Selective oxidation of alkyl-substituted polyaromatics using ruthenium-ion-catalyzed oxidation

Article abstract of DOI:10.1002/chem.201405831

Ruthenium-ion-catalyzed oxidation of a range of alkylated polyaromatics has been studied. 2-Ethylnaphthalene was used as a model substrate, and oxidation can be performed in either a conventional biphasic or in a monophasic solvent system. In either case the reaction rates and product selectivity are identical. The reaction products indicate that the aromatic ring system is oxidized in preference to the alkyl chain. This analysis is possible due to the development of a quantitative NMR protocol to determine the relative amounts of aliphatic and aromatic protons. From a systematic set of substrates we show that as the length of the alkyl chain substituent on a polyaromatic increases, the proportion of products in which the chain remains attached to the aromatic system increases. Larger polyaromatic systems, based on pyrene and phenanthrene, show greater reactivity than those with fewer aromatic rings, and the alkyl chains are more stable to oxidation.

Full text of DOI:10.1002/chem.201405831

DOI: 10.1002/chem.201405831
Full Paper
&
Polynuclear Aromatic Hydrocarbons |Hot Paper|
Selective Oxidation of Alkyl-Substituted Polyaromatics Using
Ruthenium-Ion-Catalyzed Oxidation
Ewa Nowicka,[a] Meenakshisundaram Sankar,[a] Robert L. Jenkins,[a] David W. Knight,[a]
David J. Willock,[a] Graham J. Hutchings,[a] Manuel Francisco,[b] and Stuart H. Taylor*[a]
Abstract: Ruthenium-ion-catalyzed oxidation of a range of
alkylated polyaromatics has been studied. 2-Ethylnaphtha-
lene was used as a model substrate, and oxidation can be
performed in either a conventional biphasic or in a mono-
phasic solvent system. In either case the reaction rates and
product selectivity are identical. The reaction products indi-
cate that the aromatic ring system is oxidized in preference
to the alkyl chain. This analysis is possible due to the
development of a quantitative NMR protocol to determine
the relative amounts of aliphatic and aromatic protons.
From a systematic set of substrates we show that as the
length of the alkyl chain substituent on a polyaromatic
increases, the proportion of products in which the chain
remains attached to the aromatic system increases. Larger
polyaromatic systems, based on pyrene and phenanthrene,
show greater reactivity than those with fewer aromatic
rings, and the alkyl chains are more stable to oxidation.
Introduction
ported to oxidize hydrocarbons; however, RICO stands apart
due to its high reactivity and functional group tolerance.[12] It
has been shown that RICO occurs regioselectively at the most
electron-rich CꢀH bond with retention of configuration.[13] This
high regioselectivity is remarkable as it is rare amongst catalyt-
ic oxidation systems. The regioselectivity of RICO has been ex-
ploited in two different research areas. The first is the oxidative
fragmentation of aromatic compounds to carboxylic acids in
synthetic organic chemistry,[9b,14] and the second is in the selec-
tive oxidation of alkylated aromatics, with aromatics oxidized
to CO2 and H2O leaving the aliphatic chain intact terminated
by a carboxylic acid group.[2] This method is used extensively
to characterize the aliphatic chain present in highly complex
mixtures of substituted polyaromatic compounds.[15] A number
of papers have been published using this methodology to
characterize the aliphatic chains in coal and asphaltenes.[16] The
approach is based on three hypotheses: 1) the aromatic hydro-
carbons are oxidized to CO2 and H2O in high yield, 2) oxidation
of the aliphatic portion of the compounds occurs exclusively at
the carbon atoms attached to the aromatic fragment and
3) the aliphatic chain is not further oxidized. Although these
hypotheses have been tested using a limited range of sub-
strates, a detailed quantitative investigation of the hypotheses
outlined above has not been reported in the literature.[17]
In spite of the reputation for high reactivity, many
researchers have tried to utilize RICO chemistry and improve
its selectivity. Harris and co-workers[11c] have reported that,
using low temperatures and limiting the level of oxidant, alkyl
substituted pyrenes can be converted with good selectivity to
the corresponding 4,5-diones or 4,5,9,10-tetraones, although
yields were generally less than 50% and no information on the
oxidation of the alkyl substituents was presented.
In 1953, Djerassi reported a new ruthenium tetroxide system
that could be used in place of the more toxic OsO4 for the oxi-
dation of various organic substrates.[1] This seminal discovery
led to the introduction of an entirely new chemistry for the
oxidation of organic compounds, which was later named as
ruthenium-ion-catalyzed oxidation (RICO), in which RuO4 is
generated in situ by the combination of a ruthenium ion and
an oxidizing agent.[2] RICO was further developed into a practi-
cal catalytic oxidation system by Sharpless et al.,[3] who report-
ed that by adding acetonitrile as a co-solvent, the selectivity,
as well as the activity, for the oxidation of alkenes to carboxylic
acids could be improved with a much lower metal loading.
