M. C. Putnam, S. W. Boettcher, M. D. Kelzenberg, D. B. Turner
-Evans, J. M. Spurgeon, E. L. Warren,
R. M. Briggs, N. S. Lewis, and H. A. Atwater,
Si Microwire
-Array Solar Cells
, 2010.
1
Si Microwire
Array Solar Cell
s Supporting Infor
mation
Ag Back Reflector:
Figure
S1 provides
scanning electron microscope
(SEM ) images that document the fab
rication of a
Ag back reflector
. Follo wing two 500 nm Ag
evaporations
Ag uniformly coat
ed the substrate
and the
wire sidewalls (Fig. S
1A). PDMS was then
deposited
and continuously coated the Ag
‐coated
substrate
(Fig. S
1A,B ). (Because the SEM images
shown are from the edge of a
wire array, the
PDMS is thinner than in the center of the wire
array and
there exists a small area at the
immediate
wafer edge where no PDMS coating
exists.
) A Ag etch
was then used to remove any
Ag that was
not protected by the PDMS film at the
base of the wire array (Fig. S
1B). After PRS cell
fabrication, t
he PDMS
‐protected Ag
back reflector
was revealed by cell cro
ss‐sectioning
(Fig. S
1C).
Separately, the textured nature of the mounting
wax , which
results from the presence of the Al
2
O
3
scattering particles
, was visible
above the
protective PDMS layer (Fig.
S1C ).
a
SiN
x
:H Layer
:
Figure
S2 is an SEM image of a wire array after
selective removal of the
a
‐SiN
x
:H layer from the
wire tips. The bright tip is the c
‐Si wire, while the
darker base is the
a
‐SiN
x
:H‐coated c
‐Si w ire. The
difference in the extent of the exposed tip relates
to vari
ations in the wire height and variations in
the height of the mounting wax etch barrier
(removed prior to imaging.)
Figure S1. Tilted scanning electron microscope (SEM)
images illustrating the fabrication of a Ag back
reflector. A) SEM ima
ge post Ag and protective
polydimethylsiloxane (PDMS) deposition. B) SEM image
of the wire array from A) after a Ag
‐etch. C) Cross
‐
sectional SEM image of a PRS microwire solar cell.
Supplementary Material (ESI) for Energy & Environmental Science.
This journal is © Royal Society of Chemistry 2010
M. C. Putnam, S. W. Boettcher, M. D. Kelzenberg, D. B. Turner
-Evans, J. M. Spurgeon, E. L. Warren,
R. M. Briggs, N. S. Lewis, and H. A. Atwater,
Si Microwire
-Array Solar Cells
, 2010.
2
Figure S2. SEM image of a wire array after selective
removal of
a
‐SiN
x
:H from the wire tips. The mounting
wax, which was used as an etch barrier, has been
removed from the wire array for clarity.
Cell Area
:
Scanning photocurrent microscopy (SPCM)
image
s ( 90
μ
m x 90
μ
m) ( Fig. S
3A) were overlaid
to produce a
photocurrent
map of the cell
perimeter (Fig. S
3B), which was then analyzed to
calculate the cell
area (Fig. S3C
).
Area analysis was performed using the
‘thresholding’ feature in Image J. Thresholding
was done in such a way that all
of the
wires
withi
n the cell perimeter (defined by the
photoactive wires) were selected.
T
he indent on
the left side of the cell resulte
d from contact
shadowing and an appropriat
e correction to the
cell area was
made. A small photocurrent signal
was present outside of the cell perimeter (Fig.
S3A) and is presumed to arise from light that was
scattered
/reflected
into the active area.
Thou
gh
this additional collection area was
accounted for
during the thresholding process, no correction
should
have been
necessary given that an
equivalent amount of light would have also been
scattered/reflected out of the cell.
As discussed in the text t
he dark spots
(Fig.
S3A,B ) indicate wires that are not electrically
contacted by the indium tin oxide (ITO).
Comparing Fig. S
3B with Fig.
4C, the fraction of
electrically inactive wires
was higher near the c
ell
perimeter
(2 ‐20%), which is not unexpected
given the decreased ITO thickness at the device
edge.
Figure S3. Measuring PRS C4R5’s active area.
A) 90
μ
m x 90
μ
m scanning photocurrent
microscopy (SPCM) image along the cell
perimeter. B) Twenty
‐six SPCM images ove
r‐
laid to map out the cell perimeter. C) Image of
B) after thresholding. The blue line is the cell
perimeter from which the cell area was
calculated.
