of 23
1
Near
Unity
Absorption
in
Van
der
Waals
Semiconductors
for
Ultrathin
Optoelectronics
Deep Jariwala,
1,2
Artur R. Davoyan,
1,2,3
Giulia Tagliabue,
1,4
Michelle C. Sherrott,
1,2
Joeson
Wong
1
and Harry A. Atwater
1,2,3,4*
1
Department of Applied Physics a
nd Materials Science, California
Institute of Technology, Pasadena,
CA-91125, USA
2
Resnick Sustainability Institute
, California Institute of Techn
ology, Pasadena, CA-91125, USA
3
Kavli Nanoscience Institute, Calif
ornia Institute of Technology
, Pasadena, CA-91125, USA
4
Joint Center for Artificial Photosynthesis, California Institut
e of Technology, Pasad
ena, CA-91125, USA
*Corresponding author:
haa@caltech.edu
Abstract:
We demonstrate near unity, broadb
and absorbing optoelectronic
devices using sub-15 nm thick
transition metal dichalcogenide
s (TMDCs) of molybdenum and tung
sten as van der Waals
semiconductor active layers. Specifically, we report that near-
unity light absorption is possible in
extremely thin (< 15 nm) Van der Waals semiconductor structures
by coupling to strongly
damped optical modes of semiconductor/metal heterostructures. W
e further fabricate Schottky
junction devices using these highl
y absorbing heterostructures
and characterize their
optoelectronic performance. Our w
ork addresses one of the key c
riteria to enable TMDCs as
potential candidates to achieve h
igh optoelectronic efficiency.
Keywords:
Transition metal dichalcogenide
s, heterostructures, light-trap
ping, broadband, near-
unity absorption, photovoltaics
TOC:
2
Advances in synthesis, proces
sing and nanofabrication of low-di
mensional materials over the last
two decades have enabled significant progress towards thin semi
conductor layers for high
efficiency optoelectronics
1-4
and for solar energy c
onversion applications.
5-8
For established
crystalline inorganic semiconducto
r absorbers, light management
structures such as microwire
arrays,
9, 10
, Mie resonators
11
, photonic crystals
12, 13
and plasmonic metal nanostructures
14, 15
enable enhanced absorption in the active layers, and reduced re
flection. In conventional
crystalline semiconductors, achie
ving the necessary surface pas
sivation while incorporating such
light management structures is a considerable challenge, since
an increasing surface/volume ratio
typically results in reduced radiative efficiency. The emergenc
e of two dimensional (2D)
semiconducting atomic layers namely TMDCs of molybdenum and tun
gsten
16
has opened up a
new class of high radiative e
fficiency semiconductors that can
be synthesized in ultrathin form.
Several reports have demonstrated the use of TMDCs as active la
yers in optoelectronic and
photovoltaic devices. Most reports have utilized TMDCs in a bac
k-gated van der Waals Schottky
junction geometry with graphene
17, 18
, a van der Waals p-n heterojunction
19, 20
or in an
electrostatically split-gated p-n homojunction
19, 21
geometry. In spite of recent theoretical and
experimental advances in light trapping in ultrathin 2D layers,
22-26
in most approaches to date,
the absorption in active layer is far from optimal and often na
rrowband or highly sensitive to the
angle of incidence.
Metallic rear surfaces are commonly used for enhancing
light absorption in optoelectronic
devices. For structures whose thickness is greater than wavelen
gth scale, the performance of the
metallic rear surface can be interpreted as a simple ray optica
l specular reflector. However when
the semiconductor absorber/refle
ctor heterostructure thickness
is at or below the wavelength
3
scale, a different conceptual approach is needed. Prior computa
tional investigations have shown
that thin absorber/metal heterostructures result in light absor
ption enhancement due to an
increase in the local density of states (LDOS) near the semicon
ductor/metal interface.
27, 28
If the
heterostructure is thin, then lig
ht absorption can be enhanced
in a broadband manner,
corresponding to enhanced absorption close to the interface, wh
en a thin semiconductor is placed
in intimate planar contact on a reflecting metal substrate.
27
This concept was then demonstrated
experimentally in ultrathin (< 25 nm) germanium (Ge) on gold (A
u) and silver (Ag).