Consequently, after this discovery, RICO has been reported to
be effective for many reactions, including dihydroxylation of
olefins,[4] dehydrogenation of alcohols and amines,[5] selective
monooxidation of vicinal diols,[6] ketohydroxylation,[7] oxidative
cyclization of polyenes,[8] oxidative cleavage of double and
triple bonds,[9] oxidation of heteroatoms[10] and oxidation of
saturated hydrocarbons and aromatic hydrocarbons.[11]
Initially, RICO was used mainly for the oxidative fragment-
ation of C=C bonds and aromatic compounds, which has
resulted in a reputation for the chemistry being very reactive
and difficult to control. Various oxidizing agents have been re-
[a] E. Nowicka, M. Sankar, R. L. Jenkins, D. W. Knight, D. J. Willock,
G. J. Hutchings, S. H. Taylor
Cardiff Catalysis Institute, School of Chemistry, Cardiff University
Main Building, Park Place, Cardiff, CF10 3AT (UK)
E-mail: taylorsh@cardiff.ac.uk
[b] M. Francisco
ExxonMobil, Research & Engineering Company, 1545 Route 22 East
Annandale, New Jersey 08801 (USA)
Chem. Eur. J. 2015, 21, 4285 – 4293
4285
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
In the present work, we set out to investigate the selectivity
of RICO chemistry for alkylated aromatic compounds using
NMR analysis to quantify the aromatic versus aliphatic
oxidation selectivity under various experimental conditions. To
achieve this, we have set out the following objectives: 1) to
develop methodology to quantify the extent of aromatic vs
aliphatic oxidation, 2) to establish new protocols for RICO
chemistry[6a] that allow greater control of reactivity and 3) to
understand whether it is possible to manipulate selectivity by
changing the reaction parameters. Against this background,
we report an NMR methodology to quantify the ratio of aro-
matic to aliphatic protons oxidized during a reaction. This
method is then applied to the products from RICO for a series
of alkylated aromatics giving a quantitative estimation of the
amount of aliphatic C/H oxidized relative to the amount of aro-
matic C/H oxidation. We report the results from the detailed
investigation on the RICO of a range of substrates, including
2-ethylnaphthalene, 1-decylnaphthalene, 1-octadecylnaphtha-
lene, 2-nonylphenanthrene, 9-octadecylphenanthrene, 1-butyl-
pyrene, 1-decylpyrene and 1-octadecylpyrene. To our
knowledge, this is the first quantification of the regioselectivity
of oxidation at aliphatic and aromatic carbon using an in situ
1H NMR technique for RICO.
&
Figure 1. Plot of concentration of naphthalene ( ) remaining after quench-
ing with Na2SO3 at the start of reaction (*) vs molar ratio of Na2SO3/NaIO4.
Reaction Conditions: Substrate 0.195 mmol (initial number of mmol: ꢀ),
NaIO4 1.56 mmol, RuCl3 0.012 mmol, Na2SO3 1.56 mmol, (ratio 1:1), MeCN
20 mL, H2O 10 mL, T=228C, stirring speed 500 rpm.
Oxidation of 2-ethylnaphthalene
The experimental protocol developed for large PAH systems
has to cope with the low solubility of PAHs in the aqueous
media suited for RICO. This resulted in the use of a biphasic
reaction system for long-chain alkylated aromatics. For the
smallest PAH studied, 2-ethylnapththalene,
a monophasic
approach could also be employed and so initial experiments
on alkylated polyaromatic oxidation concentrated on 2-ethyl-
napththalene and a study on the effect of the solvent system
was performed. Oxidation using a monophasic system also al-
lowed us to perform experiments in an NMR tube (details in
Experimental Section). Reaction products were present in both
the organic and the aqueous layers. Here the organic layer
analysis is presented as it gives qualitative identification of all
products.
Results and Discussion
Reaction quenching studies
RICO is an extremely active oxidation system and even at
295 K, the temperature routinely used in these studies, the
rate of reaction was rapid. Accordingly, it is important to be
able to control the reaction, especially to be able to carry out
the analysis of the substrate and the reaction products as
a function of time. Hence, a method was developed to quench
the reaction in a simple, reliable and reproducible manner.
Two strategies for effective quenching could be envisaged: de-
activating or removing the catalyst, or rapidly removing the
iodate oxidant. We decided to employ the latter approach
using the readily oxidized sulfite ion which was introduced by
adding Na2SO3 to the reaction mixture at the desired quench
time. To test the efficacy of this approach, increasing amounts
of Na2SO3 were added to reaction mixtures with naphthalene
as the substrate and the level of naphthalene remaining after
a short time was analyzed using GC. Figure 1 shows that, at
ratios of sulfite/oxidant below 1, naphthalene was still oxidized
after the introduction of sulfite, although the degree of oxida-
tion decreased as the ratio of Na2SO3 to NaIO4 was increased.
At a sulfite/oxidant ratio of 1 or greater, no oxidation of the
substrate was observed. Further experiments were also
performed in which a substrate was allowed to partially react
Figure 2 shows that changing the solvent system from
monophasic (acetonitrile+water) to biphasic (dichloro-
methane+acetonitrile+water) does not alter the rate of
before it was quenched using
a stoichiometric amount
^
Figure 2. Oxidation of 2-ethylnaphthalene in monophasic ( ) and biphasic
of sulfite. The concentration of the substrate was then
monitored over a 24 h period, and no further conversion after
the addition of sulfite occurred.
*
(
) solvent system: Kinetic studies. Reaction Conditions: Substrate 0.164
mmol, NaIO4 2.624 mmol, RuCl3 0.024 mmol, T=228C, stirring speed
500 rpm, DCM 16 mL, MeCN 7 mL, H2O 7 mL or MeCN 20 mL, H2O 10 mL.
Chem. Eur. J. 2015, 21, 4285 – 4293
www.chemeurj.org
4286
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Table 1. Oxidation of 2-ethylnaphthalene in
system: GC-MS product identification.[a]
a monophasic solvent
Reaction products[b]
Symbol
used in
Figure 3
Retention
time [min]
Name
Structure
phthalaldehyde
A
B
12.55
17.13
isobenzofuran-
1,3-dione
6-ethylisobenzofuran-
1,3H-one
C
18.10
20.06
5-ethylisobenzofuran-
1,3-dione
D
Figure 3. Comparison of GC-MS traces from 2-ethylnaphthalene oxidation in
monophasic and in biphasic solvent system. Reaction Conditions: Substrate
0.164 mmol, NaIO4 2.624 mmol, RuCl3 0.024 mmol, T=228C, stirring speed
500 rpm; A) DCM 16 mL, MeCN 7 mL, H2O 7 mL; B) MeCN 20 mL, H2O 10 mL.