Supplementary Material (ESI) for Energy & Environmental Science.
This journal is © Royal Society of Chemistry 2010
M. C. Putnam, S. W. Boettcher, M. D. Kelzenberg, D. B. Turner
-Evans, J. M. Spurgeon, E. L. Warren,
R. M. Briggs, N. S. Lewis, and H. A. Atwater,
Si Microwire
-Array Solar Cells
, 2010.
3
Table S1.
V
oc
and
FF
(All Devices)
As
-
Grown:
V
oc
(mV)
FF
(%)
C4R2
401
59.3
C4R3
209
44.9
C4R4
452
61.4
C4R5
257
42.2
C4R6
478
59.1
C3R2
419
43
C3R3
339
52
C3R4
474
66.2
C3R5
453
65.8
C3R6
485
68.4
C2R3
482
69.4
C2R4
492
70.1
C2R5
484
71.6
C2R6
429
59.1
C1R6
463
54.4
V
oc
and
FF
:
As seen in Table S1 above, the
V
oc
and
FF
were
remarkably consistent for the PRS solar cells. The
V
oc
and
FF
were also consistent between the best
Scatterer and As
‐Grown solar cells, however some
cells with lower
V
oc
and
FF
were observed. For the
As‐Grown solar cells, obvious fabrication defects
(cracking of the mounting wax prior to ITO
deposition) may have re
sulted in the larger
variation in cell performance.
Between cells with
similar performance (within each respective cell
type), we attribute much of the variation in
FF
to
the observed variations in the probe tip to ITO
contact resistance.
Scatterer
:
V
oc
(mV)
FF
(%)
C1R1
477
61.7
C2R1
429
54.8
C3R1
387
53.5
C4R1
475
61.4
C1R2
498
67.5
C2R2
503
68.6
C3R2
481
54.3
C4R2
475
65.1
C1R3
497
64.9
C2R3
486
60.4
C3R3
505
68.8
C2R4
499
68
PRS:
V
oc
(mV)
FF
(%)
C2R1
491
59.3
C3R1
487
61.2
C4R1
488
59.7
C5R1
485
61.9
C2R2
497
61
C3R2
493
60.8
C4R2
495
61.1
C5R2
489
60
C2R3
499
63.3
C3R3
497
63
C4R3
495
62.9
C5R3
493
61.5
C2R4
504
62.6
C3R4
494
64.5
C4R4
502
62.5
C5R4
501
61.5
C2R5
503
66.1
C3R5
500
67.2
C4R5
498
65.4
C5R5
497
62.6
C2R6
502
63.4
C3R6
499
63.3
C4R6
489
61
C5R6
485
64.3
Supplementary Material (ESI) for Energy & Environmental Science.
This journal is © Royal Society of Chemistry 2010
M. C. Putnam, S. W. Boettcher, M. D. Kelzenberg, D. B. Turner
-Evans, J. M. Spurgeon, E. L. Warren,
R. M. Briggs, N. S. Lewis, and H. A. Atwater,
Si Microwire
-Array Solar Cells
, 2010.
4
Indium Tin
Oxide:
Figure S4
plots the transmission as a function
of
wavelength for a glass cover
slip with and without
a 150 nm
‐thic k indium tin oxide (ITO) layer.
Transmission through the ITO was found to b
e >
80% for wavelengths > 500 nm, and at least
65%
for wave
lengths between
400 and 500
nm.
Strong oscillations in transmission were observed
as a result of Fabry
‐Pérot interference. Thus, a
5
nm running
average was used to smooth the
oscillations in transmission for wavelengths > 700
nm. As can be seen by comparing the smoothed
data below, the oscillations were inherent to the
thin nat
ure of the glass cover
slip.
Fig. S4.
Transmission as a function
of wavelength for a
glass cover
slip with and without a 150 nm
‐thick indium
tin oxide coating. A 5 nm running average was applied
to smooth
the oscillations in transmission at
wavelengths greater than 700 nm.
Supplementary Material (ESI) for Energy & Environmental Science.
This journal is © Royal Society of Chemistry 2010
M. C. Putnam, S. W. Boettcher, M. D. Kelzenberg, D. B. Turner
-Evans, J. M. Spurgeon, E. L. Warren,
R. M. Briggs, N. S. Lewis, and H. A. Atwater,
Si Microwire
-Array Solar Cells
, 2010.
5
Experimental:
Wire Array Growth
. Si microwire
arrays were grown as described previously.
5
The
growth substrates were boron
‐doped p
++
‐Si (111) wafers, having a resistivity,
ρ
< 0.001
Ω
·cm, that were coated with 450 nm of thermal oxide
(Silicon Quest International). Arrays
of 4 ‐μm‐diameter circular holes, on a square lattice with a 7 μm pitch, were defined in the
oxide by photolithographic exposure and development of a photoresist layer (Microchem
S1813), followed by a buffered HF(aq) (
BHF) etch (Transene Inc.) The holes were then
filled with 600 nm of Cu (ESPI metals, 6N) via thermal evaporation onto the patterned
photoresist, followed by liftoff. Patterned substrates approximately 1.5 cm × 1.5 cm in
dimension were then annealed in a
tube furnace for 20 min at 1000 °C under H
2
flowing at a
rate of 500 sccm. Wire growth was performed by the introduction of SiCl
4
(Strem,
99.9999+%), BCl
3
(Matheson, 0.25% in H
2
), and H
2
(Matheson, research grade) at flow rates
of 10, 1.0, and 500 sccm, r
espectively, for 30 min. Following growth, the tube was purged
with N
2
at 200 sccm and was allowed cool to ~ 650 °C over the course of ~30 min.
pn
Junction Fabrication.
Following growth the Cu catalyst was removed from the wire
arrays by etching in 5%
HF(aq) for 30
s, 6:1:1 by volume H
2
O:H
2
O
2
(30% in H
2
O):conc. HCl
(aq.) at 75
°C for 15
min, and 20
wt % KOH (aq.) at 20
°C for 60
s. A conformal SiO
2
diffusion
‐barrier that was 200
nm in thickness was grown via dry thermal oxidation at
1100
°C for 2
h. Th
e wire array samples were then coated with a solution that contained
4.4 g hexamethycyclotrisiloxane (Sigma
‐Aldrich), 1
g PDMS (Sylgard 184, Dow Corning),
and 0.10
g of curing agent in 5
ml of dicholoromethane; spun at 1000
RPM for 30
s; and
cured at 150
°C for 30
min, to produce a 10
–20 μm thick PDMS layer selectively at the base
of the wire array.
6
After a quick etch (~2 s) in a 1:1 mixture of 1.0
M tetrabutylammonium
fluoride in tetrahydrofuran (Sigma
‐Aldrich) and dimethylformamide (PDMS etch)
26
and a D
I
rinse, these partially infilled arrays were immersed for 5 min in BHF, to remove the exposed
diffusion
‐barrier oxide. The PDMS was then completely removed by etching for 30
min in
PDMS etch. A 10
min piranha etch (3:1 aq. conc. H
2
SO
4
:H
2
O
2
) was performe
d to remove
residual organic contamination. After etching the wires for 5
s in 10% HF (aq), thermal P
diffusion was performed using solid source CeP
5
O
14
wafers (Saint
‐Gobain,
PH‐900 PDS
) at
850 °C for 10 min (As
‐Grown and Scatterer) or 15 min (PRS) under
an N
2
ambient, to yield a
Supplementary Material (ESI) for Energy & Environmental Science.
This journal is © Royal Society of Chemistry 2010
M. C. Putnam, S. W. Boettcher, M. D. Kelzenberg, D. B. Turner
-Evans, J. M. Spurgeon, E. L. Warren,
R. M. Briggs, N. S. Lewis, and H. A. Atwater,
Si Microwire
-Array Solar Cells
, 2010.
6
radial
p‐n junction in the wire regions unprotected by the thermal oxide. A 30
s etch in BHF
was used to remove the surface dopant glass.
Photovoltaic
Device Fabrication.
The
AsGrown
cell was fabricated as follows. After p
‐n
junction fabrication, the wire array was heated to 150
°C on a hot plate, and mounting wax
(Quickstick 135, South Bay Tech.) was melted into the array. Excess wax was removed from
the array using a glass coverslip. The mounting wax was then etched in an
O
2
plasma (400
W, 300 mTorr) until the wire tips were sufficiently exposed for electrical contacting (30
‐90
min). After etching with BHF for 30 s, 150 nm of indium tin oxide [0.0007
Ω
·cm] was
sputtered (48 W, 3 mTorr, 20:0.75 sccm Ar:10% O
2
in Ar) throug
h a shadow mask, to serve
as a transparent contact to the n
‐type shell of the Si microwires, thereby defining the area of
the microwire solar cells. Contact to the p
‐type core of the Si microwires was established
through the p
+
‐Si substrate by scribing a
Ga/In eutectic onto the back side of the growth
wafer.
