy losses calculation is determined by the energy
loss tables available through the GEANT-4 package,
54
which is believed to be within about 1...2%. Simula-
tion has shown that threshold of the proton detection
by the SSD/CsI telescope is Tth ≈ 18MeV that cor-
responds to the photon energy Eγ ≈ 40MeV for the
d(γ, p)n, and Eγ ≈ 36MeV for the 12C(γ, p0)
11B re-
actions.
3.4.3. CsI detector energy calibration
Energy calibration assumes establishing a cor-
relation between the pulse-height of the CsI de-
tector signals and energy of the incoming protons.
The CsI pulse-height spectra, corresponding to the
events from the prompt region of the Fptdc spec-
tra, demonstrate two large maxima (if data from the
CD2 run is used) for all the FP channels energies.
They correspond to the signals, produced by pro-
tons emitted from the d(γ, p) (the first peak) and the
12C(γ, p0)
11B reactions, as one can see in Fig.8 (left).
The maxima are on a smooth random background,
resulted from the random FPtdc coincidences events,
being under prompt peak.
0 500 1000 1500 2000 25000
10
20
30
40
50
60
70
80
C
o
u
n
t
Channel CsI
0 500 1000 1500 2000 2500−10
0
10
20
30
40
50
C
ou
nt
Channel CsI
Fig.8. Left: Pulse-height spectrum of the CsI detector signals, produced by protons emitted from the CD2
target due to the d(γ, p) and the 12C(γ, p0)
11B reactions. Photon energies are Eγ = 50.4± 0.5MeV. Yellow
histogram is the constructed background. Right: Pulse-height spectrum after background subtraction. Curves
are the Gauss fits
In order to determine exact position of
the peaks, the background was subtracted.
For that the background pulse-height spec-
trum was generated, using the events in
the random region of the Fptdc spectrum.
400 500 600 700 800 900 1000 1100 1200 13000
10
20
30
40
50
60
M
eV
Channel CsI
Fig.9. Relation between the pulse-height of the CsI
detector signal (the ADC peak position) and: (i)
Energy deposited by proton in the CsI detector
(circles), average over results obtained for the
d(γ, p) and 12C(γ, p0)
11B reactions. Line is the
linear fit; (ii) Energy of the initial proton, produced
in the target (triangles). Line is the third degree
polynomial fitting
The background spectrum was normalized to the
prompt one and subtracted, Fig.8 (right). The nor-
malization factor was determined coming from re-
quirement of equality the number of events of the
both spectra in the ”background range”, being on
the right side of the peaks, e.g., above the 900-th
channel in the Fig.8 (left). The maxima positions
were determined by the fitting of the peaks by Gaus-
sian.
Initial energy of the protons, producing the ob-
served CsI pulse-height spectra, were calculated using
the tagged photon energy value of the corresponding
FP channel, and the polar angle value of the pro-
ton emission, θp = 900. Taking into account the en-
ergy losses, the energy deposited by the proton into
the CsI detector was determined. Thus, correlation
between the deposited energy and the amplitude of
the CsI detector signal (the ADC peak position) was
obtained. Such correlations were obtained using pro-
ton emission from the d(γ, p) and 12C(γ, p0)
11B reac-
tions. They were, practically, identical. The average
dependence is shown in Fig.9. It demonstrates linear
dependence between the energy of incoming into the
CsI proton and the detector CsI signal within energy
interval Tp = 10...50MeV
For missing energy spectra construction, there is
more appropriate the direct relation between the CsI
detector pulse-height signal and initial energy of the
proton, produced in the target. It is shown in Fig.9
by triangles. Due to increasing energy losses for low
energy protons, there is deviation from linearity in
the low energy range. This dependence was fitted
55
by third degree polynomial and was used for deter-
mination the initial proton energy at missing energy
spectra construction.
3.4.4. Angular capture and effective solid
angle
If not to use information about numbers of trig-
gered strips in the first (n1) and the second (n2)
silicon strip detectors, an angular aperture of the
CsI/SSD telescope is determined by the size of ac-
tive aria of the second SSD, its distance from the
target, and the size of the target active volume.