6-ethylnaphthalene-
1,4-dione
E
F
21.8
6-ethyl-2-hydroxy-
naphthalene-1,4-dione
22.35
oxidation of 2-ethylnaphthalene. It is also important to note
that, according to GC-MS measurements, the product
distribution from the monophasic system was the same as the
biphasic one (Figure 3). Some minor differences are observed
in the chromatograms above 21 min; however, these are asso-
ciated mainly with analytical artefacts as a consequence of the
high temperatures required to elute the heavy aromatic com-
pounds. The total area for peaks eluting above 21 min was
<10% of the total peak area for the major products (A–F).
Hence, simplifying the reaction solvent system did not funda-
mentally change reaction rates or selectivity, but does simplify
the product analysis. This practical advantage lead us to use
the monophasic solvent system to study the oxidation of 2-
ethylnaphthalene further.
[a] Reaction conditions: Substrate 0.164 mmol, NaIO4 1.403 mmol,
RuCl3·xH2O 0.012 mmol, CH3CN 20 mL, H2O 10 mL, 295 K and 4 h reaction
time. [b] Compounds identified by GC-MS and GC.
aromatic groups are not necessarily converted to CO2 and
water. The majority of products that have the aliphatic chain
intact also retain the substituted aromatic ring, so that selectiv-
ity after 4 h under our conditions appears to favor the
oxidation of the unsubstituted ring of the naphthalene moiety.
1
In situ H NMR analysis of 2-ethylnaphthalene oxidation
Special attention was focused on the selectivity of aromatic
CꢀH versus alkyl CꢀH oxidation. The compounds identified
using GC-MS from the oxidation of 2-ethylnapthalene are
listed in Table 1. It is clear that most of the products have un-
oxidized alkyl chains, still attached to an oxidized aromatic ring
system (Figure 3). This result is surprising for RICO, as it has
been widely reported that the alkyl chain is readily oxidatively
cleaved to form aliphatic carboxylic acids with the aromatic
moiety oxidized to CO2 and water.[17b,18]
To quantify the aromatic versus aliphatic oxidation selectivity,
an in situ time online study of the ruthenium-ion-catalyzed
oxidation of 2-ethylnaphthalene was also performed inside an
NMR tube as described in the Experimental Section. It is worth
noting that, due to the lack of solubility of some products,
reactions performed in an NMR tube were designed to give
partial conversion. This was achieved by reducing the amount
of oxidant and catalyst compared to the laboratory-scale ex-
1
To confirm these results, the mixture of reaction products
was analyzed immediately after quenching of the reaction
using 2D 1H NMR correlation spectroscopy (COSY); a typical
spectrum is shown in Figure 4. The COSY NMR spectra show
coupling between protons from the CH2 group of the alkyl
region and protons from the CH3 group, indicating that the
alkyl chain is preserved. This observation is consistent with the
reaction products identified by GC-MS (Table 1, Figure 3).
Therefore, it is clear that during the RICO of 2-ethylnapthalene,
the aromatic component can be selectively oxidized in prefer-
ence to the aliphatic ethyl substituent. Table 1 also shows that
partial oxidation of the aromatic system takes place and so the
periments. In Figure 5 the H NMR signals of the entire spectra
are grouped into four regions; A) aliphatic region (3.51–
0.53 ppm), B) olefinic region (6.98–6.43 ppm), C) aromatic
region (8.63–7.51 ppm) and D) carboxylic acid/aldehyde region
(10.82–10.63 ppm).
The integrated peak areas in the defined regions are
proportional to the number of protons present in the
reaction mixture within the assigned functionality, and so
allow the relative rates of aliphatic and aromatic carbon oxida-
tion to be measured. Integration of the peaks from a reference
sample of 2-ethylnaphthalene, without acetonitrile, gave an ar-
omatic: aliphatic ratio of 1.33 in close agreement with the 1.4
Chem. Eur. J. 2015, 21, 4285 – 4293
www.chemeurj.org
4287
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
as a general subscript for A, B, C
or D according to the peak of in-
terest. IX is the integrated area
relative to TMS for the corre-
sponding region.
1
RX ¼
ꢁ 100
ð1Þ
IA þ IB þ IC þ ID
The time-on-line profiles for
the relative percentage of pro-
tons are shown in Figure 6. It is
clear that, in the first 100 min of
reaction, there is a steady de-
crease in the proton signal in
the aromatic region and a corre-
sponding increase in the aliphat-
ic region. The first data point is
measured 15 mins after the ini-
tial mixing of reactants and we
find an aromatic: aliphatic ratio
of 0.88. Linear extrapolation of
the data from the first 90 min of
the reaction back to zero gives
an estimate for the aromatic/ali-
phatic ratio of 0.91 at the start
of the reaction, in good agree-
Figure 4. Two-dimensional NMR (COSY) spectra of 2-ethylnaphthalene oxidation products in a monophasic
solvent system. Reaction Conditions: Substrate 0.164 mmol, NaIO4 2.624 mmol, RuCl3 0.024 mmol, T=228C,
stirring speed 500 rpm, MeCN 20 mL, H2O 10 mL.
Figure 6. Normalized relative percentage amounts of protons in different
1
environments determined by H NMR spectroscopy as a function of 2-ethyl-
&
^
naphthalene oxidation reaction time. RD, acid and aldehyde; RC,
~
*
aromatic; RB, double bond; RA, aliphatic.
1
Figure 5. Overlaid H NMR spectra of 2-ethylnaphthalene oxidation with
marked regions of integration. Reaction Conditions: Substrate 0.101 mmol,
NaIO4 20.421 mmol, RuCl3 0.5 mmol, T=228C, CD3CN 2 mL, H2O 3 mL.
ment with the measured ratio for 2-ethylnaphthalene and ace-
tonitrile in the absence of oxidant and catalyst. After 105 mins
of reaction the observed integrated area for each region of the
spectra becomes constant with time, indicating that the oxida-
tion reaction has stopped. From this point forward the average
aromatic/aliphatic ratio is 0.68, indicating a significant prefer-
ence for aromatic over aliphatic oxidation in the RICO reaction.
