ArticlePDF Available

Abstract and Figures

Hydrogenated amorphous silicon (a-Si:H) is attractive for radiation detectors because of its radiation resistance and processability over large areas with mature Si microfabrication techniques. While the use of a-Si:H for medical imaging has been very successful, the development of detectors for particle tracking and minimum-ionizing-particle detection has lagged, with almost no practical implementation. This paper reviews the development of various types of a-Si:H-based detectors and discusses their respective achievements and limitations. It also presents more recent developments of detectors that could potentially achieve single particle detection and be integrated in a monolithic fashion into a variety of applications.
Content may be subject to copyright.
Topical Review for Semiconductor Science and Technology
Review of Amorphous Silicon Based Particle Detectors: The Quest for
Single Particle Detection
N. Wyrsch, C. Ballif.
Ecole Polytechnique Fédérale de Lausanne (EPFL), Institute of Microengineering (IMT),
Photovoltaics and thin-film electronics laboratory, Breguet 2, 2000 Neuchâtel, Switzerland,
Abstract. Hydrogenated amorphous silicon (a-Si:H) is attractive for radiation detectors
because of its radiation resistance and processability over large areas and with mature Si
microfabrication techniques. While the use of a-Si:H for medical imaging has been very
successful, the development of detectors for particle tracking and minimum-ionizing-particle
detection has lagged, with almost no practical implementation. This paper reviews the
development of various types of a-Si:H-based detectors and discusses their respective
achievements and limitations. It also presents more recent developments of detectors that could
potentially achieve single particle detection and be integrated in a monolithic fashion into a variety
of applications.
Keywords: amorphous silicon, particle detectors, radiation hardness, monolithic device
1. Background
1.1 Motivations
Among possible materials for detectors, hydrogenated amorphous silicon (a-Si:H) is one of
the most radiation-resistant semiconductors and is therefore very attractive for the fabrication of
particle sensors [
1
]. The fact that this material can be deposited on various types of substrates and
alloyed with other elements to modify its properties is another crucial advantage. a-Si:H has been
successfully implemented in large-area particle detectors for X-ray radiography using an indirect
detection scheme. However, despite more than 30 years of research, this technology has hardly
been used as particle detectors in experiments or systems that rely on direct detection. One of the
reasons is the failure to demonstrate detection of single, high-energy particles.
In this topical review we present the development of a-Si:H-based particle detectors and in
particular (but not only) strategies aiming at minimum ionizing particles (MIPs) and single particle
detection. To this purpose, we discuss in detail the detection capability and the fundamental limits
and bottlenecks (and the possible solution to overcome them) of this thin-film technology. In this
context, some recent developments are presented in more detail. They comprise the possibilities
of vertical integration of a thick a-Si:H diode array on top of a readout ASIC (the so-called “thin-
film on ASIC” or “TFA” technology) as well as vertically integrated microchannel plates (MCP).
The possibility of depositing a-Si:H over a large area, on various substrates, and micro-machining
the layers is also opening various opportunities in the field of particle detectors. The high radiation
hardness of a-Si:H together with the ability to process this material offers the potential for
implementation in experiments in colliders with high luminosity, e.g. future LHC (Large Hadron
Collider) upgrades [
2
].
1.2 a-Si:H material and technology
The first investigation of a-Si:H was reported by Chittik et al. in 1969 [
3
]. It was observed
that the growth of this material from plasma-enhanced chemical vapour deposition (PE-CVD, also
referred to as glow discharge) from silane (SiH4) led to a much lower defect density compared to
evaporated or sputtered amorphous silicon. The development of a-Si:H technology in the 70’s [
4
]
and especially the demonstration that it can be substitutionally doped (n-type and p-type) by Spear
and Lecomber in 1975 [
5
] spurred the development of various types of devices such as transistors
[
6
], solar cells [
7
] and memories [
8
].
a-Si:H material is a disordered semiconductor. The irregular arrangement of atoms has the
consequence that not all Si-Si bonds can be satisfied and leads to the presence of broken or
dangling bonds (DBs). Hydrogen is thus introduced into the material to passivate those DBs which
act as defects and recombination centres. The minimum amount of H necessary to passivate most
of the DBs is about 1% atomic. The H content in the material has an influence on its bandgap
(increasing the hydrogen content enlarges the bandgap) and depends on the deposition conditions
such as the deposition temperature. The typical H content of the material deposited by PE-CVD is
usually around 10% atomic. a-Si:H material is generally deposited at temperatures around 200°C
from the dissociation of silane SiH4. H is generally added to dilute the silane and improve the
resulting material properties. Deposition temperatures as low as 100°C are possible but lead to
more defective (and less dense) material while the highest possible deposition temperature is given
by the start of the desorption of H at around 350°C. This relatively low processing temperature
allows for the deposition of layers or devices on a variety of substrates [
9
] and even on top of
sensitive devices such as CMOS (Complementary Metal-Oxide Semiconductor) chips, as
discussed below in section 3.
Several deposition techniques are available for electronic-quality a-Si:H: PE-CVD with
plasma excitation at the radio frequency used for industrial processes (RF at 13.56 MHz) [
10
], at
very-high frequencies (VHF, between 27 and 150 MHz) [
11
] or at microwave frequencies [
12
], or
alternatively hot-wire (HW) deposition (also known as catalytic CVD) [
13
]. All techniques cited
above have been able to produce high-quality a-Si:H material and have been successfully used, at
least in the laboratory, for a-Si:H device fabrication. However, each one of them offers specific
advantages and disadvantages in terms of deposition area, deposition rate or material properties.
Doping is achieved by adding PH3 to the process gas (e.g. mixture of silane SiH4 and H) for n-type
material and by adding B2H6 or TMB (Trimethylboron) for p-type material.
The disordered nature of a-Si:H leads to under-coordinated Si atoms (atoms with one or more
DBs) and the resulting DBs lead to the presence of a continuous distribution of states within the
bandgap. States can be classified as extended states in the valence and conduction bands with a
finite mobility, or as localized states comprising DB states in the centre of the bandgap and
conduction and valence band tail states. These tail states exhibit an approximately exponential
shape as a function of energy, and their width (which depends on the magnitude of the disorder)
is asymmetric (larger on the side of the valence band). Strictly speaking, therefore, a-Si:H does
not have a bandgap but a mobility gap, as carriers located in localized states have zero or very low
mobility [
14
]. An optical gap is also commonly defined to characterize optical absorption, which
reflects the distribution of states in the bandgap and band edges. Often reported are Tauc optical
gaps [
15
] or E04 values. E04 is defined as the photon energy for which an optical absorption
coefficient of 10000 cm-1 is measured. Typical bandgap values depend on the metric used and on
the material properties (mainly on the H content [
16
]) and are around 1.7 to 1.8 eV.
a-Si:H has been so far treated as a continuous random network of atoms with isolated
randomly distributed DBs as the defect sites and randomly distributed hydrogen atoms. However,
this view is challenged by the observation of the predominance of hydrogenated di-vacancies in a
dense a-Si:H network and the predominance of hydrogenated nano-sized voids in less dense a-Si:H
[
17
,
18
]. Increasing the presence of atomic H in the plasma (supplied from the decomposition of
silane or added to silane) can lead to a deposition regime with the growth of crystalline phases in
the material. a-Si:H deposited at the onset of the crystalline growth is often sought for high-quality
devices [
19
]. The lowest defect density is usually obtained at low deposition rates [
20
].
Transitional material of high quality that incorporates nano-crystals embedded in an amorphous Si
tissue, so-called polymorphous material [
21
], is also getting attention for use in devices.
Given its relatively wide bandgap and its disordered nature leading to a low charge carrier
mobility, a-Si:H is a semi-isolating material with resistivity values higher than 1010 cm-1. This
resistivity can be varied by more than 7 orders of magnitude by p- or n- doping [5]. As transport
take place in extended states, conductivity is thermally activated. DBs play an important role as
the main recombination centres. Doping of the material also creates additional defects (additional
DBs) through a chemical equilibrium process, which rapidly degrades the quality and, as a
consequence, the carrier lifetime of the material [
22
]. Doped layers therefore cannot be used as
active layers in photodiodes or particle sensors.
a-Si:H exhibits metastable material properties. Upon light soaking, photoconductivity and
dark conductivity decrease with time and saturate after several hours or days. Annealing of the
material for several hours at temperatures ≥150°C reverses the change [
23
]. This effect, known as
the Staebler-Wronski effect (SWE), is due to an increase in the density of DBs created by the
breaking of Si-Si weak bonds. The creation of additional DBs is not directly linked to photon
absorption but to recombination events. A similar defect creation mechanism can therefore be
observed when electron-hole (e-h) pairs recombine following double injection (in the absence of
illumination) [
24
]. Several models have been proposed to give a microscopic description of the
SWE but so far with no satisfactory result [24]. It is, however, established that the SWE is related
to the H content and H bonding [
25
]. More detailed information on the metastability of a-Si:H and
how it influences the material or device properties can be found in reference textbooks [
26
,
27
].
As discussed in the following sections, the design of a-Si:H particle detectors may require
the deposition of rather thick layers. Deposition of dense a-Si:H usually results in the presence of
compressive mechanical stress in the material. This stress often depends on the H content in the
film [
28
,
29
] and as a consequence on the deposition conditions [29,
30
]. In addition, a thermal
expansion coefficient mismatch between the substrate and a-Si:H may play a significant role
(increasing or reducing the mechanical stress). The deposition temperature, which has a strong
influence on both the H content and the resulting stress due to thermal expansion mismatch, is a
key parameter for stress management to avoid any delamination of the films [
31
].
1.3 a-Si:H for particle detection (detection scheme and requirements)
Particle detection using an a-Si:H-based device is usually achieved either directly using thick
diodes or indirectly using a thin photodiode coupled with a scintillator. In case of direct detection,
the e-h pairs generated by ionization (interaction of the particles within the a-Si:H absorber) are
collected on the diode. A thick diode is necessary as interactions of ionizing particles with the
a-Si:H generate a small number of e-h pairs per unit length. For energetic charged particles, the
energy loss per unit distance is roughly proportional (Bethe-Bloch formula [
32
]) to the atomic
number and to the density of the material. Given the stopping power (most probable) value at
minimum ionization of 1.66 MeV cm2g-1 for a MIP in c-Si one can expect the creation of about
108 e-h pairs per micron of material crossed. Experimentally, a value of 80 e-h pairs is observed
(related to the average value of the energy loss). The values for a-Si:H are similar but not known
with confidence as complete charge collection is usually not achieved. Furthermore, as discussed
in section 2.1, a large scattering is observed in experimental values of the energy needed to create
e-h pairs.
