Full Paper

www.afm-journal.de

Liquid Phase Exfoliated Indium Selenide Based Highly

1. Introduction
Layered semiconductors of the IIIA–VIA group have attracted considerable
attention in (opto)electronic applications thanks to their atomically thin 2D semiconductor materials, including
layered transition metal dichalcogenides
structures and their thickness-dependent optical and electronic properties,
(TMDs) with formula MX2 (M = Mo, W;
which promise ultrafast response and high sensitivity. In particular, 2D X = S, Se, Te), have been intensely studied
indium selenide (InSe) has emerged as a promising candidate for the as substitutes for conventional semicon-
realization of thin-film field effect transistors and phototransistors due to its ductors in opto-electronic applications.[1,2]
high intrinsic mobility (>102 cm2 V−1 s−1) and the direct optical transitions in The β polymorph of indium selenide
(β-InSe) is a promising van der Waals
an energy range suitable for visible and near-infrared light detection. A key
layered material due to its remarkable
requirement for the exploitation of large-scale (opto)electronic applications fundamental properties,[3] such as a low
relies on the development of low-cost and industrially relevant 2D material electron effective mass (m* = 0.143 m0)[4]
production processes, such as liquid phase exfoliation, combined with and a bandgap ranging from ≈1.26 eV
the availability of high-throughput device fabrication methods. Here, a β for the bulk to ≈2.11 eV for the mono­
polymorph of indium selenide (β-InSe) is exfoliated in isopropanol and layer at room temperature,[5,6] crossing
over from an indirect to a direct semi-
spray-coated InSe-based photodetectors are demonstrated, exhibiting
conductor when the crystal thickness is
high responsivity to visible light (maximum value of 274 A W−1 under larger than six layers (≈5 nm, each layer
blue excitation 455 nm) and fast response time (15 ms). The devices is ≈0.84 nm thick).[7–9] The fundamental
show a gate-dependent conduction with an n-channel transistor behavior. properties of β-InSe indicate the poten-
Overall, this study establishes that liquid phase exfoliated β-InSe is a valid tial use of few-layer flakes in field effect
transistors (FETs) and sensitive electric
candidate for printed high-performance photodetectors, which is critical
light detectors thanks to a direct bandgap,
for the development of industrial-scale 2D material-based optoelectronic which is tunable by the crystal thickness
devices. in a wide spectral range from visible to

N. Curreli, Dr. M. Serri, Dr. E. Lago,[++] E. Petroni,[++] Dr. B. Martín-García, Dr. D. Spirito,[+] Dr. R. Krahne
Dr. A. Politano,[+++] Dr. V. Pellegrini, Dr. F. Bonaccorso Nanochemistry Department
Graphene Labs Istituto Italiano di Tecnologia
Istituto Italiano di Tecnologia Via Morego 30 ,16163 Genova, Italy
via Morego 30, 16163 Genova, Italy Dr. E. Lago, E. Petroni
E-mail: [email protected] Dipartimento di Chimica e Chimica Industriale
N. Curreli Università degli Studi di Genova
Dipartimento di Ingegneria Elettrica ed Elettronica via Dodecaneso 31, 16146 Genoa, Italy
Università di Cagliari Dr. B. Gürbulak
P.zza d’Armi, 09123 Cagliari, Italy Department of Physics
The ORCID identification number(s) for the author(s) of this article Faculty of Sciences
can be found under https://doi.org/10.1002/adfm.201908427. Atatürk University
25240 Erzurum, Turkey
[+]Present address: IHP—Leibniz-Institut für innovative Mikroelektronik, Prof. S. Duman
Im Technologiepark 25, 15236 Frankfurt (Oder), Germany Department of Basic Sciences
[++]Present address: STMicroelectronics, SMART POWER Technology
Faculty of Sciences
R&D, Via C. Olivetti 2, 20864 Agrate Brianza, Italy Erzurum Technical University
[+++]Present address: Dipartimento di Scienze Fisiche e Chimiche, 25050 Erzurum, Turkey
Università degli Studi dell'Aquila, Via Vetoio, 42, 67100 Coppito, Italy Dr. V. Pellegrini, Dr. F. Bonaccorso
BeDimensional Spa
DOI: 10.1002/adfm.201908427 via Albisola 121, 16163 Genova, Italy

Adv. Funct. Mater. 2020, 1908427 1908427 (1 of 10) © 2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.afm-journal.de

near-infrared (NIR) light.[5,10] Several groups reported n-type of costs and scalability.[41] Although, the compatibility of the
FET based on single layer and thin InSe crystals with values inks with high-throughput printing processes and the level of
of electron mobility of µe ≈ 103 cm2 V−1 s−1.[11,12] Photodetec- performance of InSe printed devices still needs to be assessed,
tors with a photoresponsivity Rph ≈ 5 × 107 AW−1 in single flake LPE might represent a viable approach to exploit InSe few-
devices,[8,13–18] surpassing TMDs[19,20] and rivalling other high layers for optoelectronic applications.
mobility (≥102 cm2 V−1 s−1) direct bandgap 2D semiconductors We remark that LPE allows the exfoliation of layered
such as black phosphorus (BP)[21,22] have also been demon- materials in a liquid medium in ambient conditions,[42]
strated. Furthermore, a lower enthalpy of oxygen chemisorp- applying an external stimulus such as ultra-sonication,[42,43]
tion on monolayers of InSe compared to single layer BP makes high-shear mixing,[44,45] ball milling,[46,47] wet jet milling,[48]
it more environmentally stable compared to the latter.[23] and microfluidization.[49] For the efficient exfoliation of layered
The high responsivity of β-InSe photodetectors has been materials, the solvent should prevent restacking and aggrega-
explained by the direct bandgap of thick (>5nm) flakes, which tion of the exfoliated layers by satisfying specific surface energy
results in efficient light absorption in the material.[24] Extrinsic requirements for the considered material, which are summa-
factors due to the photodetector design (e.g., short channel rized by the solvent surface tension and the Hansen solubility
length) may raise the responsivity as well.[18,25] Moreover, the parameters.[50,51] Usually, the dispersions produced by LPE
presence of photoconductive gain mechanisms in β-InSe also have a heterogeneous composition of exfoliated flakes both
contributes to the high responsivity, multiplying the number in lateral size and thickness,[43] which can be fractioned, for
of electrons flowing in the device per absorbed photon.[26] In example, by using sedimentation based separation (SBS).[52–55]
particular, the presence of traps that reduce the mobility of The as-produced flakes are mostly basal plane defect-free and
minority carriers (holes) in the material, results in the circu- unfunctionalized.[41,43,53–55] In addition, LPE is the ideal tool for
lation of photogenerated electrons multiple times through the the formulation of inks[50,53,56] and pastes,[57,58] as well as for the
device before recombination.[19] Unfortunately, the increased integration of the exfoliated flakes in composites,[59] and coat-
responsivity gain comes at the expense of a slower response ings.[60,61] This makes dispersions of 2D materials appealing for
time of the photodetector,[27] which might be problematic for the realization of fully printed devices.[50,53,56]
applications that require a high bandwidth, such as imaging[28] From the production point of view, the choice of the sol-
or optical communication.[29] Furthermore, trap-assisted recom- vent is a critical issue of the LPE process. In fact, common
bination results in sublinear dependence of the responsivity solvents used for the formulation of 2D material based inks (e.g.,
as a function of the light intensity, limiting the sensitivity at N-methyl-2-pyrrolidone—NMP, dimethylformamide—DMF,
high optical power and complicating the readout of the photo­ N-cyclohexyl-2-pyrrolidone—CHP), present serious health haz-
detector signal.[30] Therefore, the study of the charge trapping ards (e.g., NMP, DMF, CHP, health code ≥ 2 NFPA704).[62] Fur-
and recombination mechanisms in InSe could provide insights thermore, the low vapor pressure (Pv ≈ 0.05 kPa)[63] and high
for the optimization of the semiconducting material and the boiling point (Tb ≈ 200 °C)[64] of the aforementioned solvents
device design for different application requirements. result in surface contamination of the exfoliated flakes with an
Although promising figures of merit, FoM, (µe, Rph) insulating organic layer, which is detrimental for the electronic
have been reported for InSe devices, the aforementioned conduction of the exfoliated materials.[18,65,66] Recently, LPE of
results were obtained using single-flake devices, produced by 2D materials in low boiling point solvents such as acetone,[67]
low-throughput methods of fabrication, such as mechanical alcohols,[50] and water/alcohol co-solvent systems[18,50,68–70] has
exfoliation[14] of bulk crystals followed by manual assembly[31] of been investigated as a strategy to reduce the use of toxic chemi-
heterostructures. The latter requires the identification of suit- cals, minimizing the contamination by solvent residuals in
able flakes and a precise alignment of the electrodes by electron films of LPE flakes.
beam lithography (EBL).[18] Thus, to enable the adoption of this The LPE exfoliation of InSe has been achieved by ultra-son-
material in consumer electronics, scalable and sustainable pro- ication by using isopropyl alcohol (IPA)[69] and water-ethanol
cesses for both material production and device fabrication are (H2O/EtOH) mixtures,[18] allowing an effective removal of the
needed. To this end we recall that chemical vapor deposition solvent by heating at moderate temperatures (e.g., ≈50 °C)
of InSe films with mostly monolayer thickness was achieved thanks to their low boiling point (78.2 °C < Tb < 100 °C).[62] In
recently over areas of 1 cm2, enabling the fabrication of n-type particular, the Hersam’s group demonstrated that InSe flakes
FETs with electron mobility up to ≈30 cm2 V−1 s−1.[32–34] Wafer (average thickness of 41 nm) exfoliated by ultra-sonication in
scale growth of InSe films by pulsed laser deposition has also deoxygenated H2O/EtOH mixtures exhibit electron mobilities
been demonstrated, producing films with an electron mobility µe ≈ 19 cm2 V−1 s−1 and photoresponsivity Rph ≈ 5 × 107 A W−1
of 10 cm2 V−1 s−1 and a response of 27 A W−1 in phototransistor in single flake phototransistors.[18] On the other hand, devices
devices.[35] These techniques require a fine tuning of the growth produced with InSe flakes exfoliated in high boiling point
parameters (e.g., temperature, growth duration),[36] expensive solvents (i.e., NMP, DMF) have shown lower performance,
equipment,[37] and high temperature (>600 °C)[32,35] conditions, compared with the one produced in H2O/EtOH mixtures,
which increase the energy requirements for the device fabri- due to organic contaminants.[18] The same study reported a
cation. Therefore, they still do not represent the solution for responsivity of Rph = 10 AW−1 in devices fabricated from InSe
consumer electronics application. based films obtained by membrane filtration of the disper-
Liquid phase exfoliation (LPE)[38–40] is another process that sions in H2O/EtOH, which resulted in percolating networks of
was successfully exploited to produce few layer InSe crystals, flakes manifesting a deterioration of the photodetector perfor-
offering advantages over the aforementioned methods in terms mance compared to the single flake.[18] This phenomenon is a