50 60 70 80 90 100 110 120 1300
20
40
60
80
100
120
C
o
u
n
t
p, deg.θ
Fig.10. Simulated angular capture of the SSD/CsI
telescope
The geometrical angular capture was calculated us-
ing the simulation data base, described above. The
trajectories of the, so-cold, ”detected protons”, which
were emitted from the target active volume, passed
through the active aria of the both SSDs, and entered
into the CsI, were constructed, and the corresponding
polar angles of the particles emission were calculated
−30 −20 −10 0 10 20 30
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
C
ou
nt
s
strip∆
Fig.11. Distribution of the number of events
detected by the SSDs, as a function of difference,
∆ = n2 − n1, between numbers of triggered strips in
the first (n1) and the second (n2) SSDs. The
shaded histogram corresponds to events from the
proton band in Fig.3
The simulation has shown that in the case of detec-
tion of all particles, passing the active aria of the
SSDs, the telescope angular capture is rather large,
∆θp ≈ 280 (FWHM), and the telescope can detect
protons, emitted from the target, within the polar
angles interval θp ≈ 700...1050, as shown in Fig.10.
At the distance between the first and the second SSD
of 15mm, the angular interval of the proton regis-
tration θp ≈ ±200 restricts the maximal difference
between the triggered strip numbers in the SSDs,
∆ = n2 − n1, by value | ∆ |≤ 3, at that main bulk
of the detected events should be within | ∆ |≤ 2 for
FWHM telescope angular capture ∆θp ≈ 300. The
experimental distribution of the number of events,
detected by the SSDs, as a function of the difference,
∆ = n2 − n1, is shown in Fig.11. One can see that
there are practically no events with | ∆ |≥ 3 if the
detected particles were taken from the proton band.
The events with more difference are the background
events, and their amount is negligibly small. The
main bulk of the detected events are within | ∆ |≤ 2.
30 40 50 60 70 800
50
100
150
200
250
300
350
400
, m
sr
Ω∆
, MeVγE
Fig.12. The effective solid angle, as a function of
tagged photon energy values, corresponding to the
focal plane channels
In order to increase statistics for further analysis,
summation of the experimental data was produced
for four physically adjacent FP channels. As a result,
the FPtdc spectra were formed for twelve energy bins
with central energies Eγ,bin =37.2, 41.2, 45.1, 49.0,
52.8, 56.5, 60.2, 64.0, 67.4, 70.8, 74.2, 77.6MeV, and
the bin width ∆Eγ,bin ≈ 4MeV. The effective solid
angle of the telescope was calculated for all bins, us-
ing total numbers of the ”generated”, Np,tot(j), and
the ”detected protons”, Np,det(j), from the data base,
∆Ω(j) = 4π
Np,det(j)
Np,tot(j)
. (5)
The solid angle values are shown in Fig.12, as a
function of the photon bin energy. The statistical
uncertainty was calculated by,
σ(j) = ∆Ω
√
(
1
Ndet(j)
+
1
Ntot(j)
) (6)
and was ∼ 2.6% for all bins. The solid angle within
the statistical accuracy is, practically, constant in the
energy range Eγ = 40...80MeV, its averaged value in
this interval is
∆Ω = 249.0± 2.25msr. (7)
56
The steep decrease at photon energies less 40MeV
is due to threshold of the proton registration.
As a check, the simulation was performed for a
point source, for which the solid angle value ∆Ωp =
237.5msr was obtained. The analytical calculation
performed for rectangle detector of the same square
gave the solid angle value 240.9 sr, that is 1.4% more.
The difference can be due to a different shape of the
detectors.The systematic uncertainty due to uncer-
tainties of the measurements of the set-up dimensions
is estimated to be ∼ 2%.
3.5. The reactions selection
3.5.1. The d(γ, p)n reaction. Mission energy
spectra
There are two types of events in the prompt peak
region, µ ± 3σ, of the FPtdc spectrum: true coinci-
dences from various channels of the carbon disinte-
gration (the deuteron disintegration, as well, if the
CD2 target is used), and the random background
events. In order to separate contributions from
the background and to select yield of the d(γ, p)n
and 12C(γ, p)11B reactions, a missing energy (MisE)
method was applied. It assumes construction miss-
ing energy spectra of the reaction under study. The
missing energy is given by the relation,
Em = Eγ − Tp − Tr, (8)
where Em is the photon energy, Tp is the proton ki-
netic energy, measured by the CsI detector and cor-
rected to the energy losses on its way from the origin
point to the detector, Tr is the energy of a recoil nu-
cleus (neutron or 11B) which is calculated using the
reaction kinematics, known photon energy, and the
proton emission angle value θp = 900. Taking events
from the prompt region of the FPtdc spectra, the
prompt missing energy spectra were generated for all
eleven energy bins for the CD2 and CH2 targets data.