In case of high energy photons, the energy loss also highly depends on the particle energy
and the interacting material. The interaction by photoemission or Compton effect results in the
production of an energetic electron that leads to e-h pair generation. As a consequence higher Z
materials exhibit larger photon attenuation coefficients and thick detectors are needed for Si. At
higher photon energy interaction occurs by direct e-h pair production and the attenuation
coefficient of photon with a Si detector is well below 0.1 cm-1, meaning that any Si detector is
almost transparent to such X-ray. Only a fraction of the photon energy can thus be collected and
as thick as possible detectors are required. For indirect detection, a scintillating material is used to
convert the ionizing radiation into visible or UV photons by luminescence. In this case, a much
thinner photodiode can be used. A schematic drawing of the two detection schemes is given in
Figure 1.
Figure 1: Schematic particle detection schemes of X-rays: Direct detection using a thick diode
(left) and indirect detection using a thin diode and a scintillating layer (right).
An n-i-p (n-doped, intrinsic and p-doped layers) diode structure is used to collect e-h pairs
created either by the ionizing particle (charged particles or photons), depending on the detection
scheme. As indicated in the previous section, the intrinsic layer is the only active layer, and doped
layers should be as thin as possible to minimize parasitic absorption. In both cases, a full depletion
of the diode should be achieved as only e-h pairs generated within the depletion region can be
collected and therefore can contribute to the signal. As discussed in section 2.2, this requirement
is especially critical for direct detection where thick diodes are mandatory. High-quality material
is required but the leakage (or dark) current, which strongly increases with bias voltage (see Figure
2), usually limits the range of useful diode thickness. At low bias voltage, the dark current is
essentially given by thermal carrier emission from the bulk and depends on the material quality
and intrinsic layer thickness. As the electric field increases, injection from the doped layer and the
Poole-Frenkel effect in the high-field region (close to the p-i interface) become dominant.
Figure 2: Leakage current of a-Si:H diodes of various thicknesses deposited at a low rate (about
0.3 nm/s and two different H dilutions of silane) and at a high rate (1.5 nm/s) (from [
33
]).
10-12
10-11
10-10
10-9
10-8
10-7
10-6
-2x105-1x1050
p-i-n, 1 µm, low rate 1
p-i-n, 1 µm, low rate 2
n-i-p, 1 µm, high rate
n-i-p, 12 µm, high rate
n-i-p, 30 µm, high rate
Idark [A/cm2]
Field [V/cm]
In order to minimize this sharp increase of leakage current with growing electric field,
several solutions have been proposed in the literature: (a) An increase of the p-layer thickness
[
34
,
35
], (b) the introduction of a double p-layer [
36
] or (c) a buried p-layer [35], or (d) the
introduction of a buffer layer at the p-i interface [
37
].All these solutions can be easily implemented
and require minimal modifications of the fabrication process. However, it is not yet clear which
solution offers the lowest leakage current as interface properties (rather than the intrinsic or doped
layer properties) usually play the major role in controlling leakage. Practically, the signal current
given by the number of e-h pairs created and the electric field necessary to achieve full depletion
should be much higher than the leakage current for this electric field value. The optimum thickness
is the highest one for which this condition is satisfied. Additionally the readout electronics also
has to accommodate this DC current leakage. In order for the diode to sustain a high reverse bias
with low leakage, it is necessary to have a well-defined metal contact on one side of the diode and
to etch the doped layer around that contact [34]. In designing the detection device, one has also to
keep in mind that the defect density, and as a consequence leakage current, may increase as a
function of material irradiation due to additional defect creation.
The difficulty to achieve full collection on thick a Si:H diodes (as discussed in section 2.2)
and the higher signal-to-noise ratios that could be obtained from the combination of thin a-Si:H
photodiode coupled with CsI scintillator led to a shift of the interest to indirect detection scheme
[36]. The integration of a Si:H photodiodes in an active matrix of a Si:H based transistors here
allows for spatial detection and imaging.
For indirect detection, thin devices are sufficient to absorb all the photons emitted by the
scintillator. Low dark current is still necessary and low defect density material is required.
However, achieving full depletion at low bias voltage is not an issue here. In case of scintillators
emitting in the UV range, a critical diode design may be necessary to minimize parasitic absorption
in the contact or doped layers of the diode (layers which are not photovoltaically active). A possible
solution to overcome this problem is shown below in section 4.
Note also that a-Si:H layers may also be used as resistive or protective layers in other types
of particle detectors. Such application usually relies on the high resistivity of a-Si:H. This material
was, for example, used very successfully as spark protection layers in GridPix gaseous
proportional detectors [
38
,
39
].
2. a-Si:H diode as a particle detector
2.1 Charge particle detectors
Development of charged particle (proton and alpha) detectors based on thick a-Si:H diodes
was pioneered by the group of Perez-Mendez at the University of California, Berkeley [
40
,
41
]
which also achieved the first detection of MIPs from various radioactive sources [
42
,
43
]. Several
other groups rapidly developed similar detectors in France [
44
,
45
], in Italy [
46
], in Japan [
47
] and
in Switzerland [
48
]. Detectors of various thicknesses from a few microns to 50 microns were
investigated for the detection of various types of particles. High radiation hardness was also
demonstrated, rendering a-Si:H very attractive for radiation detectors (see section 2.4). However
the signal-to-noise ratio for MIPS detection using the direct detection scheme was not high enough
(values between 2 and 3) [43,
49
] to allow for single particle detection and to allow for particle
tracking.
While thick detectors are desirable to generate the maximum number of e-h pairs, the applied
electric field, in many cases, could not be high enough to ensure full depletion. In this case, only
the charge pairs generated in the depletion region can be collected before recombination, leading
to incomplete charge collection. Signal formation is given by the drift of the e-h pairs in the
intrinsic layer of the diode and therefore depends on the distance travelled. Full collection is
achieved when both the electron and the hole from a generated pair are collected at the diode
contacts [44]. Analysis of the signal pulse shape shows a fast contribution (due to electrons) and a
much slower contribution (due to holes) given by the carriers’ respective drift mobility values. In
practical experiments, hole collection is difficult to record as the signal is difficult to resolve [
50
].
The e-h pair generation is characterized by the mean energy W needed to create one e-h pair.
A linear relationship has been observed between W and the optical gap with a value of 3.6 for c-Si
(crystalline silicon) [
51
]. First measurements indicated a value of W=6 eV consistent with this
relationship [42]. However, more detailed analysis indicated this value to be overestimated due to
incomplete charge collection and found the mean W value to be between 3.4 and 4.4 eV [50]. W
for a-Si:H is therefore quite close to that of c-Si. The reason why a-Si:H does not follow the usual
relationship is not understood.
Energy loss in the material by a MIP is proportional to the density and atomic number. The
interaction of MIPs with Si or a-Si:H results in a small number of e-h pairs generated per unit
distance. In order to increase this number of e-h pairs generated in the intrinsic layer, one can use
a heavier metal electrode as the diode contact (on the incoming side of the diode). Interaction of
the MIP with the metal layer will create a shower of e-h pairs, leading to the injection of some of
them in the intrinsic layer, leading to an increased number of collected charges [48,
52
]. Increase
of the signal by up to a factor of 2 has been reported when replacing Al with Au contacts on state-
of-the-art a-Si:H thick diodes [48].
Recently, large-area 4×1.5 cm2 a-Si:H particle detectors deposited on glass were developed
by our group to be implemented in front of the magnetic septum of a synchrotron to monitor an
extraction beam. It is to our knowledge the first practical implementation of an a-Si:H detector in
a high-energy physics experiment. Figure 3 shows the response of such a 5-µm-thick detector
exposed to a variable fraction of a 250 MeV proton beam. A very high linearity of the signal
(collected charge) as a function of the proton flux can be observed in this fully depleted a-Si:H,
demonstrating both the linear response of the detector to the number of impinging protons
(independently of the position) and its the spatial uniformity.
Figure 3: Charge collected (from the integral of the current during one spill) as a function of a
250 MeV proton beam fraction seen by a 6 cm2 detector situated in the magnetic septum of a
synchrotron [
53
].
2.2 Depletion condition and material requirements
As already stated above, only e-h pairs generated within the depletion region can be
collected. Thus, to maximize the number of pairs collected, this region should extend throughout
the intrinsic layer thickness. Assuming a constant density of ionized defects as a function of
distance from the p-i interface, the minimal voltage VF necessary to fully deplete a diode is given
by [
54
]:
0
5x10-9
1x10-8
0 0.5 1
Collected charge [C]
Fraction of selected beam
aSi
db
FdqN
V
0
2
*
2
(1)
where Ndb* is the density of ionized dangling bonds,
0 is the vacuum permittivity,
aSi is the
permittivity of amorphous silicon and d is the intrinsic layer thickness. In a-Si:H, the ratio of
ionized defects was found to be about 3035% [
55
]. A shown in Figure 4, the voltage needed for
full depletion grows quadratically with thickness and reaches a value of about 1000 V (or a mean
electric field of 2x105 Vcm-1) for a 50-µm-thick diode. Increasing the thickness while keeping full
depletion results in an increase of the electric field that concentrates in the vicinity of the p-i
interface and in a higher risk of breakdowns. However, such a high field will result (as seen in
Figure 2) in unacceptable leakage current values. Assuming a drift electron mobility of 3 cm2V-1s-
1, a generation of 100 e-h pairs per micron and a collection of 50% the charge transit time at 1000
V is 8.3 ns, which results (if charge trapping is neglected) in a peak current of 4.8×10-8 A, a value
much lower than the leakage current. This diode is clearly not suited for MIP detection, and the
strategy to target the highest diode thickness cannot be successful. The use of moderately thick
detectors is therefore advisable.
io
Figure 4: Minimum voltage necessary to fully deplete an a-Si:H n-i-p diode as a function of
diode thickness and ionized defect density.
In order to reduce as much as possible the defect density and, as a consequence, the needed
full depletion voltage, helium dilution of silane has been proposed. Diodes with an ionized defect
density as low as 7×1014 cm-3 have been obtained [54,
56
]. However, similar defect densities have
also been obtained with hydrogen dilution (and without He) [
57
]. As an alternative way to reduce
the voltage needed for full depletion, Morosanu et al. proposed to slightly p-type dope the intrinsic
layer. The procedure was successfully used in detectors with a thickness of up to 50 µm [
58
].