29
However
it is difficult to thin down covalently bonded, isotropic 3D se
miconductors to below 100 nm
thickness without significant degradation of crystalline qualit
y, increasing defect density or
influence of surface oxides and states on electronic charge tra
nsport. This imposes limitations on
the applicability of 3D semic
onductors in ultrathin photovoltai
c devices. By contrast, TMDCs
have self-passivated, dangling bond- and oxide-free surfaces
16, 30
and are thus attractive
alternatives for ultrathin absorbers when coupled with reflecti
ve metals (Figure 1 a). Here we
report near-unity, broadband ab
sorption in ultrathin (12-15 nm)
TMDC layers and demonstrate
proof-of-concept devices as pot
ential candidates for photovolt
aic applications.
An initial look at a microm
echanically exfoliated WSe
2
structure on a template stripped
31
Ag
substrate in broadband white-light illumination shows regions o
f stark and varying color
contrasts from pale red to dark blue (Figure 1b). Observing at
higher magnification further
reveals crystalline flakes with uniform smooth, straight edges,
stepped layers and thickness
variations akin to numerous prior observations of exfoliated 2D
crystals on SiO
2
substrates.
32, 33
Measurement of thickness with atomic force microscopy (AFM) ind
icates flake thicknesses
varying from ~3 nm (pale red) to 13 nm (nearly black) (Figure 1
c-e) suggesting a highly
4
absorbing nature. The step height and surface roughness (root m
ean square roughness < 1 nm for
Ag and < 0.3 nm for WSe
2
) are also highly uniform as see
n in the AFM topography (Figure
1 f).
Absorption spectra for varying thickness WSe
2
on Ag (Figure 2 a) back reflector were
calculated using available values of refractive index and extin
ction coefficient from the
literature
34
to quantify the above observations. Strongly enhanced absorpti
on was observed with
increasing thickness of the WSe
2
with near-unity abs
orption peak occurr
ing between 500-650 nm
for varying thickness of the flakes. The peaks in absorption, e
xcept for the primary exciton peak
at the absorption edge, undergo a
red shift with increasing thi
ckness of WSe
2
suggesting
dependence on the optical path length implying thin film interf
erence effect where the reflected
light is strongly attenuated, l
eading to non-trivial interface
phase shifts.
29
Briefly, in the case of a
perfect metal/lossless dielectric (
k
=0) with refractive index n the phase shift at the metal
dielectric interface is
p
corresponding to perfect reflection. Hence a minimum dielectr
ic film
thickness of λ/4n on the metal w
ould form an optical cavity wit
h 0 or 2
p
phase shift at the
dielectric/air interface. I
f the dielectric is lossy (
k
≠0) however, even for thicknesses in the deep
subwavelength regime, the tota
l reflection and transmission pha
se shifts can be approximately 0
or 2
p
at the air/dielectric interface
giving rise to an absorbance r
esonance as seen in Figure 1 c-
e. Experimentally acquired spectra of WSe
2
on template stripped Ag surfaces (Figure 2b) shows
remarkably good qualitative and quantitative agreement with the
calculations. An interesting
observation in the above experime
nts is that a broadband perfec
t absorption only occurs for a
narrow range of WSe
2
thicknesses between 12-15 nm only with an intimate contact wit
h a metal
back reflector (See Supporting in
formation S1 for more details)
. Below or above this thickness,
there is increased reflection in the red or blue parts of the s
pectrum leading to net reduction in
5
integrated absorption. Further,
for bulk free standing or glass
supported TMDCs, the maximum
above-gap absorption is limited to a maximum of ~40%. Due to th
e large index mismatch, a
large fraction (50-60%) of the incident light is reflected back
from the surface of bulk crystals
35,
36
(See supporting information figure S1). Likewise, the absorpti
on in few layer-bulk TMDCs on
the conventionally used Si/SiO
2
substrates is also limited to a maximum between 50-60%
37, 38
The above observations are not unique to WSe
2
and can be further generalized to other TMDCs
(Figure 2 c-f) as well as Au b
ack reflectors
(See Supporting In
formation S2).
Although the absorption peaks
in our structure are depen
dent on path length, they are highly
insensitive to the angle of incidence as a can be seen for the
case of 13 nm WSe
2
on Ag (Figure
3a). The peak absorption stays over 80% even at a 60 ̊ incident
angle (Figure 3 b) suggesting
relatively low sensitivity to the angle of incident light. This
feature of TMDC/Ag
heterostructures is highly advant
ageous for off-normal light co
llection and may be of a particular
interest for photovoltaic applic
ations and solar energy harvest
ing.