−35 −30 −25 −20 −15 −10 −5 0 50
50
100
150
200
250
300
C
ou
nt
s
, MeVmisE
−35 −30 −25 −20 −15 −10 −5 0 5−20
0
20
40
60
80
100
120
140
C
ou
nt
s
, MeVmisE
−35 −30 −25 −20 −15 −10 −5 0 5−20
0
20
40
60
80
100
120
C
ou
nt
s
, MeVmisE
Fig.13. Left: missing energy spectra of the protons detected from the CD2 (black line) and CH2 targets
after normalization (red line). Middle: the missing energy spectrum of the protons after subtraction the
normalized CH2 spectrum. Line is the five order polynomial fit the remaining background. Right: the
missing energy spectrum of the d(γ, p)n reaction after the background subtraction. Photon energy
Eγ = 49.0± 2.0MeV. The width of the histogram bin is 0.3MeV
The spectra obtained from the CH2 target data
were normalized to experimental conditions, being at
corresponding measurements with the CD2 target.
That is, the corrections were made to different photon
flux and thickness of the targets, in order that yields
of the carbon disintegration processes were identical
for both targets, what is needed for subtraction the
carbon disintegration background from the deuteron
disintegration yield. As example, some MisE spec-
tra are shown in Figs.13-15 (left). The recoil nu-
cleus energy values were calculated for reaction of the
deuteron disintegration, that is Tr = Tn. Shape of the
presented spectra is typical for all bins: (i) There are
maxima resulted from the reactions of a carbon and a
deuteron (for measurements on the CD2 target) dis-
integration, being on top of a random background;
(ii) The background is distributed over a large range
and decreased with the missing energy decreasing.
−30 −20 −10 0 100
20
40
60
80
100
120
140
160
180
200
220
240
C
ou
nt
s
, MeVmisE
−30 −20 −10 0 10−20
0
20
40
60
80
100
C
ou
nt
s
, MeVmisE
−30 −20 −10 0 10−30
−20
−10
0
10
20
30
40
50
60
70
C
ou
nt
s
, MeVmisE
Fig.14. The same for photon energy Eγ,bin = 56.5± 2.0MeV
According to choice Tr = Tn, the events corre-
sponding to reaction of the deuteron disintegration
are located in a peak, position of which should be
the same for all energy bins, and should be equal
to the deuteron binding energy, Ed ≈ 2.2MeV, that
with a good accuracy is observed in the experiment,
Figs.13-18.
There is a weaker maximum to the right side from
the previous one. It can be seen more clearly in Figs.
16-18 (middle) for CH2 target spectra after back-
57
ground subtraction. This peak corresponds to the
sum of possible reactions 12C(γ, p2−5)
11B, when the
nucleus 11B is in one of higher excited states with
Eex ∼ 5.02, 6.74, 6.79 and 7.29MeV, which are also
no separated owing to large energy resolution. The
distance between the carbon and the deuteron max-
ima increases if the photon energy increases, and they
are well separated in the CD2 Emis spectra at photon
energies Eγ ≥ 52MeV. For less energy the maxima
corresponding to the above mention two-body carbon
disintegration processes coincide with the deuteron
peak.
−30 −20 −10 0 100
20
40
60
80
100
120
140
160
C
ou
nt
s
, MeVmisE
−30 −20 −10 0 10−10
0
10
20
30
40
50
60
C
ou
nt
s
, MeVmisE
−30 −20 −10 0 10−30
−20
−10
0
10
20
30
40
C
ou
nt
s
, MeVmisE
Fig.15. The same for photon energy Eγ,bin = 70.8± 2.0MeV
Effective way to remove all background contribu-
tion to the deuteron maximum, both from the car-
bon disintegration and random background, is to sub-
tract the MisE spectra, measured on CH2 target,
from the spectra measured on CD2 target after cor-
responding normalization. However, direct subtrac-
tion may give incorrect value of the d(γ, p)n reac-
tion yield because of the random background is not
identical for CD2 and CH2 spectra. So, it is im-
possible to subtract simultaneously the random and
carbon background correctly, thus, two-step proce-
dure was applied. There were two variants of its
application, different by execution sequence of the
random and carbon background subtraction. In the
first variant (Variant A), on the first step the car-
bon background was removed by subtraction of the
normalized (as was described the above) CH2 spec-
trum. The results of the subtraction are shown in
Figs.13-15 (middle). One can see that the remaining
random background is small in the left side of the
spectrum, but it reaches ∼ 20% under the deuteron
peak and has non trivial energy dependence. On the
second step, this remaining background, with the ex-
ception the deuteron peak range, was fitted by five
order polynomial to provide adequate description the
background in all energy range. The final missing
energy spectra, after the fitted background subtrac-
tion, are shown in Figs.13-15 (right). One can see
clear peak corresponding to the d(γ, p)n reaction at
the Emis ≈ 2.2MeV and, practically, full cancella-
tion the peak of the 12C(γ, p)11B01 reactions, giving
confidence in the background subtraction. Looking
at the deuteron missing energy peak, a full width at
half maximum of about ∼ 3MeV (FWHM) is ob-
served. The dominating contribution to the FWHM
width comes from the kinematical spread due to pho-
ton energy interval, large angular acceptance of the
CsI/SSD telescope and inherent CsI detector resolu-
tion.