At all times the signal from protons in an olefinic or acid/alde-
hyde environment remains negligibly small.
expected. Under reaction conditions the acetonitrile signal at
2.3 ppm will also be integrated which results in an aromatic:
aliphatic ratio of 0.916. As signals from reaction products could
possibly overlap with the acetonitrile peak, we decided to
simply integrate the whole aliphatic region rather than at-
tempt to explicitly exclude acetonitrile.
The relative percentage of protons from different regions, RX,
can be calculated from Equation (1), in which we have used X
The data presented in Figure 6 only shows quantification of
the relative proportions of protons in different environments,
Chem. Eur. J. 2015, 21, 4285 – 4293
www.chemeurj.org
4288
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
vent system introduces new complications for quantitative
analysis because of the distribution of substrate and products
between the organic and aqueous phases. We have seen that
for 2-ethylnaphthalene, the change of solvent system does not
affect the activity (Figure 2) or selectivity (Figure 3) of the
oxidation reaction. Accordingly, with higher molecular weight
polyaromatic compounds, we are able to use a biphasic
solvent system and still draw comparisons with the results
from monophasic oxidation of 2-ethylnaphthalene.
In the case of higher molecular weight polyaromatics, and
those with longer alkyl chains, we build further on the idea of
using the integrated areas from different regions of the
1H NMR spectra to quantify the degree of regioselectivity. The
electronic environment of the terminal H in a long-chain alkyl
substituent of a polyaromatic compound will be affected to
a lesser extent by changes due to the oxidation of the poly
aromatic region, than will the protons in the ethyl group of 2-
ethylnaphthalene. Consequently, the terminal proton was used
as an internal standard for the quantification of the RICO prod-
ucts of long chain alkyl aromatics in the biphasic solvent
Figure 7. Absolute numbers of protons in different environments deter-
mined by H NMR spectroscopy as a function of 2-ethylnaphthalene oxida-
1
~
&
^
tion reaction time. IC aromatic; IA, aliphatic; IA +IB +IC +ID, total.
rather than the total number of protons, so that the contribu-
tion of total oxidation is not taken into account. In order to
provide information on the change of the total number of pro-
tons, the absolute values relative to TMS, IX, are plotted against
time in Figure 7. To estimate the reduction in the total number
of protons associated with organic molecules present in the
sample we first extrapolated the data between 15 and 105 min
back to time zero (R2 =0.96) obtaining a value of 3.36 in the ar-
bitrary units of Figure 7. Comparing this against the value ob-
tained at 105 min, we estimate the total absolute number of
protons to be reduced by 16%. Given the spread in the data
post 105 min we conclude that this is probably an upper esti-
mate and so the contribution to the loss in proton signal from
total oxidation is small. Figure 7 also shows that the reduction
in total signal in the first 105 min of the experiment is due to
loss of aromatic protons, while the aliphatic region of the spec-
trum remains largely unchanged over the time period of the
experiment. This is the behavior we would expect from the
products listed in Table 1, since most products show aromatic
carbon oxidation and for some of the products the partial
oxidation of the aromatic ring actually introduces additional
aliphatic protons.
1
system using H NMR spectroscopy. Conversely, the methylene
protons of the aliphatic carbon, which are attached to the aro-
matic ring, will produce a triplet that is highly susceptible to
any changes at the point of attachment of the aliphatic chain
to the aromatic ring. This triplet signal was used for the quan-
tification of the oxidation products in which the aliphatic
chains remained intact. In addition to analysis of products
using NMR spectroscopy, the conversion of the alkylated
polyaromatic compounds was determined using GC.
To demonstrate this methodology, RICO of 1-decylnaphtha-
1
lene is used as an example. Figure 8 shows the H NMR spectra
of 1-decylnaphthalene before and after RICO for 4.5 h. The sin-
glet at 0.9 to 0.7 ppm (region D in Figure 8) is the terminal
methyl group, which is used as an internal standard. The triplet
at 3.1–2.9 ppm (region A in Figure 8) is used for quantifying
the amount of products with aliphatic chain substituent intact
after RICO. The areas of these two peaks, for the 0 and 4.5 h
samples, were calculated relative to a calibrated TMS standard
and values are given in Table 2. The ratio of peak areas [Eq. (2)]
From the data presented above it is clear that RICO of
2-ethylnaphthalene, in a monophasic solvent system, leads to
selective oxidation of aromatic carbon, whilst there is little or
no oxidation of the aliphatic ethyl side chain. GC-MS analysis
has revealed that the majority of the products identified have
the ethyl chain intact (Table 1) and no products originating
from oxidative cleavage of the alkyl chain were observed.
Table 2. Oxidation of 1-decylnaphthalene in a biphasic solvent system
integrated area values (against TMS). Molar ratio substrate/oxidant 1:2.[a]
Sample
Area A
Area D
Area ratio rx =A/D
pa [%]
0 h
4.5 h
0.28
0.17
0.43
0.43
0.65
0.39
100
60
Oxidation in a biphasic solvent system
[a] Reaction conditions: Substrate 0.0264 mmol, NaIO4 0.052 mmol, RuCl3
0.0006 mmol, CDCl3 2.4 mL, CD3CN 1 mL, H2O 1 mL, T=295 K, stirring
speed 500 rpm.
