Communications
Table 1: Oxidation of benzyl alcohol into benzaldehyde and benzoic acid
nickel). In further support of a metal-free, carbocatalyzed
under various conditions.[a]
oxidation process, no reaction was observed when natural
flake graphite (used as a starting material to prepare GO) was
examined as an oxidant under otherwise identical conditions.
To ascertain the fate of the GO material used in these
oxidation reactions, the residual catalyst was separated from
the reaction mixture by filtration and characterized. Com-
pared to the GO starting material, the FTIR spectrum (KBr)
of the carbon material isolated at the conclusion of the
reaction exhibited an attenuated signal (relative to the other
Catalyst[b] Loading
[wt%]
T
PhCHO
PhCO2H
[%][d]
TON
[e]
[8C] [%][c]
(ꢀ10À2
)
GO
20
20
100 24
100 22
0
0
0
0
0
0
0
0
0
1.1
1.0
–
GO[f]
graphite[g] 20
100
75
100
150
0
5
5
6
GO
GO
GO
GO
GO
GO
GO
GO
GO
GO
GO
GO
5
5
5
0.93
0.93
1.1
0.56
1.1
1.3
0.56
1.1
20
20
20
50
50
50
200
200
200
75 12
100 24
150 27
75 30
100 61
150 85
75 73
peaks in the spectrum) at n = 3150 cmÀ1 (O H stretch), as
À
well as the disappearance of diagnostic signals at 1685 cmÀ1
À1
and 1140 cmÀ1 (attributed to
=
(C O stretch), 1280 cm
0
epoxide absorbances[15]). The FTIR spectrum also revealed
new signals at 1650 cmÀ1 and 1500 cmÀ1, which were attrib-
uted to the presence of aromatic and olefinic species. In
addition, the isolated material exhibited a significantly higher
powder conductivity and C/O ratio than the GO starting
material (15 SmÀ1 versus 2.2 ꢀ 10À5 SmÀ1 and 7:1 versus 2:1,
respectively).[16] Collectively, these results suggested to us that
the GO catalyst underwent partial reduction during the
conversion of PhCH2OH into PhCHO and afforded a carbon
product that was similar to the r-GOs and CMGs that had
been previously synthesized by other methods.[3a,b] This
reduction process appeared to be concomitant with the
release of water, as determined by NMR spectroscopy and
coulimetric Karl Fischer titration of the product mixture.
To determine whether the GO was directly oxidizing the
alcohol or functioning as a catalyst with ambient oxygen as
the terminal oxidant,[17] the aforementioned oxidation reac-
tion was performed under an atmosphere of nitrogen. After
24 hours at 1008C, an aliquot removed from the reaction
mixture was found to contain less than 5 mol% PhCHO, as
determined by NMR spectroscopy. Underscoring the impor-
tance of oxygen, continued heating of this reaction mixture
under ambient atmosphere afforded 23% conversion into
PhCHO after another 24 hours. Notably, upon separation of
the catalyst from the substrate by filtration, the GO was found
to retain its oxidative properties and was successfully reused
for multiple cycles.[18]
0.1
12
4
7
51
1.6
0.36
0.43
0.46
100 92
150 49
[a] Unless otherwise noted, all reactions were performed in neat benzyl
alcohol using the catalyst loading and temperature indicated for 24 h.
[b] Unless otherwise noted, the GO was prepared using the Hummers
method.[5] [c] Refers to the conversion of benzyl alcohol into benzalde-
hyde, as determined by 1H NMR spectroscopy. [d] Refers to the
conversion of benzyl alcohol into benzoic acid, as determined by
1H NMR spectroscopy. [e] The turnover number (TON) was calculated as
a ratio of the mol of oxidized product/mass GO. [f] Catalyst was
prepared using the Staudenmaier method.[6] [g] Natural flake graphite
was purchased from Bay Carbon, Inc. or Alfa Aesar and used without
further purification.
studied under various conditions, including catalyst loading
(5–200 wt%), temperature (25–1508C), and reaction time (3–
144 h). At temperatures ꢀ 758C, the percentage conversion
peaked at 73%, even after long reaction periods (144 h) and
high catalyst loading (200 wt%). However, temperatures
above 1008C were found to increase the conversion of
PhCH2OH into PhCHO to 85%, although relatively high
catalyst loadings (ꢁ 50 wt%) were still required.[21] At
elevated temperatures (ꢁ 1008C), an appreciable amount of
PhCO2H was observed in the NMR spectra of the reaction
mixtures. Moreover, the acid content appeared to increase
with temperature, catalyst loading, and reaction time. Ulti-
mately, we found that heating PhCH2OH in the presence of
200 wt% GO at 1008C for 24 hours afforded high conversions
(> 98%) with good selectivity for the aldehyde versus the
acid (92:7).
The turnover numbers (TON; expressed as a ratio of mol
product produced/mass catalyst because of the non-Berthol-
loid nature of the GOs) for the various oxidation reactions
were calculated and summarized in Table 1. The measured
values remained relatively constant at 10À2 molgÀ1, irrespec-
tive of catalyst loading or reaction temperature, which
suggested to us that the catalyst consistently reached its
maximal activity under these conditions.
To determine if radical species were involved in the
aforementioned oxidation processes, neat PhCH2OH was
treated with GO (20 wt%) and butylated hydroxytoluene
(BHT; 20 wt%), a known radical inhibitor.[19] After 24 hours
at 1008C, less than 5% conversion into PhCHO was observed
by NMR spectroscopy. However, when 1:1:1 (by mass) of GO/
PhCH2OH/BHT were heated at 1008C for 14 hours, a 26%
conversion was observed, indicating BHT does not inhibit the
ability of GO to oxidize the alcohol to a significant extent, but
rather impedes the oxidation of the reduced catalyst. In
support of this assessment, heating GO in the presence of
cyclopropylcarbinol resulted in the formation of a number of
products, including olefinic species, as determined by NMR
spectroscopy; no aldehyde or carboxylic acid products were
observed. Similar results were obtained when chromic acid
was used to oxidize this same substrate, a process known to
proceed via radical species.[20]
Next, the scope of reactivity of the aforementioned
carbocatalyst was explored using a variety of primary and
secondary benzylic and aliphatic alcohols under conditions
that resulted in the highest conversion of PhCH2OH into its
oxidized products (200 wt% GO, 1008C, 24 h). As summar-
ized in Table 2,[22] benzylic alcohols (Table 2, entries 1–3)
As summarized in Table 1 (see Table S1 for additional
optimization studies), the oxidation properties of GO were
6814
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 6813 –6816