The reaction yield, Ypd, was obtained from the fi-
nal d(γ, p)n MisE spectra, Figs.13-18 (right), using
two ways:
(i) The d(γ, p) peak was fitted by Gaussian,
Y = y0 +
A
σ
√
2π
e−
(x−µ)2
2σ2 , (9)
where the peak position µ, width of the maximum
σ, and constant y0 are the fitting parameters. The
reaction yield is determined by relation
Ypd =
A
w
, (10)
where A =
√
2πσYm is the square of the fitted
Gaussian, Ym is the Gaussian maximum height, and
w = 0.3MeV is the width of the step in the MisE
spectrum construction.
(ii) The yield was also obtained by summation the
counts in the peak regions, µ±3σ. The parameters µ
and σ were taken from the Gauss fit. Both methods
gave practically the same results of the yields, differ-
ing no more 5%. The statistical error of the yields is
given by
∆Ypd ≈
√
NCD +NCH +Nfit , (11)
where NCD, NCH and Nfit are the number of counts
in the peak-region, µ ± 3σ, for the CD2 and the
CH2 missing energy spectra and the fitted back-
ground spectra, respectively. The background terms
Nfit is small, ∼ 10% of the sum NCD + NCH , thus
∆Ypd ≈
√
NCD +NCH .
3.5.2. 12C(γ, p01)
11B reaction. The random
background subtraction
Information on carbon disintegration processes
have been obtained from measurements on both tar-
gets. The targets provide identical conditions for
data taking with exception the low energy bins,
Eγ,bin < 49MeV, in which the peaks of the reactions
12C(γ, p01)
11B in missing energy spectra shifted to
the range of the deuteron peak, and correct deter-
mination of the reaction yield from the CD2 target
data is imposable. As was shown above, the max-
ima corresponding to the carbon disintegration are on
58
top of smooth random background, Figs.16-18 (left),
which strongly increases with Emis energy increasing.
For the background subtraction, special background
spectra were generated for every energy bin, taking
events from the random range of the corresponding
FPtdc spectra, shown in Fig.4. They were normal-
ized to the prompt Emis spectra by requiring the
same number of the events for the prompt Npt and
the random Nr spectra in the selected interval of the
background range on the left side of the spectra be-
fore the range of the maximum of the 12C(γ, p01)
11B
reactions.
−35 −30 −25 −20 −15 −10 −5 0 50
50
100
150
200
250
300
C
ou
nt
s
, MeVmisE
−35 −30 −25 −20 −15 −10 −5 0 5−40
−20
0
20
40
60
80
100
120
C
ou
nt
s
, MeVmisE
−35 −30 −25 −20 −15 −10 −5 0 5
−20
0
20
40
60
80
100
C
ou
nt
s
, MeVmisE
−35 −30 −25 −20 −15 −10 −5 0 50
20
40
60
80
100
120
140
160
180
200
220
C
ou
nt
s
, MeVmisE
−35 −30 −25 −20 −15 −10 −5 0 5−20
0
20
40
60
80
100
C
ou
nt
s
, MeVmisE
Fig.16. Left: missing energy spectra for CD2 (up)
and CH2 (down) targets (black lines) and random
background spectra after normalization (red lines).
Middle: the missing energy spectra of after the
background subtraction. Right: the d(γ, p)n reaction
missing energy spectrum obtained due to the CD2
and CH2 spectra subtraction. Photon energy
Eγ = 49.0± 2.0MeV. Lines are the Gauss fit of the
spectra
The normalization coefficient value was deter-
mined as,
kbg =
Npt
Nr
. (12)
The kbg values are ∼ 0.1 for all energy bins due to
wider intervals of the FPtdc spectra which are used
for the background missing energy spectra genera-
tion, than for the prompt spectra. The normalized
background spectra well agree with the prompt ones
in the background range, Figs.16-18 (left). Thus af-
ter the background subtraction, the spectra both for
the CD2 and CH2 targets were flat within the sta-
tistical errors and consistent with zero in the back-
ground range, Figs.16-18 (middle), giving confidence
in the background subtraction. On the next step the
above CD2 and CH2 spectra were subtracted, and
the d(γ, p) reaction Emis spectra (variant B) were
obtained, Figs.16-18 (right). Both variants gave co-
incident values of the d(γ, p) reaction yield within the
statistical accuracy.
