One-step hydrothermal amino-grafting of graphene oxide as an efficient solid base catalyst

Article abstract of DOI:10.1039/c3cc49529a

A one-step hydrothermal route is developed to prepare amino-grafted graphene oxide as an environmentally benign heterogeneous solid base catalyst.

Full text of DOI:10.1039/c3cc49529a

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
One-step hydrothermal amino-grafting of
graphene oxide as an efficient solid base catalyst†
Cite this: Chem. Commun., 2014,
50, 4305
Yicheng Zhang,a Chunlin Chen,b Guangjun Wu,a Naijia Guan,a Landong Li*a and
Jian Zhang*b
Received 16th December 2013,
Accepted 2nd March 2014
DOI: 10.1039/c3cc49529a
www.rsc.org/chemcomm
A one-step hydrothermal route is developed to prepare amino-grafted conditions.1c Another route is oxidation–amination but it suffers
graphene oxide as an environmentally benign heterogeneous solid base from the remaining acidic functional groups being prone to catalyze
catalyst.
side-reactions.3b In addition, the use of a hazardous reagent like
n-BuLi or SOCl2 inevitably results in a high capital cost and
With the developments in metal-free carbon catalysis and graphene complicated procedures. Herein we report a convenient and cheap
materials, the catalytic applications of graphene and its precursor route to graft the as-synthesised graphene oxide (GO) with large
graphite oxide have been widely investigated in a variety of reactions.1 amounts of nitrogen-containing groups. In principle, the GO was
The most attractive property of a graphene-based material as the produced from graphite according to the Hummers method with
catalyst is its controllable surface properties allowing the function- some modifications.4 The GO was dispersed into the aqueous
alization of active groups. For instance, the sulfonated graphene was solution of primary amines and then transferred into an autoclave
applied as a water-tolerant solid acid catalyst and showed a high for hydrothermal treatment at 333–453 K. The amino-grafted GO
activity in hydrolysis of ethyl acetate.2 It is possible to graft nitrogen materials were labeled as MAGO (methylamine), EAGO (ethylamine),
atoms on the matrix of graphene sheets and thus endue the carbon PAGO (propylamine) and BAGO (butylamine).
solid with the surface basicity. More examples of N-incorporated
The morphological properties of GO and the representative
nanocarbons for base catalysis can be related to carbon nanotubes amino-functionalized GO (PAGO) were determined by means of
(CNTs).3 Bitter and his co-workers used N-doped CNTs to catalyze the AFM and TEM. The AFM image in Fig. 1a reveals that the lateral
Knoevenagel condensation between benzaldehyde and ethylcyano- dimensions of the as-prepared GO ranged between several hundred
acetate, and found that the activity was determined by the amount of
pyridinic N components.3a Tessonnier et al. applied amino-grafted
CNTs as a solid base catalyst for transesterification of glyceryl
tributyrate with methanol and observed a superior activity to the
commercial hydrotalcite.3b A survey of the literature shows that the
use of N-incorporated graphene as a base catalyst has been still
limited in several specific reactions, i.e. hydrolysis of ethyl acetate,1c
and, importantly, there is still a lack of fundamental research into the
working mechanism of graphene base catalysis.
Amino-grafting of carbon materials is performed via the
consecutive deprotonation–carbometalation by n-BuLi followed by
an electrophilic substitution with bromotriethylamine or 2-diethyl-
aminoethylbromide, which is usually conducted under the dry
a Key Laboratory of Advanced Energy Materials Chemistry (Ministry of Education),
College of Chemistry, Nankai University, Tianjin 300071, P.R. China.
E-mail: lild@nankai.edu.cn; Tel: +86-22-23500341
b Department of New Energy Technology, Ningbo Institute of Materials Technology
& Engineering, Chinese Academy of Sciences, Ningbo 315201, P.R. China.
E-mail: jzhang@nimte.ac.cn; Fax: +86-574-87615956
† Electronic supplementary information (ESI) available: Experimental details and
Fig. 1 AFM (a, b) and TEM (c) images of GO and TEM images of PAGO (d, e).