10
100
1000
10000
010 20 30 40 50
5x1014 cm-3
1x1015 cm-3
2x1015 cm-3
Full depletion voltage [V]
Diode thickness [µm]
2.3 Other particle detection
a-Si:H detectors have also been designed for the direct detection of X-rays, gamma rays,
electrons, neutrons and ions (in addition to protons and alpha particles). Detection of X-rays or
gamma rays could be achieved with thick diodes as designed for MIP detection [
59
]. However the
detection of such particles using an indirect scheme was found to more effective. Detection of
neutrons requires a reactive layer. Different designs have been proposed in the literature with, for
example, a combination a-Si:H diodes with Gd [
60
] or with 10B layers [
61
]. One could also
incorporate a 10B-rich semiconductor directly into an a-Si:H-based device.
2.4 Radiation resistance of a-Si:H
The high radiation hardness of a-Si:H was recognized very early, and this property was one
of the driving forces behind the development of detectors for high-energy physics experiments or
for medical imaging [
62
]. Several irradiation tests were performed using proton [47,
63
,
64
,
65
],
gamma [
66
], neutron [42], or heavy ions irradiation [
67
] on thin or thick p-i-n devices. For thin
devices, in most cases, experiments were carried out to study the effect of irradiation on the
photovoltaic properties for space applications. Figure 5 shows the effect of exposure to a
displacement dose on solar cell efficiency for various semiconductors. a-Si:H is found to be one
of the most radiation-resistant semiconductors.
Figure 5: Effect of irradiation, expressed as displacement damage dose, on the normalized solar
cell efficiency of several PV technologies (from Bätzner et al. [
68
], Srour [63] and Klaver [
69
]).
Proton irradiation tests were performed at CERN on thick a-Si:H diodes to get more insight
into defect formation and metastability. 32-µm-thick a-Si:H diodes were exposed to a 24 GeV
proton beam in the “IRRAD1” facility up to fluences of 2×1016 protons/cm2 [
70
]. Figure 6 shows
the current induced by the proton spills (120-ms-long periods of “beam on” with an average
fluence of 1.3×1011 protons). One can observe a slow degradation of the detector response with
fluence which tends to saturate at higher fluences (≥1015 protons/cm2 [70]) at a value
approximately equal to half of the initial value. An interesting self-annealing effect, related to the
metastability of the material (similar to SWE), is also observed when the irradiation was stopped
for 20 hours. Within the same study, similar samples were also irradiated with lower energy
protons of 405 keV. It was observed that these protons are much more effective at creating defects.
These additional defects then reduce the internal field and charge collection. At these energy levels,
protons i.e. hydrogen atoms are implanted in thick diodes and full recovery of the initial
properties by annealing is no longer possible (in contrast to thinner diodes where protons are not
stopped inside the device) [70]. In contrast to proton irradiation, electron irradiation creates
metastable defects that can be fully annealed out [
71
,
72
]. Note that effects of electron irradiation
at energies much above 1 MeV are not known.
Figure 6: Current induced by the 24 GeV proton spills on a 32.6-µm-thick a-Si:H diode as a
function of proton fluence within the CERN IRRAD1 facility. During the beam interruption at a
fluence of 4.5×1014 protons/cm2, the sample was kept under the same reverse bias voltage (300
V). The noise visible during the experiment originates from the variation of the beam intensity
from spill to spill.
An in-depth understanding of the effect of irradiation on a-Si:H is still missing.
Characterisation in terms of defect density or defect distribution is very incomplete due to
difficulties to perform this type of investigation on a-Si:H in general and on radioactive sample
(after irradiation) in particular. While the influence of the type of particle and its energy on the
material or diode properties can be deduced from various studies found in the literature, the
evolution of those properties with time as a function of the irradiation history cannot yet be
forecasted with enough confidence.
3. Vertically integrated a-Si:H detectors
Because of its performance as a particle detector and its high radiation resistance, a-Si:H
may be attractive for medical imaging, in addition to particle tracking in high-energy physics
experiments. For these applications, vertically integrated a-Si:H detectors were fabricated
following an approach pioneered by the University of Siegen for visible light imagers [
73
]. This
approach, known as “thin film on ASIC” (TFA) or “thin film on CMOS” (TFC), consists in a
vertical integration of the detecting layer on top of the readout electronics. It is obtained by the
direct deposition of an a-Si:H diode layer stack on the CMOS readout chip (see Figure 7). The
back electrodes (in general, the pads present on the readout chip) define the active area and the
individual diodes. Such a monolithic device facilitates the integration of the detector with the
readout electronics, reduces the related noise and can almost eliminate dead areas between pixels.
TFA sensors were successfully developed, aiming at the detection of singly charged particles
(including MIPs) and X-rays [
74
,
75
,
76
,
77
]. Examples of TFA detectors are shown in Figure 8.
Successful detections of β particles from 63Ni (with a maximum energy of 67 keV) and of beta
particles from 90Sr (with a maximum energy of 546 keV) were achieved as seen in Figure 9 [76].
Figure 7: Schematic view of a thin film on ASIC (TFA) particle detector.
Figure 8: Photographs of (left) an “AFP chip” with a linear array of 32 pixels (only one row of
the central pixels is used) consisting of a 32-µm-thick a-Si:H diode connected to an active
feedback preamplifier (AFP) and (right) of a 8×6 pixel particle detector in TFA technology with
a 15-µm-thick a-Si:H diode array. Each octagonal pixel has a lateral size of 150 µm with a pitch
of 380 µm and is connected to an AFP and shaper. The top common Indium tin oxide (ITO)
electrode is connected with a wire glued with Ag paste (left) or with a wire bonded to a gold tab
glued with Ag paste.
Figure 9: Signal as a function of β energy recorded with a 63Ni β source (left) and distribution of
amplitudes obtained with a 90Sr β sources (right) with a 32-µm-thick TFA “AFP chip”.
During the development of TFA imagers, it was observed that chip surface morphology and
pixel geometry can affect the leakage current of the pixels [
78
]. A test chip was designed by CERN
to study the effect of pixel geometry and edges on the performance of TFA sensors [
79
]. This chip
included pixels of various sizes and shapes, micro-strips of various width and pitch values and two
different geometries for the openings in the passivation layer of the chip. As for most TFA devices,
the back contacts of the individual pixels are provided by the top metal layer of the ASIC through
openings in the ASIC passivation. These openings can be local (for a single pixel) or global (for a
group of pixels or the entire chip). The local openings correspond to the size of the metal pads, but
can be slightly larger or smaller than the pads. For the TFA imagers [78], the edge of the
passivation openings has a detrimental effect and may lead to large additional leakage. A global
opening is therefore necessary to keep low leakage current values [79,
80
]. Diodes without an n-
layer (i.e. an i-p layer structure deposited directly on the metallic back contact, metal-i-p
configuration) even further reduce the leakage current [78,80].
The same test chip was also used to study spatial response and carrier collection as a function
of diode and micro-strip geometries. From electron-beam-induced current (EBIC) measurements,
lateral charge collection can be evaluated. For a-Si:H, lateral collection extends to only a few
microns, as transport is driven mainly by drift. Micrometre-size high spatial resolution can
therefore be achieved using pixelated devices or micro-strips [
81
]. Figure 10 shows an EBIC image
of a set of micro-strips connected in parallel covered with a 5-µm-thick a-Si:H n-i-p diode. Each
micro-strip (1.5 µm wide and spaced by 3.5 µm) is clearly resolved. Almost no cross-talk between
micro-strips located only 4 µm apart was observed in another experiment with a radioactive beta
source [79]. In contrast to c-Si based tracking devices where position is determined by the charge
sharing between neighbouring channels, spatial resolution is here essentially determined by the
shape of the a-Si:H diode electrodes. With limited lateral charge collection devices with feature
size and pitch smaller than the diode thickness (typically of less than 20 µm) can be fabricated
with almost no dead area. a-Si:H-based detectors could therefore be applied to beam or particle
tracking with a resolution of a few microns.
-50
0
50
100
150
-200 0 200 400 600 800
Signal amplitude [mV]
Time [ns]
46.8 keV
36.0 keV
15.6 keV
0
20
40
60
80
100
120
140
160
0 0.02 0.04 0.06 0.08 0.1
Signal
Noise
Entries
Signal (mV)
90Sr
Figure 10: (Left) EBIC image of a set of micro-strips with a width of 1.5 µm and spacing of 3.5
µm covered with 5-µm-thick a-Si:H n-i-p diode and (right) corresponding signal from EBIC line
scans through the micro-strip set. Measurement was performed at an applied voltage of 30 V and
beam energy of 20 keV. All strips are connected in parallel. The two vertical dark bands in the
EBIC image correspond to areas degraded by previous EBIC line scans.
As observed for a-Si:H particle detectors deposited on glass, full depletion was here also
difficult to attain in diodes thicker than 15 µm. This thickness value seems for the time being to
be an optimal value in terms of ease of fabrication and depletion region width [79]. The maximum
signal-to-noise ratio was on the order of 5 (higher than previous results from detectors deposited
on glass) which is still insufficient for clear single MIP detection [79,
82
]. However, as discussed
in the previous section a clear room for improvement of this noise-to-signal ratio exists.
Optimizing the growth of the material for low defect density (deposition using low growth rates
such as the ones used for state-of-the-art solar cell fabrication with a reduction of the growth rate
by at least a factor of 10 with respect to the one used for the present detectors) could reduce the
defect density by up to one order of magnitude and could then allow for much thicker device with
full depletion. Such deposition conditions at low rates have not yet been considered as they
substantially increase the processing time and result in more dense and mechanically stressed
layers (requiring additional optimization to avoid layer delamination). Optimization of the device
p-i interface for low leakage current and deposition of the detector on a readout chip with a flat
surface morphology are also proven solutions which permit higher device thicknesses and bias
electric field values without increasing the leakage current. These two approaches (separately of
jointly) should enable a higher signal-to-noise ratio, possibly allowing for clear single MIP
detection.