Adv. Funct. Mater. 2020, 1908427 1908427 (2 of 10) © 2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.afm-journal.de

consequence of the high contact resistance between flakes com- few layer β-InSe flakes with no defects, the efficacy of the exfo-
pared to the intrinsic resistance of the flakes themselves and liation of the crystal was evaluated by morphological, chemical,
is exacerbated in the case of percolating films due to electrical and structural analyses. Raman spectroscopy measurements,
conduction occurring through an increased number of inter- shown in Figure 1a, were performed in order to demonstrate
flake boundaries. Although these results highlight the potential the absence of other chemical species, (e.g., In2Se3, In2O3)[75,77]
of LPE InSe based optoelectronics from a fundamental point of that could have been formed during the exfoliation process or
view, the feasibility of scalable and repeatable processes used in subsequent exposure to atmosphere. β-InSe has six Raman
industrial production has not yet been investigated. active vibrational modes, which are assigned to the peaks in the
In this work, we produce β-InSe inks by ultrasound-assisted spectra in Figure 1a[74,77] in both the bulk and exfoliated β-InSe
LPE in IPA and use spray coating deposition of the inks on samples. The data confirms the absence of polymorph crystals.
silicon substrates to fabricate highly responsive photodetectors. Peaks related to other In and Se compounds or oxidized phases
Spray coating is an industrially compatible process that can be are not present, suggesting that the exfoliated flakes keep their
optimized to produce highly uniform films of electronically crystalline integrity.[75,77] These conclusions are also supported by
active materials and has already been applied to the fabrication X-ray photoelectron spectroscopy (XPS) spectra of the In 3d and
of perovskite based solar cells.[71,72] Our results show that the Se 3d regions (Figure S5, Supporting Information), showing only
photodetectors exhibit high responsivities to light in a broad a 0.2 eV broadening of the peaks in the exfoliated material, which
spectral range (450–900 nm), with a maximum responsivity is indicative of a slight increase in defects and oxide species com-
Rph = 274 AW−1 at 0.53 mW cm−2 irradiance and a rise time pared to the bulk, and the XRD patterns shown in Figure 1b.
of ≈15 ms. Our photodetector outperforms, in term of respon- These data are in agreement with the diffraction of β-InSe
sivity, previous devices based on percolating networks of solu- (ICDD 98-018-5172). The XRD patterns demonstrate, both for
tion processed 2D flakes[18] by more than one order of magni- bulk and for exfoliated samples, the occurrence of a hexagonal
tude. By studying the optical power dependence of the respon- β-InSe structure with lattice parameters of a = b = 4.005 ± 0.004 Å
sivity and the time-dependent profiles of the photocurrent and c = 16.660 ± 0.004 Å.[69] Moreover, the β-InSe bulk crystals
under light modulation, we conclude that the recombination exhibit only reflections belonging to the (001) family, indicating
rate of the photocarriers is a cubic function of the concentration strong texture of crystalline flakes along the c axis. In the exfoli-
of charge carrier density. We suggest that the third order rate ated β-InSe flakes, we notice also reflections from other orien-
dependence could be related to a recombination mechanism tations (e.g., 010, 110, and 011), due to disordered arrangement
assisted by doubly charged Se vacancies. of the nanocrystals, although a strong preferential orientation
of the (001) plane is still present. Figure 1c,d reports repre-
sentative transmission electron microscopy (TEM) and atomic
2. Results and Discussion force microscopy (AFM) images of the exfoliated β-InSe flakes,
respectively. Statistical analysis using a log-normal distribution[78]
2.1. Material Production and Characterization (Figure 1e) indicates that the exfoliated β-InSe flakes have a typical
lateral dimension (mode) of ≈113 nm (σ = 0.84), corresponding
The synthesis of β-InSe monocrystals was performed with to an average surface area of ​​≈6.1 × 10−3 µm2, and a thickness
the Bridgman–Stockbarger method.[73] InSe can exist in three of ≈4 nm (σ = 0.54) as measured by AFM in Figure 1f. There-
crystal polymorphs, β, γ, and ε, that differ only in the layers fore, based on the thickness of monolayer β-InSe (≈0.9 nm),[79,80]
stacking.[3,74,75] The crystal structure obtained by this method flakes with single/few layers are effectively produced.
has the β form, as determined by X-ray diffraction (XRD).[73]
Based on the study of the LPE of β-InSe in IPA reported by
Petroni et al.,[69] we performed sonication and ultracentrifu- 2.2. Phototransistor Fabrication and Characterization
gation (see Section 4) in order to achieve a good compromise
between ink stability, the lateral size, and thickness of the We investigated the application of β-InSe based inks in printed
flakes. The concentration of the exfoliated β-InSe flakes was photodetectors using a spray coating deposition method. The
evaluated by means of optical extinction measurements. The inks exhibit low viscosity (η25° ≈ 3.2 mPa × s at 25 °C), with
concentration of the obtained dispersion is 0.11 g L−1, as esti- Newtonian behavior at shear rates between 0.2 and 300 s−1
mated by the Beer–Lambert law (Figure S1, Supporting Information), which makes them suit-
able for spray coating deposition.[81] An interdigitated gold
A electrode array was fabricated using EBL and lift-off (see
αc = (1)
l Section 4) on highly p-doped (boron) silicon (100) with 100 nm
thermal oxide (SiO2) on the surface. The interdigitated design
A allows to achieve a low electrical resistivity, by reducing the
where is the absorbance per length, c is the concentration
l gap between electrodes L and increasing the total width of the
and α is the extinction coefficient with α ≈ 580 L g−1 m−1 at conduction channel W, in a more compact format compared to
600 nm.[69] As reported in ref. [69], a typical extinction spec- linear electrodes, by using a parallel array of finger electrodes.
trum of the β-InSe dispersion (Figure S4, Supporting Infor- Additionally, the reduction of L leads to a faster response of the
mation) exhibits features at ≈275 nm and at ≈360 nm, which photo­detector[82] by reducing the transit time of the carriers.
correspond to direct electronic transitions between valence The electrodes used in our devices consist of 65 interdigitated
and conduction band.[76] In order to confirm the production of fingers with an overlap of 40 µm and a gap L = 1 µm, which