9, 39
Based on the above discussion,
it is evident that the TM
DC/metal stack is a suitable ultrathin
absorber for a light-harvesting
device. To demonstrate this con
cept, we fabricated a simple
device, as shown in Fig. 3a with a metal ring electrode on top
using standard photolithography
and metal evaporation. The back reflector combined with a patte
rned metal electrode on top of
the flake creates a metal
1
/TMDC/metal
2
sandwich structure (Figure 4 a-b) that can effectively
function as a Schottky barrier device if there is sufficient di
fference between work functions of
metal
1
and metal
2
(Figure 4c). Considering the sm
all size of the top ring electr
ode and a
conductive metallic back substrate, the devices can only be pro
bed accurately while being
viewed under a high magnification
(50 x), long working distance
objective. Upon broadband,
white light illumination, (Hg vapor lamp, X-Cite 120 Q) the dev
ices show a pronounced
6
photovoltaic response (Figure 4d).
To deduce the collection are
a and current density, a spatial
photocurrent map is acquired usi
ng scanning photocurrent micros
copy (Figure 4e). The
photoexcited carriers diffuse and
get collected from approximat
ely1-3 μm region in the vicinity
of the inner and outer metal ring contact boundary (see Support
ing information S3). Based on
this, we estimate photocurrent density values in Figure 4f. Whi
le, the incident light on the device
is focused owing to the nature of the measurement and the small
size of the device, it is still
noteworthy that for ~20 x concentrations (2.1 W/cm
2
), the short circuit current density (J
SC
) is >
10 mA/cm
2
. Considering that semiconducting
TMDCs are still in the early
research phase in
terms of material quality and cr
ystal defect con
trol, these pho
tocurrent values are promising in an
un-optimized device structure. The van der Waals interlayer bo
nding in TMDCs induces some
level of electron-hole confinement at all thicknesses. Thus, ex
citon binding energies even in bulk
TMDCs are ~70-80 meV.
40
To investigate if the photocurrent is limited by lack of excit
on
dissociation or free carrier rec
ombination, the exponential dep
endence of photocurrent on
incident light intensity was investigated (Figure 4f, inset). A
n exponent close to unity points to
monomolecular recombination
41
suggesting excitons recombini
ng at neutral impurity or one of
free carriers reacting with a
n oppositely charged impurity.
Finally, we investigate the spectral dependence of photo
current by illuminating with a laser
focused on a fixed spot generating photocurrent in a 12 nm WS
2
/Ag device (Figure 5 a). For
input powers of 1.6 μW at 633 nm corresponding to the primary e
xciton peak of WS
2
, we
observe pronounced photovoltaic effect with open circuit voltag
es (V
OC
) approaching 0.2 V and
I
SC
> 100 nA, resulting in a single-wavelength power conversion ef
ficiency ~ 0.5 % (Figure 5b).
At this power, the external quantum efficiency (EQE) is ~ 13 %
comparable with previously
reported values in multilayer devices. At higher input power,
the efficiency drops down to
7
below 8% (Figure 5c) suggesting increasing recombination with i
ncreasing carrier density,
indicating a carrier density dependent recombination mechanism
such as Auger recombination.
The EQE also remains relatively
constant between 8-12 % above t
he absorption edge as seen in
WSe
2
on Au (Figure 5d) and its spect
rum roughly corresponds to the
absorption one. The
resulting above-gap IQE is a modest 10% across the absorption s
pectrum. The lack of high
quantum efficiency can be attributed to several factors. Primar
y among them is the device
geometry which prohibits optical
excitation of the TMDC directl
y beneath the top metal
electrode which results in in-pla
ne diffusion of carriers for c
ollection. Second, the Schottky
junction leads to recombination of all excitons and free electr
on hole pairs that reach the metal
electrode. Finally, the semiconductor quality remains far from
optimal as evidenced from the
exponent of power dependence of
photocurrent suggesting monomol
ecular recombination. The
carrier collection and EQE may be improved by use of transparen
t top contact such as
graphene
17, 18
in addition to a type-II hete
rojunction between two TMDCs
42
(See Supporting
Information S5).
In summary, we have shown an ultrathin, near-unity, broa
dband semiconducting absorber
system using TMDC/metal heterostructure and have applied it in
Schottky junction
optoelectronic devices. It is also worth noting that most ligh
t trapping techniques in thin
optoelectronics involve the integ
ration of a patte
rned nanostru
cture which could
significantly add
to the total cost and complexity of the resulting device. In co
ntrast, the above presented results
avoid the use of any nanopatternin
g to enhance light absorption
. With further development of
the presented structure to int
roduce a p-n junction and carrier
selective contact layers, we expect
that it might be possible to engineer V
OC
> 1 V and thus eventually obtain meaningful power
conversion efficiencies. The e
fficient light absorption results
reported here, combined with the
8
recent demonstration of near-unity luminescence quantum yield i
n MoS
2
,
43
and advances to
improve the TMDC material quality
44
hold promise for future h
igh-efficienc
y, ultrathin
optoelectonics and photovoltaics
with TMDC active layers.