Due to decreasing accuracy of the normalization
coefficient determination with photon energy increas-
ing, resulted from decreasing both level of the back-
ground of the prompt spectra in the background
range and the background range decreasing (normal-
ization interval), as well, an additional control of the
background subtraction was applied at data process-
ing for energy bins Eγ,bin ≥ 56.5MeV, using the miss-
59
ing energy interval of the spectra behind the deuteron
peak which becomes enough broad for these energies.
The spectra in this interval for the CD2 and CH2
targets are determined by the same processes of the
carbon disintegration, thus after the background sub-
traction they have to be, in principle, identical within
the statistical accuracy. So, the additional control
consisted in requiring the same number of the events
for the CD2 and CH2 Emis spectra in the selected
interval above the d(γ, p) peak after the background
subtraction. As a rule, the determined kbg values pro-
vided also and this control requirement, and after the
CD2 and CH2 Emis spectra subtration, the d(γ, p)
spectra were obtained with flat parts, consistent with
zero within the statistical errors, below and above the
d(γ, p) maximum, as shown in Figs.17,18 (right).
−30 −20 −10 0 100
20
40
60
80
100
120
140
160
180
200
220
240
C
ou
nt
s
, MeVmisE
−30 −20 −10 0 10−20
−10
0
10
20
30
40
50
60
70
C
ou
nt
s
, MeVmisE
−30 −20 −10 0 10−30
−20
−10
0
10
20
30
40
50
60
70
C
ou
nt
s
, MeVmisE
−30 −20 −10 0 100
20
40
60
80
100
120
140
160
180
C
ou
nt
s
, MeVmisE
−30 −20 −10 0 10−20
−10
0
10
20
30
40
50
60
C
ou
nt
s
, MeVmisE
Fig.17. The same for photon energy
Eγ = 56.5± 2.0MeV
The yield of the d(γ, p)n reaction for the variant
B data processing was obtained by the same way, as
for the previous variant A, using both the Gauss fit
of the peak and summation events under the peak.
The statistical errors of the yield for variant B are
calculated by
∆Ypd =
√
NCD2 + k2bgCD2
NBgCD2 +NCH2 + k2bgCH2
NBgCH2 , (13)
where NBgCD2 and NBgCH2 are the number of counts
in the peak region, µ ± 3σ, for the background the
CD2 and CH2 missing energy spectra, respectively.
Because normalization coefficient is kbg ∼ 0.1, contri-
bution of the background terms is small, ∼ 10%, and
the statistical errors were determined by statistics of
the prompt peak yields for the CD2 and CH2 targets,
∆Ypd ≈
√
NCD2 +NCH2 , as for the variant A. The
statistical accuracy of the yields varied from ∼ 10%
at Eγ = 41.2MeV bin to ∼ 15% for Eγ = 70.8MeV.
60
The results of the A and B variants of the data pro-
cessing within the data accuracy practically coincide
for all photon energy bins that also gave confidence
of the 12C(γ, p01)
11B reaction yield obtaining.
−30 −20 −10 0 100
20
40
60
80
100
120
140
160
180
200
C
ou
nt
s
, MeVmisE
−30 −20 −10 0 10−20
−15
−10
−5
0
5
10
15
20
25
30
C
ou
nt
s
, MeVmisE
−30 −20 −10 0 10−30
−20
−10
0
10
20
30
40
C
ou
nt
s
, MeVmisE
−30 −20 −10 0 100
20
40
60
80
100
120
140
C
ou
nt
s
, MeVmisE
−30 −20 −10 0 10−15
−10
−5
0
5
10
15
20
C
ou
nt
s
, MeVmisE
Fig.18. The same for photon energy
Eγ = 70.8± 2.0MeV
The yields of the carbon disintegration were ob-
tained by the same way, both Gauss fit and summa-
tion the events under 12C(γ, p01)
11B peak. At some
photon energies we used two Gaussians for correct
separation the contributions resulted from the higher
exited states or the deuteron disintegration process to
the peak of the 12C(γ, p01)
11B reaction. The Gauss
fit and summation gave the same results within the
statistical accuracy. The statistical error of the reac-
tion yields is given by
∆YCD(CH) =
√
NCD(CH) + k2
bgNBgCD(CH) , (14)
where NCD(CH) and NBgCD(CH) are the number of
counts in the peak region, µ ± 3σ, for the CD2 (or
CH2) prompt and background missing energy spec-
tra, respectively. Because, as stated the above, the
normalization coefficient is kbg ∼ 0.1, contribution
the second term is ∼ 10% and the statistical errors
were determined mostly by statistics of the prompt
peak, ∆YC ≈
√
NCD(CH). The statistical accuracy
of the yields for both targets varied from ∼ 3% at
Eγ ∼ 40MeV to ∼ 20% for end of the tagging in-
terval Eγ ∼ 70MeV due to strong decreasing of the
12C(γ, p01)
11B reaction cross section.
