advances.sciencemag.

org/cgi/content/full/7/23/eabe4206/DC1

Supplementary Materials for

Strongly enhanced and tunable photovoltaic effect in

ferroelectric-paraelectric superlattices

Yeseul Yun, Lutz Mühlenbein, David S. Knoche, Andriy Lotnyk, Akash Bhatnagar*

*Corresponding author. Email: akash.bhatnagar@physik.uni-halle.de

Published 2 June 2021, Sci. Adv. 7, eabe4206 (2021)

DOI: 10.1126/sciadv.abe4206

This PDF file includes:

Figs. S1 to S8
 A B
SB55 BC55
7.939 C 4.10
STO
a

Lattice parameter (Å )
4.05 c'
7.74
qz (Å -1)

* * 4.00

7.542
3.95

7.344 3.90

BTO SB55 BC55 SBC555

-2.763 -2.565 -2.368 -2.763 -2.565 -2.368
qx (Å -1) qx (Å -1)

Fig. S1. Structural characterization of two-color superlattices. Reciprocal space

mapping around (103) plane of (A) SB55 and (B) BC55. Star and yellow arrows indicate

STO substrate and satellite peaks from SL, respectively. (C) Lattice parameters of single,

2- and 3-color SLs. Solid line indicates the lattice parameter of STO substrate.
 A B
-5 RT
10 RT
Current density (A/cm2)

Current density (A/cm2)

-5
10
10-6

10-7 10-6 77 K
77 K

10-8
10-7

10-9
-0.1 0.0 0.1 -0.1 0.0 0.1
Voltage (V) Voltage (V)
C D
RT RT
Current density (A/cm2)

Current density (A/cm2)

10-5

77 K 10-5
-6
10

10-7
10-6
77 K
10-8

10-7
10-9
-0.1 0.0 0.1 -0.1 0.0 0.1
Voltage (V) Voltage (V)

Fig. S2. Temperature dependent photo IV characteristics. Current-voltage (IV)

characteristics measured with 3.06 eV of (A) BTO, (B) SB55, (C) BC55 and (D) SBC555.
The dip in the IV curve represents open circuit voltage (VOC ). As the temperatures are

lowered, the magnitude of VOC gradually increases and eventually goes beyond the range
of voltage used for the photoelectrical measurements. The IV curves therefore appear to
be much more linear at lower temperatures.
 A B
400 Cycle 1 90
Cycle 2
Cycle 3
Amplitude (pm)

300
45

Phase ()
200 0

100 -45 Cycle 1

Cycle 2
0 -90 Cycle 3

-6 -4 -2 0 2 4 6 -6 -4 -2 0 2 4 6
Bias (V) Bias (V)

Fig. S3. Piezoresponse from superlattice. (A) Amplitude-Voltage and, (B) Phase-

Voltage characteristics acquired with PFM from sample SBC222 at room temperature.

The measurements were conducted with the PFM in DART (dual amplitude resonance

tracking) mode.
A 500 SBC
B 1.5 SBC
Laser 555 Laser 555
0
ON OFF ON ON OFF ON
0.0
Current density (mA/cm2)

Current density (mA/cm2)

50 0.4

ON OFF ON BC55 ON OFF ON BC55

0 0.0

10 0.5

ON OFF ON SB55 ON OFF ON SB55

0 0.0

0.03
0.5 BTO BTO
OFF
ON OFF ON 0.00
0.0 ON ON
-0.03
0 30 60 90 120 0 30 60 90 120
Time (s) Time (s)

Fig. S4. Photo-response measured with time. Current-time response acquired

with the illumination ON and OFF at (A) RT and (B) 77K.

 A C
θ
[001]
Half-wave
plate
θ

[100]
SL film SL film
Nb:STO [010] Nb:STO

B D
1.1 1.1
BPV Responsivity (a.u.)

1.0 BPV Responsivity (a.u.) 1.0

0.9 0.9

0.8 0.8
0 90 180 270 360 0 90 180 270 360
Light polarization (°) Light polarization (°)

Fig. S5. The angle-dependent photovoltaic current of SBC222. (A) Schematic

of the measurement geometry with illumination along the direction of [001] and, (B)

the resultant BPV responsivity. The vaules are normalized to the maximum value. (C)

Schematic of the measurement geometry with light source tilted by 45° and, (D) the

resultant BPV responsivity. The BPV responsivity was subtracted from the obtained and

normalized PV current and absorbed intensity of light.

The linear BPV current can be written as:

Ji = Iβijk ej ek (S1)

where I is the intensity of the light, βijk is the third-rank BPV tensor and ej and ek are

the components of the light polarization vector. The space group of BTO is P4mm and
corresponding third-rank tensor is
 
0 0 0 0 β15 0
βijk = 0 0 0 β15 0 0 (S2)
β31 β31 β33 0 0 0

When the propagation direction of the incident light is along [001], the resultant BPV

current will be
cos2 θ
 
2
 sin θ 
   
0 0 0 0 β15 0  0  0
Ji = I 0 0 0 β15 0 0  0 =I 0  (S3)
    
β31 β31 β33 0 0 0  0  β31
 
1
 
sin 2θ
2
wherein θ represents the angle which light makes with the direction of current flow. It

becomes apparent from the above equation that the resultant photocurrent flows only

along [001] and is independent of θ, which is also consistent with the experimental result

(Fig. S5B).
 A B
1.8

Current density (mA/cm2)

Current density (A/cm2)

10-3
1.6

1.4
10-4
9th Oct. 2019 9th Oct. 2019
30th Mar. 2020 30th Mar. 2020
1.2
-5
10

-0.1 0.0 0.1 -0.1 0.0 0.1

Voltage (V) Voltage (V)

Fig. S6. Long-term photocurrent reproducibility in SBC222. IV characteristics

measured with 3.06 eV at (A) RT and (B) 77K.

 1200

Relative permittivity (r)

1000

SBC222
800

600

400

BTO
200

-6 -4 -2 0 2 4 6
Voltage (V)

Fig. S7. Relative permittivity measured as a function of dc bias. The measure-

ments are performed at 100 kHz with a 100 mV ac signal.

 A

Fig. S8. Tetragonality of BTO and within superlattice. Evaluation of pro-

jected lattice distances in (A) BTO and (B) SBC222 SL. (Left) Atomic-resolution STEM-
HAADF image. The inset show magnified images with unit cells marked by rectangles.

(Right) Structure models derived from the experimental HADDF image. The underlined

italic values show lattice distances extracted from theoretical structure. The values in ta-
ble represent averaged interatomic distances calculated using at least 1000 measurement

points for a particular crystallographic axis. The STDW is the standard deviation.