METHODS:
Sample preparation.
TMDC flakes were deposited on template stripped Au and Ag via
mechanical exfoliation of bulk crystals (HQ Graphene). The resu
lting flakes were identified by
optical microscopy and later characterized by AFM to determine
the flake thickness. The Au and
Ag films were deposited by electron beam and thermal evaporatio
n respectively without any
adhesion layers on Si wafers with native oxide only. Standard s
olvent and plasma cleaning
procedures were used for cleanin
g Si wafers prior to deposition
. The substrate was heated to 100
̊C during thermal evaporation of Ag and the deposition rates we
re maintained at ~0.1 Å/sec for
the first 30 nm in the case of both Au and Ag followed by rampi
ng up to ~1 Å /sec till the final
thickness reached 120 nm. The metal films were then template st
ripped using a thermal epoxy
(Epo-Tek 375, Epoxy Technology) using a
procedure described in
ref.
31
Device fabrication, absorbance and photocurrent measurements.
Devices were fabricated
using standard photolithography and thermal or e-beam metal eva
poration. All absorbance
measurements and the EQE spectrum measurements were performed u
sing a home built
absorption measurement setup. Tunable, monochromatic light (400
-1800 nm) was obtained by
coupling a supercontinuum laser
(Fianium) to a monochromator. T
he collimated, monochromatic
beam, was then focused on the sample with a long working distan
ce (NA = 0.55), 50x objective
and the reflection was measured with a Si detector. The used ob
jective ensures close-to-normal
incidence illumination of the device. The reflection spectrum w
as then normalized to the
reflections from a silver mirror (Thorlabs). In the absence of
transmission, absorption was
9
obtained as 1-Normalized Reflec
tion (see Supporting information
S4). Electrical measurements
were performed using Keithley 2400 and 236 source meters and cu
stom LabView programs. The
spatially varying photocurrent measurements and global broadban
d illumination measurements
were performed on a scanning confocal microscope (Zeiss, LSM 71
0) and the incident laser
power was measured using power meter (ThorLabs). The devices we
re probed using piezo
controlled microbot manipulators (Imina Technologies) and all m
easurements were performed
under ambient temperature and pressure conditions.
ASSOCIATED CONTENT:
Supporting Information
:
Experimental methods, additional experimental data, calculation
s and analysis accompany this
paper. This material is availabl
e free of charge via the Intern
et at http://pubs.acs.org
AUTHOR INFORMATION
Corresponding authors:
*Harry A. Atwater, E-mail:
haa@caltech.edu
NOTES:
Competing financial interests
: The authors declare no com
peting financial interests.
ACKNOWLEDGEMENTS
This work is part of the 'Light-Material Interactions in Energy
Conversion' Energy
Frontier Research Center funded by the U.S. Department of Energ
y, Office of Science, Office of
Basic Energy Sciences under Award Number DE-SC0001293.
D.J., A.R.D and M.C.S.
acknowledge additional support from Resnick Sustainability Inst
itute Graduate and Postdoctoral
Fellowships. A.R.D also acknowle
dges support in part from the K
avli Nanoscience Institute
Postdoctoral Fellowship. G.T. ac
knowledges support in part from
the Swiss National Science
Foundation, Early Postdoc Mobility Fellowship n. P2EZP2_159101.
J.W. acknowledges support
from the National Science Founda
tion Graduate Research Fellowsh
ip under Grant No. 1144469.
10
Author contributions:
D.J. prepared the samples and fabricated the devices. A.R.D. p
erformed
all the calculations. D.J., G.T.
and J.W. performed the electri
cal and photocurrent measurements.
M.C.S. assisted with sample preparation and fabrication. H.A.A.
supervised over all the
experiments, calculations and da
ta collection. All authors cont
ributed to data interpretation,
presentation and writing
of the manuscript.
FIGURES:
Figure 1. Absorbing dielectrics on metals:
a.
Schematic diagram of a thin, multilayer TMDC
film on a Au/Ag back r
eflecting substrate.
b.
Low magnification optical m
icrograph of exfoliated
WSe
2
flakes of on template stripped Ag substrate. (Scale bar = 50 μ
m)
c-e.