3.5.3. Cross section
The cross section was calculated using the formula
dσ
dΩ
=
Yp,i
ND(C)∆ΩNγ(i)εst,i
, (15)
where
- Yp,i is the reaction yield for i-th energy bin;
- ND(C) is the number of deuterons (carbons)
nuclei per cm2 for the target located under angle
θm = 600 to the photon beam direction. They are
61
for deuteron ND = 1.541 × 1022D/cm2, and for car-
bon NC(CD2) = 0.771× 1022C/cm2 for the CD2 and
NC(CH2) = 0.885× 1022C/cm2 for the CH2 target.
- ∆Ω = 249.00 ± 2.25msr is the effective solid
angle of the CsI/SSD telescope;
- Nγ(i) = NFP (i)εtag(i) is the tagged photon flux
incident on the target. NFP (i) is the number of post-
bremsstrahlung electrons, corresponding to the i-th
energy bin, εtag(i) is the tagging efficiency. We use
the experimental value of the tagging efficiency, av-
eraged over tagged energy range, εtag(i) = 0.35;
-εst,i is the stolen correction to the cross section
for the for the i-th energy bin. This correction re-
sulted from the fact that the uncorrelated electron
can be registered in the region to the left of the
prompt peak, thus a focal plane TDC can be stopped
by a random electron, arriving earlier than a corre-
lated. If the random events are Poisson-distributed in
time, the stolen-coincidence correction may be writ-
ten as [8],
εst,i = e−
t0nFP,i
df , (16)
where nFP,i is the efficient rate of the FP counters,
corresponding to the i-th energy bin, t0 is the position
of the lower limit of the prompt region in the FPtdc
spectrum, df is the average duty factor of the beam
during the run. The stolen correction is proportional
to the count rate in the focal-plane detectors and in-
versely proportional to the duty factor. The rate of
the focal plane detectors corresponding to the i-th
energy bin is ranged from nFP,i ∼ 0.4 to 1.2MHz.
The prompt peak position was in the 750 channel for
all FPtdc spectra, the lower limit of the prompt re-
gion was taken at 727 channel that corresponded to
t0 ≈ 128 ns. The average value of the duty factor
over all beam runs was df = 0.5. The average FP
counting rate nFP for CD2 and CH2 runs differed
no more 10%, thus the stolen-coincidence corrections
were practically identical for both runs and varied
with energy from εst,i ∼ 24%, at Eγ = 41.2MeV, to
εst,i ∼ 11% at Eγ = 78MeV.
The differential cross sections of the d(γ, p)n re-
action obtained for variants A and B of the data pro-
cessing are shown in Fig.19. They are in agreement
within the data accuracy with each other and with
the data of other laboratories. However, if to con-
sider ratio of the cross section, obtained for A and B
variants of the data processing, averaged over whole
energy interval of the measurements, there is a sys-
tematic ∼ 10% exceeding of the cross sections for
variant A above the data for variant B. This value
can be considered as systematic error of the measured
cross section.
The differential cross sections of the
12C(γ, p01)
11B reaction are shown in Fig.20.
As can be seen, the data obtained from mea-
surements on the CD2 and CH2 targets are
in a good agreement within the data accuracy.
20 30 40 50 60 70 800
5
10
15
20
25
30
35
40
45
50
, m
kb
/s
r
Ω
/dσd
, MeVγE
Fig.19. Differential cross section of the d(γ, p)n
reaction at θp = 900 for variants A (full squares)
and B (empty squares) data processing. The
literature data: [9] (circles), [10] (rhombus), [11]
(down triangles), [12] (triangles)
The ratio of the cross sections average over energy
interval of the measurements is R = 1.03 ± 0.08 for
yields obtained by summation and R = 0.97 ± 0.06
for yields obtained by Gauss fit. The average differen-
tial cross sections over these runs are shown in Fig.20
(right). Due to large angular capture of the telescope
and strong angular dependence of the cross section,
the effective angle of the proton detection was less
the geometrical angle of the telescope position rela-
tively the photon beam. It was calculated using the
cross section angular dependences from [1]. The ob-
tained effective angle values were ∼ 30 till ∼ 50 less
the angle of the telescope position (900) at photon
energy increasing from Eγ ∼ 40MeV to ∼ 62MeV.
