Preparation of Nafion Membranes for Reproducible Ammonia Quantification in Nitrogen Reduction Reaction Experiments

Article abstract of DOI:10.1002/anie.202007998

This study highlights the importance of following a strict protocol for Nafion membrane pretreatment for electrochemical nitrogen reduction reaction experiments. Atmospheric ammonia pollution can be introduced to the experimental setup through membranes and interpreted falsely as catalysis product from N₂. The sources of ammonia contamination vary drastically between locations worldwide and even within the same location between days depending on temperature, wind direction, fertilizer use, and manure accumulation in its vicinity. The study shows that significant amounts of ammonium is accumulated in the membranes after commonly practiced pretreatment methods, where the amount depends on the ammonia concentration in the surrounding of the experiment. Therefore, we introduce a new pretreatment method which removes all the ammonium in the membrane. The membranes can be stored for several days but a short final step in the method needs to be carried out right before NRR experiments.

Full text of DOI:10.1002/anie.202007998

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Preparation of Nafion® Membranes for Reliable Ammonia
Quantification in Nitrogen Reduction Reaction Experiments
Authors: Fatemeh Hanifpour, Arnar Sveinbjörnson, Camila Pía
Canales, Egill Skúlason, and Helga Dögg Flosadóttir
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202007998
Link to VoR: https://doi.org/10.1002/anie.202007998

10.1002/anie.202007998
Angewandte Chemie International Edition
COMMUNICATION
Preparation of Nafion® Membranes for Reproducible Ammonia
Quantification in Nitrogen Reduction Reaction Experiments
Fatemeh Hanifpour,[a] Arnar Sveinbjörnson,[b] Camila Pía Canales,[a] Egill Skúlason[a] and Helga Dögg
Flosadóttir*[b]
[a]
[b]
F, Hanifpour, Dr. C. P. Canales, Prof. E. Skúlason
Science Institute & Faculty of Industrial Engineering, Mechanical Engineering and Computer Science,
University of Iceland,
VR-III, Hjardarhaga 2, 107 Reykjavík, Iceland
E-mail: helgadogg@atmonia.com
A, Sveinbjörnson, Dr. H. D. Flosadóttir
Innovation Center Iceland
Árleyni 2-8, 112 Reykjavík, Iceland
Supporting information for this article is given via a link at the end of the document.
commonly exists as a point source from industry or agriculture,
and its concentration at each location can therefore vary quite
drastically between days and seasons within the exact same
location depending on temperature, wind direction, fertilizer use
and manure accumulation and use.[6] These variations can cause
false positives in NRR experiments. A study on ammonia
concentrations in the atmosphere in Taiwan shows that NH3
levels can reach as high as 100 ppb in such a highly populated
and polluted location.[7] Atmospheric contamination of such levels
could interfere with the ammonia quantification in NRR
experiments done world-wide today. The ventilation system
together with the fume hood in laboratories should be capable of
providing indoor air quality according to acceptable standards.
For such ventilation to be effective, large volumes of external
atmosphere are pumped through the laboratory during the day, in
most cases not considering purifying gaseous atmospheric
pollution from the inlet air. Poor ventilation and infiltration are
found to be the major sources of laboratory indoor air pollution.[8]
To minimize the variable effects of environmental pollutants,
it is necessary to start the NRR experiments as clean as possible
and maintain that throughout the experiment by minimizing
handling of samples. A typical setup for NRR experiments
includes a double compartment electrochemical cell utilizing a
proton exchange membrane (PEM) to separate the two
compartments. The role of the PEM in aqueous-acidic electrolyte
is to crossover protons from the anodic chamber to the cathodic
chamber and, at the same time, to prevent diffusion of O2 and the
Abstract: This study highlights the importance of following a strict
protocol for Nafion® membrane pretreatment for electrochemical
nitrogen reduction reaction experiments. It is shown how
atmospheric ammonia pollutions can be introduced to the
experimental setup through the membranes, therefore,
interpreted falsely as catalysis product from N2. The sources of
ammonia contamination vary drastically between different
locations worldwide and even within the same location between
days depending on temperature, wind direction, fertilizer use and
manure accumulation in its vicinity. The study shows that
significant amounts of ammonium is accumulated in the
membranes after commonly practiced pretreatment methods,
where the amount depends on the ammonia concentration in the
surrounding of the experiment. Therefore, we introduce a new
pretreatment method which removes all the ammonium in the
membrane. The membranes can be stored for several days but a
short final step in the method needs to be carried out right before
NRR experiments.