Fabrication of the
Scatterer
cell was performed identically to that of the As
‐Grown cell,
except that prior to infilling with wax, Al
2
O
3
light ‐scattering particles (0.08 μm nominal
‐
diameter, South Bay
Technology) were added to the wire array. The wire
‐array was placed
face ‐up in a flat
‐bottomed glass centrifuge tube and ~ 3 ml of an ethanolic dispersion of the
particles (~0.3 mg/ml) were added. Centrifugation (~3000 RPM) for 5 min was used to
drive t
he particles to the base of the wire
‐array.
Fabrication of the
PRS
cell was performed identically to that for the Scatterer cell, except
that prior to the addition of the Al
2
O
3
particles, an
a
‐SiN
x
:H passivating layer and a Ag back
reflector were added t
o the cell.
After p
‐n junction fabrication, the
wire arrays were etched
for 5 min in BHF, to completely remove the remaining oxide diffusion barrier. A standard
clean was then performed (10 min in 5:1:1 by volume H
2
O:H
2
O
2
(30% in H
2
O): NH
4
OH(15%
in H
2
O) at 75
°C, 30 s in BHF, 10 min in 6:1:1 by volume H
2
O:H
2
O
2
(30% in H
2
O):conc. HCl
(aq.) at 75
°C, 30 s in BHF), prior to deposition of an
a
‐SiN
x
:H layer (~ 140 nm thick at the
wire tip and ~ 60 nm thick at the wire base) using plasma
‐enhanced chemical vapor
deposition, as described previously.
1
The
a
‐SiN
x
:H was then etched for 15 s in BHF, prior to
the deposition of a total of 1
μ
m planar
‐equivalent of Ag via thermal evaporation (two
successive 500 nm evaporations at two different specimen
‐tilt angles (± ~5
degrees) with
Supplementary Material (ESI) for Energy & Environmental Science.
This journal is © Royal Society of Chemistry 2010
M. C. Putnam, S. W. Boettcher, M. D. Kelzenberg, D. B. Turner
-Evans, J. M. Spurgeon, E. L. Warren,
R. M. Briggs, N. S. Lewis, and H. A. Atwater,
Si Microwire
-Array Solar Cells
, 2010.
7
sample rotation, to ensure continuous coverage of the growth substrate). The array was
then infilled with
~5 μm of PDMS using a process similar to the one described above.
This
PDMS etch barrier allowed the Ag at the wire tips and sidewalls
to be selectively removed
by etching for 6.5 min in 8:1:1 methanol: NH
4
OH(15% in H
2
O): 30 wt.% aq. H
2
O
2
. A thin
layer (~40 nm) of SiO
2
was then sputtered to improve the incorporation of the Al
2
O
3
particles. The Al
2
O
3
scattering elements, mounting wax, a
nd ITO were then added as
described above.
Characterization.
Dark and light current
‐voltage measurements were performed on a
probe station with a 4
‐point source
‐measure unit (Keithley 238). Contact to the ITO top
contact was made with a micromanipulator
‐controlled Au
‐coated tungsten probe tip.
Simulated solar illumination was provided by a 1000 W Xe arc lamp with air mass (AM
1.5G) filters (Oriel), calibrated to 1
‐sun illumination by an NREL
‐traceable Si reference cell
(PV Measurements, Inc.). Spectral
response measurements were performed in an overfilled
geometry using chopped (30 Hz) illumination from a 300 W Xe arc lamp coupled to a 0.25 m
monochromator (Oriel) that provided ~2 nm spectral resolution.
The specimen
photocurrent was normalized (by area) to that of a 3 mm
‐diameter calibrated photodiode,
to determine the external quantum yield.
The signals were measured with independent
lock ‐in detection of the sample and calibration channels. Scanning ph
otocurrent
microscopy measurements were performed using a confocal microscope (WiTEC) in a light
‐
beam
‐induced current (LBIC) configuration described previously.
9
Scanning photocurrent
microscopy (SPCM) images were formed by rastering each device beneath a
~1.0 μm
‐
diameter laser spot (λ = 650 nm) while recording the short
‐circuit current (0 V bias) under
otherwise dark conditions. Multiple 90
μ
m x 90
μ
m SPCM images were manually stitched
together and post
‐processed to determine the active cell area using i
mage processing
software (Image J) (see Supporting Information Fig. S3.)
Supplementary Material (ESI) for Energy & Environmental Science.
This journal is © Royal Society of Chemistry 2010