Such change of the proton emission angle increases
the cross section from 10% to 30% in this energy in-
terval, respectively. If to take into account the effec-
tive angle of the proton registration, our data are in a
reasonable agreement with the data [1], presented in
the Fig.19 for angle of the proton emission θp = 900.
4. SUMMARY
The ∆E − E CsI/SSD telescope, constructed in
MAX-lab, has been tested, in order to evaluate its
characteristics and capabilities for measurements the
(γ, p) reactions on atomic nuclei at intermediate en-
ergies, ranging from the Giant Dipole Resonance
and up to several tens of MeV below threshold for
pion photoproduction. The telescope consists of two
single-sided silicon strip detectors, with effective strip
width 2mm, and CsI counter which function as (∆E)
and (E) detectors, respectively. The Monte Carlo
simulation of the experimental set up have been per-
formed, including generation of the photon beam and
passing it to the target through a collimator, the
beam tagging efficiency and the telescope angular
capture and the effective solid angle value.
62
35 40 45 50 55 60 65 70 75 80
−110
1
10
210
310
, m
kb
/s
r
Ω
/dσd
, MeVγE
35 40 45 50 55 60 65 70 75 80
−110
1
10
210
310
, m
kb
/s
r
Ω
/dσd
, MeVγE
Fig.20. Left: The differential cross section for the 12C(γ, p01)
11B obtained from the CD2 (squares) and
CH2 (circles) runs. Right: the cross section averaged over CD2 and CH2 runs obtained by summation (full
squares) and Gauss fit (empty squares) of the reaction missing energy peak. Triangles are the data [1] for
θp = 900
The simulation has shown that the existing con-
struction of the telescope provided rather large geo-
metrical angular capture ∆θp ≈ 300 (FWHM) if do
not use the triggered strip information from the SSD
detectors. The angular resolution can be improved
by factor of two if to select the particles trajectories,
passing through the strips with identical numbers.
The possibility of ∆E−E method of proton iden-
tification and the (γ, p) reaction selection by the miss-
ing energy method were studied using reactions of
a deuteron and a carbon disintegration. The en-
ergy calibration of the CsI detector was performed
which has demonstrated a linear dependence between
the energy of incoming proton and the CsI signal
within energy interval Tp = 10...50MeV, and thresh-
old of the proton registration Tp ≈ 18MeV. So, the
CsI/SSD telescope at present construction provides
measurements of the (γ, p) reactions on atomic nu-
clei in photon energy range Eγ > 40MeV, where
quasi-deuteron mechanism of the nuclear disintegra-
tion is important. For testing the missing energy
method for the reaction selection and background
subtraction, cross sections of the d(γ, p)n and the
12C(γ, p01)
11B reactions were measured in the range
Eγ ≈ 40...70MeV, which agreed with literature data.
In order to extend investigations in the Giant Dipole
Resonance region it is necessary to decrease the
threshold of the proton registration, using thinner
coordinate detectors and nuclear targets, and plac-
ing telescope into special vacuum chamber. It will
allow one to improve energy resolution and resolve
the exited state of the final of nucleus. The angular
resolution of the telescope can be improved if to use
coordinate information on the triggered strips of the
silicon strip detectors.
The obtained results give the possibility, using
the existing at MAX-lab technique, to extend the
(γ, p) processes investigation involving the polarized
photon beam, produced at MAX-lab. Such exper-
iments will allow one to get new physical observa-
tion and open new possibility for investigation of
nuclear structure and photonuclear reaction mech-
anisms in this energy range. Analysis of results of
the 12C(γ, p01)
11B reaction measurements with the
polarized photon beam will be presented in the next
paper.
ACKNOWLEDGEMENTS
This work is supported by Swedish Research
Council, the Craaford Foundation, the Wennergren
Foundation, the Royal Physiographic Society in Lund
and the Knut and Alice Wallenberg Foundation, by
the European Community - Research Infrastructure
Action under the FP6 ”Structuring the European Re-
search Area” Programme (through the Integrated In-
frastructure Initiative ”Hadron Physics”) and partly
supported by STCU project 3239. The authors ac-
knowledge the large support of the MAX IV Labora-
tory staff which made this experiment successful.
References
1. H. Ruijter et al. Angular distribution for the
12C(γ, p)11B reaction // Phys.Rev. C54. 1996,
p.3076.
2. V. Ganenko et al. Linearly polarized photon
beam at MAX-lab // Nuclear Inst. and Methods
in Physics Research, A. 2014, p.137-149.