The monophasic solvent system has advantages for ease of
quantification and analysis of products. However, it also has
limitations, as the solubility of higher molecular weight
polyaromatics is low. One of the main aims of this study is to
investigate a range of alkyl-substituted polyaromatics, and
study the selectivity pattern for their ruthenium-ion-catalyzed
selective oxidation. To overcome the solubility problem, the
more conventional biphasic solvent system, containing water,
acetonitrile and dichloromethane was used. This biphasic sol-
for protons A and D between the 0 and 4.5 h samples gives
a measure of the percentage of products in which the aliphatic
chain remains intact (pa)
pa ¼ ðr4:5=r0Þ ꢁ 100 %
4289
ð2Þ
Chem. Eur. J. 2015, 21, 4285 – 4293
www.chemeurj.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
significantly alter the conversion
or the extent of oxidation of the
alkyl chain. However, re-charging
the system with additional oxi-
dant and catalyst after 6 h
during a 10 h reaction does lead
to complete conversion of the
substrate and practically total
loss of the alkyl chain, presuma-
bly as carboxylic acid (Table 3,
entry 3).
Having successfully demon-
strated the quantification meth-
odology for the RICO of an alkyl
1
aromatic using H NMR spectros-
copy, our next objective was to
use the methodology to under-
stand the effect of the size of
the aromatic ring moiety, the ali-
phatic chain length and substitu-
tion position on product selec-
tivity. For this purpose, in addi-
tion to 1-decylnaphthalene, six
other substrates with different
Figure 8. 1H NMR spectra of 1-decylnaphthalene before and after RICO reaction showing regions integrated as
part of the NMR methodology for determining selectivity.
In Equation (2) rx denotes the ratio of areas of region A to
the area of region D for the xth hour sample. For 1-decylnaph-
thalene oxidation it was found that 60% of the products still
had the aliphatic chain attached to the ring system. This
demonstrates that, through careful control of the amount of
oxidant present in the reaction and the reaction temperature,
the aromatic part of the molecule can be oxidized while main-
taining the attached alkyl group intact (Table 3, entry 1). This is
aliphatic chain lengths and varying numbers of fused aromatic
rings were chosen for investigation using RICO. These sub-
strates were 1-octadecylnaphthalene, 2-nonylphenanthrene, 9-
octadecylphenanthrene, 1-butylpyrene, 1-decylpyrene and 1-
octadecylpyrene for which structures are given in Table 4.
1H NMR spectra of each substrate were measured independ-
ently before and after reaction and the peaks were grouped as
Table 4. Structures of alkyl substituted PAH compounds studied in this
work.
Table 3. Effect of chain length and ring size on the RICO activity and
selectivity for alkyl substituted polyaromatics.[a]
Substrates
Structure
Substrate
Reaction
time [h]
Conver-
sion [%]
Preserved
alky [%]
1
1-decylnaphthalene
1
2
3[b]
4
5
6
7
8
1-decylnaphthalene
1-decylnaphthalene
1-decylnaphthalene
1-octadecylnaphthalene
2-nonylphenanthrene
9-octadecylphenanthrene
1-butylpyrene
4.5
6
10
4.5
4.5
4.5
4.5
4.5
4.5
42
46
99
76
73
62
80
90
81
60
59
4
87
69
46
68
86
83
2
3
1-octadecylnaphthalene
2-nonylphenanthrene
1-decylpyrene
1-octadecylpyrene
9
4
5
6
7
9-octadecylphenanthrene
1-butylpyrene
[a] Reaction Conditions: 295 K, 1000 rpm stirring speed, substrate
0.0276 mmol, NaIO4 0.2065 mmol, RuCl3·xH2O 2 mmol. [b] After reaction
time 10 h, addition of extra 0.0276 mmol NaIO4 and 0.2065 mmol RuCl3
·xH2O after 6 h.
in sharp contrast to the usual situation with RICO chemistry in
which an excess of oxidant oxidizes aromatic groups complete-
ly, leaving the aliphatic chain as a carboxylic acid.[19] Using 1-
decylnaphthalene as a model compound the catalytic activity
was investigated for longer reaction times (entries 2 and 3,
Table 3). Increasing the reaction time from 4.5 to 6 h did not
1-decylpyrene
1-octadecylpyrene
Chem. Eur. J. 2015, 21, 4285 – 4293
www.chemeurj.org
4290
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
before; 8.50–7.50 ppm (aromatic protons), 3.40-3.20 ppm
(a-Hs, CH2) of the aliphatic chain, 2.25–2.00 ppm (b-CH2) of the
aliphatic chain, 1.91–1.90 ppm (CH2 of the remaining chain)
and 1.40–1.20 pm (terminal CH3 group). Using the
methodology discussed above, the percentage of compounds
present for which the alkyl chain remains attached to a polyaro-
matic system after 4.5 h reaction time is summarized for each
substrate in Table 3. Results for 1-decylnaphthalene after 6 h
and after a 10 h experiment, in which catalyst and oxidant
were recharged after 6 h, are also presented. It is evident that
the amount of products with intact alkyl chain after 4.5 h of
RICO, ranges from 46% for 9-octadecylphenanthrene to 87%
for 1-octadecylnaphthalene. This clearly suggests that the reac-
tivity of the alkyl chain in the RICO of alkyl aromatics is sub-
strate dependent. We have also found that control of selectivi-
ty in RICO is possible beyond that usually seen under more
forcing conditions for which, irrespective of the substrate, all
of the aromatic ring system is oxidized, leaving the aliphatic
chains as carboxylic acids.[20] In our experiments this only
occurs after the longer 10 h reaction times and with catalyst
and oxidant refreshed after 6 h (Table 3, entry 3).
Table 5. Products observed by GC-MS following the oxidation of 1-decyl-
naphthalene.