Adv. Funct. Mater. 2020, 1908427 1908427 (3 of 10) © 2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.afm-journal.de

Figure 1. Morphological, chemical, and structural characterization of β-InSe flakes. a) Raman spectra for bulk β-InSe (blue trace) and exfoliated β-InSe
(red trace) samples. b) XRD diffractogram for bulk β-InSe (blue trace) and exfoliated β-InSe (red trace) samples. c) Representative TEM image of an
isolated β-InSe flake. d) Representative AFM image of an isolated β-InSe flake. Height profile (solid white line) of the indicated section (white dashed
line) is also shown. e) Lateral size statistical analysis for β-InSe flake dispersion. f) Statistical analysis of the thickness of β-InSe flakes in the dispersion.

results in an overall channel width W = 5160 µm, providing deposited by e-beam evaporation on the same substrate to
an active area of W × L = 5760 µm2 for photodetection. During exploit the large area deposition capability of the spray coating
the same lithographic process, several Au electrode arrays were process. A volume of ≈30 mL of the as-produced β-InSe ink in

Adv. Funct. Mater. 2020, 1908427 1908427 (4 of 10) © 2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.afm-journal.de

Figure 2. a) Schematic illustration of the β-InSe photodetector. b) Optical microscope image of the complete photodetector and c) magnified detail
of the conducting channel region. d) SEM image of the β-InSe film in the conducting channel region of the device. e) Dark source–drain current (ID)
versus source–drain voltage (VD), at bottom gate voltages between −40 and 40 V. In the inset, a log–log plot of ID versus VD (VG = 0 V) is depicted,
showing the transition from ohmic to space charge limited conduction. f) Source–drain current (ID) versus gate voltage (VG) curve at VD = 2 V, obtained
from the data in (e).

IPA was deposited by spray-coating onto the patterned substrate, to avoid the influence of moisture and atmospheric gases. The
which was heated to 60 °C to favor film drying during the dep- doped silicon substrate was electrically contacted to apply a gate
osition process. The coated sample was placed under vacuum bias VG. As shown in Figure 2e, the drain-source current (ID)
overnight at room temperature and then annealed at 200 °C varies with drain voltage VD under different VG (from −40 to
for 30 min in argon atmosphere to stabilize the deposited film 40 V) in dark condition, displaying a gate-tuneable behavior,
and remove residues of adsorbed solvent and moisture. The indicating that the electrical characteristics of β-InSe device can
film thickness was 515 ± 107 nm, as measured by profilometry be effectively controlled by electrostatic doping. In the inset to
(Figure S2, Supporting Information). The basic structure of the Figure 2e, a quasi-linear regime (slope of the curve r = 1.2) for
β-InSe photo­detector is illustrated by the scheme in Figure 2a 0 < VD < 0.5 V suggests ohmic contacts between the gold elec-
and optical microscope images in Figure 2b,c. trode and β-InSe. At larger bias with 0.5 < VD < 2 V, a quad-
Figure 2d shows the top-view scanning electron micros- ratic dependence of the drain current is observed, which can be
copy (SEM) image of the β-InSe film deposited on the elec- explained by a space-charge limited current (SCLC) regime.[83]
trodes, presenting a percolating network of flakes of lateral size SCLC becomes dominant when the injected electron concentra-
<500 nm. Energy-dispersive X-ray (EDX) spectroscopy (Figure S3, tion exceeds that of the thermally generated one. Charge-carrier
Supporting Information) was performed during SEM imaging, traps within the semiconducting film, located, for example,
confirming the presence of β-InSe flakes in the film. From the in the partially oxidized surface of the flakes, are effective in
EDX analysis on the device, we determined a ratio of 1:0.87 immobilizing most of the injected electrons.[84,85] For shallow
between In and Se, in agreement with previous characteriza- traps located at an energy Et below the conduction band edge,
tion of the exfoliated β-InSe performed by our group.[69] The the SCLC current density is given by a quadratic relation with
elemental composition, within the experimental uncertainty of respect to the voltage (I D ∝ VD2).[83] Figure 2f shows the ID–VG
EDX, indicates the presence of selenium vacancies in the semi- transfer curves in the dark at a drain bias voltage of 2 V. The
conductor film.[69] The signal from oxygen can be explained device exhibits an ambipolar behavior with a minimum ID cur-
mostly by the SiO2 layer on the surface of the substrate, rent at VG ≈ −10 V and a distinct asymmetry in electron and hole
although a superficial oxidation of the β-InSe flakes is also conduction, showing an average mobility of the film (defined in
possible due to the processing in air. the linear regions of the ID–VG curve) of 3 × 10−5 cm2 V−1 s−1
The current–voltage (ID–VD) curves of the β-InSe photode- for the electrons and 1 × 10−5 cm2 V−1 s−1 for the holes. We
tector were measured at room temperature and under vacuum attribute the decrease in mobility compared to single flake

Adv. Funct. Mater. 2020, 1908427 1908427 (5 of 10) © 2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.afm-journal.de

Figure 3. a) Photocurrent response for varying incident light power on the active area of the device measured at three different wavelengths. The
dashed lines represent a power-law fit to the data. The inset plots the photoresponsivity Rph versus the irradiance Ir. b) Spectral responsivity of β-InSe
photodetector. The dashed line is a linear fit of the tail of the spectrum; the onset of the photoresponse at ≈900 nm (1.38 eV) is highlighted. The inset
plots the response to 0.1 Hz pulsed light (530 nm) c) Detail of photocurrent rise and decay during pulsed (2 Hz) 625 nm light excitation. d) Plot of
the time derivative of the photocurrent versus the photocurrent after switching off the light. Inset: Responsivity versus light chopping frequency under
illumination at 625 nm. Dashed lines indicate the power laws f−0.5 and f−0.2.

values to the presence of traps and the surface oxidation during (in which Ir is the irradiance, S is the active area of the device),
spray deposition processing in air, which increases the con- is decreasing as the power increases. The maximum measured
tact resistance between the flakes.[86] The transistor is n-doped responsivity at 455 nm is ≈274 AW−1, at ≈0.53 mW cm−2 irradi-
at VG = 0 V and, for this reason, no off-state condition can be ance. The device maintained its functionality even in ambient
measured. This phenomenon could be explained by the pres- conditions and after 1 year (Figure S7, Supporting Informa-
ence of the defect states in the gap of β-InSe. The n-doping tion). To assess the minimum detectable light intensity, we
in InSe nanoflakes was also confirmed by the position of the calculated the detectivity, defined as[87]
Fermi level compared to the valence and conduction bands,
as determined through ultraviolet photoelectron spectroscopy S
(UPS) (Figure S6, Supporting Information). D* = (3)
NEP
In order to reveal the photodetection behavior of the device,
the electrical characteristics were measured under the exposure Si
In which NEP = is the noise equivalent power and Si is
to light of different wavelength and intensity. The photodetector Rph
response is measured as photocurrent (Iph), corresponding the spectral density of the noise current. We obtained a max-
to the difference between the current measured (at VD = 2 V, imum value (corresponding to the maximum responsivity) of
VG = 0 V) under light and in the dark, for light intensities (Ir) NEP ≈1.31 × 10−15 W Hz−0.5 and D* ≈ 5.49 × 1012 Jones. These
between 0.5 and 7.5 mW cm−2. Figure 3a reports the photo- values are obtained estimating Si by the sum of two noise
current as a function of the light power on the device, using mechanisms: the thermal noise Sc = (4kT/Rdark)0.5, calculated
three different LEDs with emission peaks at 455, 530, and considering the dark resistance Rdark of the device, and the shot
625 nm (FWHM = 17, 42, 20 nm, respectively). The depend- noise, Ss = (2eIdark)0.5, from the dark current Idark.
ence on light power is sublinear; therefore, the photorespon- These FoM values compare favorably with those reported in
sivity (Rph, inset of Figure 3a), defined as literature.[88–91] The responsivity is at least one order of mag-
nitude higher than those of photodetectors based on solution
processed 2D crystals, and InSe in particular.[18] There are
I ph several factors that can cause the increased responsivity of
R ph = (2)
Ir ⋅ S our device. The short length of the channel (L) increases the

Adv. Funct. Mater. 2020, 1908427 1908427 (6 of 10) © 2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.afm-journal.de

photoresponsivity of the device, since Rph is proportional to Here, F is the generation rate due to incoming light, and Rr
L−2.[25] Moreover, the highly uniform coverage of β-InSe on is a constant controlling the recombination rate. This depend-
the Si/SiO2 substrate over the active area could promote the ence is further confirmed by the photocurrent vs light power
extraction efficiency of the photogenerated carriers as already dependence (Figure 3a). Assuming that steady state is reached
suggested in literature for perovskite-,[92] thin-film CdTe[93] and for these measurements (dn/dt = 0), by solving Equation (5) we
graphene based devices.[94] find a dependence as Iph ∝ n ∝ Fϑ, with ϑ = 1/3. In fact, the fits
The spectral response of the device is shown in Figure 3b shown in Figure 3a yield ϑ625 = 0.324 ± 0.003, ϑ530 = 0.318 ±
from the visible to the NIR range, measured with mechanically 0.004, ϑ455 = 0.293 ± 0.013, confirming this prediction.
chopped light at 23 Hz (see Section 4) at bias voltages VD = 2 V The observed power-law dependence is related to the
and gate voltage VG = 0 V. The responsivity is high at short wave- dynamics of the traps and recombination centers that enable
length (≈400 nm) and decreases towards longer wavelength with the photoconduction in nanostructured materials. A nonlinear
a pronounced drop at ≈800 nm. If the wavelength dependence dependence of the photocurrent on the light intensity can arise
is dominated by the absorption coefficient, it is possible to esti- from the distribution of traps and recombination centers within
mate the bandgap from the intercept with the wavelength axis the band gap[95,96] and the saturation of these states under
of a linear fit of the spectral responsivity in the region near the strong light excitation,[13,18,97–99] or even from the presence of
photoresponse threshold wavelength (Figure 3b). The resulting a space charge region.[100,101] The particular case of the third-
value of ≈900 ± 65 nm corresponds to a bandgap of ≈1.4 ± 0.1 eV, order equation, as suggested in ref. [30], can be related to the
which is in the range between the values reported for single recombination assisted by doubly charged Se vacancies, which
layer and bulk β-InSe (2.11–1.26 eV respectively), confirming have been observed in our sample by EDX analysis and by pre-
that the film is composed by few layer flakes. These results are vious studies on β-InSe.[69,84,85]
also supported by the observation of a broad photoluminescence We remark that doubly charged vacancies in CdS have been
(PL) peak at 959 ± 50 nm (Figure S8, Supporting Information). related to more complex processes involving not only charge
Finally, in Figures 3b (inset) and 3c we report the time response excitation, but also charge-lattice coupling that lead to a ther-
of the device, obtained by measuring the current during pulsed mally activated recombination behavior,[30] suggesting that fur-
light excitation. Excellent long-term stability is observed with a ther studies of these properties can lead to interesting physics
fast switching and reproducible transition from “ON” to “OFF” of the material.
conditions. The measured rise (fall) time (Figure 3c), defined
as the time to pass from 10% (90%) to 90% (10%) of the max-
imum photocurrent, is τR = 15.1 ms (τF = 63.7 ms), while other 3. Conclusion
photo­detectors made with films of LPE flakes of β-InSe reported
response times of the order of seconds.[10,18] In the inset to We demonstrated that spray-coated liquid phase exfoliated
Figure 3d, we report the dependence of the photocurrent ampli- β-InSe photodetectors have good photosensitive responsivity
tude on the modulation frequency f. The amplitude decreases to visible and NIR (<900 nm) light and can be used as semi-
with the increase of the frequency, and the signal is measurable conductor channel material in phototransistors, obtaining
up to 1 kHz; the functional dependence of the frequency does a high-throughput and low-cost method to implement β-
not follow a simple low-pass behavior but exhibits richer fea- InSe photodetectors with high sensitivity (maximum D*
tures with complex power-law trends that follow a f−0.5 and f−0.2 ≈5.49 1012 Jones, Rph ≈ 274 A W−1 at 455 nm illumination) and
dependence at low and high modulation frequency respectively, fast response time (τR ≈ 15 ms, τF 64 ms). These FoM represent
as shown by dashed lines in the panel. Such behavior could indi- an improvement of over one order of magnitude with respect
cate an energy distribution of trap states in the material. to reported photodetectors based on percolating networks of
The time decay of photocurrent in Figure 3c can be investi- solution processed 2D flakes[18] and are competitive with the
gated more in detail to understand the mechanism of charge values reported for transition metal dichalcogenide single
recombination in our device. In general, the time derivative of flake devices.[91] The use of spray coating and environmentally
the photoexcited charge density n, after turning off the light, is friendly solvents in the fabrication process makes this material
a function of n itself. an interesting alternative for photodetectors in the visible and
NIR spectral range, also for applications on flexible substrates.
dn The gate tuneable transport and the cubic photo-charge recom-
= g (n ) (4)
dt bination law behavior observed in β-InSe nanoflakes films
prompt further fundamental studies of this mesoscopic system.
Assuming that the photocurrent is proportional to n, the
function g can be determined from the time dependence of Iph
during the decay measurement. In Figure 3d we plot the photo- 4. Experimental Section
current derivative (dIph(t)/dt) versus Iph after light was switched
off. A third-order polynomial yields an excellent fit to the data Exfoliation of Bulk β-InSe: β-InSe single crystals were grown by the
modified Bridgman–Stockbarger method, as described elsewhere.[73] A
(dashed line in Figure 3d), suggesting that the differential equa-
quantity of 40 mg of β-InSe was exfoliated in 20 mL IPA (ACS Reagent,
tion controlling generation–recombination under light is ≥99.8%, Sigma-Aldrich) through LPE[42] in a sonic bath (6 h, 25–35 °C,
Branson 5800 cleaner, Branson Ultrasonics), followed by SBS[102]
dn (1000 × g for 30 min at 15 °C, Beckman Coulter Optima XE-90) and
= F − Rr n 3 (5)
dt collection of the supernatant (≈80%), as described in ref. [69].

