Graphene Oxide: Carbocatalyst or Reagent?

Article abstract of DOI:10.1002/anie.201809979

Metal-free heterogeneous carbon-based materials have the potential to facilitate a wide range of organic transformations in an economical and environmentally-friendly manner. However, the mechanism of their action is often obfuscated by their ill-defined nature, so careful analysis of the reaction products is essential before they can be labelled “catalysts” in accordance with the IUPAC definition. We present here our findings on the archetypal “carbocatalytic” conversion of benzyl alcohol to benzaldehyde, in which graphene oxide is reduced and a major side product, dibenzyl ether, is also formed. Thus, our work reclassifies graphene oxide as a reagent and presents an important step in the proper evaluation of reactions in the nascent field of carbocatalysis.

Full text of DOI:10.1002/anie.201809979

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
International Edition Chemie
www.angewandte.org
Accepted Article
Title: Myths and Realities of Graphene Oxide: Carbocatalyst or
Reagent
Authors: Stanislav Presolski and Martin Pumera
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.201809979
Angew. Chem. 10.1002/ange.201809979
Link to VoR: http://dx.doi.org/10.1002/anie.201809979
http://dx.doi.org/10.1002/ange.201809979

Angewandte Chemie International Edition
10.1002/anie.201809979
COMMUNICATION
Myths and Realities of Graphene Oxide: Carbocatalyst or Reagent
[a]
[b]
Stanislav Presolski and Martin Pumera*
Abstract: Metal-free heterogeneous carbon-based materials have
the potential to facilitate a wide range of organic transformations in an
economical and environmentally-friendly manner. Often times,
however, the mechanism of their action is obfuscated by their ill-
defined nature, so careful analysis of the reaction products is essential
before they can be labelled ‘catalysts’ according to IUPAC’s definition.
We present here our findings on the archetypal ‘carbocatalytic’
conversion of benzyl alcohol to benzaldehyde, in which graphene
oxide is reduced and a major side product – dibenzyl ether, is also
formed. Thus, our work re-classifies graphene oxide as a reagent and
presents an important step in the proper evaluation of reactions in the
nascent field of carbocatalysis.
the original article[7] and performed a number of in-depth
spectroscopic studies. We prepared Hummer’s graphene oxide
(GO)[9] and evaluated its capacity to oxidize benzyl alcohol
(BnOH) as outlined in Scheme 1. Reactions were performed
according to the method described by Bielawski,[7] namely adding
50 μL of neat benzyl alcohol and the appropriate amount of GO
(10 mg for 20 wt% and 100 mg for 200 wt%) in a 4.5-mL glass
vial, then heating the mixture overnight at 100 °C before diluting
with chloroform and analysing the filtrate.[7]
Graphene and related carbonaceous materials, such as graphene
oxide (GO), have had a significant impact on the fields of solid-
state physics and chemistry over the past two decades with their
mechanical and electronic properties extensively studied.[1] Their
chemical reactivity, especially the ability to catalyse reactions, on
the other hand, has only recently become a subject of detailed
investigations.[2] Carbocatalysis is a term that encompasses a
number of organic transformations that are mediated or greatly
accelerated by the various functional groups at the defect sites
present on graphene oxide.[3] The ill-defined nature of these sites
that have been shown to vary depending on the method of GO
preparation,[4] as well as the difficulty in excluding metal ion
participation have complicated the mechanistic elucidation of
Scheme 1. Reaction of benzyl alcohol with graphene oxide (GO) produces
benzaldehyde, as well as dibenzyl ether.
We used X-ray photoemission spectroscopy (XPS) to determine
the carbon-to-oxygen composition of the freshly prepared GO and
found it to be in excellent agreement with the combustion analysis
data reported by Bielawski (Table 1).[7] Similarly, we analysed the
material after it was used as 20 wt% catalyst in the oxidation of
benzyl alcohol, which resulted in the apparent loss of more than
2/3 of the oxygen content. While this implies that GO was a
reagent in the reaction, the fact that it was used in sub-
stoichiometric amounts allows for the possibility that it also acted
as a catalyst. Moreover, there was no attempt to exclude oxygen
from the reaction vial.
[
5]
reaction pathways. Yet the allure of a valuable, low price, largely
carbon-based catalyst has drawn many to this emergent field with
often times little regard given to reproducibility and in-depth
understanding of the underpinnings of ‘carbocatalysis’.[6]
We set ourselves to investigate the mechanism of the archetypal
carbocatalytic oxidation of benzyl alcohol to benzaldehyde
published in 2010 in this journal.[7] While chemists are familiar with
the term “catalyst”, it is important to remind ourselves that
according to IUPAC a “catalyst is both a reactant and product of
the reaction”,[8] whose structure and elemental composition
should thus remain unchanged. In the course of our studies, we
discovered that despite the original claims,[7] graphene oxide does
not fulfil this definition in the case of the abovementioned reaction.
This revelation should encourage researchers in the field to align
their “carbocatalyst” claims with the IUPAC definition of catalyst.
In order to critically assess one well-documented case of
carbocatalysis, where graphene oxide was utilized to convert
benzyl alcohol to benzaldehyde, we followed the procedures in
Table 1. Oxygen-to-carbon ratio in GO samples before and after being used
in BnOH oxidation reactions.
Combustion Analysis[a]
XPS[b]
O:C ratio of fresh GO
0.52
0.14
73%
0.54
0.15
72%
O:C ratio of recovered GO
Fraction of oxygen loss
[
a] 50 wt% GO, as described by Bielawski.[7] [b] 20 wt% GO, this work.
1
However, H NMR spectrum of the product mixture (Fig. 1 and
[
a]
Prof. Dr. S. Presolski
Division of Science
Yale-NUS College
S11) was dominated by a singlet at 4.56 ppm, rather than the
benzaldehyde (PhCHO) peak at 10.01 ppm or even the starting
material one at 4.67 ppm. These results were corroborated by
GC-MS, which revealed that only half of the total ion count could
be ascribed to the starting material or the expected product (10%
and 40% respectively, see SI). Among the more than a dozen
impurities present in the chromatogram, the most prominent one
was a compound with a mass spectrum consistent with dibenzyl
ether (DBE). Heating GO in phenol under the same reaction
conditions did not result in the release of small molecules into the
solvent,[ suggesting that the impurities were products of heating
16 College Ave West, Singapore 138527
[b]
Prof. Dr. Martin Pumera
Center for Advanced Functional Nanorobots
Department of Inorganic Chemistry, Faculty of Chemical Technology
University of Chemistry and Technology Prague
Prague 6, Czech Republic
E-mail: pumera.research@gmail.com; www.nanorobots.cz; Twitter:
@PumeraGroup
Supporting information for this article is given via a link at the end of
the document.
10]
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.201809979
COMMUNICATION
BnOH in the presence of GO. Both NMR and GC-MS data agree
that the condensation of two benzyl alcohol molecules was the
major reaction that took place and mechanistically, it could be
attributed to catalysis by the acidic groups present on GO. While
the same side product has been observed with other benzyl
alcohol oxidation catalysts,[11] further investigations on the precise
reaction pathway in the case of GO is needed. As far as the
oxidation reaction is concerned, only about 40% of the starting
material was converted to PhCHO with the concomitant loss of
~40 mol% oxygen from the GO with respect to the starting
material (see Supporting Information for detailed calculations).
Therefore, it was reasonable to hypothesise that GO acted as a
simple oxidant and we set further tests to ascertain whether the
carbonaceous material is a catalyst, a catalytic precursor, or a
reagent.
Figure 2. Product mixture composition as determined by GC-MS with 5%
uncertainty.
In the original “carbocatalysis” report graphene oxide is touted to
be a catalyst for alcohol oxidation,[7] yet later the same group
stated that alcohols can be used as potent reductants of GO in a
chemically irreversible reaction.[12] Our analysis shows that
indeed, GO is a reagent for benzyl alcohol oxidation, not a catalyst
as previously claimed.