4. Detectors using an indirect detection scheme
The combination of a Si detector with a scintillating layer can be used to avoid the limitations
of Si to detect MIPs. Such a detection scheme has been successful for X-ray radiography using
a-Si:H photodiodes [
83
,
84
,
85
]. With the ability of a-Si:H to be deposited on large areas, X-ray
imagers of various sizes have been developed and deployed for medical applications. These
imagers comprise a thin-film transistor (TFT) backplane used to address the individual a-Si:H
photodiodes, which are integrated on the same substrate as the TFTs. A scintillating layer such as
CsI is deposited on top of the photodiode array. Readout is performed by sequentially addressing
each row of photodiodes, switching on the respective TFTs and extracting the collected charge
stored on these photodiodes. Benefiting from the possibility to deposit a-Si:H on large substrate,
large-area X-ray imagers able to perform chest radiography, for example, have been developed
[
86
]. Besides radiography, such imagers can also be used for other types of medical imaging such
as fluoroscopy and angiography, reducing the X-ray dose for the patient [
87
]. While MIP detection
can be performed with large-area devices, their application and design is for imaging and they are
not capable of single MIP detection. Detection is here limited by the TFT leakage, readout
frequency and imager size.
50 µm
0
5
10
15
20
020 40 60 80 100 120 140
Current [µA]
Position [µm]
Dummy strips
Chip passivation opening
Positron emission tomography (PET) scanning [
88
] is also attracting attention for the
application of a-Si:H photodiodes [
89
]. This medical imaging technique is based on the use of a
radioactive positron emitter injected into the patient’s blood. Shortly after the emission, the
positron annihilates itself with a nearby electron leading to the emission of two 511 keV photons
in opposite directions. A synchronous detection is required to reconstruct the emission region. For
this purpose, fast-decaying scintillating materials are needed. Within the Crystal Clear
Collaboration [
90
], several fast scintillators have been developed and studied. However, most have
their peak emission at UV or deep blue wavelengths, which are more difficult to record using
a-Si:H photodiodes [89]. However, a-Si:H n-i-p photodiodes with thinner p-doped layers and an
optimized front transparent electrode (transparent conductive oxide, TCO), or photodiodes in the
reverse configuration (p-layer at the back with respect to illumination) without an n-layer, can
achieve high external quantum efficiency at wavelengths of interest, as shown in Figure 11. While
an integrated device combining a fast scintillator, a-Si:H diodes and readout electronics has not
yet been tested, the performance of the individual components seems sufficient to allow for gamma
detection in PET scanners [89].
0
0.2
0.4
0.6
0.8
1
400 500 600 700 800
nip- <p> 1'+2.5'/4'/8'/9' - ITO 45"
<p> 1'+2.5'
<p> 1'+4'
<p> 1'+8'
<p> 1'+9'
EQE @ -3 V
Wavelength [nm]
0
0.2
0.4
0.6
0.8
1
400 500 600 700 800
pi - ITO 45"
-5.00 V (170ºC/170ºC)
-9.00 V (170ºC/170ºC)
-5.00 V (170ºC/200ºC)
-9.00 V (170ºC/200ºC)
EQE
Wavelength [nm]
Figure 11: External quantum efficiency (EQE) of several a-Si:H diodes: (left) in the metal-n-i-p-
TCO configuration with various p-layer thickness and (right) in the metal-p-i-TCO
configuration. The p-layer is formed of a stack of two individual layers: the first one is the same
for all diodes (1’ deposition time) while the deposition time, and thickness, of the second layer
are variable. Indium tin oxide (ITO) is used here as the TCO layer.
An innovative particle detector using an indirect detection scheme was recently proposed. It
comprises a microfluidic device filled with a scintillating liquid where the microchannels are also
used to guide the light to the photodiodes [
91
]. a-Si:H is attractive for this application, as
photodiodes can be deposited directly on the microfluidic device for monolithic integration. Also
in this case the device geometry, light yield of the scintillating liquid and light propagation through
the micochannels should not allow for single MIP detection when combined with a-Si:H
photodiodes. Applications to beam monitoring are nevertheless possible [
92
].
5. a-Si:H-based microchannel plate detectors (AMCP)
5.1 AMCP concept
Microchannel plates (MCPs) are electron-multiplier devices [
93
] that were initially
developed for night vision [93] and astrophysics applications [
94
,
95
] and then used in a variety of
other fields. Such detectors consist in thick glass plates with very narrow microchannels (or pores)
drilled throughout the device. When a primary electron hits the channel wall, secondary electrons
are emitted. Since a high electric field is applied between the two faces of the plate, the primary
electrons multiply by impact ionization forming an avalanche within the microchannels. A
schematic view of such a device and its functioning is shown in Figure 12. Gains in excess of 1000
can be achieved from a single plate. To enhance the gain, microchannels are not perpendicular to
the surface but at a certain angle (referred as bias angle). Much higher values are obtained when
stacking several devices in a chevron arrangement [Error! Bookmark not defined.]. When
combined with a low-noise readout chip, detection of MIPs can be achieved. State-of-the-art MCPs
are fabricated from lead glass, and the channel walls are coated with a conductive material to
ensure charge replenishment.
Figure 12: Schematic cross section of two channels of a MCP device stacked on a readout
electronic chip.
We recently proposed MCP detectors based on a-Si:H as an alternative to particle detection
using thick diodes [
96
,
97
]. The properties of a-Si:H could overcome some of the limitations
introduced by the lead glass or the c-Si usually used for MCP fabrication [
98
,
99
]. The bulk
resistivity of a-Si:H (see section 1.2) is high enough to sustain a high bias voltage and small enough
to provide charge replenishment after the emission of secondary electrons while keeping
reasonably low leakage current and to avoid the need of coating the channel walls for charge
replenishment. With a more efficient charge replenishment along the length of the channel, a-Si:H-
based MCPs (AMCPs) are expected to have much shorter dead time (time needed for charge
replenishment) and larger gain under continuous operation.
Another key advantage of AMCPs is the possibility to vertically integrate MCPs on the
readout electronics, similar to the approach used for TFA detectors (see section 3). All fabrication
steps are at processing temperatures 200°C (see below) and fully compatible with CMOS
electronics. Such monolithic integration should provide low noise and low power consumption
and ease single MIP detection. An additional advantage of AMCPs is that channel micromachining
is performed using deep reactive ion etching (DRIE), which allows the channel geometry to be
fully customized and adapted to the readout electronics. The combination of photolithography and
DRIE can provide aspect ratios comparable to lead-glass MCPs [
100
].
5.2 AMCP fabrication
All AMCPs thus far were fabricated on thermally oxidized Si wafers (500 µm thick, single
or double side polished) with patterned Al or Cr electrodes. This substrate was chosen to mimic as
much as possible the device structure of a fully integrated AMCP on its readout electronics. The
main part of the AMCP consists in a thick (up to 120 m) a-Si:H layer covered with a doped a-Si:H
layer used as the top electrode. After the deposition of the layer stack, a first patterning/etching
step is performed to define the individual device (test reticles) and to uncover the bonding pads,
followed by the DRIE step to drill the microchannels. The critical aspects of the fabrication reside
primarily in the deposition of the thick a-Si:H layer. This is performed by PE-CVD (as for other
a-Si:H devices) under conditions that must minimize stress in the layer to avoid delamination of
the stack or formation of blisters. Details on the fabrication can be found in [
101
] and [
102
].
The first generation of AMCPs consisted in simple two-terminal devices: a bottom electrode
deposited on the oxidized wafer and a top doped layer. While avalanche events could be
demonstrated using EBIC measurements [97], this configuration did not allow us to separate the
current induced by the avalanche from the leakage current through the AMCP structure. A new
generation was designed featuring an additional electrode at the back, separated from the bottom
readout electrode (anodes) by a thin insulating layer [101]. The overall device structure and
detailed scanning electron microscope (SEM) images of the device are given in Figure 13. In this
configuration, the bias electric field used to induce avalanches and electron multiplication is
applied between the top and intermediate electrode, while the bottom electrode is used for readout
and is therefore insulated from the leakage current. In the same figure, detailed SEM images from
the channel entrance and bottom of the channel are also presented.
Figure 13: (a) Schematic view of a vertically integrated AMCP on chromium readout anodes.
The electric field for electron multiplication is applied between the top and intermediate
electrodes. (b) SEM image of the AMCP channel openings and (c) of the intermediate and
bottom readout electrodes separated by an insulating layer stack. From [101].
AMCPs with channel lengths of up to 100 µm have been fabricated with a variety of
microchannel diameters (or channel apertures) between nominally 4 and 6 µm. Microchannels
with parallel walls could be obtained for all geometries; however, DRIE led to a slight increase in
the channel aperture (by 1 to 1.5 µm from the nominal diameter). Aspect ratio values of up to 15
were demonstrated.
5.3 AMCP performance
AMCP devices were characterized by means of EBIC measurements and in quasi-steady
state mode with electron irradiation from a photocathode illuminated by UV light. Excitation was
here modulated by chopping the illumination at low frequency (< 1 kHz). Gain was obtained from
the difference between the signals with and at 0 V bias voltage applied. Comparisons of the gain
for different channel geometries (effect of channel length and aspect ratio) are plotted in Figure
14.
At this stage the observed gain values and their dependency on channel geometry is
consistent with the performance observed in state-of-the-art lead-glass MCPs. According to the
model developed by Eberhardt [
103
], we could expect the gain values of AMCPs to increase by a
factor of 102103 for an aspect ratio of 30, a value which is achievable with DRIE of Si [
104
].
From the modelling of the present AMCP, a secondary electron yield the number of electrons
generated per collision of 1.51.7 was inferred. This yield (and, as a consequence, the AMCP
gain), which is quite low compared to values achieved in state-of-the-art photomultiplier tube
dynodes (values of 4), should be increased by coating the channel wall with Al2O3 as done in state-
of-the-art MCPs [
105
]. A first test with such a coating in an AMCP led to an increase of the gain
by a factor of 2 [102]. Possible alternatives to explore are SiO2, Si3N4 [
106
] or diamond-like layers
[
107
].
Figure 14: AMCP gain dependency on aspect ratio. (a) Gain versus bias voltage for aspect
ratios of 10.3:1 and 12.5:1, with a channel length of 76 µm and channel apertures of 7.6 and 6.1
µm, respectively. (b) Gain versus electric field for aspect ratios of 12.5:1 and 8:1, with a channel
aperture of 6.1 µm and channel lengths of 76 and 53 µm, respectively. From [101].
By increasing the aspect ratio of AMCPs to values around 30 and improving the secondary
emission yield (with an optimized coating), gain values approaching104 can be expected. With a
vertical integration of AMCPs on low-noise readout electronics, the detection of a single MIP
should be possible at very high repetition rate. Nevertheless, besides the demonstration of such
gain values, a comprehensive analysis of AMCPs is still to be carried out. As observed during the
development of AMCPs, AMCPs exhibit drift of their performance (marked reduction of the gain)
under steady-state operation (AMCP continuously exposed to electrons emitted from a
photocathode). While a charging effect of the SiOx layers present in the layer stack has been
observed, the metastability of the material (see section 1.2) could also be involved. Finally no
performance analysis in transient and low excitation conditions (towards single particle detection)
has yet been completed.