Molecular
formula
Mass
detected (EI)
Possible
structure
1
2
3
C18H24O3
C18H26O2
C19H26O2
288
274
286
4
C19H26O3
304
The data in Table 3 also begins to give some insight into this
selective oxidation chemistry. Comparison of results from 1-
decylnaphthalene and 1-octadecylnaphthalene shows that the
longer alkyl chain is preserved to a greater extent after oxid
ation (Table 3, entries 1 and 4). This observation is further
substantiated by the results of the RICO of 1-butylpyrene and
1-octadecylpyrene (Table 3, entries 7 and 9). In this case, the
conversion of both molecules was practically identical, but the
longer alkyl chain showed a lower susceptibility to oxidation. It
is also interesting to compare the catalytic activity for a range
of polyaromatic compounds that have similar lengths of ali-
phatic chain attached to different sizes of polyaromatic ring
systems. This comparison can be made by considering the
data for 1-decylnaphthalene, 2-nonylphenanthrene and
1-decylpyrene. Both the conversion and percentage of the pre-
served alkyl chain increase with increasing molecular weight of
the ring system. Therefore, it appears that increasing the size
of the ring system decreases the reactivity at the a-position for
RICO.
was small in comparison to other products found in the organ-
ic layer.
The position of substitution of the alkyl chain on the ring
system also exerts an influence on reactivity and product
selectivity. This influence has been probed by oxidation of
9-octadecylphenanthrene for comparison with results from 2-
nonylphenanthrene. The alkyl chain in the case of 9-octadecyl-
phenanthrene is attached to the position that is most likely to
be attacked by RuO4, according to the idea of a preference for
the double bond which leads to the minimum loss of aroma-
ticity on oxidation.[12b,21] The percentage of preserved a-H for
the 9-substituted phenanthrene is considerably lower than
that for the substituent at the 2-position (Table 3, entries 5 and
6). Indeed, it appears from the conversion data that for 9-octa-
decylphenanthrene the pa number can be accounted for
purely on the basis of unreacted substrate. Nevertheless, it has
been found that the product from the oxidation of this
substrate still contains preserved alkyl chain oxidized in the
a-position as shown by MS analysis (Table 6).
However, we have found that the selectivity for oxidation of
the alkyl group is highly dependent on the chain length, sub-
stitution position on the polyaromatic ring system and the
number of fused aromatic rings. When the product distribution
following the oxidation of 1-decylnaphthalene was analyzed in
further detail most of the products have preserved alkyl
chains. Based on the molecular formula information, obtained
by GC-MS analysis, a list of possible oxidation products was
proposed (Table 5). Here, two molecules are of special interest
(entries 3 and 6), as they show oxidized carbon in the
a-position without loss of the alkyl substituent from the
aromatic ring system. This preservation of the substituent was
Using an electron ionization MS method molecular ion
peaks at m/z 462 and 444 were detected and assigned to
structures of compounds having oxidized but preserved alkyl
chains. Moreover, a strong peak from unreacted substrate was
also detected. Using electrospray ionization, positive peaks at
m/z 463 and 445 were determined, which can be assigned to
structures in entries 2 and 4 in Table 6, respectively. 9-Bromo-
phenanthrene was found even in the starting material and it is
probably a substrate from the 9-octadecylphenanthrene syn-
thesis. These data suggest that there are only two products
obtained in this reaction. Using the developed NMR methodol-
ogy it was only possible to quantify unreacted substrate, be-
cause the signals from products corresponding to the a-H are
not present for these a-carbon oxidized products. In addition
b-H atoms that become a-H atoms after oxidation are
significantly shifted in the NMR spectra due to the presence of
the C=O group and are outside the integration region.
1
confirmed by the H NMR methodology. In the aqueous layer,
decanoic acid and phthalic acid were also detected. It is be-
lieved that these products are obtained after CꢀC bond cleav-
age between the aromatic ring carbon and the a-carbon of
the alkyl chain. However, the concentration of these products
Chem. Eur. J. 2015, 21, 4285 – 4293
www.chemeurj.org
4291
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
polynuclear aromatics while preserving aliphatic chain
substituents. These findings may also lead to the improvement
in the processing of heavy fractions of oil to preserve
carbon number while increasing process ability.
Table 6. Possible products of 9-octadecylphenanthrene oxidation and
contamination.
Molecular
formula
Mass
detected (EI)
Possible
structure
Experimental Section
1
2
3
4
C14H9Br
C32H44O2
C32H46
256
462
430
444
Materials and equipment
All the chemicals used were of high purity, purchased from com-
mercial sources or supplied by ExxonMobil and were used without
further purification. Commercial materials were acetonitrile
(99.95% Fisher Scientific), dichloromethane (99.5% Fisher Scientif-
ic), deuterated chlorobenzene (99.8%D), deuterated acetonitrile
(98%D, Goss), H2O (HPLC grade Sigma-Aldrich), NaIO4 (>99%
Sigma-Aldrich), RuCl3·xH2O (98% Aldrich), 2-ethylnaphthalene
(99%, Sigma-Aldrich). 1-Octadecylnaphthalene, 1-decylnaphtha-
lene, 2-nonylphenanthrene, 9-octadecylphenanthrene, 1-octadecyl-
pyrene, 1-butylpyrene and 1-decylpyrene were of >95% purity
and were supplied by ExxonMobil.
C32H44O
Catalytic oxidation reactions were performed at atmospheric pres-
sure in a 50 mL glass round-bottomed flask with a jacket connect-
ed to a thermostat (Julabo F25-ME Refrigerated/Heating Circulator)
fitted with a water pump to maintain the desired temperature of
295 K. The reaction mixture was stirred using a magnetic bar
(500 rpm). For reactions performed on a smaller scale, a 5 mL
round bottom flask maintained at 295 K was used with the same
stirring speed.
From the above results it is clear that the amount of alkyl
chain that is preserved under RICO depends on the ring size,
the length and the location of the attached aliphatic chain.
This is an important result from this work, because in the
characterization of coal and asphaltenes, it has always been as-
sumed that all of the aromatic components are consumed
with alkyl chains ending up as aliphatic acids. This assumption
may be true for long alkyl chains in an excess of oxidant, but
not necessarily so for shorter alkyl chains, as has been
presented in the case of 2-ethylnaphthalene oxidation. Mecha-
nistic investigations are in progress to find the rationale
behind this trend.
Product identification was carried out by GC-MS analysis with
reference to the standard library of compounds supplied with the
instrument.