The development of a sustainable and efficient process for
ammonia (NH3) production from nitrogen (N2) is now a significant
research and development objective. Many researchers are
focusing on NH3 synthesis at ambient conditions by means of N2
reduction reaction (NRR) using heterogeneous catalysts as an
alternative for the conventional Haber-Bosch process. This
interesting new field is at the same time quite challenging as there
are numerous possible sources of contamination introduced to
the experiments, causing false positive results, i.e. the produced
NH3 may not in fact be the result of a catalytic reaction but come
from different sources of contamination. In this sense, several
researchers have recently published articles stressing the
importance of rigorous and efficient experimental protocols to
conduct NRR as reliable as possible to minimize the external
sources of contamination. These sources include, but are not
limited to, the inherent contamination in the setup, the reactant
gas (14N2 and 15N2) and electrolyte, human breath and nitrile
rubber gloves.[1-5] For this reason, as reported by Chorkendorff
+
decomposition of the NH4 product on the counter electrode.
However, PEMs – with Nafion® 211 and Nafion® 117 being the
most common membranes in such applications – can accumulate
and partially crossover NH4+ ions resulting in gradual decrease of
membrane conductivity during experiments.[1-2,9] Ren and co-
workers have concluded that the use of Nafion® membranes in
NRR experiments may be inappropriate, as ammonium can pass
through the membrane, be adsorbed on the membrane or, even,
interact with it, which is unfavorable for reliable quantification of
the product.[10] Although long-term application of Nafion® (7-10
days) is not suitable due to loss of conductivity, Nafion®
membranes are nevertheless widely used in short term (~2 h)
NRR experiments today and deserve further investigation on
pretreatment protocols in order to obtain as reliable data as
possible.
and co-workers, it is important to consider
a systematic
benchmarking protocol to eliminate the false and retain the true
results in NRR.[5]
However, environmental effects on NRR experiments have
not yet been seriously considered. Atmospheric NH3 pollution
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202007998
Angewandte Chemie International Edition
COMMUNICATION
In this work, we report measurements on ammonium
contamination in pretreated Nafion® membranes using the
common pretreatment methods reported in the literature and
recommend a modification to this process. Two of the most
common Nafion® membranes (117 and 211) are used as an
dissolved in any aqueous solution open to air, and protonated to
form ammonium. A piece of Nafion® submerged in the aqueous
solution is then exposed to ammonium ions, that are then
exchanged with the acid sites of the membrane, in equilibrium
+
with the NH4 concentration in the solution. Following the ion
+
example. It is observed that the accumulation of NH4 in the
exchange, the Nafion® adsorbed ammonium ions self-diffuse into
and through the membrane.[16] This brings forth doubts in the
actual quantification of the NRR product.
membrane highly depends on the local atmospheric levels of
NH3(g) on the specific day when the pretreatment process takes
place or when the membranes are used. As atmospheric levels of
ammonia vary between locations, this can be an important factor
to consider when doing NRR experiments in highly populated and
polluted cities, or close to industries or agriculture. This study is
performed in the outskirts of Reykjavík, Iceland, a city where
atmospheric pollutants are usually very low.[11] The main
repeating sources of atmospheric NH3 pollution is from the
geothermal wells in Hengill area and a few agricultural activities
such as a pig farm in 25 km radius around the research site.