Adv. Funct. Mater. 2020, 1908427 1908427 (7 of 10) © 2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.afm-journal.de

Material Characterization: The extinction spectra of the exfoliated InSe Source (S) and drain (D) contacts were patterned by EBL with
flake dispersions in IPA diluted 1:10 was measured using a Cary Varian poly(methyl methacrylate) (PMMA) resist, metal e-beam evaporation
5000 UV–vis, subtracting the extinction of pure solvent. (5 nm Ti/30 nm Au) using magnets to not expose the PMMA, followed
TEM analysis was carried out on a JEOL JEM 1400Plus microscope by lift-off. The device was completed by spraying the entire surface of the
(120 kV), with a LaB6 source, a Gatan CCD camera Orius 830. Samples chip with ≈30 mL of the β-InSe ink in IPA, while heating the substrate
were prepared by drop-casting the flake dispersions onto carbon-coated to 60 °C in air. The coated sample was placed under vacuum overnight
Cu grids and dried overnight in vacuum. at room temperature and then annealed at 200 °C for 30 min in argon.
The XPS analysis was performed on a Kratos Axis UltraDLD Device Characterization: Electrical characterization was performed at
spectrometer at a vacuum better than 10−8 mbar, using a monochromatic room temperature in vacuum, using a shielded probe station equipped
Al Kα source operating at 20 mA and 15 kV and collecting photoelectrons with a Keithley 2612 source meter with two channels to control both
from a 300 × 700 µm2 sample area. Wide spectra were acquired at pass the drain and the gate bias potential with respect to the source terminal
energy of 160 eV and energy step of 1 eV, while high-resolution spectra of the device. To keep its stability and reduce the effect from surface
were acquired at pass energy of 10 eV and energy step of 0.1 eV. The adsorbates, the device was measured under vacuum (≈10−5 mbar). The
samples were prepared by drop-casting the dispersion of InSe flakes backside of the doped silicon substrate was electrically contacted to a
onto Au-coated Si chip in N2 atmosphere, while heating the substrate gold pad with silver paint to work as the gate electrode. The measurement
to 60 °C. Bulk InSe crystals were sticked onto conductive carbon tape procedure was controlled by a PC using a program witten in LabVIEW
and cleaved prior analysis. The samples were transferred to the XPS (National Instruments). For the photoresponse measurements, LEDs with
chamber in inert atmosphere. Data analysis was carried out with nominal wavelength of 455, 530, and 625 nm (M455L3, M530L3, M625L3
CasaXPS software (version 2.3.19PR1.0). The energy scale was calibrated Thorlabs) were used. Light was focused on the device with a collimating
by setting the C 1s peak at 284.8 eV. lens. The LED output power was controlled by the power supply (LEDD1B
UPS was performed with He I (hν = 21.2 eV) radiation to obtain the Thorlabs). Before measuring the device, the irradiance at the center of the
workfunction and the position of the valence band maximum of the focused light spot was determined with the following procedure: the light
materials under investigation. The experiments were carried out on the was focused on a pinhole (⌀ = 3.37 mm) and the power of the transmitted
samples after the XPS analysis using the same equipment. A −9.0 V bias light was measured with a silicon photodiode (S120VC, Thorlabs)
was applied to the sample in order to determine the low kinetic energy connected to a power meter (PM100D, Thorlabs), allowing the calculation
cut-off. of the irradiance at the center of the spot by dividing the power by the area
AFM images of the samples were acquired using a Nanowizard of the pinhole. A 10% loss due to the reflection of the optical window of
III (JPK Instruments, Germany) mounted on an Axio Observer D1 the vacuum probe station was taken into account for the calculation.
(Carl Zeiss, Germany) inverted optical microscope, using PPP-NCHR Spectral measurements were performed with a Xe lamp coupled to
cantilevers (Nanosensors, USA). Samples were prepared by drop- a Spectral Products CM110 monochromator. The light was modulated
casting the exfoliated InSe flake dispersions onto mica sheets (G250-1, by a mechanical chopper at 23 Hz and the AC current was measured
Agar Scientific Ltd., Essex, UK). The images (512 × 512 data points) of with a Signal Recovery DSP 7265 lock-in after amplification by a DL1211
5 × 5 µm2, 2.5 × 2.5 µm2, and 500 × 500 nm2 were collected at a scan rate current amplifier. The same setup was used for frequency modulation
of 0.6 Hz in intermittent contact mode by keeping the set point above measurements.
65% of the free oscillation amplitude. Height profiles of ≈100 flakes were
analyzed with JPK Data Processing software (JPK Instruments, Germany).
Raman spectroscopy measurements were carried out with excitation
wavelength of 514.5 nm and incident power of 1 mW, using a Renishaw Supporting Information
micro-Raman InVia system equipped with a 100× objective. The samples Supporting Information is available from the Wiley Online Library or
were prepared by drop-casting InSe flake dispersions on Si/SiO2 from the author.
substrates and analyzed after drying in vacuum by collecting 20 spectra
in different positions. The spectra were fitted with Lorentzian functions.
For micro-photoluminescence spectra, a Renishaw inVia was used,
equipped with a 50× (0.75 N.A.) objective and laser excitation wavelength Acknowledgements
of 785 nm. The PL spectra were recorded on films deposited on quartz
with a time exposure of 10 s and laser power ≈1 mW. The bulk sample N.C. and M.S. contributed equally to this work. N.C. gratefully
was measured with a time exposure of 1 s and laser power of ≈50 µW. acknowledges the financial support of a Ph.D. scholarship from Region
The crystal structure was characterized by XRD using a PANalytical Sardinia (P.O.R. Sardegna, European Social Fund 2007–2013—Axis IV
Empyrean diffractometer with Cu Kα radiation. The samples were Human Resources, Line of Activity I.3.1). The authors thank the Clean
obtained by deposition of InSe dispersions on Si substrates. Room Facility of the Italian Institute of Technology for support with
The viscosity of the InSe ink in IPA was measured with a Discovery device fabrication and Dr F. De Angelis (Plasmon Nanotechnologies
HR-2 Hybrid Rheometer (TA instruments), using a double-wall concentric group) for the access to the Raman equipment.
cylinders geometry (inner diameter of 32 mm and outer diameter of
35 mm), designed for low-viscosity fluids. The temperatures of the inks
were set and maintained at 25 °C throughout all the measurements. Conflict of Interest
The SEM/EDX measurements on the devices were performed using
a Helios Nanolab 600 (FEI Company) combined with an X-Max detector The authors declare no conflict of interest.
and INCA s system (Oxford Instruments) for the EDX spectra acquisition
and analysis. For SEM measurements an accelerating voltage of 10 kV
and beam current of 0.2 nA were used, while for EDX measurement Keywords
these values were set to 10 kV and 0.8 nA. The sample was imaged as it
was, without any conductive coating applied to the surface.
Device Fabrication: A set of nine interdigitated source/drain 2D semiconductors, field effect transistors, indium selenide, liquid phase
electrodes for FETs were patterned on a highly p-doped Si chip covered exfoliation, photodetectors, solution processed, spray coating
with a 100 nm layer of thermal oxide. Each source/drain electrode
pair consisted of 65 pairs of interdigitated fingers with an overlap of Received: October 12, 2019
40 µm and a gap L = 1 µm, which resulted in an overall channel width Revised: January 5, 2020
W = 5160 µm and an active area of W × L = 5160 µm2 for photodetection. Published online:

Adv. Funct. Mater. 2020, 1908427 1908427 (8 of 10) © 2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.afm-journal.de