Ruthenium-ion-catalyzed oxidation in monophasic and bi-
phasic solvent systems
Reactions using a monophasic solvent system employed a glass
reactor that was charged with 2-ethylnaphthalene (25.6 mg,
0.164 mmol) dissolved in acetonitrile (20 mL). An aqueous solution
of NaIO4 (300 mg, 1.403 mmol in 10 mL H2O) and solid RuCl3·xH2O
(2.5 mg, 0.012 mmol) were added to this solution, and this was
considered as the starting point of the reaction. For GC-MS
analysis, reactions were quenched by adding aqueous Na2SO3
(0.1 mL, 0.176 g, 1.403 mmol in 6 mL H2O) to a 0.5 mL sample of
the reaction liquor.
Conclusion
Historically, it has always been assumed that the aromatic
regions of alkylated polyaromatic compounds are oxidized
mainly to CO2 and H2O by RICO chemistry and the aliphatic
chain remains as the corresponding carboxylic acid. Based on
the data reported in this paper this assumption does not
always hold, and RICO chemistry may be used in a more selec-
tive manner to oxidize the aromatic regions of these molecules
leaving the alkyl substituents unmodified and still attached to
the ring system. A quantitative NMR protocol has been
developed to determine the relative amount of aliphatic and
aromatic protons resulting from the RICO of alkyl-substituted
polyaromatic compounds, which may also find applications
with other oxidation approaches to substituted aromatic
systems. The reaction products under the relatively mild
conditions employed here are highly dependent on the
aliphatic chain length, substituent position and size of the
polyaromatic ring system which also affect the degree of
conversion.
For reactions in
a biphasic solvent system, dichloromethane
(16 mL) acetonitrile (5 mL) and water (5 mL) were used.[6a] The ratio
of substrate/oxidant/catalyst was the same as in the monophasic
solvent system (0.164:1.403:0.012). For GC-MS analysis, samples
were withdrawn and analyzed from both immiscible layers. For
NMR analysis, all products were extracted using DMSO after
previous evaporation of solvents.
Gas chromatography/mass spectrometry analysis
For qualitative analysis, GC-MS was used. Analyses were performed
using a Waters GCT Premier instrument fitted with an Agilent HP-
5 MS column (0.25 mmꢁ0.25 mmꢁ30 m) with helium as the carrier
gas (1 mLminꢀ1) and a temperature program from 303 K to 563 K.
Mass spectra of the unknown compounds were compared against
the NIST mass spectral database. Standards of the compounds
identified were then injected into the GC-MS so that comparison
of their retention time with those of the components in the reac-
tion mixture could be used to confirm assignments. Reaction mix-
Here we have shown that RICO chemistry can be used in
a
selective way to reduce the aromaticity of alkylated
4292
Chem. Eur. J. 2015, 21, 4285 – 4293
www.chemeurj.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[2] a) O. P. Strausz, T. W. Mojelsky, E. M. Lown, Fuel 1992, 71, 1355–1363;
b) L. M. Stock, K.-T. Tse, Fuel 1983, 62, 974–976.
[3] P. H. J. Carlsen, T. Katsuki, V. S. Martin, K. B. Sharpless, J. Org. Chem.
1981, 46, 3936–3938.
[4] a) V. Piccialli, D. M. A. Smaldone, D. Sica, Tetrahedron 1993, 49, 4211–
4228; b) T. K. M. Shing, E. K. W. Tam, V. W. F. Tai, I. H. F. Chung, Q. Jiang,
Chem. Eur. J. 1996, 2, 50–57; c) T. K. M. Shing, V. W. F. Tai, E. K. W. Tam,
Angew. Chem. Int. Ed. Engl. 1994, 33, 2312–2313; Angew. Chem. 1994,
106, 2408–2409.
[5] a) H. Yamaoka, N. Moriya, M. Ikunaka, Org. Process Res. Dev. 2004, 8,
931–938; b) A. M. Felix, J. V. Earley, R. I. Fryer, L. H. Sternbach, J. Hetero-
cycl. Chem. 1968, 5, 731.
[6] a) K. Morikawa, J. Park, P. G. Andersson, T. Hashiyama, K. B. Sharpless, J.
Am. Chem. Soc. 1993, 115, 8463–8464; b) N. Momiyama, H. Yamamoto,
J. Am. Chem. Soc. 2003, 125, 6038–6039.
[7] a) B. Plietker, Eur. J. Org. Chem. 2005, 1919–1929; b) B. Plietker, J. Org.
Chem. 2004, 69, 8287–8296.
tures were also analyzed quantitatively using a Varian gas chroma-
tography, system fitted with a VF-5HT column (0.25 mmꢁ30 m)
and a flame ionization detector (FID). Helium was used as the
carrier gas. The GC oven temperature was increased from 343 to
683 K at a heating rate of 20 Kminꢀ1
.
In situ NMR studies on the oxidation of 2-ethylnapthalene
All 1H NMR spectra were acquired on a Bruker 500 MHz DPX
system equipped with a 5 mm autotune broad band probe. Water
suppressed NMR spectra were acquired over 32 scans using the
Bruker pulse sequence zgpr. A power level of 53.5 db was em-
ployed for water pre-saturation, in conjunction with a relaxation
delay of 5 s and an acquisition time of 1.638 s.
An aqueous solution of NaIO4 (90 mg, 0.421 mmol in 3 mL H2O)
was added to
a solution of 2-ethylnaphthalene (15.8 mg,
[8] a) G. Bifulco, T. Caserta, L. Gomez-Paloma, V. Piccialli, Tetrahedron Lett.
2002, 43, 9265–9269; b) G. Bifulco, T. Caserta, L. Gomez-Paloma, V. Pic-
cialli, Tetrahedron Lett. 2003, 44, 5499–5503.
[9] a) D. Yang, C. Zhang, J. Org. Chem. 2001, 66, 4814–4818; b) D. Yang, F.
Chen, Z. M. Dong, D. W. Zhang, J. Org. Chem. 2004, 69, 2221–2223.
[10] a) H. Petride, C. Dr¼ghici, C. Florea, A. Petride, Cent. Eur. J. Chem. 2004,
0.101 mmol) in CD3CN (2 mL), and the mixture stirred for 1 min.
0.7 mL of this reaction mixture was transferred to an NMR tube
fitted with a calibrated glass insert containing a TMS standard.
After recording the 1H NMR spectra of the solution (0 h sample),
RuCl3·xH2O (20 mL, 0.105 mg, 0.5 mmol) was added to the NMR tube
and 1H NMR spectra were recorded every 15 min over an 8 h
period.
ˇ
2, 302–322; b) H. Petride, O. Costan, C. Draghici, C. Florea, A. Petride,
ARKIVOC (Gainesville, FL, U.S.) 2005, 2005, 18–32; c) R. S. Glass, J. L.
Broeker, Tetrahedron 1991, 47, 5077–5086; d) W. Su, Tetrahedron Lett.
1994, 35, 4955–4958.
[11] a) D. M. Piatak, G. Herbst, J. Wicha, E. Caspi, J. Org. Chem. 1969, 34, 116–
120; b) S. Imajo, H. Kuritani, K. Shingu, M. Nakagawa, J. Org. Chem.
1979, 44, 3587–3589; c) J. Hu, D. Zhang, F. W. Harris, J. Org. Chem.
2005, 70, 707–708.
[12] a) A. Tenaglia, E. Terranova, B. Waegell, J. Chem. Soc. Chem. Commun.
1990, 1344–1345; b) A. Tenaglia, E. Terranova, B. Waegell, J. Org. Chem.
1992, 57, 5523–5528; c) J. M. Bakke, J. E. Brænden, Acta Chem. Scand.
1991, 45, 418–423.
[13] J. L. Coudret, B. Waegell, Inorg. Chim. Acta 1994, 222, 115–122.
[14] a) U. A. Spitzer, D. G. Lee, J. Org. Chem. 1974, 39, 2468–2469; b) D. G.
Lee, U. A. Spitzer, J. Cleland, M. E. Olson, Can. J. Chem. 1976, 54, 2124–
2126.
[15] a) O. P. Strausz, E. M. Lown, Fuel Sci. Technol. Int. 1991, 9, 269–281;
b) O. P. Strausz, 1989, pp. 395–400.
Small-scale ruthenium-ion-catalyzed oxidation in a biphasic
solvent system for NMR analysis
An aqueous solution of NaIO4 (44 mg, 0.206 mmol in 1 mL H2O)
was added to a solution of substrate (0.028 mmol in 2 mL CDCl3
and 1 mL CD3CN) were placed in a round-bottomed flask, and the
mixture was stirred for 1 min. A 0.6 mL portion of the organic layer
was transferred to an NMR tube containing a calibrated glass insert
and 0.3 mL of CD3CN. The content of the tube was subsequently
analyzed using 1H NMR spectroscopy and the spectra labelled as
reaction time 0 h. An aqueous solution of RuCl3·xH2O (20 mL,
2 mmol) was then added to the round-bottomed flask and stirring
was commenced. After the requisite time, a 0.6 mL portion of the
organic layer was transferred to an NMR tube containing a
calibrated TMS glass insert, 0.3 mL of CD3CN was added and the
1H NMR spectra recorded. For conversion data, the sample (0.2 mL)
was mixed with a 20 mL aliquot of 2-butanol and injected into the
GC immediately after NMR analysis.
[16] a) Y.-G. Huang, Z.-M. Zong, Z.-S. Yao, Y.-X. Zheng, J. Mou, G.-F. Liu, J.-P.
Cao, M.-J. Ding, K.-Y. Cai, F. Wang, W. Zhao, Z.-L. Xia, L. Wu, X.-Y. Wei,
Energy Fuels 2008, 22, 1799–1806; b) S. Murata, K. U-esaka, H. Ino-ue,
M. Nomura, Energy Fuels 1994, 8, 1379–1383; c) A. Ma, S. Zhang, D.
Zhang, Org. Geochem. 2008, 39, 1502–1511.
[17] a) L. M. Stock, S. H. Wang, Fuel 1986, 65, 1552–1562; b) L. M. Stock, S. H.
Wang, Fuel 1985, 64, 1713–1717.
[18] L. M. Stock, K. T. Tse, S. H. Wang, Abstr. Pap. Am. Chem. Soc. 1985, 190,
113.
[19] O. P. Strausz, T. W. Mojelsky, E. M. Lown, The Chemistry Of Alberta Oil
Sand Bitumen, International Academic Publishers, Beijing 1992, pp. 29–
35.
[20] a) T. W. Mojelsky, T. M. Ignasiak, Z. Frakman, D. D. McIntyre, E. M. Lown,
D. S. Montgomery, O. P. Strausz, Energy Fuels 1992, 6, 83–96; b) K. E.
Singleton, R. G. Cooks, K. V. Wood, K. T. Tse, L. Stock, Anal. Chim. Acta
1985, 174, 211–223.
Acknowledgements
The Cardiff authors would like to thank ExxonMobil Corporate
Strategic Research (CSR) for financial support. The authors
would like to acknowledge Christiane Reece and Peter
Miedziak for their help in designing the front cover.
[21] G. M. Badger, Aromatic Character and Aromaticity, Cambridge University
Press, Cambridge, 1969.
Keywords: homogeneous catalysis
·
oxidation
·
NMR
spectroscopy · polynuclear aromatic hydrocarbons · ruthenium
Received: October 27, 2014
[1] C. Djerassi, R. R. Engle, J. Am. Chem. Soc. 1953, 75, 3838–3840.
Published online on January 7, 2015
Chem. Eur. J. 2015, 21, 4285 – 4293
www.chemeurj.org
4293
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim