Supplemental Information:
Near-Infrared Photodetector Consisting of J-Aggregating
Cyanine Dye and Metal Oxide Thin Films
Timothy P. Osedach†§, Antonio Iacchetti‡¶, Richard R. Lunt$, Trisha L. Andrew†, Patrick R.
Brown&, Gleb M. Akselrod& and Vladimir Bulović†*
[*]
Prof. Vladimir Bulović
Building 13-3138, Department of Electrical Engineering and Computer Science,
Massachusetts Institute of Technology, Cambridge, Massachusetts 02139
E-mail: bulovic@mit.edu
[†]
Timothy P. Osedach, Trisha L. Andrew, Prof. Vladimir Bulović
Department of Electrical Engineering and Computer Science,
Massachusetts Institute of Technology, Cambridge, Massachusetts 02139
[§]
Timothy P. Osedach
School of Engineering and Applied Sciences, Harvard University, Cambridge, Massachusetts
02138
[‡]
Antonio Iacchetti
Center for Nano Science and Technology of IIT@PoliMI, Via Pascoli 70/3, 20133 Milano, Italy
[¶]
Antonio Iacchetti
Politecnico di Milano, Dip. Elettronica e Informazione, P.za L. da Vinci 32, 20133 Milano, Italy
[$]
Richard R. Lunt
Department of Chemical Engineering and Materials Science, Michigan State University, East
Lansing, MI 48824, USA
[$]
Patrick R. Brown, Gleb M. Akselrod
Department of Physics, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139
Keywords: J-Aggregate, Cyanine, Photodetector, Photovoltaic, Infrared, Exciton Diffusion
Fabrication of U3 Photodiode Structures
To fabricate the U3 photodiode structures, glass substrates with pre-patterned ITO electrodes
(obtained from Thin Film Devices, Inc.) are first cleaned, treated in oxygen plasma and then transferred
into a sputtering chamber. A 50 nm thick film of ZnO is then deposited at a rate of 0.05 nm/s by RFsputtering at a power of 150 W and in an argon atmosphere of 20 mTorr. Films of U3 (purchased from
Ryan Scientific and used without any further purification; CAS Number 202135-09-9) are prepared by
dissolving U3 powder in ultrapure water and spin-casting at 4000 RPM for 60 seconds on top of the ZnO
layer. The ZnO surface is pre-treated in oxygen plasma prior to spin-casting. The thickness of the U3
film is varied from 1 nm to 15 nm by varying the concentration of the U3 solution from 2 mg mL-1 to 10
mg mL-1. Excess material is swabbed away from the edges of the substrate in order to ensure good
electrical contact to the ITO electrodes. All of the subsequent processing and testing steps are conducted
under rigorously air-free conditions. The MoO3 hole transporting layer is thermally sublimed at a rate of
~0.1 nm s-1 and at a base pressure of 1×10-6 Torr. The 25 nm ITO top-contact is deposited via RFsputtering in an Argon atmosphere of 5 mTorr at a rate of 0.005 nm s-1. A low sputtering power of 7 W
was used in order to minimize damage to the MoO3 and the U3 layers, as described elsewhere [18, 16].
Finally, a silver mirror (200 nm thick) is sputtered on top of the ITO at a rate of 0.05 nm s-1 to complete
the device.
