Reviewers' comments:
Reviewer #1 (Remarks to the Author):
This paper reported efficient light
-generated production of hydrogen on InP
-Rh photoelectrodes
in
microgravity environment by nanostructured catalyst. And the behavior of the applied system in
terrestrial and microgravity environment is simulated using a kinetic transport model. The
microgravity experiment adds valuable information to the solar hyd
rogen generation research field.
However, it would be better if the questions as below could be addressed before publication.
1. The whole theory is based on the assumption that the formation of a forth layer by the evolved
hydrogen bubbles on the electr
ode surface that hinders mass transfer. However, how to quantify the
forth layer remains unclear. Simply judging by eye, under microgravity condition there are also a lot
of bubbles attach to the surface of the nanostructured photoelectrode. The photocurre
nt of the thin
film sample was reduced by 4 times, but it seems only a small portion of the electrode surface was
covered by bubbles under microgravity. Is there a better way to characterize the properties
(thickness? coverage?) of the forth layer?
2. Wh
y is the open
-circle voltage of the nanostructured photoelectrodes greatly (~100mV) improved
under microgravity environment, as shown in Fig. 2a?
3. Comparing the fabrication processes of thin film and nanostructured photoelectrode, it seems that
the nanostructured sample has an additional O2
-plasma treatment process. This probably explains
why the InOx/POx signals are stronger in the nanostructured sample. I am wondering if the O2
-
plasma treatment would change the surface chemistry of the photoelectrode that affects its
microgravity behavior? To confirm/exclude this effect, O2
-plasma treated thin film photoelectrode
should also be tested.
4. AFM images in Figure 3c should be given in the same scale for easy comparison.
5. Page 10, the open circuit vol
tage in SI Fig. 5b is only reduced by 50 mV, not 0.5 V.?
6. Electrochemical impedance spectroscopy (EIS) of the photoelectrodes in microgravity environment
could be provided useful information on mass trasportation. However, is that technically challenging
given that the free fall time is only 9.3 s?
7. Chronoamperometry that capture the whole free fall process (before-
during
-after) should give some
interesting dynamic results.
8. Equation 1, jL was not difined.
9. Page 13, NSL should be SNL.
10.
The addition of IPA to the HClO4 electrolyte was not mentioned in the experimental section. What
is the purpose of adding IPA? Would that change the PEC properties of the photoelectrodes?
11. Since pressure also plays a significant role in the bubble evo
lution, is the PEC cell under ambient
pressure during the experiment?
Reviewer #2 (Remarks to the Author):
This work is appropriate to be publish in the Nature Communications Journal but needs few re
arrangements.
The bibliography is rather complete concerning PhotoElectrochemistry but sufficient concerning
general electrochemistry for hydrogen production under zero gravity.
In my opinion, 2 references might be added:
Y. Fukunaka, T. Homma, R. Hagiwara, T. Nohira, K. Hachiya, T. Matsuoka, Yunfeng L
iang, T. Goto, H.
Yasuda, H. Matsushima, N. Kishimoto, T. Ito, Y. Takahashi, K. Kinoshita, M. Takayanagi, Y. Sone, T.
Ishikawa, T. Wakatsuki, K. Nishikawa, S. Yoda, M. Rosso, R. C. Alkire, W. Schwarzacher, O.
Magnussen, S. Kjelstrup, Ph. Mandin, D. R. Sado
way, R. C. Miranda, Non
-equilibrium Electrochemical
Processing of Nano
-structured Energy Conversion & Storage Devices , Space Utiliz. Res., 27, 227-
230
(2011)
Z. Derhoumi, Ph. Mandin, H. Roustan and, R. Wuthrich, Experimental investigation of two-
phase
electrolysis processes: comparison with or without gravity, J Appl Electrochem (2013) 43:1145
–1161
In a concrete way:
Page7: Figure 2a and b are not clear.
Figure2a is not readable: the legend and caption for the 4 different cases :1 g, 10-
6g, thin film
,
nanostructured is not clear at all.. It seems that different shade of greys might allow in your opinion a
easy reading? In fact With figure 2a I deduce that there is no influence of zero gravity on
performance.. but I must be wrong?
Concerning Fig2b, t
he 1g thin film photograph is not planar? Difficult to see anything with these
figures.
Page8: The influence of Surface Nanotopography might be given first in my opinion...
Generally it is clearer to explain the experimental set up and the design of exper
iments. In this paper,
the experimental set
-up is given after...
Figure3b is really strange for me: the line “thin film” gives information at “500 nm scale” whereas the
line “nanostructured” gives information at “2 and 1 μm”.. I might have think that it sh
ould be the
contrary? Am I wrong?
Page9: you pass directly from fig3 to fig5 instead of the fig4 introduction? Why?
At the end of Page9 “Output characteristic simulation” it is introduced an other experimental set-
up (in
a numerical part?)... no
scheme is given?
Page10-
11: I don’t know if it is my own pdf editor but all equations (1), (2), (3), (4) page20 ...were
not correctly printed
Page12: Figure4 has the same problem as Figue2: no clear legend and caption..
Impossible to determine the 4 c
ases with only grey shades... In text you speak of yellow or cyan lines
which are not allowed with usual grey scale printers.
I think these simulations should have been compared with experimental results of figure 2..But not in
a clear way: difference theo
ry/experiments...?
Page15: after the part “conclusion” arrives a big experimental set
-up “materials and methods” part..
I think one part might be put at the beginning and perhaps the rest in a correctly named “Annexe”...?
Reviewer #3 (Remarks to the Aut
hor):
The manuscript “Efficient Solar Hydrogen Generation in Microgravity Environment” explores how
microgravity conditions affect the photoelectrochemical performance of InP/Rh photocathodes during
Hydrogen evolution reaction. Concretely, the authors ob
served that flat films were highly sensitive to
the microgravity conditions experiencing a drastic decrease in photocurrent and photovoltage, whereas
the performance of nanotextured films remained unaltered. As the authors highlighted in the
introduction,
several studies have addressed and examined the performance of electrolyzers under
these so
-called microgravity conditions, but none of them dealt with photoelectrodes for direct water
splitting. These reports indicated that the lack of buoyancy in these c
onditions was preventing from
achieving a performance close to that obtain in “standard conditions” given the limitations in mass
transfer imposed in the surface of the electrode when the bubbles of evolving gas blocked the catalytic
surface. Obviously, wh
en moving from “dark electrocatalyst” (electrolyzer) to “photoelectrodes” the
authors found that, again, the lack of buoyancy in microgravity was the main responsible for the
decreased performance, although the authors offer an interesting approach to addr
ess this known issue,
nanotexturing the catalytic surface to promote the desorption of the gas bubbles, and beside they
develop a simple model to simulate the results. The manuscript is well-
written and structured, and
provides an interesting route to addr
ess the formation of “froth layers” on the catalytic surface under
microgravity. The novelty and impact of this work certainly deserves further consideration for
publication in Nature Communications after a minor revision:
- In Figure 2b the authors incl
uded some pictures to demonstrate that the formation of a “froth
layer” is the main cause for the detrimental performance in the “thin film” with respect to the
“nanostructured”. However, if we have to consider as a direct proof the pictures corresponding to
10-
6g, it is not really evident that the “froth layer” of the thin film is more dense or blocking more
surface area than the one depicted for the nanostructured one. Could the authors provide
another picture where the differences are more evident?
- Regarding Figure 2a, the authors describe how the J-
V curve decreased significantly for the thin
film but, they should include some comments on why the Voc of the nanostructured one is further
increase under microgravity conditions (it shifts about > 100 mV
under microgravity).
- Regarding the electrocatalyst, the authors indicate in page 8 that in both “flat” and
“nanostructured” Rh electrocatalyst deposition the coverage is similar “the Rh 3d3/2 and 3d5/7
signal intensities are almost identical, suggestin
g a similar coverage of Rh on both electrodes”.
This sentence could be misleading since clearly the coverage of the underlying InP is different
according to the AFM and SEM images. I guess the authors refer to the same catalyst loading?
Expectedly this cou
ld also be correlated to the integrated current during the electrochemical
deposition of Rh.
- The authors established simple models that, quite successfully, describe the performance in
microgravity. However, it would certainly add more impact to this work if the authors modify the
ir
m
odel/equations to incorporate a parameter that takes into account/evaluate the loss of voltage
caused by the froth layer.
- Regarding the better bubble desorption in the nanostructure the authors could include the
reference
Angew. Chem. Int. Ed. 2012, 51, 10760 where the use of nanopillars of InP
demonstrated a fast desorption of H2 bubbles, or others.
Response to reviewer comments
Reviewer #1
1.
The
photocurrent
of the
thin
film
sample
was
reduced
by
4
times,
but
it seems
only
a
small portion of the electrode surface was covered by
bubbles under microgravity. Is
there a better way to characterize the properties (thickness? coverage?) of the forth
layer?
This is a valuable comment made by the reviewer; we include a different image of the
thin film photoelectrode in microgravity envir
onment in the revised manuscript,
which shows the described phenomena in a clearer way. It is, however, difficult to
completely describe the thickness of the froth layer and its coverage of the electrode in
the recorded images; recently, Zhang et al. (2006
) have demonstrated the formation of
so-
called “nanobubbles” on the electrode surface, which are also formed during
electrochemical gas bubble formation in terrestrial environments (Zhang et al.
Langmuir
22
, 8109
-8113, 2006). These “nanobubbles” can e.g.,
be detected by in
- situ
electrochemical AFM measurements where bubble nucleation and growth can be
investigated on a nm scale and the total surface coverage of gas bubbles on the
electrode can be estimated. These measurements would certainly be of great
advantage here and terrestrial experiments with thin film and nanostructured
photoelectrodes are planned. In microgravity environment, we are unfortunately
limited to the camera resolution and can therefore only determine the gas bubble froth
layer thicknes
s on the
μ
m to mm
scale.
2.
Why is the open
-circuit voltage of the nanostructured photoelectrodes greatly
(~100mV)
improved
under
microgravity
environment,
as
shown
in
Fig.
2a?
We appreciate the comment and include a study based on five different thin film and
nanostructured photoelectrodes in the SI, where we measured in independent cyclic
voltammetry experiments the V
OC
of the samples. As Table 1 (SI) shows, the V
OC
of
the
nano
structured and the thin film electrode is almost identical under terrestrial
conditions and the V
OC
value of the nanostructured photoelectrode measured under
terrestrial conditions in Figure 2a is in the error
range.
3.
I am wondering if the O
2
-plasma treatme
nt would change the surface chemistry of
the photoelectrode that affects its microgravity behavior? To confirm/exclude this
effect,
O
2
-plasma
treated
thin
film
photoelectrode
should
also
be
tested.
This is an interesting question, however, the O
2
plasma treatment is likely to improve
the stability of the open InP spots on the nanostructured photoelectrode which is
exposed directly to the electrolyte in the experiments. The formed InO
x
is likely to
protect the surface, although the catalytic event occurs at the Rh catalyst and the
directly underlying InP. Therefore, the photoelectrocatalytic event is very unlikely to
be affected by the O
2
plasma treatment.
4.
AFM
images
in Figure
3c
should
be
given
in
the
same
scale
for
easy
comparison.
The reason for the two resolutions of the AFM images -
500nm for the thin film
photoelectrode and 1
μ
m for the nanostructured sample -
is the better visibility of the
fine structure of the deposited Rh of the thin film sample at this resolution and the
peri
odic arrangement of the nanostructured photoelectrode at 1
μ
m. Nevertheless, we
included an AFM image with a 500nm resolution of the nanostructured sample in the
SI.
5.
Page
10,
the
open
circuit
voltage
in
SI Fig.
5b
is only
reduced
by
50
mV,
not
0.5
V.?
We
appreciate the comment, this is correct. We changed the value accordingly in the
revised manuscript.
6.
Electrochemical impedance spectroscopy (EIS) of the photoelectrodes
in
microgravity environment could be provided useful information on mass
transportation. However, is that technically challenging given that the free fall
time
is only 9.3
s?
EIS could be an interesting experiment, provided that more experimental time is
available as rightfully addressed by the reviewer. This would be an experiment to be
planned for parabolic flights or stationary microgravity (ISS).
7.
Chronoamperometry that captures the whole free fall process (before-during
-after)
should give some intere
sting dynamic
results.
This is an interesting comment of the reviewer; indeed, we recorded
chronoamperometric data during the free fall experiment, which we now mention in
the manuscript. We also include the corresponding data in the SI.
8.
Equation 1, j
L
wa
s not
defined.
Yes, this is correct, we only defined
j
L
in the
Materials and Methods
part. In the
revised manuscript, we included a definition with equation 1.
9.
Page 13, NSL should be
SNL.
This is correct. We changed it in the revised manuscript.
10.
The addition of IPA to the HClO
4
electrolyte was not mentioned in the experimental
section. What is the purpose of adding IPA? Would that change the PEC properties of
the
photoelectrodes?
We greatly appreciate the comment; this is correct, we only include
d information on
the addition of IPA in the figure caption of Figure 2 and in the text. In the revised
manuscript, this information is also added in the experimental section and an
explanation is included: IPA reduces the surface tension of the electrolyte
, which
favors gas bubble desorption according to Vogt H.
Electrochim. Acta
56
2404
-2410
(2011). XPS measurements confirm that the surface composition of the photoelectrode
was not affected by this addition to the electrolyte and also terrestrially, the
photoelectrochemical properties were not influenced. In microgravity environment,
however, we observed in a parallel study that the addition of 1% IPA leads to an
increased gas bubble release and therefore, an improved J
-V behaviour for both, thin
film and n
anostructured electrode.
11.
Since pressure also plays a significant role in the bubble evolution, is the PEC cell
under ambient pressure during the
experiment?
Yes, the PEC cell is under ambient pressure during the experiment in microgravity
conditions whic
h is also mentioned in the revised manuscript (p. 19).
Reviewer #2
1.
In my opinion, 2 references might be
added.
We appreciate the comment; the two suggested references are added to the
manuscript.
2.
Page 7: Figure 2a and b are not
clear.
Figure 2a is not readable: the legend and caption is not clear. Concerning Figure 2b,
the 1g thin film photograph is not
planar?
For a better visibility, independently of the employed color code, we introduced a
dotted line for the JV behaviour of the thi
n film sample and a dashed line for the
nanostructured sample in terrestrial conditions and a dashed
-dotted line for the thin
film sample and a straight line for nanostructured electrode in microgravity
environment.
Yes, it is correct, the so
-called “thin
film” electrode is not strictly planar and the
photoelectrochemically deposited Rh shows an AFM height profile with variations in
the range of 13nm, which is also mentioned in the manuscript on p. 15. In comparison
to the “nanostructured” photoelectrode,
however, the “thin film” electrode
exhibits
a
continuous
thin
film
of deposited
rhodium.
3.
Page 8: The influence of Surface Nanotopography might be given first in my opinion.
We appreciate the suggestion made by the reviewer, we followed, however, the order of
the experimental investigations, which started with the experiments at the drop tower
and the observation of the differences in the JV behaviour of the thin film and
nanostructured photoelectrodes in microgravity environment. Subsequently, we
investi
gated and characterized the electrodes afterwards to further understand the role
of the electrocatalyst surface topography for the photoelectrochemical experiments in
microgravity.
4.
Generally, it is clearer to explain
the experimental set up and the design of
experiments.
In this
paper,
the
experimental
set
-up
is given
after.
This is a reasonable suggestion, we followed, however, the
Nat. Comm.
manuscript
guidelines, which also accounts for comment 11. Nevertheless, the first part of the
results section, includes a detailed description of the experiment in microgravity
environment. Additional information related to the experimental part can also be
found in Figure 1.
5.
Figure 3b, the line “thin film” gives information at “500 nm scale” whereas the line
“nanostructured” gives information at “2 and 1 μm”. I might have think that it should
be the
contrary?
We are grateful for the comment and would li
ke to refer the reviewer to our answer to
the comment #4 of reviewer
#1.
6.
Page 9: you pass directly from Figure 3 to Figure 5 instead of the Figure 4
introduction?
Since we included more SI Figures in the revised manuscript, the numeration of the
figures changed slightly. Nevertheless, we appreciate the comment, although we kept
the original structure of the manuscript. After the introduction of Figure 3, showing
the surface characteristics of the two photoelectrodes, these characteristics are
compared to t
he surface compositions determined by XPS, the spectra are shown in
Figure 5 in the SI. Thereafter, in the discussion part, the drop tower results and the
theoretical assumption of mass transfer limitations influencing the JV characteristics
of the thin fi
lm sample are compared to an experimental series under terrestrial
conditions, where the influence of the mass transfer limitation on the JV
characteristics is demonstrated. The results are presented in the SI Figure 6. In the
following, Figure 4 in the ma
in paper is introduced, where the observed JV behaviors
under terrestrial and microgravity conditions is simulated.
7.
At the end of Page 9 “Output characteristic simulation” it is introduced another
experimental
set
-up
(in
a
numerical
part?).
No
scheme
is g iven?
We appreciate the comment, however, since the theoretical simulations are based on
previous publications which illustrate the applied method and procedures well, we
would like to refer the reviewer to the cited references for further details of the
theoretical modelling.
8.
Page 10
-11: I don’t know if it is my own pdf editor but all equations (1), (2), (3), (4)
page 20 were not correctly
printed.
We apologize for any inconveniences caused by the display of the equations.
Hopefully, the pdf
version of the revised manuscript allows the complete display of the
equations.
9.
Page 12: Figure 4 has the same problem as Figure 2: no clear legend and caption.
Impossible to determine the 4 cases with only grey shades. In the text, you speak of
yellow
or
cyan
lines
which
are
not
allowed
with
usual
grey
scale
printers.
We would like to refer the reviewer to our reply to comment #2; we introduced
additional graphical features which should allow the distinction of the different JV
curves independently from
the color code.
10.
I think these simulations should have been compared with experimental results of
Figure
2. But
not
in a
clear
way:
difference
theory/experiments...?
We appreciate the comment and included an additional discussion part in the revised
manusc
ript, referring to differences in the theoretical simulation and the experiment.
11.
Page 15: after the part “conclusion” arrives a big experimental set
-up “materials and
methods” part. I think one part might be put at the beginning and perhaps the rest in
a correctly named
“Annex”?
We would like to refer the reviewer to our reply to comment #4.
Reviewer #3
1.
In Figure 2b the authors included some pictures to demonstrate that the formation of
a “froth layer” is the main cause for the detrimental performance in the “thin film”
with respect to the “nanostructured”. However, if we have to consider as a direct
proof the pictures corresponding to 10
-6
g, it is not really evident that the “froth layer
”
o
f the thin film is denser or blocking more surface area than
the one depicted for the
nanostructured one. Could the authors provide another picture where the differences
are more
evident?
We are grateful for the comment and would like to refer the reviewer to our reply t
o
r
eviewer #1, comment #1.
2.
Regarding Figure
2a, the authors describe how the J
-V curve decreased significantly
for the thin film but, they should include some comments on why the
V
OC
of the
nanostructured
one
is further
increased
under
microgravity
conditions.
We greatly appreciate the comment and would also like to refer the reviewer again t
o
o
ur reply to reviewer #1, comment #2.
3.
Regarding the electrocatalyst, the authors indicate in page 8 that in both “flat” and
“nanostructured” Rh electrocatalyst deposition the coverage is similar “the Rh 3d
3/2
and 3d
5/2
signal intensities are almost identical, suggesting a similar coverage of
R
h
o
n both electrodes”. This sentence could be misleading since clearly the coverage of
the
underlying
InP
is different
according
to the
AFM
and
SEM
images.
I
guess
the
authors refer to the same catalyst loading? Expectedly this could also be correlated
to the integrated current during the electrochemical deposition of Rh.
This is a valuable comment which is very much appreciated. It is true that the
indicated sentence is misleading; the signal intensities are similar, although the local
Rh coverage is distinctively different due to the different surface topographies of the
Rh catalyst. We changed the sentence in the revised manuscript to: “Despite the fact
that Rh was d
eposited through the polystyrene sphere mask, the Rh 3d
3/2
and 3 d
5/2
signal intensities are almost identical, suggesting a similar overall coverage of Rh on
both electrodes with distinctive differences in the local coverage due to the different
surface topographies.” (p. 9)
4.
The authors established simple models that, quite successfully, describe the
performance in microgravity. However, it would certainly add more impact to this
work if the authors modify their model/equations to incorporate a parameter t
hat
takes
into
account/evaluate
the
loss
of voltage
caused
by
the
froth
layer.
We are grateful for the comment; the voltage loss due to the froth layer is already
taken into account by the extension of the Butler
-Volmer
equation for mass transfer
limitation. For the description of the voltage loss due to recombination processes,
however, the so-
called “saturation current” or “dark current” plays a crucial role for
photodiodes. We succeeded in modelling the JV behaviour o
f the thin film electrode in
microgravity environment considering mass transfer limitations and resulting
increased surface recombination of charge carriers by lowering the saturation current.
Please compare Figure 4 in the revised manuscript and the
Materials and Methods
part for further details.
5.
Regarding the better bubble desorption in the nanostructure the authors coul
d
i
nclude the reference Angew. Chem. Int. Ed. 2012, 51, 10760 where the use of
nanopillars
of InP
demonstrated
a
fast
desorption
of H
2
bubbles,
or
others.
We appreciate the suggestion and include the additional reference in the revised
manuscript.
REVIEWERS' COMMENTS:
Reviewer #1 (Remarks to the Author):
I feel that the points raised in the previous round of review have been satisfactorily addressed.
Reviewer #2 (Remarks to the Author):
Thank you for all the answers and correction made. On my side all is better.
regards
Reviewer #3 (Remarks to the Author):
The authors have addressed compellingly all the comments, and I would suggest to accept the
manuscript for publication.