Measurements of atmospheric NH3(g) are not commonly
performed in Iceland where the experiments reported here are
performed. Geothermal wells, however, have been shown to emit
NH3 to some extent along with many other gases such as H2S,
CO2, H2 and CH4.[12,13] Air pollution stations in Reykjavík
commonly monitor H2S as the main pollutant. The studies
presented here are performed at the Innovation Center Iceland
that is within 1 km distance from the nearest monitoring station,
Lambhagi. In this study, we use environmental data of
atmospheric H2S for the purpose of estimating the change in the
local, proportional atmospheric level of NH3, as these two tend to
co-exist and correlate in atmosphere as emissions from
geothermal wells. The geothermal well at Nesjavellir is estimated
to be the main source of atmospheric NH3 pollution, but there may
also be other temporary sources. The source of atmospheric
NH3(g) can vary between locations and the level of concentration
can vary within location between seasons and days. It should be
noted that the levels of H2S and NH3 can vary significantly
between days in Iceland, depending on wind direction and
location with regards to the geothermal areas, but these levels are
commonly very low compared to most cities.[7] This study is in all
cases performed at the same location but different days. We
observe a great variation in repeated experiments depending on
the day. These results can in principle be extrapolated to different
In this study, multiple versions of pretreatment methods are
performed to account for many possible variations within
laboratories carrying out NRR experiments to date. These are
referred to as Methods 1-4 within this text. These different
procedures are shown in detail in Table 1 and compared with a
modified procedure (Method 5) we introduce here to minimize the
amount of ammonium ions in Nafion® membranes before they
are used in NRR experiments.
Dry Nafion® sheets of size 4.5 cm2 are used in all cases. All
containers are acid washed with 20% HCl at least 12 h before the
experiments. The effect of container type (Method 1 vs. 2) and the
use of fresh container in each step vs. using same container all
through the pretreatment procedure (Method 3 vs. 4) are studied.
Ammonia uptake by Nafion® is measured at the end of the
pretreatment processes by sonicating the membrane in 5 ml of
fresh 0.05 M H2SO4 for 10 minutes and measuring the ammonia
content in the acidic solution by means of Flow Injection
Analysis®,
a selective quantification method of aqueous
ammonium ions as low as 1 ppb (see the full description in SI). In
all experiments, blanks were collected by sonicating 5 mL of 0.05
M sulfuric acid for 10 minutes under the same condition as the
samples. In all cases the ammonia detected in the blanks was
lower than the limit of detection.
Method 1 and 2 used the same procedure but different
containers; plastic beakers with screw caps (Method 1) and glass
beakers covered with Al foil (Method 2). The study showed that
plastic beakers with screw caps were not suitable containers for
Table 1. Variations in Nafion® pretreatments tested in this study. Polypropylene
beaker with screw cap is used in Method 1 but glass beakers covered with Al
foil are used in Methods 2-5. In Method 3 a fresh baker is used in each step of
the procedure while Methods 4 and 5 uses the same beaker in all steps.
Method
NO.
Method Name
Container
Procedure
locations world-wide where
a large variation of ammonia
concentration may exist in atmosphere depending on both the
date and location of the experiment.
1
Polypropylene
w/screw cap
1. 80°C, 1h, 5% H2O2
2. 80°C, 1h, DI water
3. 80°C, 1h, 0.5 M H2SO4
4. 80°C, 1h, DI water
Short
treatment
Therefore, there is a need for a standard procedure for
Nafion® pretreatment to ensure that the membrane is not a
source for ammonia. This study introduces a method for
expansion and purification of Nafion® membranes that provides
ammonium-free membranes to conduct NRR experiments with.
The commercial expansion method for Nafion® membranes
is to use water (room temperature or up to 90 ˚C) or dilute acids
(0.2% to 0.5% HNO3 or HCl), depending on application.[14] The
common Nafion® pretreatment method as practiced in most of the
published papers in the NRR field, however, is to treat Nafion®
first using H2O2 (3-5)% at 80 ˚C to remove organic impurities, next
using 0.1-1 M H2SO4 at 80 ˚C to eliminate metallic impurities and
to protonate the membrane, and finally leave Nafion® in milli pore
water at 80 ˚C to rinse the excess acid and reach the maximum
expansion.[15] The commonly used pretreatment process could
leave the membrane polluted with significant amounts of
ammonium because atmospheric and gaseous ammonia is
2
3
Glass, Al foil
Glass, Al foil
(Fresh)
Long
1. 80°C, 1h, 5% H2O2
2. 80°C, 1h, DI water
3. 80°C, 3h, 0.5 M H2SO4
4. 80°C, 6h, DI water
treatment
(common
practice)
4
5
Glass, Al foil
(Same)
Modified
method
Glass, Al foil
(Same)
In ultrasonic bath:
1. RT[a], 20 min, 5% H2O2
2. RT, 20 min, DI water
3. RT, 20 min, 0.5 M H2SO4
4. RT, 20 min, DI water
5. 80°C, > 12h, DI water
6. RT, Sonication in 0.05 M
H2SO4 and DI water
directly before use
[a] RT: Room Temperature
2
This article is protected by copyright. All rights reserved.

10.1002/anie.202007998
Angewandte Chemie International Edition
COMMUNICATION
this high temperature method, as the caps popped up randomly
due to high pressure inside, and the contents were, on such
occasions, totally exposed to the surrounding environment. Thus,
plastic beakers were disregarded in later experiments.
glass until its application. However, directly prior to use, the
membrane needs to be sonicated in 0.05 M H2SO4 and DI water.
Immersing Nafion® in H2SO4 solution provides excessive access
to protons, reversing the cation exchange and extracting 푁퐻4+
from Nafion® into the solution and thus, fully protonating the
membrane prior to use in experiments. Ammonia content of
Nafion® prepared using this procedure was measured several
times (either using one beaker or different beakers in the
pretreatment procedure) and the resulting amount was always
lower than the limit of detection of the method (5 ng NH3). As
shown in Fig. 1, H2S contaminant was present in the atmosphere
during all the tests that were performed using Method 5, and in
many cases the levels were even higher than those of the tests
done using Methods 1 – 4, yet no ammonia contamination was
found in the membrane after such pretreatment. It is worth to note
that all measurements of each sample were repeated at least
twice in a row. The measurements were in all cases reproducible
within measurement error. For a detailed description of the
ammonia measurement method, refer to the supporting
information (SI).
It is common laboratory practice to pretreat several Nafion®
membranes simultaneously for a use-period of the following 5-7
days. Nafion® is then stored in water until use, during which time
it may accumulate NH3 to higher contamination levels. This highly
depends on the storage solution which is constantly in dynamic
equilibrium with the surrounding atmosphere. An experiment to
demonstrate the effect of atmospheric NH3(g) concentration on
extended storage of Nafion® in water, was performed with
Nafion® 211 (as no distinct difference was observed between
Nafion® 211 and 117 in previous experiments). Here, a 10-day
study was performed where a fully prepared membrane (4.5 cm2
of Nafion® 211 pretreated initially using Method 5) was kept in
mQ water and analyzed for NH3 concentration on various days
(Fig. 2a). Six samples were taken: the first one, right after
pretreatment which as expected did not contain any ammonia
contamination, and the rest of the samples were taken after 2, 3,
7, 8 and 10 days after the initial pretreatment. The results
Method 3 and 4 were performed according to the common
method practiced for Nafion® pretreatment as described in Table
1. There, the comparison was made between using a new and
clean beaker for every new preparation step (Method 3) and using
only one and single beaker for all preparation steps (Method 4).
Ammonia contents of pretreated Nafion® 211 and 117
undergoing these methods are shown in Fig. 1. Apparently, the
use of pretreatment methods 1 - 4 does not remove ammonium
from the membranes in any of the 18 tests performed. The
ammonia content is measured from around 0.1-2 g whereas the
limit of detection in this study is much lower, or 5 ng. This external
ammonium contamination would therefore significantly affect the
results obtained in NRR experiments. There is no clear difference
between Methods 1-4 tested here when it comes to ammonium
removal. However, fluctuations between replicates of the same
tests could only be attributed to the atmospheric concentrations
of ammonia (estimated from the H2S atmospheric level in this
study) or other external contamination factors. This underpins the
demand for a reliable method for contaminant-free preparation of
Nafion® membrane.
Figure 1. Ammonia content of Nafion® after treatment methods listed in Table
1. Solid columns in each method represent NH3 data on Nafion® 211 and empty
columns represent the data on Nafion ® 117. H2S data is shown as dots. Limit
of Detection (LOD) of the method for ammonia analysis is typically between 5-
15 ng (dashed line at 5 ng). Method 5 was repeated 7 times on different days
for both Nafion® types where atmospheric H2S levels were varying but the
ammonia content was always under the LOD.
demonstrated
a high variation of NH3 concentration in
membranes on the specified days.
In Fig. 2a, the ammonia concentration in the membrane is
plotted against the number of days passed initial pretreatment,
using the modified procedure (Method 5). The results appear to
follow the local atmospheric concentration of NH3 (estimated from
the H2S atmospheric level). It should be noted that here the
ammonia content was measured in the membranes after storing
them in water for a few days without performing the final step of
Method 5 directly before analysis. Therefore, similar sets of tests
were repeated, but sample collection and analysis were done
after performing the final step of Method 5 (sonication in 0.05 M
H2SO4 and DI water) directly before sample collection, see Fig.
2b. Despite the various H2S levels on various days within this test,
the results showed ammonia concentrations bellow limit of
detection in all the samples.
Note in Fig. 2a that although the average of H2S level on the
pretreatment day (day 0) was significant, there was no ammonia
detected in the sample, demonstrating the methods
reproducibility and the need for this pretreatment. Fig. 2b also
demonstrates that pretreatment can be done several days before
experiments if the final step of the method is performed directly
before use. Furthermore, the variable concentrations observed
depending upon the local and short-term atmospheric levels of
In order to reach the maximum expansion and remove
ammonium from Nafion® membrane before applying it in the
electrochemical cell, we suggest a pretreatment process referred
to Method 5 in Table 1. Nafion® is first sonicated for 20 minutes
at room temperature (RT) in 5% H2O2. The analysis of total
organic content (TOC) in Vario TOC cube (Elementar
Analysensysteme GmbH, 63505 Langenselbold, Germany)
shows that this step can remove all the possible organic impurities
in the membrane (see Table S1 in SI). Next, Nafion® is sonicated
for 20 minutes at RT in ultrapure water (milliQ Quantum TEX)
before undergoing the same sonication step in 0.5 M H2SO4. ICP
analysis (Agilent Technologies 7800 ICP-MS) confirms that this
step can remove metallic impurities inside the membrane (see
Table S2 in SI). Nafion® is next sonicated again for 20 minutes at
RT in milliQ water to rinse the excessive acid from the previous
step. In this step, Nafion® membrane is expanded and protonated
and may be kept at room temperature in water inside an amber
3
This article is protected by copyright. All rights reserved.

10.1002/anie.202007998
Angewandte Chemie International Edition
COMMUNICATION
NH3 implies that the comparison of an NRR experiment and its
associate blank test done on different days may not represent the
true results without ensuring a clean and closed experimental
environment.
The sources of atmospheric H2S in Iceland are geothermal
wells and drill-holes. Ammonia emissions from geothermal wells
in Iceland have not been widely studied. One study performed in
Iceland observed NH3 in geothermal steam condensate of
Hellisheiði.[17] On the other hand, H2S, a well-known geothermal
emission gas, is extensively monitored.[18] Furthermore, a study
on geothermal wells in Cerro Prieto IVin Mexico measured CO2,
H2S, H2, CH4 and NH3 from geothermal wells, which demonstrates
the co-existence of NH3 and H2S.[12] Intermittent and short-term
increase in local atmospheric levels of H2S and NH3, therefore,
simultaneously implies the short-term increase of geothermal well
pollution often observed in specific wind direction. The direct
relation between H2S and NH3 concentration levels in atmosphere
has also been reported in a study by Blunden et al. where they
monitor the diurnal variations for H2S and NH3 emission rates from
the ventilation exhaust of a pig barn in North Carolina.[13] The
geothermal wells are the main source of atmospheric NH3 in
Reykjavík, Iceland, but there are many other possible sources
elsewhere causing much higher atmospheric concentrations.
It is important to notice that the concentration of ammonium
in Nafion® reported here are significant, yet not very high. We
show that the concentration of ammonium in the membrane highly
depends on atmospheric levels of NH3, which we observe by
performing the experiments on different days using commonly
utilized pretreatment methods and compare it with pollutant data
for those days. Since the atmospheric levels of ammonia can vary
greatly between locations world-wide,[19] the ammonium content
in the membrane will highly depend on the location of the
experiment. The atmospheric concentration of ammonia also
highly depends upon temperature and season (fertilizer
application).[20] Measurements of aerosol NH3 in Heimaey, an
active volcano island south of Iceland, demonstrated 0.05-0.12
ppb NH3 in air.[21] Atmospheric concentrations have been
measured in various cities worldwide and reported between 3 ppb
and 100 ppb (up to 800 times the reported concentration in
Heimaey) depending on location and time.[7] These variations
further support the demand to ensure that Nafion® used for each
experiment is not a source of ammonia. The very low atmospheric
concentrations of NH3 in Iceland can still cause observable
concentrations of NH3 in the membranes, unless the proposed
method in this study is used. Variations of atmospheric NH3 may
in many cases cause differences in NH3 concentrations that can
be falsely attributed to catalysis.
Figure 2. Ammonia content in Nafion® 211 (4.5 cm2) as a function of days after
pretreatment and atmospheric H2S levels. (a) Membranes were pretreated
initially using the modified method and kept in water in a glass beaker covered
with Al foil. Significant amount of ammonia was measured, except at day zero,
right after pretreatment (b) Membranes were pretreated initially using the
modified method and kept in water in a glass beaker covered with Al foil and
step 6 in Method 5 was carried out directly before ammonia measurements.
LOD of the method for ammonia analysis is typically between 5-15 ng (dashed
line at 5 ng). Here the ammonia content was always lower than the LOD.
Further comparison between Nafion® 211 pretreatment
using Method 4 and Method 5 was done in a double-compartment
H-cell. Two cells were used for each pretreatment method, i)
blank cell: filled with 20.0 mL of 0.05 M H2SO4 in both cathodic
and anodic chambers and ii) sample cell: filled with 20.0 mL of
0.05 M H2SO4 containing 0.3 ppm of ammonia (60 µg) in the
cathodic chamber and no concentration of ammonia in the anodic
chamber. N2 was bubbled through the cathodic chambers for 2 h.
The Nafion® membranes were collected and sonicated in 5.0 mL
of 0.05 M H2SO4. The ammonia contained in Nafion® after the
test is shown in Table 2. Similar results as shown before in Fig. 1
were obtained for the blank tests; Nafion® pretreated using
Method 4 included significant amount of ammonium whereas with
Method 5 this amount was under detection limit. For the sample
tests it is clear that in both cases (methods 4 and 5) more
ammonia was adsorbed into the Nafion® compared with the
“blank” tests, demonstrating the fast equilibrium between
electrolyte and membrane. Similar amount of ammonium was
absorbed into the membranes with both methods (65-78 ng),
which is around 0.001% of the ammonia in the electrolyte (60 µg)
during the 2-hour experiments. This simulated NRR experiments
shows that our proposed method can be used for the pretreatment
process and then ammonia from the Nafion® can be accounted
for in the end of the experiment if needed.
In conclusion, this study reports an observation of
ammonium contamination in expanded and protonated Nafion®
membranes ready for application, as one of the sources of false
positive results in NRR experiments. An improved pretreatment
procedure is suggested to ensure ammonium-free membranes
prior to use. The contamination levels are demonstrated to be
dependent upon the NH3 concentrations in local atmosphere that
can vary quite drastically between locations world-wide and on
daily basis. The modified method of Nafion® pretreatment is both
more practical than previous methods and rules out the chance of
Table 2. Ammonia content in Nafion® 211 undergone common and modified
pretreatment methods in a simulated NRR experiment.
Nafion® being
experiments.
a source for ammonia detected in NRR
Method No.
Ammonia in Nafion® in
blank cell / ng
Ammonia in Nafion® in
sample cell / ng
4
5
150±8
< 18
228±13
65±1
Acknowledgements
4
This article is protected by copyright. All rights reserved.

10.1002/anie.202007998
Angewandte Chemie International Edition
COMMUNICATION
This work was supported by the Icelandic Research Fund (grant
nos. 196437-051 and 185404-051) and the Icelandic Technology
Development Fund (grant no. 175350-0611). We wish to thank
the Icelandic environmental agency for providing us with the H2S
data.
Keywords: Nafion • Pretreatment method • Ammonia •
Contamination • Nitrogen reduction reaction
[1]
G. F. Chen, S. Ren, L. Zhang, H. Cheng, Y. Luo, K. Zhu, L. X. Ding, H.
Wang, Small Methods 2019, 3, 1800337.
[2]
[3]
C. Tang, S.-Z. Qiao, Chemical Society Reviews 2019, 48, 3166-3180.
B. H. Suryanto, H.-L. Du, D. Wang, J. Chen, A. N. Simonov, D. R.
MacFarlane, Nat. Catal. 2019, 2, 290-296.
[4]
[5]
S. L. Foster, S. I. P. Bakovic, R. D. Duda, S. Maheshwari, R. D. Milton,
S. D. Minteer, M. J. Janik, J. N. Renner, L. F. Greenlee, Nat. Catal. 2018,
1, 490-500.
S. Z. Andersen, V. Čolić, S. Yang, J. A. Schwalbe, A. C. Nielander, J. M.
McEnaney, K. Enemark-Rasmussen, J. G. Baker, A. R. Singh, B. A. Rohr,
M. J. Statt, S. J. Blair, S. Mezzavilla, J. Kibsgaard, P. C. K. Vesborg, M.
Cargnello, S. F. Bent, T. F. Jaramillo, I. E. L. Stephens, J. K. Nørskov, I.
Chorkendorff, Nature 2019, 570, 504-508.
[6]
[7]
[8]
[9]
C. Krauter, D. Goorahoo, C. Potter, S. Klooster, Emission Inventories-
Partnering for the Future 2002, 11, 15-18.
S. Wang, J. Nan, C. Shi, Q. Fu, S. Gao, D. Wang, H. Cui, A. Saiz-Lopez,
B. Zhou, China. Scientific reports 2015, 5, 15842.
T. Ugranli, E. Gungormus, A. Sofuoglu and S.C. Sofuoglu,
Comprehensive Analytical Chemistry, Vol. 73, 2016, pp. 859-878.
K. Hongsirikarn, J. G. Goodwin Jr, S. Greenway, S. Creager, J. Power
Sources 2010, 195, 30-38.
[10] Y. Ren, C. Yu, X. Tan, X. Han, H. Huang, H. Huang, J. Qiu, Small
Methods 2019, 3, 1900474.
[11] K. Steinecke, Atmos. Environ. 1999, 33, 4157-4162.
[12]
R. M. Barragán, V. M. Arellano, M. H. Rodríguez, A. Pérez, N. Segovia,
Geofluids 2010, 4, 511-524.
[13] J. Blunden, V. P. Aneja, P. W. Westerman, Atmos. Environ. 2008, 42,
3315-3331.
[14] Fuel Cell Store, "NAFION Hydrogen-Form Membrane Expansion
Technical Information Bulletin 93-01, 2019.
[15] R. Kuwertz, C. Kirstein, T. Turek, U. Kunz, J. Membr. Sci. 2016, 500,
225-235.
[16] R. M. Jung, H. S. Cho, S. Park, J. W. Van Zee, J. Power Sources 2015,
275, 14-21.
[17] A. I. Remoroza, Calcite Mineral Scaling Potentials of High-Temperature
Geothermal Wells (Unpublished MSc Thesis), University of Iceland,
Reykjavik, 2010.
[18] Measurements on air quality in Iceland, Environmental agency,
https://loftgaedi.is/?zoomLevel=7&lat=64.894972&lng=-18.675028
[19] M. Crippa, D. Guizzardi, M. Muntean, E. Schaaf, F. Dentener, J. A. van
Aardenne, S. Monni, U. Doering, J. G. Olivier, V. Pagliari, Earth Syst. Sci.
Data 2018, 10, 1987-2013.
[20] A. D. Larios, S. Godbout, S. K. Brar, J. H. Palacios, D. Zegan, A. Avalos-
Ramírez, F. Sandoval-Salas, Biosyst. Eng. 2018, 169, 165-174.
[21] J. M. Prospero, D. L. Savoie, R. Arimoto, H. Olafsson, H. Hjartarson, Sci.
Total Environ. 1995, 160, 181-191.
5
This article is protected by copyright. All rights reserved.

10.1002/anie.202007998
Angewandte Chemie International Edition
COMMUNICATION
A utilizable method of Nafion® pretreatment for nitrogen reduction reaction (NRR) is proposed. Using this method, one rules out one
of the critical sources of false positive results in NRR experiments. Failing to follow these steps can introduce variable amounts of
atmospheric ammonia to the setup which may be interpreted falsely as catalysis product. This observation questions the results of the
otherwise interesting published reports in this field.
6
This article is protected by copyright. All rights reserved.