Lastly, it is important to point out that while we replicated the
amounts and conditions of GO oxidation based on the initial report
[
7], the reactions described here are not representative of how
most organic chemists perform synthetic transformations. Notice
that neat benzyl alcohol has a concentration of 9.6 M, which is
100–1,000 times greater than what is typical for small-scale
organic synthesis. And for industrial-scale operations a catalyst
loading of 200 wt% is orders of magnitude greater than the norm.
To illustrate the departure from the standard conditions generally
employed in an organic chemistry laboratory, we photographed
the reaction vials before and after overnight heating, as well as
their chloroform filtrates (Figure 3).
1
Figure 1. H NMR spectrum of oxidation products of the 20 wt% reaction (50 μL
BnOH, 10 mg GO).
Increasing the amount of GO to 200 wt% (540 mol% oxygen to
benzyl alcohol) resulted in nearly 90% benzaldehyde yield, with
no benzoic acid detected by either of the two analytical techniques
we employed. The condensation product DBE, however, was still
present in appreciable amounts as shown in Figure 2.
Nevertheless, just like in the substoichiometric reaction, we could
not reject the hypothesis that GO was a simple oxidizing reagent,
given the high catalyst loading. Hence we reused the GO, but
contrary to the original report which states that “At high initial
loadings (≥50 wt%), no change in activity was observed for up to
A
B
C
0 cycles”[7] the relative amount of benzaldehyde we observed
Figure 3. Appearance of vials containing (from left to right): 10 mg GO, 50μL
BnOH on 10 mg GO (20 wt%), 50μL BnOH, 50μL BnOH on 100 mg GO (200
wt%) before the reaction (A), after 24 hours at 100°C (B), and the respective
chloroform filtrates (C).
1
steadily decreased with each subsequent run. Interestingly,
however, the amount of unreacted BnOH remained low, while the
concentration of dibenzyl ether increased. This is most likely a
consequence of two different reaction rates, where the
condensation prevails only when the GO is depleted of oxygen.
At 20 wt% loading, the GO flakes were barely covered by benzyl
alcohol, and when excess “carbocatalyst” was used, the starting
material was completely adsorbed on it and could not facilitate
stirring. Furthermore, the chloroform filtrate of the 200 wt%
reaction was orange, clearly indicating that in addition to PhCHO
1
and DBE, which were identified by GC-MS and H NMR, more
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.201809979
COMMUNICATION
complex colourful impurities also formed in what appears to be a Acknowledgements
heterogeneous gas-phase reaction.
In conclusion, we were able to reproduce the synthesis of
graphene oxide that was reported to be an oxidation
carbocatalyst.[ However, we demonstrated that it primarily acts
as a reagent if we must adhere to the IUPAC definition. We find
that given the extensive loss of oxygen from GO, the low yields of
benzaldehyde formed under substoichiometric conditions, and
the poor recyclability of the graphene oxide even when employed
in large excess, using the word “catalyst” in the context of benzyl
alcohol oxidation is unsubstantiated. Furthermore, despite many
attempts we were unable to obtain quantitative conversion of
starting material to desired product due to the formation of
dibenzyl ether, as well as other impurities. The procedure also
does not offer a practical synthetic route to either PhCHO or DBE,
owing to the difficulty of filtering the heterogeneous
oxidant/dehydration reagent.
SP thanks Yale-NUS College for start-up grant funding. This work
was supported by the project Advanced Functional Nanorobots
7]
(
reg. No. CZ.02.1.01/0.0/0.0/15_003/0000444 financed by the
EFRR).
Keywords: graphene oxide • catalysis • carbocatalyst •
oxidation • critical analysis
[
[
1]
2]
M. J. Allen, V. C. Tung, R. B. Kaner, Chem. Rev. 2010, 110, 132-145.
a) D. R. Dreyer, C. W. Bielawski, Chem. Sci. 2011, 2, 1233-1240; D. S.