Determination of Frontier Energy Levels with Cyclic Voltammetry
Electrochemical measurements on solutions of U3 dissolved in either ultrapure water (monomeric
U3) or in 0.01M aqueous NaCl (J-aggregated U3) were made with an Autolab PGSTAT 20 potentiostat
(EcoChemie) using a quasi-internal Ag wire reference electrode (BioAnalytical Systems) submersed in
0.1M aqueous NaCl. Cyclic voltammograms were recorded using a platinum button electrode as the
working electrode and a platinum coil counter electrode. The ferrocene/ferrocenium (Fc/Fc+) redox
couple in dichloromethane was used as an external reference. A quasi-reversible oxidation peak and an
irreversible reduction peak are observed (see Fig. S1). The half potential of the first oxidation peak is
measured to be 1.04 V vs. Ag/AgCl, or equivalenly 0.64 V vs. Fc/Fc+ (the ferrocene redox couple occurs
at 400 mV vs. Ag/AgCl). To convert this oxidation potential to the highest occupied molecular orbital
(HOMO) energy level referenced to the vacuum level, we employ the linear relationship between
oxidation levels established by cyclic voltammetry and HOMO levels obtained from ultraviolet
photoemission spectroscopy that was established in ref. 1. We arrive at a HOMO level of 5.5 ± 0.1 eV.
In order to determine the lowest occupied molecular orbital (LUMO), we first use the empirical relation
described in ref. 2 to relate optical bandgap (corresponding to the optical transition at λ = 790 nm) to the
transport bandgap (at energy Et) of U3. We then add Et to the HOMO level. Et is calculated to be 1.7 ±
0.6 eV and the LUMO level is calculated to be 3.8 ± 0.7 eV.
Figure S1. Cyclic voltammogram of J-aggregated U3 in 0.01 M aq. NaCl, measured with a platinum
button electrode and a Ag/AgCl reference electrode at a scan rate of 100 mV/s under ambient conditions.
Modeling J-V Characteristics to Ideal Diode Equation
J-V characteristics were fitted to the generalized Shockley equation3, given below:
𝐽=𝑅
𝑅𝑝
𝑠 +𝑅𝑝
{𝐽𝑠 [exp (
𝑒(𝑉−𝐽𝑅𝑠 )
)−
𝑛𝑘𝐵 𝑇
𝑉
1] + 𝑅 }
𝑝
(1)
Js is the reverse-bias saturation current, n is the diode ideality factor, and Rs and Rp are the series and shunt
resistances of the diode.
Figure S2. J-V characteristics for devices consisting of U3 films of different thicknesses. The MoO3
thickness is 60 nm. Fits to the ideal diode equation are shown in black.
Table I. Parameters extracted from fits to the ideal diode equation for devices consisting of U3 films of
different thicknesses.
U3 Thickness
Rp
Rs
Js
n
[nm]
[Ω cm2]
[Ω cm2]
[A cm-2]
15.0
3793900
28
1.4×10-7
2.9
10.5
607650
11
1.1×10-7
2.4
8.1
35083
5
1.5×10-6
2.8
4.0
41940
17
7.7×10-7
3.0
1.0
338
3
2.7×10-7
1.2
Measurement of Absorption and Internal Quantum Efficiency
Optical reflection was measured with a Varian Cary 5000 UV-Vis-NIR Spectrophotometer for
structures identical to the device structure shown in Fig. 1 and also in control structures in which the U3
film was replaced by a 20 nm-thick film of sputter-deposited Al2O3 (index of refraction of n = 1.76 at λ =
790 nm). The thickness of the Al2O3 layer is selected to approximately match its optical path length with
that of the U3 film at its absorption peak. This ensures that, apart from its lack of optical absorption, the
Al2O3 layer will mimic the presence of the U3 film so that non-active layers will be subjected to nearly
identical optical fields in both structures. By taking the difference between the magnitudes of the optical
absorption in these structures, the fraction of light absorbed within the U3 film can be determined. The
MoO3 thickness is varied in these structures in order to determine the effect of optical spacer thickness on
total absorption within the U3 film. For a MoO3 thickness of 68 nm, only 11% of the light is reflected at
the absorption peak (λ = 766 nm), corresponding to 89% absorption within the structure, as shown in Fig.
S3(a). In the control structure containing a thin film of Al2O3, however, we measure reflection of 70%,
indicating that 30% of the light impinging on the device at this wavelength is being reflected, scattered, or
absorbed in the metal oxide layers or silver mirror. Subtracting this loss from the absorption measured in
the structure containing U3, we find that 59  3% of the light impinging upon the device is absorbed
within the 8.1  0.3 nm-thick dye film. We note that the FWHM of this absorption curve (84.9 nm) is in
good agreement with the FWHM of the EQE spectrum measured for a comparable device (85.6 nm).
Fig. S3(b) shows the peak absorption (black squares) as well as the internal quantum efficiency
(red circles), as a function of MoO3 thickness. Internal quantum efficiency (IQE) is calculated by
dividing the EQE measured in completed photodetector devices by the fraction of absorbed light
determined as described above. The peak in IQE of 25.8  3 % suggests that there is a substantial loss of
photocurrent in the device. This loss of photocurrent is attributed to fast non-radiative recombination
processes in the U3 film that result in a low exciton diffusion length, LD. Photo-generated excitons
generated too far from the narrow photoactive region of the film will recombine rather than contribute to
photocurrent.
Figure S3. (a) Optical absorption calculated from 1-R in structures consisting of spun-cast U3 and
sputter-deposited Al2O3 (reference) films. Here, R is the measured percentage of reflected light. (b)
Absorption and internal quantum efficiency in structures consisting of 8.1 nm thick U3 films, for different
MoO3 thicknesses.
Optical Constants for U3
Figure S4. Optical constants for spun-cast U3 films derived from spectroscopic ellipsometry.
Optical Modeling of the EQE Spectrum
The considerable broadening of the EQE peak relative to the absorption spectrum of a neat film of U3
(Fig. 2(b)) can be explained by optical interference effects within the device. T-matrix modeling of the
structure reveals that the shape of the EQE curve is highly sensitive to the particular thickness of the U3
film and its position within the cavity (see Fig. S5). The normalized absorption of a U3 film (1-T-R), the
experimental EQE, and EQE data simulated using a T-matrix model are shown in Fig. S6. The broad
EQE feature that is observed experimentally is clearly predicted using this model.
Figure S5. (a) Normalized simulated EQE for devices with the following structure: ITO(120
nm)/ZnO(50 nm)/U3(X)/MoO3(60 nm)/ITO(25 nm)/Ag. The U3 thickness, X, ranges from 2 nm or 10
nm. (b) Simulated FWHM versus U3 thickness.
Figure S6. Normalized absorption (1-T-R), simulated EQE, and experimental EQE. The simulated EQE
curves are for devices with the following structure: ITO(120 nm)/ZnO(50 nm)/U3(X)/MoO3(60
nm)/ITO(25 nm)/Ag. The U3 thickness, X, is 1.05 nm or 10.5 nm.
Dependence of ZnO Thickness
The thickness of ZnO was varied in order to determine its effect on the performance of the device. EQE
spectra for devices with ZnO thicknesses of 35 nm, 50 nm and 65 nm are shown in Fig. S7. The U3
thickness in these structures was 8.1 nm. Very little variation in the peak performance is observed.
Figure S7. External quantum efficiency as a function of ZnO thickness.
Capacitance Measurements
The capacitance of photodetector structures was measured with a Solartron 1260 Impedance Analyzer at a
frequency of f = 37 Hz. Capacitance at zero-bias for a device consisting of an 8.1 nm thick film was
measured to be 112 nF cm-2. The roll-off associated with the RC time constant of the structure is
calculated to occur at fRC=1.6 MHz, suggesting that the response speed in the present case is limited by
the slow diffusion of carriers or excitons within the structure.
References:
[1]
B. W. D'Andrade, S. Datta, S. R. Forrest, P. Djurovich, E. Polikarpov, M. E. Thompson, Org.
Electron. 2005, 6, 11.
[2]
P. I. Djurovich, E. I. Mayo, S. R. Forrest, M. E. Thompson, Org. Electron. 2009, 10, 515.
[3]
R. H. Bube, A. L. Fahrenbruch, Eds., Advances in Electronics and Electron Physics, Academic,
New York 1981.