[1] F. Bonaccorso, Z. Sun, T. Hasan, A. C. Ferrari, Nat. Photonics A. Morpurgo, J. N. Coleman, V. Nicolosi, L. Colombo, A. Fert,
2010, 4, 611. M. Garcia-Hernandez, A. Bachtold, G. F. Schneider, F. Guinea,
[2] Q. H. Wang, K. Kalantar-Zadeh, A. Kis, J. N. Coleman, C. Dekker, M. Barbone, et al., Nanoscale 2015, 7, 4598.
M. S. Strano, Nat. Nanotechnol. 2012, 7, 699. [26] H.-W. Yang, H.-F. Hsieh, R.-S. Chen, C.-H. Ho, K.-Y. Lee,
[3] D. Boukhvalov, B. Gürbulak, S. Duman, L. Wang, A. Politano, L.-C. Chao, ACS Appl. Mater. Interfaces 2018, 10, 5740.
L. Caputi, G. Chiarello, A. Cupolillo, Nanomaterials 2017, 7, 372. [27] G. Konstantatos, Nat. Commun. 2018, 9, 5266.
[4] N. Kuroda, Y. Nishina, Solid State Commun. 1980, 34, 481. [28] S. Goossens, G. Navickaite, C. Monasterio, S. Gupta,
[5] M. Brotons-Gisbert, D. Andres-Penares, J. Suh, F. Hidalgo, J. J. Piqueras, R. Pérez, G. Burwell, I. Nikitskiy, T. Lasanta, T. Galán,
R. Abargues, P. J. Rodríguez-Cantó, A. Segura, A. Cros, E. Puma, A. Centeno, A. Pesquera, A. Zurutuza, G. Konstantatos,
G. Tobias, E. Canadell, P. Ordejón, J. Wu, J. P. Martínez-Pastor, F. Koppens, Nat. Photonics 2017, 11, 366.
J. F. Sánchez-Royo, Nano Lett. 2016, 16, 3221. [29] T. Mueller, F. Xia, P. Avouris, Nat. Photonics 2010, 4, 297.
[6] J. Camassel, P. Merle, H. Mathieu, A. Chevy, Phys. Rev. B 1978, 17, [30] H. Yin, A. Akey, R. Jaramillo, Phys. Rev. Mater. 2018, 2, 084602.
4718. [31] G. Iannaccone, F. Bonaccorso, L. Colombo, G. Fiori, Nat. Nano-
[7] G. W. Mudd, M. R. Molas, X. Chen, V. Zólyomi, K. Nogajewski, technol. 2018, 13, 183.
Z. R. Kudrynskyi, Z. D. Kovalyuk, G. Yusa, O. Makarovsky, L. Eaves, [32] H.-C. C. Chang, C.-L. L. Tu, K.-I. I. Lin, J. Pu, T. Takenobu,
M. Potemski, V. I. Fal’ko, A. Patanè, Sci. Rep. 2016, 6, 39619. C.-N. N. Hsiao, C.-H. H. Chen, Small 2018, 14, 1802351.
[8] S. Lei, L. Ge, S. Najmaei, A. George, R. Kappera, J. Lou, [33] Y.-H. Lee, X.-Q. Zhang, W. Zhang, M.-T. Chang, C.-T. Lin,
M. Chhowalla, H. Yamaguchi, G. Gupta, R. Vajtai, A. D. Mohite, K.-D. Chang, Y.-C. Yu, J. T.-W. Wang, C.-S. Chang, L.-J. Li, T.-W. Lin,
P. M. Ajayan, ACS Nano 2014, 8, 1263. Adv. Mater. 2012, 24, 2320.
[9] G. W. Mudd, S. A. Svatek, T. Ren, A. Patanè, O. Makarovsky, [34] Y. Zhan, Z. Liu, S. Najmaei, P. M. Ajayan, J. Lou, Small 2012, 8,
L. Eaves, P. H. Beton, Z. D. Kovalyuk, G. V. Lashkarev, 966.
Z. R. Kudrynskyi, A. I. Dmitriev, Adv. Mater. 2013, 25, 5714. [35] Z. Yang, W. Jie, C.-H. Mak, S. Lin, H. Lin, X. Yang, F. Yan, S. P. Lau,
[10] Z. Li, H. Qiao, Z. Guo, X. Ren, Z. Huang, X. Qi, S. C. Dhanabalan, J. Hao, ACS Nano 2017, 11, 4225.
J. S. Ponraj, D. Zhang, J. Li, J. Zhao, J. Zhong, H. Zhang, Adv. [36] B. Liu, M. Fathi, L. Chen, A. Abbas, Y. Ma, C. Zhou, ACS Nano
Funct. Mater. 2018, 28, 1705237. 2015, 9, 6119.
[11] W. Feng, W. Zheng, W. Cao, P. Hu, Adv. Mater. 2014, 26, 6587. [37] T. T. Kodas, The Chemistry of Metal CVD, Wiley-VCH, Weinheim,
[12] D. A. Bandurin, A. V. Tyurnina, G. L. Yu, A. Mishchenko, V. Zólyomi, Germany 1994.
S. V. Morozov, R. K. Kumar, R. V. Gorbachev, Z. R. Kudrynskyi, [38] V. Nicolosi, M. Chhowalla, M. G. Kanatzidis, M. S. Strano,
S. Pezzini, Z. D. Kovalyuk, U. Zeitler, K. S. Novoselov, A. Patanè, J. N. Coleman, Science 2013, 340, 1226419.
L. Eaves, I. V. Grigorieva, V. I. Fal’ko, A. K. Geim, Y. Cao, Nat. [39] E. D. Grayfer, M. N. Kozlova, V. E. Fedorov, Adv. Colloid Interface
Nanotechnol. 2017, 12, 223. Sci. 2017, 245, 40.
[13] S. R. Tamalampudi, Y.-Y. Lu, U. R. Kumar, R. Sankar, C.-D. Liao, [40] N. Dhar, N. Syed, M. Mohiuddin, A. Jannat, A. Zavabeti,