High magnification
11
micrographs of yellow, red and blue square regions on
(b)
respectively with increasing flake
thickness from (
c)
to (
e)
. The sharp blue shift in color
and rising contrast with increa
sing
thickness can be seen (Scale bar = 10 μm).
f.
AFM topography of the flake region in (
e).
denoted
by the green dashed square.
12
Figure 2. Absorption spectra of ultrathin TMDCs on Ag back refl
ector: a.
Calculated
absorption spectra of varying thicknesses of WSe
2
on an optically thick Ag film. The solid lines
represent total absorption in the WSe
2
/Ag stack while the dashed lines represent absorption only
13
in the WSe
2
.
b.
Experimentally measured absorption spectra of WSe
2
flakes exfoliated on
template stripped Ag films.
c-d.
Same as a-b except for WS
2
on Ag. e-f Calculated (e) and
measured (f) absorption spectra
for varying thicknesses of MoS
2
on Ag.
Figure 3. Angle dependence of absorption in TMDC/Ag heterostruc
tures:
a.
Contour plot of
calculated absorption spectra at varying angles for 13 nm WSe
2
on Ag back reflector. The
insensitivity of the absorption as a function of incident angle
is apparent.
b.
Line cut from
(a)
at
520 nm showing the angle depende
nce of peak absorption.
14
Figure 4: Device structure and characteristics a.
Optical micrograph of a representative
device comprising of 13 nm WSe
2
on Ag with a Pd/Au ring electrode on the top (Scale bar = 10
μm)
b.
Schematic representation of side view of the device in
(a).
c.
Schematic band diagram
showing Schottky contact on the Ag side and ohmic contact on th
e Pd side with a depleted WSe
2
in between.
d.
Current-voltage characteristics of a representative device (13
nm WSe
2
/Ag) under
dark and broadband white light illumination from a Hg-vapor lam
p source.
e.
Spatially varying
photocurrent map of the device acquired at 16 μW incident power
. Inset shows the line profile of
photocurrent magnitude along the white line in the map. The pho
tocurrent profile suggests
carrier diffusion length of ~ 1.5 μm.
f.
Current density vs voltage (J-V) curves estimated based
on the active area determine from (
e
) and I-V plots from (
d
). Inset shows circuit current density
proportional to input power
with an exponent α = 0.99.
15
Figure 5. Monochromatic illumina
tion and external quantum effic
iency: a.
Absorbance
spectrum of 12nm WS
2
/Ag stack. A near-unity absorbance is observed at the primary e
xciton
peak. The red line denotes the
633 nm excitation wavelength. In
set shows the optical micrograph
of the device along wit
h electrical probes (Scale bar = 10 μm).
b.
I-V characteristics of the
device in
(a)
with the 633 nm laser focused on a photocurrent producing spot
.
c.
Power
dependence of EQE for 633 nm incident laser. Inset shows an exp
onent of ~0.9 for power
dependence of photocurrent for this device.
d.
EQE (red) and IQE (blue) spectra for a 19 nm
16
WSe
2
on Au device showing power generation across the absorption sp
ectrum (green). The laser
power is ~ 1 μW for 650 nm wit
h about 10% variation across the
spectrum.
References:
1.
Sukhovatkin,
V.;
Hinds,
S.;
Brzozowski,
L.;
Sargent,
E.
H.
Science
2009,
324,
1542
1544.
2.
Konstantatos,
G.;
Howard,
I.;
Fischer,
A.;
Hoogland,
S.;
Clifford,
J.;
Klem,
E.;
Levina,
L.;
Sargent,
E.
H.
Nature
2006,
442,
180
183.
3.
Peumans,
P.;
Bulovi
ć
,
V.;
Forrest,
S.
Appl.
Phys.
Lett.
2000,
76,
2650
2652.
4.
Yang,
Y.;
Zheng,
Y.;
Cao,
W.;
Titov,
A.;
Hyvonen,
J.;
Manders,
J.
R.;
Xue,
J.;
Holloway,
P.
H.;
Qian,
L.
Nat.
Photon.
2015
,
10.1038/nphoton.2015.36.
5.
Polman,
A.;
Knight,
M.;
Garnett,
E.
C.;
Ehrler,
B.;
Sinke,
W.
C.
Science
2016,
352.
6.
Atwater,
H.
A.;
Polman,
A.
Nat.
Mater.
2010,
9,
205
213.
7.
Polman,
A.;
Atwater,
H.
A.