6. Future prospects and conclusions
While a-Si:H-based detectors have been successfully deployed for digital radiography, they
have not (yet) found their place in other applications. Despite the attractive features of a-Si:H, such
as large-area deposition, relatively easy processing and high radiation resistance, the difficulty to
achieve a high signal-to-noise ratio for MIP detection and the unclear benefits of this technology
compared to other materials have discouraged most of the actors. The performance of the devices
fabricated in our laboratory or reported in the literature is clearly not at the level of what could be
possible. This is not too surprising given the much smaller effort invested for particle-detection
applications compared to the effort invested for solar cell or transistor applications. The need for
thick devices also adds more difficulties in the mastering of the technology. Diode leakage current
could be reduced by an improvement in material quality and diode design. The design and
fabrication of CMOS chips dedicated to TFA would also permit a gain in performance.
The very high spatial resolution of a-Si:H permitted by TFA technology has not yet been
exploited. This resolution is basically given by the geometry of the back electrode(s) of the diode
and could be as small as a few micrometres. High-precision beam positioning, beam shaping or
particle tracking/imaging could be achieved with this technology.
Indirect detection using scintillators could possibly achieve single MIP detection.
Application to PET scanners was targeted by a few groups but little effort has been invested and
little progress made for any successful application. New detector concepts using scintillating liquid
and microfluidic devices offer new perspectives, but the detection volume strongly reduces the
number of available photons and completely prohibits single MIP detection. However the
technology could allow for flexible devices or geometries that are not achievable by other means
and thus could possibly initiate new applications.
MCPs are an exciting new application of a-Si:H. Achieving single particle detection by
improving the gain would pave the way for many applications. Vertical integration on readout
electronics developed for nuclear physics (such as Medipix or NINO chips) would allow for
photon-counting imagers. AMCPs could be applied to a wide range of applications that need both
a high spatial resolution (given by the pitch of the microchannel) and ultra-fast time resolution.
Comprehensive studies are still needed to understand the material behaviour in AMCP and the
fundamental limits of its operation. Furthermore the gain is still too low and must be improved by
at least an order of magnitude to permit any practical application.
The potential for a-Si:H-based detector deployment remains considerable as the
requirements for radiation-resistant detectors become more stringent. An increase in the luminosity
of modern accelerators is calling for alternative solutions for the detectors. There are several
candidates that could offer the required properties. However, all materials have benefits and
drawbacks, and a-Si:H is no exception.
Acknowledgments
Part of the results presented were obtained with the financial support of CERN, ETHZ and
of the Swiss National Science Foundation projects 126926/1 and 144375/1.
Appendix. List of abbreviations
a-Si:H Hydrogenated amorphous silicon
AFP Active feedback amplifier
AMCP a-Si:H-based MCP
ASIC Application-specific integrated circuit
CMOS Complementary metal-oxide semiconductor
CVD Chemical vapour deposition
DB Dangling bond
DRIE Deep reactive ion etching
EBIC Electron-beam induced current
e-h electron-hole
ITO Indium tin oxide
LHC Large Hadron Collider
MCP Microchannel plate (detectors)
MIP Minimum ionizing particle
SEM Scanning electron microscope
TCO Transparent conductive oxide
TFA Thin film on ASIC
TFC Thin film on CMOS
TFT Thin-film transistor
UV Ultraviolet
References
[
1
] Milbrath B D Peurrung Bliss M Weber W J 2008 Radiation detector materials: An overview
J. Mater. Res. 23 2561-2581
[
2
] Casse G 2013 Recent developments on silicon detectors Nucl. Instr. and Meth. in Phys. Res.
A732 1620.
[
3
] Chittik R C Alexander J H Sterling H E 1969 The Preparation and Properties of Amorphous
Silicon J. Electrochem. Soc. 116 77-81.
[
4
] Fritzsche H 1980 Characterization of Glow-Discharge Deposited a-S:H Sol. Ener. Mat. 3
447-501
[
5
] Spear W Lecomber P G 1976 Electronic properties of substitutionally doped amorphous Si
and Ge Phil. Mag. 33 935-949
[
6
] Tuan H C 1984 Amophous Silicon Thin Film Transistor and its Application s to Large-Area
Electronics Mater. Res. Soc. Proc. 33 247
[
7
] Carlson D E Wronski C R 1976 Amorphous silicon solar cell Appl. Phys. Lett. 28 671-673
[
8
] Owen A E Le Comber P G Spear W E Hajto J 1983 Memory switching in amorphous silicon
devices J. of Non-Cryst. Sol.59-60 12731280
[
9
] Rech B Wagner H 1999 Potential of amorphous silicon for solar cells Appl. Phys. A 69 155-
167
[
10
] Rech B Wieder S Siebke F Beneking C Wagner H 1966 Material Basis of Highly Stable
a-Si:H Solar Cells Mater. Res. Soc. Symp. Proc. 420 33
[
11
] Shah S Dutta J Wyrsch N Prasad K Curtins H Howling A Hollenstein C 1992 VHF plasma
deposition: A comparative study Mater. Res. Soc. Proc. 258 15-26
[
12
] Paquin L Masson D Wertheimer M R Moisan M 1985 Amorphous silicon for photovoltaics
produced by new microwave plasma-deposition techniques Canadian Journal of Physics
63 831-837
[
13
] Mahan A 2003 Hot wire chemical vapor deposition of Si containing materials for solar cells
Sol. Ener. Mat. and Sol. Cells 718 299-327
[
14
] Wronski C R Lee S Hicks M Kumar S 1989 Internal photoemission of holes and the mobility
gap of hydrogenated amorphous silicon Phys. Rev. Lett. 63 1420-1423
[
15
] Tauc J Grigorovici R Vancu A 1966 Optical properties and electronic structure of amorphous
germanium Phys. Stat. Sol. (b) 15 627-637
[
16
] Zanzucchi P Wronski C R Carlson D E 1977 Optical and photoconductive properties of
dischargeproduced amorphous silicon J. of Appl. Phys. 48 5227-5236
[
17
] Smets A H M Kessels W M M van de Sanden M C M 2003 Vacancies and voids in
hydrogenated amorphous silicon Appl. Phys. Lett. 82 1547-1549
[
18
] Smets A H M Wronski C R Zeman M van de Sanden M C M 2010 The Staebler-Wronski
Effect: New Physical Approaches and Insights as the Route to Reveal its Origin Mater.
Res. Soc. Symp. Proc., 1245 A.14.02
[
19
] Lee Y Jiao L Liu H Lu Z Collins R Wronski C W 1996 Stability of a-Si:H solar cells and
corresponding intrinsic materials fabricated using hydrogen diluted silane Proc. of the
25th IEEE photovolt. Spec, conf., pp 1165-1168
[
20
] Kondo M Matsuda A 2002 An approach to device grade amorphous and microcrystalline
silicon thin films fabricated at higher deposition rates Curr. Opinion in Solid State and
Mat. Sci. 6 445453
[
21
] Roca i Cabarrocas P Fontcuberta i Morral A Poissant Y 2002 Growth and optoelectronic
properties of polymorphous silicon thin films Thin Solid Films 403 39-46
[
22
] Street R A 1982 Doping and the Fermi energy in amorphous silicon Phys. Rev. Lett. 49 1187-
1190
[
23
] Staebler D L Wronski C W 1977 Reversible conductivity changes in dischargeproduced
amorphous Si Appl. Phys. Lett. 31 292-294
[
24
] Kolodziej A 2004 Staebler-Wronski effect in amorphous silicon and its alloys Opto-
Electron. Rev. 12 21-32
[
25
] Stradins P 2010 Staebler-Wronski Defects: Creation Efficiency , Stability, and Effect on
a-Si:H Solar Cell Degradation Proc. of the 25th IEEE photovolt. Spec. Conf. pp 142-145
[
26
] Street R A 2005 Hydrogenated Amorphous Silicon (Cambridge, Cambridge University
Press)
[
27
] Shah A 2010 Thin-Film Silicon Solar Cells (Lausanne, EPFL Press)
[
28
] Paduschek P Höpfl C Mitlehner H 1983 Hydrogen-Related Mechanical Stress in Amorphous
and Plasma-deposited Silicon Nitride Thin Solid Films 110 291-304
[
29
] Kroll U Meier J Shah A Mikhailov S Weber J 1996 Hydrogen in amorphous and
microcrystalline silicon films prepared by hydrogen dilution J. of Appl. Phys. 80 4971
[
30
] Chabloz P Keppner H Fischer D Link D Shah A 1996 Amorphous silicon p-i-n diodes,
deposited by the VHF-GD process: new experimental results J. of Non-Cryst. Sol. 198-
200 1159
[
31
] Wyrsch Miazza C Ballif C Shah A Blanc N Kaufmann R Lustenberger F Jarron P 2005
Vertical integration of hydrogenated amorphous silicon devices on CMOS circuits Mater.
Res. Soc. Proc. 869 3-14
[
32
] Leo W R 1994 Passage of Radiation Through Matter Techniques for Nuclear and Particle
Physics Experiments (Berlin, Springer-Verlag) pp 17-68
[
33
] Wyrsch N et al. 2003 Development of Vertically Integrated Imaging and Particles Sensors
Mater. Res. Soc. Proc. 762 205
[
34
] Pochet T Dubeau J Equer B Karar A Hamel L A 1990 High reverse voltage amorphous
silicon p-i-n diodes J. of Appl. Phys. 68 1340-1344
[
35
] Fujieda I et al. 1990 Field Profile Tailoring in a-Si:H Radiation Detectors Mat. Res. Soc.
Symp. Proc. 192 399-404
[
36
] Perez-Mendez V et al. 1991 Amorphous Silicon Based Radiation Detectors J. of Non-Cryst.
Sol. 137-138 1291-1296
[
37
] Morrison S Servati P Vygranenko Y Nathan A Madan A 2002 Reduction of Dark Current
Under Reverse Bias in a-Si:H p-i-n Photodetectors Mat. Res. Soc. Symp. Proc. 715 701
[
38
] van der Graaf H 2010 Gaseous detectors Nucl. Instr. and Meth. in Phys. Res. A 628 27-30
[
39
] van der Graaf H et al. 2013 The gridpix detector: History and perspective Mod. Phys. Lett.
A 28 1-7.
[
40
] Perez-Mendez V Morel J Kaplan S N Street R A 1986 Detection of Charged Particles in
Amorphous Silicon Layers Nucl. Instr. and Meth. in Phys. Res. 252 478-482
[
41
] Perez-Mendez V Kaplan S N Ward W Qureshi S Street R A 1987 Signal, Recombination
Effects and Noise in Amorphous Silicon Detectors Nucl. Instr. and Meth. in Phys. Res.
A260 195-200
[
42
] Perez-Mendez V Kaplan S N Cho G Fujieda Qureshi S Ward W Street R A 1988
Hydrogenated Amorphous Silicon Pixel Detectors for Minimum Ionizing Particles Nucl.
Instr. and Meth. in Phys. Res. A273 127-134
[
43
] Xi J et al. 1991 Minimum ionizing particle detection using amorphous silicon diodes Nucl.
Instr. and Meth. in Phys. Res. A301 219-222
[
44
] Equer B Karar A 1988 Effect of Primary Ionization in Amorphous Silicon Detectors Nucl.
Instr. and Meth. in Phys. Res.A271 574 584
[
45
] Hordequin C Brambilla Bergonzo P Foulon F 2001 Nuclear radiation detectors using thick
amorphous-silicon Nucl. Instr. and Meth. in Phys. Res. A456 284-289
[
46
] Aglietti U Bacci C Evangelisti F Falconieri M Fiorini P Meddi F 1989 a-Si:H p-i-n Junction
as Ionizing Particle Detectors J. of Non-Cryst. Sol. 115 177-179
[
47
] Kishimoto N Amekura H Kono K LeeC G 1998 Stable photoconductivity in metastable a-
Si:H under high-energy proton irradiation J. of Non-Cryst. Sol. 227-230 238-242.
[
48
] Chabloz P et al. 1992 X-Ray Detectors Based on “Thick” a-Si:H Layers Deposited by the
VHF-GD Process Mater. Res. Soc. Proc. 258 1057-1062
[
49
] Aleksan R Bolognese T Equer B Karar A Reymond J M 1991 Observation of single
minimum ionising particles with amorphous silicon diodes Nucl. Instr. and Meth. in Phys.
Res. A305 512-516
[
50
] Dubeau J Pochet T Hamel L A Equer B Karar A 1991 Response of amorphous silicon p-i-n
detectors to ionizing particles Nucl. Instr. and Meth. in Phys. Res. B54 458 471
[
51
] Klein C A 1968 Bandgap Dependence and Related Features of Radiation Ionization Energies
in Semiconductors J. of Appl. Phys. 39 2029-2038 IEEE Trans. Nucl. Sci. 36 1347-1352
[
52
] Naruse Y Hatayama T 1989 Metal/amorphous Silicon Multilayer Radiation Detectors
[
53
] Caracciolo H, Balbinot G, Bazzano G, Bosser J, Caldara M, Parravicini A, Pullia M, Viviani
C 2011 Beam Diagnostics Commissioning at CNAO Proc. of the 2nd Int. Particle
Accelerator Conf. (IPAC2011) pp 2848-2850
[
54
] Hong W S et al. 1995 Charged particle detectors based on high quality amorphous silicon
deposited with hydrogen or helium dilution of silane", IEEE Trans. Nucl. Sci. 42 240-246
[
55
] Quereshi S Perez-Mendez V Kaplan S N Fujeda I Gho G Street R A 1989 Material
Parameters in Thick Hydrogenated Amorphous Silicon Radiation Detectors J. of Non-
Cryst. Sol. 114 417-419
[
56
] Pochet T Illie A Foulon F Equer B 1994 Characterization of new a-Si:H Detectors fabricated
fiom Amorphous Silicon deposited at High Rate by Helium enhanced PECVD IEEE
Trans. on Nucl. Sci, 41 1014 1018
[
57
] Despeisse M et al. 2008 Hydrogenated Amorphous Silicon Sensor Deposited on Integrated
Circuit for Radiation Detection. IEEE Trans. on Nucl. Sci. 55 802-811
[
58
] Morasanu C et al. 1993 a-Si:H based particle detectors with low depletion voltage J. of Non-
Cryst. Sol. 164-166 801-804.
[
59
] Perez-Mendez V Cho G Fujieda I Kaplan S N Qureshi S Street R A 1989 The Application
of Thick Hydrogenated Amorphous Silicon Layers to Charge Particle and X-ray
Detection Mater. Res. Soc. Proc. 149 621-630
[
60
] Mireshghi A et al. 1994 High Efficiency Neutron Sensitive Amorphous Silicon Pixel
Detectors IEEE Trans. on Nucl. Sci. 41 915 921
[
61
] Foulon F Bergonzo P Brambilla C Jany C Guizard B Marshall R D 1998 Neutron Detectors
Made from Chemically Vapour Deposited Semiconductors Mater. Soc. Res. Proc. 487
591-
[
62
] Antonuk L E Boudry J Yorkston J Wild C F Long M J Street R A 1990 Radiation damage
studies of amorphous-silicon photodiode sensors for applications in radiotherapy X-ray
imaging Nucl. Instr. and Meth. in Phys. Res. A 299 143-146
[
63
] Srour J R Vendura G J Lo D H Toporow C M C Dooley M Nakano R P King E E 1998
Damage Mechanisms in Radiation-Tolerant Amorphous Silicon Solar Cells IEEE Trans.
on Nucl. Sci. 45 2624-2631
[
64
] Schwarz R Kolodzey J Alijishi S Wagner S 1985 Radiation Damage by 12 MeV Protons and
Annealing of Hydrogenated Amorphous Silicon Proc. of the 18th IEEE Photovolt. Spec.
Conf. 903
[
65
] Kuendig J Goetz M Shah A Gerlach L Fernandez E 2003 Thin film silicon solar cells for
space applications: Study of proton irradiation and thermal annealing effects on the
characteristics of solar cells and individual layers Sol. Ener. Mat. & Sol. Cells 79 425-
438
[
66
] Boudry J M Antonuk L E 1994 Radiation Damage of Amorphous Silicon Photodiode
Sensors IEEE Trans. Nucl. Sci. 41 703-707
[
67
] Bacci C et al. 1991 Detection of strongly ionizing particles with a hydrogenated amorphous
silicon detector Nucl. Instr. and Meth. in Phys. Res. A306 182-186
[
68
] Bätzner D L Romeo A Terheggen M Döbeli M Zogg H Tiwari A N 2004 Stability aspects
in CdTe/CdS solar cells Thin Solid Films 451452 536543
[
69
] Klaver A 2007 Irradiation-induced degradation of amorphous silicon solar cells in space ,
TU Eindhoven Ph.D. Thesis http://repository.tudelft.nl/file/768999/375461
[
70
] Wyrsch N et al. 2006 Radiation hardness of amorphous silicon particle sensors J. of Non-
Cryst. Sol. 352 1797-1800
[
71
] Schneider U Schröder B Finger F 1987 The creation of metastable defects in a-Si:H films
by high dose irradiation with keV-electrons J. of Non-Cryst. Sol. 97&98 795-798
[
72
] Diehl F Herbst W Bauer S. Schröder B Oechsner H 1996 Metastability in p- and n-type a-
Si:H investigated by keV-electron irradiation J. of Non-Cryst. Sol. 198-200 436-440
[
73
] Schneider B Rieve P Böhm M Jähne B Haußecker H Geißler P 1999 Handbook on Computer
Vision an Applications (Boston, Academic Press) pp 237-270
[
74
] Wyrsch N Miazza C Dunand D Shah A Moraes D Anneli G Despeisse M Jarron P Dissertori
G Viertel G 2004 Vertically Integrated Amorphous Silicon Particle Sensors Mater. Soc.
Res. Proc. 808 441
[
75
] Anelli G et al. 2004 A new concept of monolithic silicon pixel detectors: Hydrogenated
amorphous silicon on ASIC Nucl. Instr. and Meth. in Phys. Res. A518 366-372
[
76
] Wyrsch N et al. 2004 Thin-film silicon detectors for particle detection Physica Status Solidi
(c) 1 1284-1291
[
77
] Jarron P et al. 2006 TFA pixel sensor technology for vertex detectors Nucl. Instr. Meth. in
Phys. Res. A560 122-126
[
78
] Miazza C et al. 2006 Image Sensors Based on Thin-film on CMOS Technology: Additional
Leakage Currents due to Vertical Integration of the a-Si:H Diodes Mater. Res. Soc. Proc.
Symp. 910 A17-03.
[
79
] Despeisse M Anelli G Kaplon J Moraes D Nardulli A Powolny F Wyrsch N 2008
Hydrogenated Amorphous Silicon Sensor Deposited on Integrated Circuit for Radiation
Detection IEEE Trans. Nucl. Sci. 55 802-811
[
80
] Wyrsch N Choong G Miazza C Ballif C 2008 Performance and Transient Behavior of
Vertically Integrated Thin-film Silicon Sensors Sensors 8 4656-4668
[
81
] Franco A Riesen Y Despeisse M Wyrsch N Ballif C 2012 High Spatial Resolution of Thin-
Film-on-ASIC Particle Detectors IEEE Trans. Nucl. Sci. 59 2614-2621
[
82
] Moraes D et al. 2004 A novel low noise hydrogenated amorphous silicon pixel detector J.
of Non-Cryst. Sol. 2004 729-731
[
83
] Kim H K Cunningham I A Yin Z Cho G 2008 Int. J. of Prec. Engin. and Manuf. 9 86-100
[
84
] Moy J-P 1999 Large area X-ray detectors based on amorphous silicon technology Thin Solid
Films 337 213-221.
[
85
] Nathan A et al. 2000 Amorphous silicon detector and thin film transistor technology for
large-area imaging of X-rays Microelectronics J. 31 883891.
[
86
] Kroft L J M Veldkamp W J H Mertens B J A Boot M V Geleijns J 2005 Comparison of
Eight Different Digital Chest Radiography Systems: Variation in Detection of Simulated
Chest Disease Amer. J. of Roentgenology 185 339-346
[
87
] Spahn M 2005 Flat detectors and their clinical applications Eur Radiol. 15 19341947
[
88
] Peng H Levin C S 2010 Recent Developments in PET Instrumentation Curr Pharm
Biotechnol. 11 555571
[
89
] Nardulli A et al. 2008 Quantum efficiency measurement of nip a-Si:H photodiode arrayon
CMOS circuit for positron emission tomography (PET) J. of Non-Cryst. Sol. 354 2603
2605
[
90
] http ://crystalclear.web.cern.ch
[
91
] Mapelli A et al. 2010 Scintillation particle detection based on microfluidics Sensors and
Actuators A162 272275
[
92
] Maoddi P 2015 Microfluidic Scintillation Detectors for High Energy Physics EPFL PhD
thesis 6620
[
93
] Wiza J L 1979 Microchannel plate detectors Nucl. Instr. and Meth. in Phys. Res. 162 587-
601
[
94
] Siegmund O, Malina R, Coburn K, Werthimer D 1984 Microchannel plate EUV detectors
for the Extreme Ultraviolet Explorer IEEE Trans. Nucl. Sci. 31 776-779
[
95
] Fraser G, Barstow M, Whiteley M, Wells A 1982 Enhanced soft x-ray detection efficiencies
for imaging microchannel plate detectors Nature 300 509-511
[
96
] Powolny F 2009 Characterization of time resolved photodetector systems for Positron
Emission Tomography Ph.D. thesis, University of Neuchâtel
[
97
] Wyrsch N, Powolny F, Despeisse M, Dunand D, Jarron P, Ballif C 2010 Micro-Channel
Plate Detectors Based on Hydrogenated Amorphous Silicon Mater. Res. Soc. Proc. Symp.
1245 193
[
98
] Beetz C P et al. 2000 Silicon-micromachined microchannel plates Nucl. Instr. and Meth. In
Phys. Res. A 442 443-451
[
99
] Duanmu Q et al. 2001 Preparation and Performance of Si Microchannel Plate Proc. of SPIE
4601 284-287
[
100
] Marty F et al. 2005 Advanced etching of silicon based on deep reactive ion etching for silicon
high aspect ratio microstructures and three-dimensional micro-and nanostructures
Microelectr J. 36 673677
[
101
] Franco A, Geissbühler J, Wyrsch N, Ballif C 2014 Fabrication and characterization of
monolithically integrated microchannel plates based on amorphous silicon Scientific
Reports 4 4597
[
102
] Franco A 2014 Monolithic Particle Detectors based on Hydrogenated Amorphous Silicon
EPFL PhD thesis 6290
[
103
] Eberhardt E H 1981 An operational model for microchannel plate devices IEEE Trans. Nucl.
Sci. 28 712717
[
104
] Yeom J Wu Y Selby J C Shannon M A 2005 Maximum achievable aspect ratio in deep
reactive ion etching of silicon due to aspect ratio dependent transport and the
microloading effect J. Vac. Sci. Technol. B23 2319
[
105
] Mane A U et al. 2012 An atomic layer deposition method to fabricate economical and robust
large area microchannel plates for photodetectors Phys. Procedia 37 722732
[
106
] Lapington J S et al. 2009 Investigation of the secondary emission characteristics of CVD
diamond films for electron amplification Nucl. Instr. & Meth. in Phys. Res.A610 253 -
257
[
107
] Fijol J J Then A M Tasker G W 1991 Secondary electron yield of SiO2 and Si3N4 thin films
for continuous dynode electron multipliers Appl. Surf. Sci. 4849 464471
... However, doping also introduces many additional defects, and for this reason direct p-n junction cannot be used as active material in particle detectors and solar cells. Therefore, the simplest detector structures that have been fabricated and successfully tested are p-i-n diodes or Schottky diodes [10]. ...
... Many different kinds of particles have been detected using planar diode devices, including MIPs [10], x-rays [15], neutrons using both boron [16] or gadolinium converters [17], alpha particles [18], and heavier ions [19]. ...
... The results of this test are shown in Figure 3 for a set of micro-strips (1.5 μm wide and spaced by 3.5 μm) on a 5 μm thick a-Si:H n-i-p diode. From this figure, it is evident that the signals from the strips are clearly separated because the lateral charge spread is in the order of few microns [10]. Cross talk (i.e., induced signal in neighbor strips) has been measured in another experiment with a beta source and was found to be negligible [22]. ...
Article
Full-text available
Hydrogenated amorphous silicon (a-Si:H) particle detectors have been considered as alternatives to crystalline silicon detectors (c-Si) in high radiation environments, due to their excellent radiation hardness. However, although their capability for particle flux measurement in beam monitoring applications is quite satisfactory, their minimum ionizing particle (MIP) detection has always been problematic because of the poor signal-to-noise ratio caused by a low charge collection efficiency and relatively high (compared to crystalline silicon) leakage current. In this article, after a review of the status of technological research for a-Si:H detectors, a perspective view on MIP detection and beam flux measurements with these detectors will be given.
... The minimum amount of H necessary to passivate most of the dangling bonds is about 1% atomic. The increase of H content enlarges the bandgap hence reducing the background current of the device, and ∼ 14% is the typical value to obtain a detector grade device Wyrsch and Ballif (2016). The bandgap depends also on the deposition conditions such as the temperature. ...
Article
Full-text available
The characteristics of a hydrogenated amorphous silicon (a-Si:H) detector are presented here for monitoring in space solar flares and the evolution of strong to extreme energetic proton events. The importance and the feasibility to extend the proton measurements up to hundreds of MeV is evaluated. The a-Si:H presents an excellent radiation hardness and finds application in harsh radiation environments for medical purposes, for particle beam characterization and, as we propose here, for space weather science applications. The critical flux detection limits for X rays, electrons and protons are discussed.
... This demonstrates the reversible nature of radiation damage on a-Si:H detectors, and its potential for beam monitoring and tracking. For a given thickness (thin) of material, a-Si:H does show superior radiation hardness compared to materials such as crystalline Si, InP/GaAs/Ge, CIGS, and CdTe [41]. Lower energy protons (405 keV) have a higher possibility of being stopped by the material, and the hydrogen atom becomes embedded inside the detector. ...
Article
Full-text available
Silicon tracking detectors have grown to cover larger surface areas up to hundreds of square meters, and are even taking over other sub-detectors, such as calorimeters. However, further improvements in tracking detector performance are more likely to arise from the ability to make a low mass detector comprised of a high ratio of active sensor to inactive materials, where dead materials include electrical services, cooling, mechanical supports, etc. In addition, the cost and time to build these detectors is currently large. Therefore, advancements in the fundamental technology of tracking detectors may need to look at a more transformative approach that enables extremely large area coverage with minimal dead material and is easier and faster to build. The advancement of thin film fabrication techniques has the potential to revolutionize the next-to-next generation of particle detector experiments. Some thin film deposition techniques have already been developed and widely used in the industry to make LED screens for TVs and monitors. If large area thin film detectors on the order of several square meters can be fabricated with similar performance as current silicon technologies, they could be used in future particle physics experiments. This paper aims to review the key fundamental performance criteria of existing silicon detectors and past research to use thin films and other semi-conductor materials as particle detectors in order to explore the important considerations and challenges to pursue thin film detectors.
... Hydrogenated amorphous silicon thin films (a-Si:H) 35 with different compositions, structures, and dopants may present a broad range of properties depending on the goal. They are, therefore, potential candidates for use in a diverse set of fields, 35 including photovoltaic studies, 36−38 particle detectors, 39 and thin-film transistors (TFTs). 40 Recently, 41 this material was also reported as a stable, potential solid lubricant at the macroscale and under high temperatures (up to 600°C), with results close to superlubricity. ...
Article
Silicon-based materials are widely applied in micro- and nanoscale devices, such as micro- and nano-electromechanical systems (MEMs and NEMs, respectively). However, the nanofriction behavior of such materials is still an issue that influences or hinders some applications. Recently, a sliding-dependent friction mechanism was accessed, by simulation of hydrogenated silicon surfaces, in which friction forces increase during sliding. The experimental evaluation of this phenomenon is still lacking, as well as the confirmation of such a behavior in ambient air conditions, related to the current application. Here, the nanotribology of hydrogenated amorphous silicon (a-Si:H) was experimentally studied under repeated scanning. As an overall analysis, friction increases during sliding, without wear. After separation and new contact, the friction force recovers an intermediate value and increases again, until reaching an intermediate steady state. The mechanism may be related to the successive creation and breaking of bonds at the interface. When the contact is ceased, the dangling bonds created at both surfaces after separation may be repassivated. Moreover, the relationship of this behavior with the photoactivity of the material was tested. If an external light source is added during the scanning, it only changes the results before the stabilization of the interface. This detailed experimental study might promote the broader nanoapplication of the material, addressing the current failures and suggesting new devices based on the anomalous behavior.
... The hydrogenation saturates most of the dangling bonds, lowering the density of the defects to 10 15 cm −3 . The typical amount of hydrogen required to obtain a-Si:H quality for detector applications is in the order of 10% atomic hydrogen [4] The inclusion of hydrogen has the additional effect of increasing the band gap to 1.7-1.9 eV [5]. ...
Article
Full-text available
In this paper, by means of high-resolution photoemission, soft X-ray absorption and atomic force microscopy, we investigate, for the first time, the mechanisms of damaging, induced by neutron source, and recovering (after annealing) of p-i-n detector devices based on hydrogenated amorphous silicon (a-Si:H). This investigation will be performed by mean of high-resolution photoemission, soft X-Ray absorption and atomic force microscopy. Due to dangling bonds, the amorphous silicon is a highly defective material. However, by hydrogenation it is possible to reduce the density of the defect by several orders of magnitude, using hydrogenation and this will allow its usage in radiation detector devices. The investigation of the damage induced by exposure to high energy irradiation and its microscopic origin is fundamental since the amount of defects determine the electronic properties of the a-Si:H. The comparison of the spectroscopic results on bare and irradiated samples shows an increased degree of disorder and a strong reduction of the Si-H bonds after irradiation. After annealing we observe a partial recovering of the Si-H bonds, reducing the disorder in the Si (possibly due to the lowering of the radiation-induced dangling bonds). Moreover, effects in the uppermost coating are also observed by spectroscopies.
... From the data presented in figure 2 it is observed that AZO devices have a lower leakage current compared to the TiO 2 device, and that thinner devices at the same field have a lower current in comparison to their thicker counterparts in agreement with what reported in ref. [3] for doped contacts. The overall value of the leakage current normalized to the area for an 8.2 μm device with AZO selective contact at 5 V/μm bias is in the order of 9.2 nA/cm 2 for sample 1 and 14.8 nA/cm 2 for sample 2, while for the TiO 2 device is 24.4 nA/cm 2 . ...
Article
Hydrogenated Amorphous Silicon (a-Si:H) is a well known material for its intrinsic radiation hardness and is primarily utilized in solar cells as well as for particle detection and dosimetry. Planar p-i-n diode detectors are fabricated entirely by means of intrinsic and doped PECVD of a mixture of Silane (SiH 4 ) and molecular hydrogen. In order to develop 3D detector geometries using a-Si:H, two options for the junction fabrication have been considered: ion implantation and charge selective contacts through atomic layer deposition. In order to test the functionality of the charge selective contact electrodes, planar detectors have been fabricated utilizing this technique. In this paper, we provide a general overview of the 3D fabrication project followed by the results of leakage current measurements and X-ray dosimetric tests performed on planar diodes containing charge selective contacts to investigate the feasibility of the charge selective contact methodology for integration with the proposed 3D detector architectures.
... Hydrogenated amorphous silicon (a-Si:H) is a disordered semiconductor obtained via plasma-enhanced chemical vapor deposition (PECVD) of a mixture of silane (SiH 4 ) and hydrogen at temperatures of 250-300 • C [1]. The resulting material has an irregular arrangement of atoms resulting in not all Si-Si bonds being saturated, leading to the presence of dangling bonds (DBs) that are related to the presence of intermediate states between the valence and the conduction bands. ...
Article
Full-text available
Hydrogenated amorphous silicon (a-Si:H) can be produced by plasma-enhanced chemical vapor deposition (PECVD) of SiH4 (silane) mixed with hydrogen. The resulting material shows outstanding radiation hardness properties and can be deposited on a wide variety of substrates. Devices employing a-Si:H technologies have been used to detect many different kinds of radiation, namely, minimum ionizing particles (MIPs), X-rays, neutrons, and ions, as well as low-energy protons and alphas. However, the detection of MIPs using planar a-Si:H diodes has proven difficult due to their unsatisfactory S/N ratio arising from a combination of high leakage current, high capacitance, and limited charge collection efficiency (50% at best for a 30 µm planar diode). To overcome these limitations, the 3D-SiAm collaboration proposes employing a 3D detector geometry. The use of vertical electrodes allows for a small collection distance to be maintained while preserving a large detector thickness for charge generation. The depletion voltage in this configuration can be kept below 400 V with a consequent reduction in the leakage current. In this paper, following a detailed description of the fabrication process, the results of the tests performed on the planar p-i-n structures made with ion implantation of the dopants and with carrier selective contacts are illustrated.
... Hydrogenated amorphous silicon (a-Si:H) is a disordered semiconductor obtained from PECVD (plasma-enhanced chemical vapour deposition) of a mixture of Silane (SiH4) and Hydrogen at temperatures of 250-300 °C [1]. The resulting material has an irregular arrangement of atoms where not all Si-Si bonds are actually saturated, leading to the presence of dangling bonds (DBs) that are related with the presence of intermediate states between the valence and the conduction bands. ...
Preprint
Full-text available
Hydrogenated amorphous silicon (a-Si:H) can be produced by plasma-enhanced chemical vapour deposition (PECVD) of SiH4 (Silane) mixed with Hydrogen. The resulting material shows outstanding radiation resistance properties and can be deposited on a wide variety of different substrates. These devices have been used to detect many different kinds of radiation namely: MIPs, x-rays, neutrons and ions as well as low energy protons and alphas. However, MIP detection using planar diodes has always been difficult due to the unsatisfactory S/N ratio arising from a combination of high leakage current, high capacitance and a limited charge collection efficiency (50% at best for a 30 µm planar diode). To overcome these limitations the 3D-SiAm collaboration proposes to use a 3D detector geometry. The use of vertical electrodes allows for a small collection distance to be maintained while conserving a large detector thickness for charge generation. The depletion voltage in this configuration can be kept below 400 V with consequent reduction in the leakage current. In this paper, following a detailed description of the fabrication process, the results of the tests performed on the planar p-i-n structures made with ion implantation of the dopants and with carrier selective contacts will be illustrated.
Article
Hydrogenated amorphous silicon is a well-known detector material for its radiation resistance. For this reason it has been used in particle beam flux measurements and in solar panels designed for space applications. This study concern 10 μm thickness, p-i-n and charge selective contacts planar diode detectors which were irradiated with neutrons to two fluence values: 1016 neq/cm2 and 5 × 1016 neq/cm2. In order to evaluate their radiation resistance, detector leakage current and response to x-ray photons have been measured. The effect of annealing for performance recovery at 100 ◦C for 12 and 24 h has also been studied. The results for the 1016 neq/cm2 irradiation show a factor 2 increase in leakage current that is completely recovered after annealing for p-i-n devices while charge selective contacts devices show an overall decrease of the leakage current at the end of the annealing process compared to the measurement before the irradiation. X-ray dosimetric sensitivity degrades, for this fluence, at the end of irradiation, but partially recovers for charge selective contact devices and increases for p-i-n devices at the end of the annealing process. Concerning the 5 × 1016 neq/cm2 irradiation test (for p-i-n structures only), due to the activation that occurred during the irradiation phase, the measurements were taken after 146 days of storage at around 0 ◦C, during this period, a self-annealing effect may have occurred. Therefore, the results after irradiation and storage show a noticeable degradation in leakage current and x-ray sensitivity with a small recovery after annealing.
Chapter
Thin-film solar cell technologies based on large areas are particularly well-suited to applications in the building industry. These applications of BIPV consist mainly of photovoltaic roofs and facades, and generally are grid-connected.
Article
Thin a-Si:H-films deposited by different methods have been irradiated with 20 keV-electrons. The DOS distribution or spin-density, respectively, have been investigated with various methods (CPM, SCLC, ESR). For irradiation doses between 0 and about 60 J/cm² a linear correlation between dose and number of created defects is found, for higher doses the characteristic is increasingly sublinear. The metastability of the defects is confirmed by annealing experiments.
Chapter
This chapter concerns the basic reactions which occur when radiation encounters matter and the effects produced by these processes. For the experimental nuclear or particle physicist, knowledge of these interactions is of paramount importance. Indeed, as will be seen in the following chapters, these processes are the basis of all current particle detection devices and thus determine the sensitivity and efficiency of a detector. At the same time, these same reactions may also interfere with a measurement by disturbing the physical state of the radiation: for example, by causing energy information to be lost, or deflecting the particle from its original path, or absorbing the particle before it can be observed. A knowledge of these reactions and their magnitudes is thus necessary for experimental design and corrections to data. Finally, these are also the processes which occur when living matter is exposed to radiation.
Article
The succesful development of radiation detectors made from amorphous silicon coul offer the possibility for relatively easy construction of large area position-sensitive detectors. We have conducted a series of measurements with prototype detectors, on signals derived from alpha particle. The measurement results are compared with simple model calculations, and projections are made of potential applications in high-energy and nuclear physics.
Article
Amorphous silicon thin films were first deposited by plasma-enhanced chemical vapor deposition (PECVD). The amorphous silicon layers deposited from silane by PECVD could be doped by adding to the plasma discharge either phosphine to form n-type layers, or diborane to form p-type layers. The conductivity of these thin amorphous silicon layers could be increased by several orders of magnitude. Amorphous silicon solar cells at first found only niche applications, especially as the power source for electronic calculators. For 15 years or so, they have been increasingly used for electricity generation: they seem particularly well-suited for wide applications in building-integrated photovoltaics. One of their main advantages is that they are available in the form of monolithically integrated large-area modules. In amorphous silicon thin films, both the bond angles and the bond lengths vary in a random fashion: There is a whole distribution of values. If the amorphous silicon layer has just a low "amount of disorder," then the distributions for bond angles and bond lengths will be very narrow.
Article
This paper will review amorphous silicon imaging technology in terms of the detector operating principles, electrical and optoelectronic characteristics, and stability. Also, issues pertinent to thin film transistor stability will be presented along with optimization of materials and processing conditions for reduced V/sub T/-shift and leakage current. Selected results are shown for X-ray and optical detectors, thin film transistors, and integrated X-ray pixel structures. Extension of the current fabrication processes to low (<100/spl deg/C) temperature, enabling fabrication of thin film electronics on flexible (polymer) substrates, will also be discussed along with preliminary results.
Article
This thesis deals with the development and study of microfluidic scintillation detectors, a technology of recent introduction for the detection of high energy particles. Most of the interest for such devices comes from the use of a liquid scintillator, which entails the possibility of changing the active material in the detector, leading to increased radiation resistance. A first part of the thesis focuses on the work performed in terms of design and modelling studies of novel prototype devices, hinting to new possibilities and applications. In this framework, the simulations performed to validate selected designs and the main technological choices made in view of their fabrication are addressed. The second part of this thesis deals with the microfabrication of several prototype devices. Two different materials were studied for the manufacturing of microfluidic scintillation detectors, namely the SU-8 photosensitive epoxy and monocrystalline silicon. For what concerns the former, an original fabrication approach based on successive bonding and selective release steps of resin layers patterned over sacrificial metal films is detailed. This approach was used to fabricate monolithic, free-standing devices embedding one or two layers of microfluidic channels, with a material budget corresponding to only 0.03% and 0.06% of the radiation length of SU-8. A first experimental validation of these devices is presented as well. Concerning silicon devices, studies on the fabrication of microchannel arrays by both dry and wet etching are reported. Adaptations of these standard techniques to the specific needs of microfluidic scintillation detectors are addressed, specifically the smoothing of scalloped sidewalls resulting from deep reactive ion etching as well as the mask design methodology applied to KOH etching in order to yield microchannels with smooth vertical sidewalls on wafers with the standard crystalline orientation. The anisotropic characteristics of wet etching were also exploited to demonstrate the fabrication of arrays of microfluidic channels having slanted reflective facets at their extremities, which can act as micromirrors that deviate the scintillation light in the out of plane direction, thus introducing new possibilities for the planar integration of the devices. Experimental results on the characterization of the light yield and attenuation length of silicon prototype devices performed using electrons from a radioactive source are presented. A brief study on the accelerated ageing of the detector in which the liquid scintillator was damaged by intense UV irradiation is reported. Such study provides encouraging results on how the capability of recirculating the active material in microfluidic scintillation detectors can be used to extend their lifetime or increase the stability of their performance in time.
Article
The review gives the introductory information on metastable light-induced degradation of amorphous silicon called the Staebler-Wronski effect (SWE). The paper highlights some recent developments in our understanding of the effect as well as some successes in reducing the SWE. The latest results of the author's experiments on the metastable effects and the degradation in thin film transistors and solar cells are also presented.