B. K. Moorthy, C.-H. Cheng, F. C. Chou, Y.-T. Chen, Nano Lett. B. Y. Zhang, R. S. Datta, P. Atkin, N. Mahmood, D. Esrafilzadeh,
2014, 14, 2800. T. Daeneke, K. Kalantar-Zadeh, ACS Appl. Mater. Interfaces 2018,
[14] G. W. Mudd, S. A. Svatek, L. Hague, O. Makarovsky, 10, 42603.
Z. R. Kudrynskyi, C. J. Mellor, P. H. Beton, L. Eaves, [41] Y. Hernandez, V. Nicolosi, M. Lotya, F. M. Blighe, Z. Sun, S. De,
K. S. Novoselov, Z. D. Kovalyuk, E. E. Vdovin, A. J. Marsden, I. T. McGovern, B. Holland, M. Byrne, Y. K. Gun’Ko, J. J. Boland,
N. R. Wilson, A. Patanè, Adv. Mater. 2015, 27, 3760. P. Niraj, G. Duesberg, S. Krishnamurthy, R. Goodhue, J. Hutchison,
[15] W. Luo, Y. Cao, P. Hu, K. Cai, Q. Feng, F. Yan, T. Yan, X. Zhang, V. Scardaci, A. C. Ferrari, J. N. Coleman, Nat. Nanotechnol. 2008,
K. Wang, Adv. Opt. Mater. 2015, 3, 1418. 3, 563.
[16] Z. Chen, J. Biscaras, A. Shukla, Nanoscale 2015, 7, 5981. [42] J. N. Coleman, M. Lotya, A. O’Neill, S. D. Bergin, P. J. King,
[17] Y.-R. Chang, P.-H. Ho, C.-Y. Wen, T.-P. Chen, S.-S. Li, J.-Y. Wang, U. Khan, K. Young, A. Gaucher, S. De, R. J. Smith, I. V. Shvets,
M.-K. Li, C.-A. Tsai, R. Sankar, W.-H. Wang, P.-W. Chiu, F.-C. Chou, S. K. Arora, G. Stanton, H.-Y. Kim, K. Lee, G. T. Kim,
C.-W. Chen, ACS Photonics 2017, 4, 2930. G. S. Duesberg, T. Hallam, J. J. Boland, J. J. Wang, J. F. Donegan,
[18] J. Kang, S. A. Wells, V. K. Sangwan, D. Lam, X. Liu, J. Luxa, J. C. Grunlan, G. Moriarty, A. Shmeliov, R. J. Nicholls, J. M. Perkins,
Z. Sofer, M. C. Hersam, Adv. Mater. 2018, 30, 1802990. E. M. Grieveson, K. Theuwissen, D. W. McComb, P. D. Nellist,
[19] M. Buscema, J. O. Island, D. J. Groenendijk, S. I. Blanter, V. Nicolosi, Science 2011, 331, 568.
G. A. Steele, H. S. J. van der Zant, A. Castellanos-Gomez, Chem. [43] F. Bonaccorso, A. Bartolotta, J. N. Coleman, C. Backes, Adv. Mater.
Soc. Rev. 2015, 44, 3691. 2016, 28, 6136.
[20] N. Huo, G. Konstantatos, Adv. Mater. 2018, 30, 1801164. [44] K. R. Paton, E. Varrla, C. Backes, R. J. Smith, U. Khan, A. O’Neill,
[21] J. Kim, S. S. Baik, S. H. Ryu, Y. Sohn, S. Park, B.-G. Park, C. Boland, M. Lotya, O. M. Istrate, P. King, T. Higgins, S. Barwich,
J. Denlinger, Y. Yi, H. J. Choi, K. S. Kim, Science 2015, 349, 723. P. May, P. Puczkarski, I. Ahmed, M. Moebius, H. Pettersson,
[22] Y. Yi, X.-F. Yu, W. Zhou, J. Wang, P. K. Chu, Mater. Sci. Eng.: E. Long, J. Coelho, S. E. O’Brien, E. K. McGuire, B. M. Sanchez,
R 2017, 120, 1. G. S. Duesberg, N. McEvoy, T. J. Pennycook, C. Downing,
[23] K. J. Xiao, A. Carvalho, A. H. C. Neto, Phys. Rev. B 2017, 96, A. Crossley, V. Nicolosi, J. N. Coleman, Nat. Mater. 2014,
054112. 13, 624.
[24] M. J. Hamer, J. Zultak, A. V. Tyurnina, V. Zólyomi, D. Terry, [45] S. Biccai, S. Barwich, D. Boland, A. Harvey, D. Hanlon, N. McEvoy,
A. Barinov, A. Garner, J. Donoghue, A. P. Rooney, V. Kandyba, J. N. Coleman, 2D Mater. 2018, 6, 015008.
A. Giampietri, A. Graham, N. Teutsch, X. Xia, M. Koperski, [46] Y. Yao, Z. Lin, Z. Li, X. Song, K.-S. Moon, C. Wong, J. Mater. Chem.
S. J. Haigh, V. I. Fal’ko, R. V. Gorbachev, N. R. Wilson, ACS Nano 2012, 22, 13494.
2019, 13, 2136. [47] L. H. Li, Y. Chen, B.-M. Cheng, M.-Y. Lin, S.-L. Chou, Y.-C. Peng,
[25] A. C. Ferrari, F. Bonaccorso, V. Fal’ko, K. S. Novoselov, Appl. Phys. Lett. 2012, 100, 261108.
S. Roche, P. Bøggild, S. Borini, F. H. L. Koppens, V. Palermo, [48] A. E. D. R. Castillo, V. Pellegrini, A. Ansaldo, F. Ricciardella, H. Sun,
N. Pugno, J. A. Garrido, R. Sordan, A. Bianco, L. Ballerini, L. Marasco, J. Buha, Z. Dang, L. Gagliani, E. Lago, N. Curreli,
M. Prato, E. Lidorikis, J. Kivioja, C. Marinelli, T. Ryhänen, S. Gentiluomo, F. Palazon, M. Prato, R. Oropesa-Nuñez,