Nat.
Mater.
2012,
11,
174
177.
8.
Vandamme,
N.;
Chen,
H.
L.;
Gaucher,
A.;
Behaghel,
B.;
Lemaitre,
A.;
Cattoni,
A.;
Dupuis,
C.;
Bardou,
N.;
Guillemoles,
J.
F.;
Collin,
S.
IEEE
Journal
of
Photovoltaics
2015,
5,
565
570.
9.
Kelzenberg,
M.
D.;
Boettcher,
S.
W.;
Petykiewicz,
J.
A.;
Turner
Evans,
D.
B.;
Putnam,
M.
C.;
Warren,
E.
L.;
Spurgeon,
J.
M.;
Briggs,
R.
M.;
Lewis,
N.
S.;
Atwater,
H.
A.
Nat.
Mater.
2010,
9,
239
244.
10.
Garnett,
E.;
Yang,
P.
Nano
Lett.
2010,
10,
1082
1087.
11.
Spinelli,
P.;
Verschuuren,
M.
A.;
Polman,
A.
Nat.
Commun.
2012,
3,
692.
12.
Bermel,
P.;
Luo,
C.;
Zeng,
L.;
Kimerling,
L.
C.;
Joannopoulos,
J.
D.
Optics
Express
2007,
15,
16986
17000.
13.
Zeng,
L.;
Yi,
Y.;
Hong,
C.;
Liu,
J.;
Feng,
N.;
Duan,
X.;
Kimerling,
L.
C.;
Alamariu,
B.
A.
Appl.
Phys.
Lett.
2006,
89,
111111.
14.
Ferry,
V.
E.;
Sweatlock,
L.
A.;
Pacifici,
D.;
Atwater,
H.
A.
Nano
Lett.
2008,
8,
4391
4397.
15.
Ferry,
V.
E.;
Munday,
J.
N.;
Atwater,
H.
A.
Adv.
Mater.
2010,
22,
4794
4808.
16.
Jariwala,
D.;
Sangwan,
V.
K.;
Lauhon,
L.
J.;
Marks,
T.
J.;
Hersam,
M.
C.
ACS
Nano
2014,
8,
1102–
1120.
17.
Britnell,
L.;
Ribeiro,
R.
M.;
Eckmann,
A.;
Jalil,
R.;
Belle,
B.
D.;
Mishchenko,
A.;
Kim,
Y.
J.;
Gorbachev,
R.
V.;
Georgiou,
T.;
Morozov,
S.
V.;
Grigorenko,
A.
N.;
Geim,
A.
K.;
Casiraghi,
C.;
Neto,
A.
H.
C.;
Novoselov,
K.
S.
Science
2013,
340,
1311
1314
18.
Yu,
W.
J.;
Liu,
Y.;
Zhou,
H.;
Yin,
A.;
Li,
Z.;
Huang,
Y.;
Duan,
X.
Nat.
Nanotechnol.
2013,
8,
952–958.
19.
Pospischil,
A.;
Furchi,
M.
M.;
Mueller,
T.
Nat.
Nanotechnol.
2014,
9,
257
261.
20.
Jariwala,
D.;
Howell,
S.
L.;
Chen,
K.
S.;
Kang,
J.;
Sangwan,
V.
K.;
Filippone,
S.
A.;
Turrisi,
R.;
Marks,
T.
J.;
Lauhon,
L.
J.;
Hersam,
M.
C.
Nano
Lett.
2016,
16,
497–503.
21.
Baugher,
B.
W.;
Churchill,
H.
O.;
Yafang,
Y.;
Jarillo
Herrero,
P.
Nat.
Nanotechnol.
2014,
9,
262
267.
22.
Liu,
X.;
Galfsky,
T.;
Sun,
Z.;
Xia,
F.;
Lin,
E.
c.;
Lee,
Y.
H.;
Kéna
Cohen,
S.;
Menon,
V.
M.
Nat.
Photon.
2015,
9,
30
34.
23.
Wang,
W.;
Klots,
A.;
Yang,
Y.;
Li,
W.;
Kravchenko,
I.
I.;
Briggs,
D.
P.;
Bolotin,
K.
I.;
Valentine,
J.
Appl.
Phys.
Lett.
2015,
106,
181104.
17
24.
Lien,
D.
H.;
Kang,
J.
S.;
Amani,
M.;
Chen,
K.;
Tosun,
M.;
Wang,
H.
P.;
Roy,
T.;
Eggleston,
M.
S.;
Wu,
M.
C.;
Dubey,
M.;
Lee,
S.
C.;
He,
J.
H.;
Javey,
A.
Nano
Lett.
2015,
15,
1356
1361.
25.
Piper,
J.
R.;
Fan,
S.
ACS
Photonics
2016,
3,
571
577.
26.
Bahauddin,
S.
M.;
Robatjazi,
H.;
Thomann,
I.
ACS
Photonics
2016,
3,
853
862.
27.
Callahan,
D.
M.;
Munday,
J.
N.;
Atwater,
H.
A.
Nano
Lett.
2012,
12,
214
218.
28.
Yu,
Z.;
Raman,
A.;
Fan,
S.
Proc.
Nat.
Acad.
Sci.
USA
2010,
107,
17491
17496.
29.
Kats,
M.
A.;
Blanchard,
R.;
Genevet,
P.;
Capasso,
F.
Nat.
Mater.
2013,
12,
20
24.
30.
Grigorieva,
I.
V.;
Geim,
A.
K.
Nature
2013,
499,
419
425.
31.
McMorrow,
J.
J.;
Walker,
A.
R.;
Sangwan,
V.
K.;
Jariwala,
D.;
Hoffman,
E.;
Everaerts,
K.;
Facchetti,
A.;
Hersam,
M.
C.;
Marks,
T.
J.
ACS
Appl.
Mater.
Interfaces
2015,
7,
26360
26366.
32.
Novoselov,
K.
S.;
Jiang,
D.;
Schedin,
F.;
Booth,
T.
J.;
Khotkevich,
V.
V.;
Morozov,
S.
V.;
Geim,
A.
K.
Proc.
Nat.
Acad.
Sci.
USA
2005,
102,
10451
10453.
33.
Wang,
Q.
H.;
Kalantar
Zadeh,
K.;
Kis,
A.;
Coleman,
J.
N.;
Strano,
M.
S.
Nat.
Nanotechnol.
2012,
7,
699
712.
34.
Li,
Y.;
Chernikov,
A.;
Zhang,
X.;
Rigosi,
A.;
Hill,
H.
M.;
van
der
Zande,
A.
M.;
Chenet,
D.
A.;
Shih,
E.
M.;
Hone,
J.;
Heinz,
T.
F.
Phys.
Rev.
B
2014,
90,
205422.
35.
Beal,
A.
R.;
Liang,
W.
Y.;
Hughes,
H.
P.
Journal
of
Physics
C:
Solid
State
Physics
1976,
9,
2449.
36.
Beal,
A.
R.;
Hughes,
H.
P.
Journal
of
Physics
C:
Solid
State
Physics
1979,
12,
881.
37.
Nayak,
P.
K.;
Yeh,
C.
H.;
Chen,
Y.
C.;
Chiu,
P.
W.
ACS
Appl.
Mater.
Interfaces
2014,
6,
16020
16026.
38.
Zhao,
W.;
Ghorannevis,
Z.;
Chu,
L.;
Toh,
M.;
Kloc,
C.;
Tan,
P.
H.;
Eda,
G.
ACS
Nano
2013,
7,
791
797.
39.
Grandidier,
J.;
Weitekamp,
R.
A.;
Deceglie,
M.
G.;
Callahan,
D.
M.;
Battaglia,
C.;
Bukowsky,
C.
R.;
Ballif,
C.;
Grubbs,
R.
H.;
Atwater,
H.
A.
Phys.
Status
Solidi
A
2013,
210,
255
260.
40.
Komsa,
H.
P.;
Krasheninnikov,
A.
V.
Phys.
Rev.
B
2012,
86,
241201.
41.
Lakhwani,
G.;
Rao,
A.;
Friend,
R.
H.
Annu.
Rev.
Phys.
Chem.
2014,
65,
557
581.
42.
Lee,
C.
H.;
Lee,
G.
H.;
van
der
Zande,
A.
M.;
Chen,
W.;
Li,
Y.;
Han,
M.;
Cui,
X.;
Arefe,
G.;
Nuckolls,
C.;
Heinz,
T.
F.;
Guo,
J.;
Hone,
J.;
Kim,
P.
Nat.
Nanotechnol.
2014,
9,
676
681
43.
Amani,
M.;
Lien,
D.
H.;
Kiriya,
D.;
Xiao,
J.;
Azcatl,
A.;
Noh,
J.;
Madhvapathy,
S.
R.;
Addou,
R.;
KC,
S.;
Dubey,
M.;
Cho,
K.;
Wallace,
R.
M.;
Lee,
S.
C.;
He,
J.
H.;
Ager,
J.
W.;
Zhang,
X.;
Yablonovitch,
E.;
Javey,
A.
Science
2015,
350,
1065
1068.
44.
Kang,
K.;
Xie,
S.;
Huang,
L.;
Han,
Y.;
Huang,
P.
Y.;
Mak,
K.
F.;
Kim,
C.
J.;
Muller,
D.;
Park,
J.
Nature
2015,
520,
656
660.
Supporting
Information
Near
Unity
Absorption
in
Van
der
Waals
Semiconductors
for
Ultrathin
Optoelectronics
Deep Jariwala,
1,2
Artur R. Davoyan,
1,2,3
Giulia Tagliabue,
1,4
Michelle C. Sherrott,
1,2
Joeson
Wong
1
and Harry A. Atwater
1,2,3,4*
1
Department of Applied Physics a
nd Materials Science, California
Institute of Technology, Pasadena,
CA-91125, USA
2
Resnick Sustainability Institute
, California Institute of Techn
ology, Pasadena, CA-91125, USA
3
Kavli Nanoscience Institute, Calif
ornia Institute of Technology
, Pasadena, CA-91125, USA
4
Joint Center for Artificial Photosynthesis, California Institut
e of Technology, Pasad
ena, CA-91125, USA
*Corresponding author:
haa@caltech.edu
S1. Absorbance calculations:
Absorption in ultrathin TMDCs on metals was calculated using th
e transfer matrix method.
1
The
optical constants used for the calculation are the bulk crystal
values for each TMDCs from
previously published reports.
2, 3
Permittivities of Ag and Au were taken from Jhonson and
Christie
4
.
The maximum integrated absorbance in ultrathin TMDCs on a refle
ctive metallic substrate is only
achieved for a critical TMDC thickness which lies somewhere bet
ween 12-15 nm. Below this
critical thickness, the absorpti
on in the red part of the spect
rum is reduced due to high reflection
from the underlying metal. Above this thickness, the absorption
in the blue part of the spectrum is
reduced due to increased reflection from the TMDC owing to the
mismatch in refractive index
between air and the TMDC. Figure 2 c-d in the manuscript can be
seen for experimental
verification. The coupling of TMDCs with reflective metals is c
rucial for this resonantly enhanced
absorption to occur. In case of a free standing WSe
2
of similar thickness, the total absorption is far
lesser approaching ~40% between 6-12 nm thickness before fallin
g down again due to increased
reflection owing to index mismatch at the air /TMDC interface (
Figure S1).
Figure S1: Calculated absorption in free standing WSe
2
with varying thickness. The total
absorption increases with increasing thickness upto 12-14 nm an
d approaches ~40% following
which it steadily drops down with f
urther increase in the thick
ness.
S2. Absorption with Au back reflectors:
Figure S3 below shows calculated and measured absorption spectr
a for varying thicknesses of
WSe
2
and WS
2
on template stripped Au back reflectors. A good qualitative an
d quantitative
agreement between the measured and calculated spectra is appare
nt once again. However in Au,
interband absorption starts dominating below 550 nm in waveleng
th. Therefore, the useful
absorption in the TMDC layer (da
shed lines) dras
tically reduces
at λ < 500 nm. Ag back reflector
is thus more suitable from an op
tical standpoint of maximizing
useful absorption.
Figure S3: Calculated (left) and
measured (right) absorption sp
ectra of WSe
2
on Au (a-b) and WS
2
on Au(c-d).
S3. Estimation of minority c
arrier diffusion length.
Minority carrier diffusion length can be estimated from the spa
tial photocurrent profile. In cases
where the diffusion length is lar
ger than the spot size, a sing
le exponential model
ൌܫ
ܫ
ܮ/ݔሺെ݌ݔ݁
where I is the photocurrent,
ܫ
is the peak photocurrent,
ݔ
is the distance and
ܮ
is the diffusion length can explain the photocurrent profile an
d provide an estimate of diffusion
length. Considering that we are using a 633 nm laser for spatia
lly resolving the photocurrent, the
diffraction limited resolution (g
iven by 0.61λ/N.A.), where N.A
. is the numerical aperture of the
objective, = 0.8 in our case) is
~
500 nm in our measurement. Based on that, we can fit our
photocurrent data to the above expo
nential decay equation and e
xtract minority carrier diffusion
lengths. We have estimated diffusion length varying from about
1.35 μm in Figure S4 a to about
3 μm in Figure S4 b.