3. S. Al. Jebali, et al. Summary of the MAX-lab
Run Period 2008.02.18 - 2008.03.17
4. J.-O. Adler et al. The upgraded photon tagging
facility at MAX-lab // Nucl. Instr. and Meth.
A715, 2013, p.1-10.
5. J. Brudvik, et al. Summary of the MAX-lab Run
Period 2008.04.14 - 2008.04.28
6. E. Aghassi, et al. Summary of the MAX-lab Run
Period 2008.06.02 - 2008.06.30
7. http://geant4.cern.ch
8. R.O.Owens // Nucl. Instrum. and Methods. 1990,
A288, p.574.
63
9. D. Babusci, V. Bellini, M. Capogni, et al.
Deuteron photo-disintegration with polarized
photons in the energy range 30-50 MeV // Nucl.
Phys. 1998, A633, p.683-694.
10. K.-H. Krause, J. Sobolewski, J. Ahrens, et al.
Photodisintegration of the deuteron by linearly
polarized photons // Nucl. Phys. 1992, A 549,
p.387-406.
11. M.P. De Pascale, G. Giordano, G. Matone, et al.
// Phys. Rev. 1985, C 32, p.1830
12. B. Weissman and H.L. Schultz // Nucl.Phys.
1971, A 174, p.129.
ÈÇÌÅÐÅÍÈÅ (γ, p) - ÐÅÀÊÖÈÉ ÍÀ ÓÑÒÀÍÎÂÊÅ ÌÀÕ-ëàá Ñ ÏÎÌÎÙÜÞ
∆E-E ÒÅËÅÑÊÎÏÀ
Ä.Ä.Áóðäåéíûé, J.Brudvik, Â.Á.Ãàíåíêî, K.Hansen, K.Fissum, L. Isaksson,
K.Livingston, M.Lundin, B.Nilsson, B.Schr�oder
Ðàññìàòðèâàþòñÿ ðåçóëüòàòû ýêñïåðèìåíòîâ ïî èçó÷åíèþ ðåàêöèé 12C(γ, p)11B è d(γ, p)n â îáëàñòè
ýíåðãèé ìå÷åíûõ ôîòîíîâ 35...80ÌýÂ. Ïîêàçàíà âîçìîæíîñòü èäåíòèôèêàöèè ïðîòîíîâ ìåòîäîì ∆Å-
Å ñ ïîìîùüþ CsI/SSD - òåëåñêîïà. Èñïîëüçóÿ ñïåêòðû íåäîñòàþùèõ ýíåðãèé, îïðåäåëåíû çíà÷åíèÿ
äèôôåðåíöèàëüíûõ ñå÷åíèé ðàññìàòðèâàåìûõ ðåàêöèé â óêàçàííûõ äèàïàçîíàõ ýíåðãèé ôîòîíîâ. Ïî-
êàçàíî õîðîøåå ñîãëàñèå ðåçóëüòàòîâ ýêñïåðèìåíòîâ ñ èìåþùèìèñÿ ëèòåðàòóðíûìè äàííûìè.
ÂÈÌIÐÞÂÀÍÍß (γ, p) - ÐÅÀÊÖIÉ ÍÀ ÓÑÒÀÍÎÂÖI ÌÀÕ-ëàá ÇÀ ÄÎÏÎÌÎÃÎÞ
∆E-E ÒÅËÅÑÊÎÏÀ
Ä.Ä.Áóðäåéíèé, J.Brudvik, Â.Á.Ãàíåíêî, K.Hansen, K.Fissum, L. Isaksson,
K.Livingston, M.Lundin, B.Nilsson, B.Schr�oder
Ðîçãëÿíóòî ðåçóëüòàòè åêñïåðèìåíòiâ ç âèâ÷åííÿ ðåàêöié 12C(γ, p)11B i d(γ, p)n â îáëàñòi åíåðãié ìi-
÷åíèõ ôîòîíiâ 35...80ÌåÂ. Ïîêàçàíà ìîæëèâiñòü iäåíòèôiêàöi¨ ïðîòîíiâ ìåòîäîì ∆Å-Å çà äîïîìîãîþ
CsI/SSD - òåëåñêîïà. Âèêîðèñòîâóþ÷è ñïåêòðè íåäîñòàòíiõ åíåðãié âèçíà÷åíi äèôåðåíöiéíi ïåðåðiçè
ðîçãëÿíóòèõ ðåàêöié â çàçíà÷åíèõ äiàïàçîíàõ åíåðãié ôîòîíiâ. Îòðèìàíi åêñïåðèìåíòàëüíi ðåçóëüòàòè
äîáðå óçãîäæóþòüñÿ iç ëiòåðàòóðíèìè äàíèìè.
64
|