Adv. Funct. Mater. 2020, 1908427 1908427 (9 of 10) © 2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advancedsciencenews.com www.afm-journal.de

P. S. Toth, E. Mantero, M. Crugliano, A. Gamucci, A. Tomadin, [72] Y. Zheng, J. Kong, D. Huang, W. Shi, L. McMillon-Brown,
M. Polini, F. Bonaccorso, Mater. Horiz. 2018, 5, 890. H. E. Katz, J. Yu, A. D. Taylor, Nanoscale 2018, 10, 11342.
[49] P. G. Karagiannidis, S. A. Hodge, L. Lombardi, F. Tomarchio, [73] B. Gürbulak, M. Şata, S. Dogan, S. Duman, A. Ashkhasi,
N. Decorde, S. Milana, I. Goykhman, Y. Su, S. V. Mesite, E. F. Keskenler, Phys. E 2014, 64, 106.
D. N. Johnstone, R. K. Leary, P. A. Midgley, N. M. Pugno, F. Torrisi, [74] S. Jandl, C. Carlone, Solid State Commun. 1978, 25, 5.
A. C. Ferrari, ACS Nano 2017, 11, 2742. [75] K. Kambas, C. Julien, M. Jouanne, A. Likforman, M. Guittard, Phys.
[50] A. Capasso, A. E. Del Rio Castillo, H. Sun, A. Ansaldo, V. Pellegrini, Status Solidi 1984, 124, K105.
F. Bonaccorso, Solid State Commun. 2015, 224, 53. [76] L. Debbichi, O. Eriksson, S. Lebègue, J. Phys. Chem. Lett. 2015, 6,
[51] S. Süß, T. Sobisch, W. Peukert, D. Lerche, D. Segets, Adv. Powder 3098.
Technol. 2018, 29, 1550. [77] M. Osman, Y. Huang, W. Feng, G. Liu, Y. Qiu, P. Hu, RSC Adv.
[52] H. Sun, A. E. Del Rio Castillo, S. Monaco, A. Capasso, A. Ansaldo, 2016, 6, 70452.
M. Prato, D. A. Dinh, V. Pellegrini, B. Scrosati, L. Manna, [78] K. Kouroupis-Agalou, A. Liscio, E. Treossi, L. Ortolani, V. Morandi,
F. Bonaccorso, J. Mater. Chem. A 2016, 4, 6886. N. M. Pugno, V. Palermo, Nanoscale 2014, 6, 5926.
[53] F. Torrisi, T. Hasan, W. Wu, Z. Sun, A. Lombardo, T. S. Kulmala, [79] J. Lauth, F. E. S. Gorris, M. S. Khoshkhoo, T. Chassé, W. Friedrich,
G.-W. Hsieh, S. Jung, F. Bonaccorso, P. J. Paul, D. Chu, V. Lebedeva, A. Meyer, C. Klinke, A. Kornowski, M. Scheele,
A. C. Ferrari, ACS Nano 2012, 6, 2992. H. Weller, Chem. Mater. 2016, 28, 1728.
[54] J. Hassoun, F. Bonaccorso, M. Agostini, M. Angelucci, M. G. Betti, [80] Z. Chen, K. Gacem, M. Boukhicha, J. Biscaras, A. Shukla, Nano-
R. Cingolani, M. Gemmi, C. Mariani, S. Panero, V. Pellegrini, technology 2013, 24, 415708.
B. Scrosati, Nano Lett. 2014, 14, 4901. [81] D. Han, Y. Khan, K. Gopalan, A. Pierre, A. C. Arias, Adv. Funct.
[55] E. Lago, P. S. Toth, G. Pugliese, V. Pellegrini, F. Bonaccorso, RSC Mater. 2018, 28, 1802986.
Adv. 2016, 6, 97931. [82] S. M. Sze, K. K. Ng, Physics of Semiconductor Devices, 3rd ed., John
[56] D. J. Finn, M. Lotya, G. Cunningham, R. J. Smith, D. McCloskey, Wiley & Sons, Hoboken, NJ 2006.
J. F. Donegan, J. N. Coleman, J. Mater. Chem. C 2014, 2, 925. [83] T. Smeeton, C. Humphreys, in Springer Handbook of Electronic and
[57] S. Bellani, E. Petroni, A. E. Del Rio Castillo, N. Curreli, Photonic Materials (Eds: S. Kasap, P. Capper), Springer Interna-
B. Martín-García, R. Oropesa-Nuñez, M. Prato, F. Bonaccorso, tional Publishing, Cham, Switzerland 2017, p. 1.
Adv. Funct. Mater. 2019, 29, 1807659. [84] G. Micocci, A. Rizzo, P. Siciliano, A. Tepore, Phys. Status Solidi
[58] W. J. Hyun, E. B. Secor, M. C. Hersam, C. D. Frisbie, L. F. Francis, 1989, 114, 253.
Adv. Mater. 2015, 27, 109. [85] G. Micocci, P. Siciliano, A. Tepore, Sol. Energy Mater. 1990, 20, 181.
[59] Z. Sun, T. Hasan, F. Torrisi, D. Popa, G. Privitera, F. Wang, [86] P.-H. Ho, Y.-R. Chang, Y.-C. Chu, M.-K. Li, C.-A. Tsai, W.-H. Wang,
F. Bonaccorso, D. M. Basko, A. C. Ferrari, ACS Nano 2010, 4, 803. C.-H. Ho, C.-W. Chen, P.-W. Chiu, ACS Nano 2017, 11, 7362.
[60] C.-H. Chang, T.-C. Huang, C.-W. Peng, T.-C. Yeh, H.-I. Lu, [87] A. Rossi, D. Spirito, F. Bianco, S. Forti, F. Fabbri, H. Büch,
W.-I. Hung, C.-J. Weng, T.-I. Yang, J.-M. Yeh, Carbon 2012, 50, 5044. A. Tredicucci, R. Krahne, C. Coletti, Nanoscale 2018, 10, 4332.
[61] P. Cataldi, I. S. Bayer, F. Bonaccorso, V. Pellegrini, A. Athanassiou, [88] M. Kielar, O. Dhez, G. Pecastaings, A. Curutchet, L. Hirsch, Sci.
R. Cingolani, Adv. Electron. Mater. 2015, 1, 1500224. Rep. 2016, 6, 39201.
[62] W. M. Haynes, CRC Handbook of Chemistry and Physics, 97th ed., [89] X. He, F. Léonard, J. Kono, Adv. Opt. Mater. 2015, 3, 989.
CRC Press, Boca Raton, FL 2016. [90] J. Wojtas, F. K. Tittel, T. Stacewicz, Z. Bielecki, R. Lewicki,
[63] K. Aim, Fluid Phase Equilib. 1978, 2, 119. J. Mikolajczyk, M. Nowakowski, D. Szabra, P. Stefanski, J. Tarka,
[64] A. L. Harreus, R. Backes, J.-O. Eichler, R. Feuerhake, C. Jäkel, Int. J. Thermophys. 2014, 35, 2215.
U. Mahn, R. Pinkos, R. Vogelsang, Ullmann’s Encyclopedia of Indus- [91] M. Long, P. Wang, H. Fang, W. Hu, Adv. Funct. Mater. 2019, 29,
trial Chemistry, Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, 1803807.
Germany 2011. [92] J.-F. Wang, L. Zhu, B.-G. Zhao, Y.-L. Zhao, J. Song, X.-Q. Gu,
[65] C. Berrueco, P. Álvarez, S. Venditti, T. J. Morgan, A. A. Herod, Y.-H. Qiang, Sci. Rep. 2017, 7, 14478.
M. Millan, R. Kandiyoti, Energy Fuels 2009, 23, 3008. [93] Y.-O. Choi, N.-H. Kim, J.-S. Park, W.-S. Lee, Mater. Sci. Eng.: B
[66] K. Lee, H.-Y. Kim, M. Lotya, J. N. Coleman, G.-T. Kim, 2010, 171, 73.
G. S. Duesberg, Adv. Mater. 2011, 23, 4178. [94] D. Periyanagounder, P. Gnanasekar, P. Varadhan, J.-H. He,
[67] A. E. Del Rio Castillo, V. Pellegrini, H. Sun, J. Buha, D. A. Dinh, J. Kulandaivel, J. Mater. Chem. C 2018, 6, 9545.
E. Lago, A. Ansaldo, A. Capasso, L. Manna, F. Bonaccorso, Chem. [95] B. Mukherjee, Y. Cai, H. R. Tan, Y. P. Feng, E. S. Tok, C. H. Sow,
Mater. 2018, 30, 506. ACS Appl. Mater. Interfaces 2013, 5, 9594.
[68] A. Harvey, C. Backes, Z. Gholamvand, D. Hanlon, D. McAteer, [96] X. Zhang, J. Jie, W. Zhang, C. Zhang, L. Luo, Z. He, X. Zhang,
H. C. Nerl, E. McGuire, A. Seral-Ascaso, Q. M. Ramasse, W. Zhang, C. Lee, S. Lee, Adv. Mater. 2008, 20, 2427.
N. McEvoy, S. Winters, N. C. Berner, D. McCloskey, J. F. Donegan, [97] M. Buscema, D. J. Groenendijk, S. I. Blanter, G. A. Steele,
G. S. Duesberg, V. Nicolosi, J. N. Coleman, Chem. Mater. 2015, 27, H. S. J. van der Zant, A. Castellanos-Gomez, Nano Lett. 2014, 14,
3483. 3347.
[69] E. Petroni, E. Lago, S. Bellani, D. W. Boukhvalov, A. Politano, [98] O. Lopez-Sanchez, D. Lembke, M. Kayci, A. Radenovic, A. Kis,
B. Gürbulak, S. Duman, M. Prato, S. Gentiluomo, Nat. Nanotechnol. 2013, 8, 497.
R. Oropesa-Nuñez, J.-K. Panda, P. S. Toth, A. E. Del Rio Castillo, [99] K. Roy, M. Padmanabhan, S. Goswami, T. P. Sai, G. Ramalingam,
V. Pellegrini, F. Bonaccorso, Small 2018, 14, 1800749. S. Raghavan, A. Ghosh, Nat. Nanotechnol. 2013, 8, 826.
[70] C. J. Zhang, S.-H. Park, O. Ronan, A. Harvey, A. Seral-Ascaso, [100] J. A. Garrido, E. Monroy, I. Izpura, E. Muñoz, Semicond. Sci.
Z. Lin, N. McEvoy, C. S. Boland, N. C. Berner, G. S. Duesberg, Technol. 1998, 13, 563.
P. Rozier, J. N. Coleman, V. Nicolosi, Small 2017, 13, 1701677. [101] M.-L. Lu, C.-W. Lai, H.-J. Pan, C.-T. Chen, P.-T. Chou, Y.-F. Chen,
[71] J. E. Bishop, J. A. Smith, C. Greenland, V. Kumar, N. Vaenas, Nano Lett. 2013, 13, 1920.
O. S. Game, T. J. Routledge, M. Wong-Stringer, C. Rodenburg, [102] F. Bonaccorso, A. Lombardo, T. Hasan, Z. Sun, L. Colombo,
D. G. Lidzey, ACS Appl. Mater. Interfaces 2018, 10, 39428. A. C. Ferrari, Mater. Today 2012, 15, 564.

Adv. Funct. Mater. 2020, 1908427 1908427 (10 of 10) © 2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim