348 | Nature | Vol 577 | 16 January 2020
Article
and of the 2.05 eV peak to a free-to-bound transition; a detailed discus-
sion may be found in Methods section ‘Notes on Extended Data Fig. 3’.
Despite these defect-related peaks, a strong PL signal from an indirect-
gap material is an indication of good optical quality.
LiInP 2 Se 6 detector devices were fabricated from large CVT-grown
single crystals using evaporated planar gold contacts. Pristine LiInP 2 Se 6
devices exhibit very low dark current, of the order of picoamperes
(Fig. 3a), corresponding to a high electrical resistivity of around
1013 Ω cm. The dark-current plot shows a non-zero current at 0 V,
though the dark current exhibits linearity in this range (Extended Data
Fig. 4a). When larger voltages are applied a diode-like current–voltage
curve is observed, whereas the low dark current is maintained at 2 nA
at 1,000 V (Fig. 3d), which is more than sufficient for effective α- and
neutron-detection measurements. This behaviour might originate from
blocked/injected current from the electrodes under a large electric
field of up to tens of kilovolts per centimetre. When exposed to ambient
light (~0.2 mW cm−2), photogenerated charge carriers in the LiInP 2 Se 6
detector increase the current by three to four orders of magnitude
(Fig. 3a). This is an exceptional photoresponse that indicates favour-
able charge transport properties in LiInP 2 Se 6 , as the photogenerated
carriers efficiently traverse the detector thickness.
As a direct-conversion semiconductor based on lithium, neutrons
(n) are captured in LiInP 2 Se 6 through the following nuclear reaction:
n + ^6 Li = ^4 He + ^3 H (energy Q = 4.78 MeV). Thus, the interaction of α parti-
cles produced by the decay of^241 Am (Q = 5.486 MeV) mimics the energy
deposition of a neutron-capture event at the detector surface.
The pulse-height spectrum of a LiInP 2 Se 6 device with a thickness of
120 μm irradiated by 5.486 MeV^241 Am α particles for electron collection
can be seen in Fig. 3b. Here the channel number reflects the charge
collection efficiency (CCE), which is the ratio of collected and gener-
ated charge from a single event (for example, neutron capture), and
under constant bias can be considered proportional to the energy of
the incident particle. The peak from the interaction of α particles shifts
to higher channel numbers with increasing voltage (Extended Data
Fig. 5), as higher CCE can be achieved with higher applied electric field.
The energy resolution for the full-energy peak at 700 V is 13.9% for the
LiInP 2 Se 6 detector, confirming that LiInP 2 Se 6 can accurately resolve the
charged particles generated by neutron interaction. To our knowledge,
this is the highest energy resolution reported for^241 Am α particles
for any direct-conversion thermal-neutron semiconductor detector,
surpassing the 23.3% resolution^15 of LiInSe 2 or the non-resolution of
the full-energy peak in h-BN due to poor CCE.
These results are reproducible for different growth sizes, and the
α-particle response of other LiInP 2 Se 6 devices with crystals grown under
identical conditions can be seen in Extended Data Fig. 6a–h. For hole
collection in a typical sample, a shoulder, instead of a full-energy peak,
is usually observed, implying relatively poor CCE of holes. Nevertheless,
several devices were able to resolve the^241 Am peak for both holes and
electrons. This is because the high electric field that this material can
withstand (at least 60 kV cm−1) allows the generated carriers to be swept
out to the contacts before trapping or recombination can occur, result-
ing in a spectroscopic response even for the relatively low-mobility
holes. For a constant applied voltage across various samples, the CCE
is higher for thinner samples because of the larger external electric
field. These results conclusively demonstrate that LiInP 2 Se 6 devices can
achieve a high-resolution spectroscopic response to^241 Am α particles.
Isotopically enriched lithium (95%^6 Li) was used to grow LiInP 2 Se 6
crystals (hereafter labelled^6 LiInP 2 Se 6 ) with maximal^6 Li to maximize
the thermal-neutron-capture cross-section. The α-particle response of
the^6 LiInP 2 Se 6 devices is similar to that of their unenriched analogues,
with full resolution of the^241 Am peak for electron collection (Extended
Data Fig. 7a–e). Hole collection of^6 LiInP 2 Se 6 devices also demonstrates
some level of α-particle response.
The thermal-neutron response of the 95%^6 Li-enriched detectors was
tested at room temperature using a moderated and very weak Pu–Be
source, which generated roughly 75 neutrons per second per square
centimetre. The absorption probability for thermalized neutrons ver-
sus the thickness of^6 LiInP 2 Se 6 is shown in Extended Data Fig. 8. The
device had an active area of 7 × 7 mm^2 and a thickness of ~90 μm, and
was subjected to a bias of 300 V. The binned pulse-height spectrum
of a^6 LiInP 2 Se 6 device under neutron irradiation shows a defined peak06 ,000 12,000200400600800200 VCountsChannel number100 V300 V400 V700 V
600 V
500 Vacbd(^10) –100 –500 50 100
–1
100
101
102
103
104
105
Current (pA)
Voltage (V)
Photocurrent
Dark current
0300 600 900 1,200
100
101
102
103
104
(^241) Am
Counts
Channel number
Background
(^57) Co
06 ,000 12,000
0
25
(^50700) V
Counts
Channel number
1 nA
2 nA
0
0
–1 kV 1 kV
700 V
Fig. 3 | Electrical characteristics and pulse-height spectra of LiInP 2 Se 6
devices irradiated by α-particle and γ-ray sources. a, Dark current and
photocurrent (absolute values) as a function of voltage (from −100 V to 100 V)
for CVT-grown LiInP 2 Se 6 (~0.3 mm thickness) with gold electrodes illuminated
by ambient light (~0.2 mW cm−2). b, Pulse-height spectra of α particles from
(^241) Am for a LiInP 2 Se 6 device (~120 μm thickness) for electron collection at
various voltages, collected for 60 s each. c, Pulse-height spectrum of a
LiInP 2 Se 6 detector (~0.5 mm thickness) under a bias of 700 V illuminated by
γ-rays from^241 Am and^57 Co, and without a source. The counts for^57 Co γ-rays and
for the natural background are negligible, as shown. The spectra were collected
for 200 s each. d, Pulse-height spectrum at 700 V with the same conditions as in
b. Inset, Dark current measured with the voltage varying from 0 to 1 kV, to −1 kV,
and to 0 V under the same conditions as in a. The contact area on the crystals in
a–d was 3 × 3 mm^2. The shaping time in b–d was 1 μs.
b
1 inch
a
0326496 128
0
20
40
60
80
Counts
Channel number
Unshielded strong source
Cd-shielded strong source
Unshielded weak source
0.95 s–1 cm–2
0.20 s–1 cm–2
0.12 s–1 cm–2
Fig. 4 | Thermal-neutron spectra resolved by^6 LiInP 2 Se 6 devices. a, Binned
pulse-height spectrum of a^6 LiInP 2 Se 6 device (shown in b) exposed to a
moderated Pu–Be source without shielding (red) and with Cd shielding
(0.125 inch thick; black) and to a source ~5 times weaker (grey) for 300 V bias,
3 μs shaping time and 30-min measurements. The counts from channel 16 and
above correspond to neutron-capture events. The count rate for the
unshielded measurement was 0.95 s−1 cm−2. b, The 7 × 7 mm2 6 LiInP 2 Se 6 device
(90 μm thickness) used for the measurements shown in a.