Chem. Commun., 2014, 50, 4305--4308 | 4305
more characterization results. See DOI: 10.1039/c3cc49529a
This journal is ©The Royal Society of Chemistry 2014

View Article Online
Communication
ChemComm
nanometers and several micrometers. The apparent thickness was to sp2 C, C–O groups, and CQO groups, respectively (Fig. 2a). For
about 0.85 nm (Fig. 1b), corresponding to the monolayer GO sheet.5 amino-grafted GO, the intensities of C–O and CQO groups decreased
The thickness of GO is much larger than that of monolayer distinctly while the signal of C–N groups appeared at 284.5 eV, clearly
graphene (0.34 nm) due to the presence of oxygen functional indicating that the functionalization of GO is mainly achieved via
groups at the periphery or in the basal plane of GO.6 The TEM reaction with oxygen groups to form C–N bonds. The N1s spectra of
image in Fig. 1c shows a typical crumpling structure, indicating amino-grafted GO samples (Fig. 2b) comprise three components, i.e.
that the GO sheet is flexible. After propylamine functionalization, pyridinic N,5,9a amine N9b,c and ammonium N9d centering at 398.4,
the morphology of GO was well preserved and a flexible PAGO sheet B400.0 and 402.1 eV, respectively. Deconvolution identified the
can be observed (Fig. 1d). The high resolution TEM image of PAGO prevailing fraction of grafted amine N species, confirming the major
in Fig. 1e reveals the existence of 3–4 layers in the PAGO nanosheet. pathway of C–N formation via the binding of amines at the C–O sites.
The interlayer spacing was calculated to be ca. 0.5 nm, which is
Thermal stability of surface functional groups is one of the most
higher than that of monolayer graphene but lower than that of important properties of a catalyst. TG-MS profiles of GO and amino-
monolayer GO, indicating that graphene oxide was partially functionalized GO are shown in Fig. 3. In agreement with the
reduced by removing oxygen-containing functional groups during previous reports,10 the major weight loss (B27%) of GO at around
the reaction. The functionalized GO layers were crumpled, inter- 473 K was assigned to the release of COx and H2O from the labile
rupted and full of defects, allowing the catalytic use with great oxygen groups. The weight loss between 500 and 950 K might be
accessibility to the reactant molecules.
related to the removal of more stable oxygen functional groups.11
We applied Raman and XPS techniques to characterize The desorption profiles of functional groups changed a lot after the
structural and surface properties of graphene materials. As seen in treatment with primary amines. First, the weight loss of amino-
Fig. S1a (ESI†), GO and grafted GO show similar strong disorder- grafted GO below 500 K was less than 6% and the amount of COx
induced D bands, in-plane vibration of the graphene lattice G bands greatly reduced, revealing the reduced number of oxygen species.
and their overtones (2D, D + G and 2G). It is accepted that the Second, the stability of functional groups was improved and the peak
intensity ratio of the D and G bands (ID/IG) is inversely proportional maximum in DTG curve shifted upwards (Fig. 3b). Desorption
to the density of defects of graphene-based materials.7 After the profiles of CO and CO2 became more dispersive and the less stable
hydrothermal treatment, the measured ID/IG ratio of GO increased components centering at around 520 K almost disappeared (Fig. 3c).
from 1.01 to 1.17, 1.22, and 1.20 for MAGO, EAGO, and PAGO, As shown in Fig. 3d, the length of the carbon chain in the primary
respectively. Such a decrease in ordering of carbon atoms can be amines can significantly affect the stability of amino groups, i.e. the
majorly related to the functionalization of amino groups via covalent shorter the carbon chain the higher the desorption temperature.
bonds,8 as confirmed by the XPS results in Fig. S1b (ESI†). The Typically, the release of methylamine, ethylamine and propylamine
intensive peak centering at ca. 400.0 eV clearly proved the existence from MAGO, EAGO and PAGO started at ca. 800 K, 600 K and 500 K,
of grafted N atoms in the treated GO samples. Elemental analysis respectively. The amine groups in functionalized GO are thermally
shows that the total contents of nitrogen are 18.0, 10.8 and 8.6% for
MAGO, EAGO and PAGO, respectively. The reason for the difference
in grafting efficiency could be attributed to the effect of chain length
in primary amines that may affect the reactivity with oxygen groups
in graphene sheets. The hydrothermal treatment also resulted in a
significant decrease of the oxygen content from 24.1% to 7.9–9.3%,
probably due to the partial reduction of GO.
The C1s and N1s spectra of GO and amino-grafted GO are
illustrated in Fig. 2. Typically, the C1s profile of GO can be fitted by
three elementary peaks, i.e. 284.8, 286.8 and 288.1 eV, corresponding
Fig. 2 High-resolution C1s (a) and N1s (b) XP spectra of GO and amine-
Fig. 3 TG-MS results of GO and amine-grafted GO: TG (a) and DTG
functionalized GO.
(b) profiles, the MS signal of released CO2 (c) and amines (d).
4306 | Chem. Commun., 2014, 50, 4305--4308
This journal is ©The Royal Society of Chemistry 2014

View Article Online
ChemComm
Communication
stable at below 500 K, allowing their applications in a variety of compare the activity of different samples. The conversion was as low
catalytic reactions. It is even possible to use MAGO as the catalyst or as 2% without the presence of any catalyst. When amino-grafted GO
inert support for some gas-phase reactions at temperatures lower was used as a catalyst, the methyl benzoate conversion ranged
than 800 K.
between 45 and 70% and PAGO exhibited the highest catalytic
Due to the environmentally friendly and sustainable properties, activity. It is known that the alkyl group can donate electrons to the
functionalized carbon materials have been applied as the solid N atom in amine and the electrophobic effect will strengthen the
acid or base catalysts. For example, amine modified carbon basicity. Generally, long carbon chains will make the electron
nanotubes exhibited a high catalytic activity in transesterification donation easy and thus enhance the basicity. BAGO should possess
reaction.3b In this work, the basicity of amino-functionalized GO the strongest basicity and exhibit the highest activity. However, with
was roughly analyzed by using different Hammett indicators. The the increasing length of the carbon chain, the steric effect becomes
results showed that all amino-grafted GO materials are super bases intensive and the transesterification reaction will be obstructed.
by processing basic sites with pKa values of 37–39. Amino-grafted PAGO showed the highest basic catalytic activity as a result of
GO is first tested as a catalyst for the Knoevenagel condensation balancing these two important factors. We note that PAGO even
reaction. The reaction between benzaldehyde and methylene exhibited a higher activity than pure n-propylamine (54% methyl
cyanide is very quick and B100% benzaldehyde conversion can benzoate conversion at 8 h), which can be explained by the
be obtained within 4 hours using all amino-grafted GO as the possibility that the linkage on the graphene matrix as a big
catalyst. If diethyl malonate was used as a reactant instead of p-conjugated system may benefit the transformation of primary
methylene cyanide, the reaction rate was lowered and benzaldehyde amine into tertiary amine to result in an enhanced basicity.
conversion of 22.1, 28.3 and 53.2% was obtained catalyzed by MAGO,
To elucidate the working mechanism of surface functional
EAGO and PAGO, respectively. The reason was attributed to the groups, we attempted to correlate the catalytic activity with the
higher stereo-hindrance of diethyl malonate than that of methylene quantitative analysis of N species by the high-resolution XPS.
cyanide when reacting with benzaldehyde. No by-products or A series of PAGO samples were prepared hydrothermally at
derivatives from amino-grafted GO can be identified by GC-MS different synthesis temperatures (denoted as PAGO-t) and
analysis, indicating that the reaction was indeed a heterogeneous evaluated in Michael addition and transesterification reactions.
catalysis process and the GO samples were effective catalysts for For the PAGO-333 sample, the deconvolution of the N1s spectrum
the Knoevenagel condensation reaction. On the basis of reaction shows two major components at 399.9 and 402.1 eV, being
data, we proposed that the order of overall basic strength should be assigned to amine and ammonium N, respectively, (Fig. S3, ESI†).
PAGO 4 EAGO 4 MAGO.
Upon increasing the hydrothermal temperature, the elementary
The activity of functionalized GO was further tested in peak at 398.4 eV corresponding to pyridinic N species became
Michael addition and transesterification reaction. In Michael intense. As seen in Table 1, the fraction of pyridinic N increases
addition reaction, nitromethane reacts with methylacrylate to from 0.39 to 1.54% with the increase in synthesis temperature from
yield methyl 4-nitrobutanoate as the main product. The methyl- 333 to 453 K. The amount of amine N species distributed rather
acrylate conversion was as low as 6% without any catalyst but randomly and the maximum of 6.12% appeared at hydrothermal
dramatically increased to 53.1% when MAGO was used as the temperature of 353 K. In contrast, the oxygen functional groups
catalyst. The conversion was 37.9 and 24.2% for EAGO and started to decompose at 373 K, which is detrimental to the reaction
PAGO, respectively. KF/Al2O3 is a kind of super strong solid between the epoxy group and amine. The molecular water may
base catalyst and its Hammett value is nearly the same as compete with amine in reacting with the epoxy group at high
amino-functionalized GO.12 When we used KF/Al2O3 (2 mmol gÀ1
equal to the N content in amino-functionalized GO) as the catalyst, and amine N species.
,
temperature, leading to the decrease in the percentage of total N
the methylacrylate conversion is 47.0% ranking behind MAGO but
For the Michael addition reaction, PAGO-333 seems to be
before EAGO or PAGO. From the research on Michael addition inactive after background subtraction, i.e. 6% in the blank
reaction, we know that an amine is apt to react with nitromethane experiment. The methylacrylate conversion gradually increased
even at room temperature and thus all amine-type N species are
totally blocked by nitromethane during the reaction. Therefore, the
catalytic activity of amino-grafted GO should mainly originate from
pyridinic N species, especially taking into account the relatively low
basicity of ammonium N species.
In transesterification, methyl benzoate reacts with excess
ethylene glycol to produce ethylene glycol acetate benzoate. It is
seen that methyl benzoate conversion increased with the reaction
time for all amine-functionalized GO samples (Fig. S2, ESI†). After
42 h, the reaction nearly completed and all grafted GO catalysts
displayed the selectivity of ca. 96%. The analysis of the reactant by
GC-MS showed the absence of leached amines or other derivatives,
confirming a good stability of active components under the reaction
conditions. The methyl benzoate conversion at 8 h was chosen to
Table 1 The contents of specific N species in PAGO and the catalytic
activities in the Michael addition and transesterification reaction
Surface N contenta/%
Conversion/%
Michael
Sample
Blank
PAGO-333 6.3
PAGO-353 8.6
PAGO-373 8.5
PAGO-393 7.3
PAGO-423 6.6
PAGO-453 6.3
Total Pyridinic Amine addition Transesterification
—
—
—
6
6
2
65
70
58
50
55
53
0.39
0.99
1.06
1.08
1.45
1.54
5.16
6.12
4.47
3.78
4.04
3.92
25
29
30
43
49
a
N content was determined by XPS.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 4305--4308 | 4307

View Article Online
Communication
ChemComm
reaction between methyl benzoate and ethylene glycol, while
pyridinic N species were responsible for the Michael addition
between nitromethane and methylacrylate. The as-obtained
amino-grafted GO is proved to be a kind of environmentally
benign heterogeneous solid base catalyst that can be applied to
a series of base-catalysed reactions.
This work was supported by the Collaborative Innovation
Center of Chemical Science and Engineering (Tianjin), the Ministry
of Education of China (NCET-11-0251 and IRT13022), 111 Project
(B12015) and the National Natural Science Foundation of China
(51202262, 50921004, and 21133010).
Notes and references
Fig. 4 Correlation between the specific N content in PAGO and the
catalytic activity.
1 (a) J. Liang, Y. Jiao, M. Jaroniec and S. Z. Qiao, Angew. Chem., Int. Ed.,
2012, 51, 11496; (b) S. Yang, X. Feng, X. Wang and K. Mu¨llen, Angew.
Chem., Int. Ed., 2011, 50, 5339; (c) C. Yuan, W. Chen and L. Yan, J. Mater.
Chem., 2012, 22, 7456; (d) S. Verma, H. P. Mungse, N. Kumar,
S. Choudhary, S. L. Jain, B. Sain and O. P. Khatri, Chem. Commun.,
2011, 47, 12673; (e) D. R. Dreyer, R. S. Ruoff and C. W. Bielawski, Angew.
Chem., Int. Ed., 2010, 49, 9336; ( f ) D. R. Dreyer and C. W. Bielawski,
Chem. Sci., 2011, 2, 1233; (g) D. R. Dreyer and C. W. Bielawski, Adv. Funct.
Mater., 2012, 22, 3247; (h) D. W. Boukhvalov, D. R. Dreyer,
C. W. Bielawski and Y. W. Son, ChemCatChem, 2012, 4, 1844;
(i) H. P. Jia, D. R. Dreyer and C. W. Bielawski, Adv. Synth. Catal., 2011,
353, 528; ( j) D. R. Dreyer, K. A. Jarvis, J. F. Paulo and C. W. Bielawski,
Macromolecules, 2011, 44, 7659; (k) D. R. Dreyer, K. A. Jarvis, P. J. Ferreira
and C. W. Bielawski, Polym. Chem., 2012, 3, 757; (l) A. D. Todd and
C. W. Bielawski, Catal. Sci. Technol., 2013, 3, 135.
from 6 to 49% with the increase in synthesis temperature, being
similar to the change in the amount of pyridinic N. A direct
correlation between the fraction of pyridinic N and the methyl-
acrylate conversion was obtained (Fig. 4a), indicating a determining
role of the pyridinic N component especially after ruling out the
effect of amine N as discussed above. For transesterification
reaction, both pyridinic N and amine N showed certain basicity,
and their contributions to the activity are difficult to be distin-
guished. To solve the problem, we passivated the amine species in
PAGO-t samples with nitromethane at the reaction temperature
(323 K for 12 h, followed by drying overnight under vacuum at
323 K) before transesterification. The nitromethane treated PAGO-t
catalysts surprisingly displayed no activity in transesterification
reaction, while the activity in Michael addition reaction was well
preserved. We therefore concluded that the amine N should be
the actual active component for transesterification reaction, as
confirmed by the proportional relationship between the content
of amine N and the conversion rate of methyl benzoate (Fig. 4b).
To sum up, we demonstrate a facile and economical route to
functionalize GO by nitrogen-containing groups via hydrothermal
treatment in the aqueous solution of primary amines. The
amount of total nitrogen atoms and the distribution of nitrogen
components can be well adjusted by controlling the synthesis
parameters like the type of amines and hydrothermal tempera-
ture. The amino-grafted GO samples displayed a super basicity
with the apparent pKa value of 37–39 and exhibited remarkable
catalytic activities in several typical homogeneous reactions, i.e.
Knoevenagel condensation, Michael addition and transesterifica-
tion. The amine N species coordinated the transesterification
2 J. Ji, G. Zhang, H. Chen, S. Wang, G. Zhang, F. Zhang and X. Fan,
Chem. Sci., 2011, 2, 484.
3 (a) S. van Dommele, K. P. de Jong and J. H. Bitter, Chem. Commun.,
2006, 4859; (b) J. P. Tessonnier, A. Villa, O. Majoulet, D. S. Suand and
¨
R. Schlogl, Angew. Chem., Int. Ed., 2009, 48, 6543.
4 W. S. Hummers and R. E. Offeman, J. Am. Chem. Soc., 1958, 80, 1339.
5 X. Li, H. Wang, J. T. Robinson, H. Sanchez, G. Diankov and H. Dai,
J. Am. Chem. Soc., 2009, 131, 15939.
´
´
6 J. I. Paredes, S. Villar-Rodil, A. Martınez-Alonso and J. M. D. Tascon,
Langmuir, 2008, 24, 10560.
7 L. G. Cançado, K. Takai, T. Enoki, M. Endo, Y. A. Kim, H. Mizusaki,
˜
A. Jorio, L. N. Coelho, R. Magalhaes-Paniago and M. A. Pimenta,
Appl. Phys. Lett., 2006, 88, 163106.
8 H. Yang, C. Shan, F. Li, D. Han, Q. Zhang and L. Niu, Chem.
Commun., 2009, 3880.
9 (a) S. R. Kelemen, M. Afeworki, M. L. Gorbaty and P. J. Kwiatek,
Energy Fuels, 2002, 16, 1507; (b) R. Pietrzak, Fuel, 2009, 88, 1871;
´
(c) G. Soto, E. C. Samano, R. Machorro, M. H. Farıas and L. Cota-
Araiza, Appl. Surf. Sci., 2001, 183, 246; (d) R. J. J. Jansen and H. van
Bekkum, Carbon, 1995, 33, 1021.
10 (a) S. Stankovich, D. A. Dikin, G. H. B. Dommett, K. M. Kohlhaas,
E. J. Zimney, E. A. Stach, R. D. Piner, S. T. Nguyen and R. S. Ruoff,
Nature, 2006, 442, 282; (b) G. I. Titelman, V. Gelman, S. Bron,
R. L. Khalfin, Y. Cohen and H. Bianco-Peled, Carbon, 2005, 43, 641.
11 J. Shen, Y. Hu, M. Shi, X. Lu, C. Qin, C. Li and M. Ye, Chem. Mater.,
2009, 21, 3514.
12 H. Kabashima, H. Tsuji, S. Nakatab, Y. Tanaka and H. Hattori, Appl.
Catal., A, 2000, 194–195, 227.
4308 | Chem. Commun., 2014, 50, 4305--4308
This journal is ©The Royal Society of Chemistry 2014