Su, G. Wen, S. Wu, F. Peng, R. Schlögl, Angew. Chem. Int. Ed. 2017,
56, 936-964; N. Morimoto, K. Morioku, H. Suzuki, Y. Nakaid, Y. Nishina,
Chem. Commun. 2017, 53, 7226-7229.
[4]
N Morimoto, T. Kubo, Y. Nishina, Sci. Rep. 2016, 6, 21715.
E. C. Su, K. P. Loh, Acc. Chem. Res. 2013, 46, 2275-2285.
S. Navalon, A. Dhakshinamoorthy, M. Alvaro, Chem. Rev. 2014, 114,
6179-6212.
[3]
5]
[
Therefore, we can conclude that despite the claims of
[
7]
carbocatalysis in the original report, GO does not act as a
convenient catalyst in the conversion of benzyl alcohol to
benzaldehyde, but as an obfuscated reagent. These findings are
in line with a recent study, which utilized GO in an oxidative C-H
coupling reaction with the concomitant production of reduced
graphene oxide (rGO).[13] We do not imply that this would be the
[
[
6]
7]
a) B. Li, Z. Xu, J. Am. Chem. Soc. 2009, 131, 16380-16382; b) L.
Pacosov, C. Kartusch, P. Kukula, J. A. van Bokhoven, ChemCatChem
2011, 3, 154-156.
D. R. Dreyer, H.-P. Jia, C. W. Bielawski, Angew. Chem. Int. Ed. 2010, 49,
6813-6816.
[8]
IUPAC Golden Book: https://goldbook.iupac.org/html/C/C00876.html
a) L. J. Cote, F. Kim, J. Huang, J. Am. Chem. Soc. 2009, 131, pp 1043-
1049; b) W. S. Hummers, R. E. Offeman, J. Am. Chem. Soc. 1958, 80,
[
[
9]
case for all “carbocatalysis” reactions – some of them may truly
be catalytic,[14] but one needs to adhere to the IUPAC definition of
catalysis and be critical of their own datasets. Because only when
the community ensures that proper control experiments,
1
339.
10] Z. Sofer, P. Simek, M. Pumera, Phys Chem Chem Phys. 2013, 15, 9257-
264.
9
analytical rigor, and reproducibility
‘
are applied to
[
[
11] F. Adam, A. E. Ahmed, S. Lih, J Porous Materials 2008, 15, 433-444.
12] D. R. Dreyer, S. Murali, Y. Zhu, R. S. Ruoff, C. W. Bielawski, J. Mater.
Chem. 2011, 21, 3443-3447.
carbocatalysis’,[ it could gain the trust of chemists outside this
15]
emergent new field.
[
[
[
13] K. Morioku, N. Morimoto, Y Takeuchi, Y. Nishina, Sci. Rep. 2016, 6,
25824.
14] N. Morimoto, K. Morioku, H. Suzuki, Y. Nakai, Y. Nishina, Chem.
Commun. 2017, 53, 7226-7229.
Experimental Section
15] D. Deng, K. S. Novoselov, Q. Fu, N. Zheng, Z. Tian, X. Bao, Nature
Nanotechnology 2016, 11, 218-230.
GO reactions with 50 μL benzyl alcohol (480 μmol) were carried out at 20
wt% (10 mg, 260 μmol oxygen) or 200 wt% (100 mg, 2.6 mmol oxygen)
loading in stirred 4.5-mL glass vials, tightly sealed with a Teflon-coated,
silicon-lined caps. After heating at 100 °C for 24 h, the vials were allowed
to cool down, 1 mL of CDCl
filtered through 0.2 μm PTFE filter. The filtrate was directly analysed by H
NMR, then 50 μL of this concentrated product mixture was diluted 30x into
3
was added in and the solid contents were
1
3
1.5 mL of CHCl for GC-MS analysis.
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.201809979
COMMUNICATION
Entry for the Table of Contents (Please choose one layout)
Layout 2:
COMMUNICATION
Stanislav Presolski, Martin Pumera*
Page No. – Page No.
Myths and Realities of Graphene
Oxide: Carbocatalyst or Reagent
Text for Table of Contents
This article is protected by copyright. All rights reserved.