Acylation of naphthalenes and anthracene on sulfated zirconia

Article abstract of DOI:10.1016/j.jcat.2005.01.024

Sulfated zirconia (SZ) exhibits a high catalytic performance in the benzoylation of 1-methoxynaphthalene. Therefore, it was used as heterogeneous catalyst in the acylation of methoxynaphthalenes, methylnaphthalenes, naphthalene, and anthracene with benzoic anhydride, benzoyl chloride, and acetic anhydride to synthesize aromatic ketones. The rate of product formation on SZ was dependent on the respective aromatic, on the solvent used, and on the ratio of aromatic to acylating agent.

Full text of DOI:10.1016/j.jcat.2005.01.024

Journal of Catalysis 231 (2005) 269–278
www.elsevier.com/locate/jcat
Acylation of naphthalenes and anthracene on sulfated zirconia
J. Deutsch a,∗, H.A. Prescott b, D. Müller a, E. Kemnitz b, H. Lieske a
a
1
Institut für Angewandte Chemie Berlin-Adlershof e.V., Richard-Willstätter-Strasse 12, 12489 Berlin, Germany
b
Humboldt-Universität zu Berlin, Institut für Chemie, Brook-Taylor-Strasse 2, 12489 Berlin, Germany
Received 19 November 2004; revised 13 January 2005; accepted 19 January 2005
Available online 17 March 2005
Abstract
Sulfated zirconia (SZ) exhibits a high catalytic performance in the benzoylation of 1-methoxynaphthalene. Therefore, it was used as
heterogeneous catalyst in the acylation of methoxynaphthalenes, methylnaphthalenes, naphthalene, and anthracene with benzoic anhydride,
benzoyl chloride, and acetic anhydride to synthesize aromatic ketones. The rate of product formation on SZ was dependent on the respective
aromatic, on the solvent used, and on the ratio of aromatic to acylating agent.
2005 Elsevier Inc. All rights reserved.
Keywords: Solid acids; Sulfated zirconia; Naphthalenes; Anthracene; Aromatic acylation; Aromatic ketones
1. Introduction
neously catalyzed acylation of bicyclic aromatics. These
publications focus mainly on the synthesis of 2-acetyl-6-
methoxynaphthalene, an intermediate for the production of
the pharmaceutical Naproxen, from 2-methoxynaphthalene
and acetic anhydride on zeolites [13–21]. Further papers
have covered the zeolite-catalyzed acetylation and benzoy-
lation of naphthalene [22,23].
In a continuation of our investigations of sulfated zirconia
(SZ) as a heterogeneous catalyst in the acylation of mono-
cyclic aromatics [24–28], we studied the acylation of naph-
thalene, methylnaphthalenes, methoxynaphthalenes, and an-
thracene on SZ. Additional aspects considered here are (i)
the catalytic activity of SZ in comparison with that of other
solid catalysts, (ii) the influence of solvents, and (iii) the
influence of the aromatic/acylating agent ratio on ketone for-
mation.
The acylation of aromatics is the most important process
in organic chemistry for the synthesis of aromatic ketones.
The target products are intermediates or final compounds
used in the production of pharmaceuticals, cosmetics, agro-
chemicals, dyes, and special chemicals. Currently, strong
economic and ecological factors demand the replacement
of corrosive Lewis acids, used stoichiometrically to activate
the acylating agent (Friedel–Crafts acylation), with hetero-
geneous acid catalysts. Because of their easy separation from
the reaction mixture by filtration, solid acids are attractive
acylation catalysts. In the case of their successful application
in large-scale processes, the extensive formation of liquid ef-
fluents can be avoided.
Numerous papers have reported on the heterogeneously
catalyzed acylation of monocyclic aromatics, such as anisole,
mesitylene, xylenes, and toluene on zeolites [1–6], clays
[7–9], Nafion-H on silica [10], and heteropoly acids [11,12].
Considerably less has been published on the heteroge-
2. Experimental
2.1. Chemicals
*
Corresponding author.
Naphthalene (99%), 1-methoxynaphthalene (98+%),
2-methoxynaphthalene (99%), 1,3-dimethylnaphthalene
(96%), anthracene (99%), acetonitrile (anhydrous, 99.8%),
E-mail address: deutsch@aca-berlin.de (J. Deutsch).
A member of the EU-funded Coordination Action of Nanostructured
Catalytic Oxide Research and Development in Europe (CONCORDE).
1
0021-9517/$ – see front matter
doi:10.1016/j.jcat.2005.01.024
2005 Elsevier Inc. All rights reserved.

270
J. Deutsch et al. / Journal of Catalysis 231 (2005) 269–278
chlorobenzene (anhydrous, 99.8%), cyclohexane (anhy-
drous, 99.5%), 1,2-dichloroethane (anhydrous, 99.8%), ni-
trobenzene (99+%), acetic anhydride (99+%), benzoyl
chloride (99%) were all purchased from Aldrich; benzoic
anhydride (97%), 1,2-dimethoxyethane (anhydrous, 99%),
and ZrOCl2 ·8H2O (> 99%) were from Fluka; 1-methyl-
cell on a Perkin-Elmer FTIR system 2000 spectrometer
(H-mordenite: on a Digilab FTS 3000 Spectrometer Excal-
ibur). We normalized the spectra by setting the transmission
at 500.35 cm−1 of each spectrum to 100% and multiply-
ing the spectrum by the corresponding factor. Brønsted acid
sites (B) are indicated by the band at 1550 cm−1. The band
at 1490 cm−1 can be assigned to PACs bonded to both Brøn-
sted and Lewis acid sites, whereas the bands at 1450 and
about 1600 cm−1 can be attributed exclusively to Lewis acid
sites (L) in the samples. When the 1490 cm−1 band intensity
is one-third that of the 1450 cm−1 band, Lewis acid sites
are exclusively present in the sample. When the 1490 cm−1
band intensity is higher than that, the sample also possesses
Brønsted acid sites.
naphthalene
(97%),
2-methylnaphthalene
(99%),
H3[PW12O40]×H2O (p.a.), and Cs2CO3 (99.5%) were from
Acros Organics. The chemicals were used without further
purification.
SZ and the following solid acids were investigated:
Nafion-H (a perfluoroalkanesulfonic acid polymer) on SiO2,
Amberlyst-15 (an arenesulfonic acid polymer), and Mont-
morillonite K-10 (a clay mineral) were purchased from
Aldrich; H-BEA (a zeolite, Si/Al = 25) and H-mordenite
(a zeolite, Si/Al = 10) were from Süd-Chemie; and
Cs2.5H0.5[PW12O40] (a heteropoly acid) was prepared ac-
cording to the method described in [29]. The catalysts were
pretreated before experiments. SZ, H-BEA, and H-mordenite
were calcined at 500 ◦C in air for 3 h; Cs2.5H0.5[PW12O40]
was calcined at 300 ◦C in N2 for 3 h; K-10 was calcined
at 200 ◦C in air for 1 h; Nafion-H on SiO2 was calcined at
120 ◦C in vacuum for 1 h; Amberlyst-15 was stored at room
temperature in vacuum over concentrated sulfuric acid.
2.3. Catalytic experiments
2.3.1. General procedure
The reaction mixture (aromatic, acylating agent, catalyst,
and solvent) was stirred (700 rpm) in a 100-ml three-necked
flask, with magnetic stirrer, thermometer, reflux condenser,
and CaCl2 tube and heated to the reaction temperature. In
some experiments the ketone formation was monitored over
the reaction time. These experiments were performed with
the use of 15 mmol aromatic, 30 mmol acylating agent,
1.125 g SZ, and 30 ml 1,2-dichloroethane. Samples (0.2 ml
of the mixture) were taken periodically after 5, 10, 20, 30,
60, 120, and 240 min via syringe, filtered to separate the
catalyst, and concentrated in vacuum to remove the solvent
and other low boiling compounds (in acetylation reactions:
acetic anhydride, acetic acid). All yields given are related to
the aromatic.
2.2. Preparation of SZ and methods of characterization
The catalyst, SZ, was prepared by the addition of aqueous
ammonia to an aqueous solution of ZrOCl2 ·8H2O until pH 8
was achieved. The precipitate was filtered, washed several
times until free of chloride, dried at 110 ◦C for 15 h, im-
pregnated with diluted sulfuric acid (10 wt%), filtered, and
calcined at 500 ◦C in air for 1 h. The resulting white solid
contained 1.64 wt% sulfur. The catalyst was stored in a des-
iccator over dry silica gel.
X-ray powder diffraction measurements of SZ were per-
formed with Cu-Kα radiation (RD 7; R. Seifert & Co.,
Freiberg, Germany). The catalyst exhibited the tetragonal
modification of ZrO2 (PDF-No 42-1164).
Specific surface areas and pore diameters were mea-
sured with nitrogen adsorption at 77 K (ASAP 2000 sys-
tem; Micromeritics). The acidities of SZ, K-10, H-BEA, and
H-mordenite were characterized by temperature-program-
med desorption (TPD) of ammonia (heat conductance de-
tection), which was pre-adsorbed at 100 ◦C. The ammonia
desorbed was quantified by reaction with 0.1 N sulfuric acid
and back-titration.
The type and amount of acid sites (Lewis, L, or Brøn-
sted, B) present in the samples relative to one another were
studied by the FTIR photoacoustic spectroscopy (FTIR-
PAS) of chemisorbed pyridine adsorbate complexes (PACs).
Samples were pretreated at 150 ◦C for 30 min under Ar
and then exposed to pyridine twice (2 × 30 µl) at 150 ◦C
with Ar flushing for 15 min in between. The measurements
were carried out before and after pyridine adsorption be-
tween 4000 and 400 cm−1 in a MTEC 300 photoacoustic
2.3.2. Analysis
The product yield was determined by 1H NMR spec-
troscopy (solvent: CDCl3). To identify and quantify the aro-
matics to be reacted and aromatic ketones formed, the fol-
lowing signals (singlets) were evaluated.
CH3O: 1-methoxynaphthalene 3.97 ppm; 1-benzoyl-4-
methoxynaphthalene 4.03 ppm; 1-acetyl-4-methoxynaphtha-
lene 4.02 ppm, (CH3CO: 2.68 ppm); 2-methoxynaphthalene
3.90 ppm; 1-benzoyl-2-methoxynaphthalene 3.80 ppm;
1-acetyl-2-methoxynaphthalene 3.94 ppm, (CH3CO: 2.63
ppm).
CH3: 1-methylnaphthalene 2.68 ppm; 1-benzoyl-4-
methylnaphthalene 2.75 ppm; 2-methylnaphthalene 2.47
ppm; 1-benzoyl-2-methylnaphthalene 2.29 ppm; 1.3-di-
methylnaphthalene 2.44 ppm, 2.63 ppm; 1-benzoyl-2,4-
dimethylnaphthalene 2.26 ppm, 2.71 ppm; 1-acetyl-2,4-
dimethylnaphthalene 2.37 ppm, 2.40 ppm, 2.59 ppm (clear
identification of the CH3CO signal was not possible).
H10: anthracene 8.41 ppm (identical to H9); 9-benzoyl-
anthracene 8.54 ppm; 9-acetylanthracene 8.47 ppm, (CH3
CO: 2.08 ppm).
1
The H NMR spectra of naphthalene and its acylation
products showed no suitable signals for the calculation of the

J. Deutsch et al. / Journal of Catalysis 231 (2005) 269–278
271
1
2
Scheme 1. SZ-catalyzed acylation of substituted naphthalenes. Aromatics: 1-methoxynaphthalene (R = H; R = CH O); 2-methoxynaphthalene
3
1
2
1
2
1
2
1
2
(R = CH O; R = H); 1,3-dimethylnaphthalene (R , R = CH ); 1-methylnaphthalene (R = H; R = CH ); 2-methylnaphthalene (R = CH ; R = H).
3
3
3
3
3
3
3
Acylating agents: benzoic anhydride (R = C H , X = OCOC H ); benzoyl chloride (R = C H , X = Cl); acetic anhydride (R = CH , X = OCOCH ).
6
5
6
5
6
5
3
3
product yield directly from the reaction mixture. Therefore,
the yields of the ketones formed from naphthalene were de-
termined directly from the weights of the isolated products.
3. Results and discussion
3.1. Textural properties, acidities, and catalytic
performances of various solid acids
2.3.3. Comparison of various solid acids as acylation
catalysts in the benzoylation of 1-methoxynaphthalene
A mixture of 15 mmol 1-methoxynaphthalene, 15 mmol
benzoic anhydride, and 30 ml 1,2-dichlorethane (solvent)
was heated to 40 ◦C. The catalyst (1.125 g) was added to the
mixture. After 2 h at 40 ◦C, a sample was taken as described
above and analyzed accordingly.
The acylation of 1-methoxynaphthalene with an equimo-
lar amount of benzoic anhydride (Scheme 1) was used as a
test reaction for comparison of the catalytic performances of
six solid acids with that of SZ. The catalytic results, textural
properties, and acidities are listed in Table 1. The results
demonstrate complex relationships between catalyst activity,
acidity, textural properties, and, probably, the accessibility of
active sites.
SZ, Cs2.5H0.5[PW12O40], Nafion-H on SiO2, Amberlyst-
15, and K-10 are more active than the acidic zeolites, H-BEA
and H-mordenite. The acid sites in the narrow zeolite chan-
nels are clearly less accessible for the sterically demanding
reactants compared with ammonia. Moreover, SZ, Nafion-H
on SiO2, and Amberlyst-15 contain chemical functions,
which seem to be especially effective as active sites. The
sulfuric and sulfonic acid groups of these catalysts are able
to react with acylating agents to give acylsulfates and acyl-
sulfonates, respectively. Acylsulfonates were synthesized
and isolated by Effenberger et al. and are very reactive
toward aromatics [30]. For SZ, Nafion-H on SiO2, and
Amberlyst-15, a formation of surface-bonded acylsulfates
or acylsulfonate surface species as acylating agents in the
catalytic cycle can be assumed. The distinct acylating ac-
tivity of these reagents may be related to their tendency
to form surface-bonded sulfate or sulfonate anions as ther-
modynamically favored leaving groups (Scheme 2). The
acylation on Cs2.5H0.5[PW12O40] (acylating species: acyl-
tungstophosphate; leaving group: tungstophosphate anion)
could follow a similar mechanism.
The number of acid sites in SZ, Nafion-H on SiO2,
Amberlyst-15, and K-10 does not correlate with their cat-
alytic performances. This is clearly demonstrated by Amber-
lyst-15, which has the highest number of acid sites (4.70
mmol/g) and exhibits only a moderate catalytic perfor-
mance. This apparent contradiction can be explained by the
exclusive accessibility of the sites in the mesopores. Most of
the acid sites are located in the micropores of the polymer
network, which are difficult for sterically demanding mole-
cules to reach.
2.3.4. Influence of the solvent and the aromatic/acylating
agent ratio on the SZ-catalyzed acetylation of
2-methoxynaphthalene
A mixture of 15 mmol 2-methoxynaphthalene and
15 mmol acetic anhydride was reacted at 70 ◦C in 30 ml ace-
tonitrile, cyclohexane, 1,2-dichloroethane, 1,2-dimethoxy-
ethane, or nitrobenzene as a solvent on SZ (1.125 g). The
reaction mixture was analyzed after a reaction time of 4 h.
To investigate the influence of the molar ratio of aromatic
to acylating agent on the product formation, the ratio was
changed stepwise from 4:1 to 1:4 in further experiments.
2.3.5. Preparative acylation experiments on SZ
The experiments were carried out in a 100-ml three-
necked flask. The freshly calcined (at 500 ◦C in air for 3 h)
SZ (1.125 g or the respective amount given in the text) was
added to a heated stirred mixture of 15 mmol aromatic,
30 mmol acylating agent (acetic anhydride or benzoyl chlo-
ride), and 30 ml 1,2-dichloroethane or chlorobenzene. The
reaction time was 4 or 20 h. After the mixture was cooled to
room temperature and the catalyst was filtered from the reac-
tion mixture, the clear filtrate was concentrated in vacuum.
The resulting crude ketones (in some cases isomer mixtures)
were purified by Kugelrohr distillation and identified by 1H
and 13C NMR spectroscopy (Unity plus 300 MHz; Varian)
and GC-MS (MD 800; Thermo). As an exception, crude 9-
benzoylanthracene obtained was purified (separation of un-
reacted anthracene) by means of flash chromatography on
silica gel (Kieselgel 60, 230–400 mesh; Merck) with the elu-
ent n-hexane/ethyl acetate (10/1 v/v).

272
J. Deutsch et al. / Journal of Catalysis 231 (2005) 269–278
Table 1
◦
Properties and catalytic performances of solid acids in the benzoylation of 1-methoxynaphthalene at 40 C according to Scheme 1
Catalyst
BET surface
Average pore
diameter (Å)
Number of acid
sites (mmol/g)
Yield of 1-benzoyl-4-methoxy-
naphthalene formed after 2 h (%)
2
area (m /g)
c
SZ
150
60
224
38
233
500
400
68
72
220
400
60
0.41
50
40
22
19
13
4
d
Cs
2.5 0.5
H
[PW
O ]
12 40
0.16
e
Nafion-H on SiO
Amberlyst-15
K-10
H-BEA
∼ 0.10
2
a
e
4.70
c
0.28
b
c
7.6 × 6.4
0.87
b
c
H-mordenite
7.0 × 6.5
0.76
3
15 mmol 1-methoxynaphthalene, 15 mmol benzoic anhydride, 1.125 g catalyst, solvent: 30 ml 1,2-dichloroethane.
a
Before reaction: swollen overnight at room temperature in the solvent, start of the reaction by simultaneous addition of acylating agent and aromatic.
b
c
d
e
Channels.
From ammonia-TPD.
Calculated from the molecular formula.
See product information.
Scheme 2. Possible mechanism of aromatic acylations on heterogeneous catalysts containing sulfate or sulfonate groups such as SZ, Amberlyst-15, and
Nafion-H on SiO . Acylating agents: benzoic anhydride (R = C H , X = OCOC H ); benzoyl chloride (R = C H , X = Cl); acetic anhydride (R = CH ,
2
6
5
6
5
6
5
3
X = OCOCH ).
3
On the whole, catalytic performance seems to depend (i)
on the chemical type and (ii) on the steric availability of the
active sites rather than on their amount.
The type of acid sites (Lewis or Brønsted or both) in the
samples has been characterized by IR experiments. Fig. 1
shows the IR spectra of the pyridine adsorbate complexes
(PACs) formed on the solid acids. The most distinct spec-
trum is that of the most acidic catalyst, Amberlyst-15, with
very broad and strong bands at 1550 and 1490 cm−1, which,
in this case, can be attributed to a very high number of
Brønsted acid sites. The number of Lewis acid sites in this
sample is much lower (weak band at 1450 cm−1). The
number of Brønsted acid sites in the other samples de-
creases in the following order: H-mordenite > SZ (very
broad band) > H-BEA > Nafion-H on SiO2 ≈ K-10, based
on the intensities of the band at 1550 cm−1. Lewis acid
sites are found in all of the samples (band at 1450 cm−1).
From the intensity ratio (1:3) of the 1490 and 1450 cm−1
band in Fig. 1, it can be concluded that Nafion-H on SiO2
Fig. 1. Photoacoustic FT-IR spectra of pyridine adsorbate complexes (PACs)
formed after pyridine adsorption on various solid acids. L, B, and B + L
indicates the bands of complexes formed on Lewis, Brønsted or both types
of acid sites, respectively.

J. Deutsch et al. / Journal of Catalysis 231 (2005) 269–278
273
Table 2
Dependence of the product formation on the nature of the solvent used in the SZ catalyzed acetylation of 2-methoxynaphthalene at 70 C according to Scheme 1
◦
Solvent
Solvent parameters
Dielectric constant
at 25 C
Yield of 1-acetyl-
2-methoxy/naphthalene
formed after 4 h (%)
Acceptor
number
Donor number
◦
a
b
(kcal/mol)
Cyclohexane
2.0
7.0
10.1
34.8
36.0
0
No electron pair donor
24.0
0
4.4
14.1
62
26
61
67
26
1,2-Dimethoxyethane
1,2-Dichloroethane
Nitrobenzene
10.2
16.7
14.8
18.9
Acetonitrile
Reaction mixture: 15 mmol 2-methoxynaphthalene, 15 mmol acetic anhydride, 1.125 g SZ, 30 ml solvent.
a
31
The acceptor number is obtained from the P-NMR chemical shift values related to that of the 1:1-adduct (C H ) PO-SbCl dissolved in 1,2-
2
5 3
5
dichloroethane.
b
The donor number has been defined as the negative ꢀH value for 1:1-adduct formation between SbCl and an electron pair donor solvent in diluted
5
solution in the noncoordinating solvent 1,2-dichloroethane.
contains a low number of Lewis acid sites (SiO2-support).
The Brønsted acid sites in this catalyst (SO3H-functions in
the surface-bonded Nafion) were not detected. Their num-
ber seems to be negligible in relation to the Lewis sites
of the support. The spectrum of PACs on K-10 shows a
band comparable in intensity to that of Nafion-H on SiO2
at 1490 cm−1, with a much weaker Lewis acid site band at
1450 cm−1. This indicates a minimal amount of Brønsted
acid sites for K-10 and practically no Lewis acid sites. The
samples SZ, H-mordenite, and H-BEA all have Lewis acid
sites (bands at 1450 cm−1) slightly higher in number than
those found for Nafion-H on SiO2. Heteropoly acids such
as Cs2.5H0.5[PW12O40] are known as pure Brønsted acid
compounds [12]. For that reason, this catalyst was not in-
vestigated by IR spectroscopy.
The interaction of water traces in the reaction mixture and
of the acids formed as co-products in the acylation (such as
hydrochloric acid, acetic acid, or benzoic acid) with Lewis
sites might generate additional Brønsted sites. Therefore, no
definitive conclusion can be drawn concerning the contribu-
tion of Lewis or Brønsted sites to the catalytic activity of the
investigated solid acids.
3.2. Aromatic acylations on SZ under varied conditions:
influence of the nature of the solvent and of the ratio of
aromatic to acylating agent
For the acylation of solid aromatics in liquid phase,
the use of solvents is necessary. To identify an appropri-
ate reaction medium for such acylations, the reaction of
the solid aromatic 2-methoxynaphthalene with acetic anhy-
dride (Scheme 1) was performed in the presence of vari-
ous solvents. As seen in Table 2 (solvent parameters from
[31]), the heterogeneously catalyzed ketone formation is
not influenced by the dielectric constant or the acceptor
number of the solvent. It was found that solvents with
a low donor number (defined as the negative ꢀH value
for 1:1 adduct formation between SbCl5 and an electron
pair donor solvent in diluted solution in the noncoordinat-
ing solvent 1,2-dichloroethane) are the most suitable re-
action media. These weak donor solvents cannot interact
with the electrophilic acylating species formed on the cat-
alyst surface and do not influence the electrophilic power
of the acylating species negatively. The same positive ef-
fect of weak donor solvents was found in the synthesis of
aromatic ketones under homogeneous (i.e., Friedel–Crafts)
conditions. 1,2-Dichloroethane, nitrobenzene, and carbon
disulfide are commonly used solvents for aromatic acy-
lation in the presence of AlCl3 or FeCl3. Consequently,
it is relevant to the success of the reaction that the sol-
vent be a weak donor, regardless of whether the reaction
is homogeneously or heterogeneously catalyzed. All further
acylations were performed in the weak donor solvent 1,2-
dichloroethane.
According to Scheme 2, the aromatic acylation on the
sulfur-containing solid acids may be catalyzed by Brønsted
acid sites.
An advantage of SZ over the other mesoporous catalysts
Cs2.5H0.5[PW12O40], Amberlyst-15, Nafion-H on SiO2, and
K-10 is its high thermal stability, which permits the removal
of deactivating carbonaceous deposits at elevated tempera-
tures. Recently we found that deactivating compounds can
be burned off completely from SZ by calcination at 550 ◦C
in air [27]. In contrast, the regeneration of spent heteropoly
acids like Cs2.5H0.5[PW12O40] is problematic [12].
The outstanding catalytic suitability of SZ for the benzoy-
lation of 1-methoxynaphthalene correlates with our experi-
ments on the benzoylation of anisole [28]. The acylations
reported in the following sections were carried out on the
catalyst with the highest performance, SZ.
An equimolar application would ensure the economic use
of expensive aromatics and acylating agents for the aro-
matic acylations. However, in practice the cheaper reactant is
used in excess to accelerate product formation and increase
yields. In view of this problem, the educt ratio was opti-
mized for the acetylation of 2-methoxynaphthalene. If the
aromatic or the acylating agent is applied in 100% excess,
the ketone yield increases from 61 to 77% or 78%, respec-
tively (Fig. 2). A further increase in the threefold amount

274
J. Deutsch et al. / Journal of Catalysis 231 (2005) 269–278
◦
Fig. 2. Dependence of the product formation on the educt ratio in the SZ catalyzed acetylation of 2-methoxynaphthalene at 70 C according to
Scheme 1. Reaction mixture: 2-methoxynaphthalene/acetic anhydride: 1/4 = 15 mmol/60 mmol, 1/3 = 15 mmol/45 mmol, 1/2 = 15 mmol/30 mmol,
1/1 = 15 mmol/15 mmol, 2/1 = 30 mmol/15 mmol, 3/1 = 45 mmol/15 mmol, 4/1 = 60 mmol/15 mmol; 30 ml 1.2-dichloroethane; 1.125 g SZ. The ketone
yield is related to the educt which was used in an amount of 15 mmol.
results in additional but lower improvements to 81% (excess
of the aromatic) and 82% (excess of the acylating agent).
An increase in the educt ratio above 3:1 leads to only mar-
ginal effects. In summary, a ratio of two molar equivalents of
acylating agent per mole of aromatic was optimal and was
applied to the acetylation and benzoylation of various aro-
matics reported in the following sections.
3.3. Acylation of different aromatics on SZ at 70 ◦C
3.3.1. Acylation of methoxynapthalenes (Scheme 1)
1- and 2-Methoxynaphthalene exhibited a similar behav-
ior in the reaction with the acylating agents used (Figs. 3
and 4). 1-Methoxynaphthalene could be acetylated with
acetic anhydride and benzoylated with benzoic anhydride
on SZ in nearly 100% yield. Benzoyl chloride reacted more
slowly and gave only a 82% yield (Fig. 3). The yields of
the acylations of 2-methoxynaphthalene did not exceed 85%
(Fig. 4). Ketone formation could not be completed by an in-
crease in the SZ amount or the reaction temperature or both
or by an extension of the reaction time. As reported, the
initial activation of the acylating agent with an equimolar
amount of AlCl3 (Friedel–Crafts conditions) before reaction
with 2-methoxynaphthaline did also not improve the ketone
yield, which remained lower than 85% [32,33]. Hence, the
acylation of 2-methoxynaphthaline with carboxylic anhy-
drides to give 1-acyl-2-methoxynaphthalenes seems to be
thermodynamically limited.
Fig. 3. Kinetics of SZ-catalyzed reactions of 1-methoxynaphthalene with
various acylating agents at 70 C. Products formed: 1-acyl-4-methoxy-
naphthalenes.
◦
3.3.2. Acylation of methylnaphthalenes (Scheme 1)
1,3-Dimethylnaphthalene reacted easily and selectively
on SZ with benzoic anhydride and benzoyl chloride to
give 1-benzoyl-2,4-dimethylnaphthalene in yields of 93
and 79%, respectively. As in the case of the methoxy-
Fig. 4. Kinetics of SZ-catalyzed reactions of 2-methoxynaphthalene
with various acylating agents at 70 C. Products formed: 1-acyl-2-meth-
oxynaphthalenes.
◦

J. Deutsch et al. / Journal of Catalysis 231 (2005) 269–278
275
with the highly stabilized benzoyl cation (delocalization of
its positive charge within the aromatic sextet) as the inter-
mediate, but insufficient in the case of the acetyl cation (in
which a delocalization of its positive charge is not possi-
ble). Therefore, methylacetophenones are not formed in true
catalytic reactions, but have to be synthesized according to
Friedel–Crafts by stoichiometric activation of the acylating
agent.
In Fig. 6, the results of the benzoylation of 1- and 2-
methylnaphthalene are compared with those of the other
aromatics. In general, 1- and 2-methylnaphthalene were ben-
zoylated considerably more slowly than 1,3-dimethylnaph-
thalene because of their known lower reactivity. The reaction
of 1-methylnaphthaline with benzoic anhydride and benzoyl
chloride resulted, respectively, in low yields of 13 and 17%
of 1-benzoyl-4-methylnaphthalene. 2-Methylnaphthalene
yielded 7 and 12% 1-benzoyl-2-methylnaphthalene with the
same two acylating agents, respectively. Neither methylaro-
matic reacted with acetic anhydride.
Fig. 5. Kinetics of SZ-catalyzed reactions of 1,3-dimethylnaphthalene
with various acylating agents at 70 C. Products formed: 1-acyl-2,4-
dimethylnaphthalenes.
◦
naphthalenes, benzoylation proceeded faster with the car-
boxylic anhydride than with the acid chloride (Fig. 5). In
contrast to that of the methoxynaphthalenes, the acetyla-
tion of 1,3-dimethylnaphthalene stagnated within minutes
at a low conversion (final ketone yield 7%). Similar phe-
nomena have been found for the acetylation of toluene on
H-BEA and heteropoly acids [34,35] and for the acetylation
of mesitylene and m-xylene on SZ [28]. To the best of our
knowledge, true catalytic synthesis of methylacetophenones
from methylaromatics on homogeneous or heterogeneous
catalysts in acceptable yields has not been reported before.
Product inhibition of the acid sites on the catalyst surface is
assumed to be responsible for the incomplete reaction [34].
Furthermore, the impossibility of a real catalytic acetylation
of methylaromatics may be related to the electronic struc-
ture of the reactants. The reactivity of methylaromatics is
increased by the weak donating field effect of the methyl
group. The resulting mild activation of the aromatic system
seems to be sufficient in the case of a catalyzed reaction
3.3.3. Acylation of naphthalene (Scheme 3) and anthracene
(Scheme 4)
Owing to the absence of activating substituents, naph-
thalene is the least reactive of the bicyclic aromatics in-
vestigated. Therefore, its acylation with benzoic anhydride
and benzoyl chloride on SZ proceeded very slowly to give
less than 10% of the target product (Fig. 6). Like 1- and
2-methylnaphthalene, naphthalene was not acylated in the
reaction with acetic anhydride.
As expected, the more reactive aromatic, anthracene, was
easily and selectively converted on SZ to 9-benzoylanthra-
cene, with yields of 61 and 67% (Fig. 6). In accordance with
the benzoylations of naphthalene, 1-methylnaphthalene,
and 2-methylnaphthalene, benzoyl chloride also revealed a
higher reactivity than benzoic anhydride. The low ketone
yield of 12% achieved in the reaction of anthracene with
◦
Fig. 6. SZ-catalyzed acylation of substituted naphthalenes, naphthalene, and anthracene at 70 C. Dependence of the ketone formation on the aromatics and
acylation agents used.

276
J. Deutsch et al. / Journal of Catalysis 231 (2005) 269–278
Scheme 3. SZ-catalyzed acylation of naphthalene. Acylating agents: benzoic anhydride (R = C H , X = OCOC H ); benzoyl chloride (R = C H , X = Cl);
6
5
6
5
6 5
acetic anhydride (R = CH , X = OCOCH ).
3
3
Scheme 4. SZ-catalyzed acylation of anthracene. Acylating agents: benzoic anhydride (R = C H , X = OCOC H ); benzoyl chloride (R = C H , X = Cl);
6
5
6
5
6 5
acetic anhydride (R = CH , X = OCOCH ).
3
3
acetic anhydride could not be improved by an increased
reaction temperature. The reasons for this may be product
inhibition of the catalyst or insufficient reactivity of an-
thracene toward the surface-bonded acetyl intermediate or
both.
4. Conclusions
Sulfated zirconia is an efficient catalyst for the reaction
of 1-methoxynaphthalene with benzoic anhydride as a ster-
ically demanding acylating agent. Here it exhibits a higher
catalytic performance than other solid acids with mesopores,
such as Cs2.5H0.5[PW12O40], Nafion-H on SiO2, Amberlyst-
15, and K-10. These results are consistent with our exper-
iments on the benzoylation of anisole. The low catalytic
performances of H-BEA and H-mordenite in the benzoy-
lation of 1-methoxynaphthalene are probably due to the
lower accessibility of their acid sites in the narrow chan-
nels of the zeolite lattices. The formation of very reactive
acylating species and thermodynamically favored leaving
groups on their surface may explain the moderate to high
catalytic performances of Amberlyst-15, Nafion-H on SiO2,
Cs2.5H0.5[PW12O40], and SZ.
A considerable influence of the solvent on the prod-
uct formation was demonstrated for the acetylation of 2-
methoxynaphthalene on SZ to 1-acetyl-2-methoxynaphtha-
lene. As found earlier in Friedel–Crafts aromatic acylations,
the use of solvents with low donor numbers is also advanta-
geous to the heterogeneously catalyzed acylation.
A molar ratio for 2-methoxynaphthalene and acetic anhy-
dride in the range between 2 and 3 (1:2 and 1:3) is sufficient
to achieve fast ketone formation with high yields.
3.4. Preparative experiments on the synthesis of aromatic
ketones on SZ
To develop practical procedures for the synthesis of ke-
tones from naphthalenes and anthracene on SZ, the reaction
conditions were optimized. These investigations were con-
centrated on acetic anhydride and benzoyl chloride as tech-
nically relevant acylating agents. The results are listed in
Table 3.
A reaction temperature of 70 ◦C was sufficient for the
acetylation and benzoylation of the more reactive aromatics.
Moreover, the commonly used catalyst amount of 1.125 g SZ
was reduced for the acylation of 1-methoxynaphthalene and
1,3-dimethylnaphthalene to 1-acetyl-4-methoxynaphthalene,
1-benzoyl-4-methoxynaphthalene, and 1-benzoyl-2,4-di-
methylnaphthalene. Based on the lower reactivity of benzoyl
chloride and to ensure a high conversion of the aromatic, the
reaction time was extended to 20 h.
The other aromatics were benzoylated at 120 or 130 ◦C in
chlorobenzene. 1-Methylnaphthalene, 2-methylnaphthalene,
and naphthalene gave isomer mixtures of the ketones ob-
tained. With the exception of anthracene, catalyst amounts of
more than 1.125 g were needed to achieve acceptable ketone
yields. The number of catalytically active sites (determined
with ammonia TPD experiments) on SZ in the reaction mix-
ture was substoichiometric (between 2 and 9% of the molar
aromatic amount) for all acylations summarized in Table 3.
Among the aromatic acylations investigated, the acety-
lation and benzoylation of 1-methoxynaphthalene and 2-
methoxynaphthalene and, furthermore, the benzoylation of
1,3-dimethylnaphthalene and anthracene can be carried out
with high ketone yields and on low SZ quantities.
In summary, the application of the heterogeneous cata-
lyst SZ for the synthesis of aromatic ketones was extended
from monocyclic aromatics to anthracene and various naph-
thalenes. SZ amounts with a substoichiometric number of

J. Deutsch et al. / Journal of Catalysis 231 (2005) 269–278
277
Table 3
SZ-catalyzed acylation of substituted naphthalenes, naphthalene, and anthracene under optimized reaction conditions according to Schemes 1, 2, and 3,
respectively
Aromatic/
acylating agent
Reaction
temperature ( C)/
solvent
Reaction
time
(h)
Amount
of SZ
(g)
Isolated and purified ketone(s)
◦
Yield
(%)
Isomer(s)
(Isomer ratio) (%)
1-Methoxynaphthalene/
Acetic anhydride
1-Methoxynaphthalene/
Benzoyl chloride
2-Methoxynaphthalene/
Acetic anhydride
2-Methoxynaphthalene/
Benzoyl chloride
1,3-Dimethylnaphthalene/
Benzoyl chloride
1-Methylnaphthalene/
Benzoyl chloride
70/
4
20
4
0.75
0.9
98
93
74
74
84
53
1-Acetyl-4-methoxy-
naphthalene (100)
1-Benzoyl-4-methoxy-
naphthalene (100)
1-Acetyl-2-methoxy-
naphthalene (100)
1-Benzoyl-2-methoxy-
naphthalene (100)
1,2-Dichloroethane
70/
1,2-Dichloroethane
70/
1,2-Dichloroethane
70/
1,2-Dichloroethane
70/
1,2-Dichloroethane
130/
1.125
1.125
0.9
20
20
20
1-Benzoyl-2,4-
dimethylnaphthalene (100)
1-Benzoyl-4-methyl-
naphthalene (92),
2.25
Chlorobenzene
a
2 further isomers
2-Methylnaphthalene/
Benzoyl chloride
130/
Chlorobenzene
20
2.25
81
1-Benzoyl-2-methyl-
naphthalene (74),
6 further isomers
b
Naphthalene/
Benzoyl chloride
Anthracene/
130/
Chlorobenzene
120/
20
4
3.375
1.125
69
89
1-Benzoylnaphthalene/
2-Benzoylnaphthalene (82/18)
9-Benzoylanthracene
(100)
Benzoyl chloride
Chlorobenzene
15 mmol aromatic, 30 mmol acylating agent, 30 ml solvent.
a
4/4.
b
13/5/4/2/1/1. The isomers formed in minor amounts were identified by their molpeaks (GC-MS analysis). A more precise characterization of these
ketones was not possible.
catalytically active surface sites can be used as an alterna-
tive to stoichiometric amounts of corrosive and ecologically
harmful Lewis acids in acylations of polycyclic aromatics.
[9] R. Selvin, B. Sivasankar, K. Rengaraj, React. Kinet. Catal. Lett. 67
(1999) 319.
[10] A. Heidekum, M.A. Harmer, W.F. Hoelderich, J. Catal. 188 (1999)
230.
[11] Y. Izumi, M. Ogawa, W. Nohara, K. Urabe, Chem. Lett. (1992) 1987.
[12] I.V. Kozhevnikov, Appl. Catal. A 256 (2003) 3.
[13] G. Harvey, G. Mäder, Coll. Czech. Chem. Commun. 57 (1992) 862.
[14] H.K. Heinichen, W.F. Hölderich, J. Catal. 185 (1999) 408.
[15] E. Fromentin, J.M. Coustard, M. Guisnet, J. Catal. 190 (2000) 433.
[16] P. Andy, J. Garcia-Martinez, G. Lee, H. Gonzalez, C.W. Jones, M.E.
Davis, J. Catal. 192 (2000) 215.
[17] E. Fromentin, J.M. Coustard, M. Guisnet, J. Mol. Catal. A 159 (2000)
377.
[18] D. Das, S. Cheng, Appl. Catal. A 201 (2000) 159.
[19] M. Casagrande, L. Storaro, M. Lenarda, R. Ganzerla, Appl. Catal.
A 201 (2000) 263.
Acknowledgments
The authors acknowledge financial support by the Min-
istry of Education and Science of the Federal Repub-
lic of Germany (project no. 03C0328). They also thank
H. Gehrmann, K. Neitzel, and W. Ziesche for valuable con-
tributions.
[20] A. Berreghis, P. Ayrault, E. Fromentin, M. Guisnet, Catal. Lett. 68
(2000) 121.
References
[21] P. Botella, A. Corma, G. Sastre, J. Catal. 197 (2001) 81.
[22] L. Cerveny, K. Mikulcova, J. Cejka, Appl. Catal. A 223 (2002) 65.
[23] D. Bhattacharya, S. Sharma, A.P. Singh, Appl. Catal. A 150 (1997) 53.
[24] V. Quaschning, J. Deutsch, P. Druska, H.-J. Niclas, E. Kemnitz,
J. Catal. 177 (1998) 164.
[25] J. Deutsch, V. Quaschning, E. Kemnitz, A. Auroux, H. Ehwald, H.
Lieske, Top. Catal. 13 (2000) 281.
[26] V. Quaschning, A. Auroux, J. Deutsch, H. Lieske, E. Kemnitz, J. Ca-
tal. 203 (2001) 426.
[27] A. Trunschke, J. Deutsch, D. Müller, H. Lieske, V. Quaschning, E.
Kemnitz, Catal. Lett. 83 (2002) 271.
[28] J. Deutsch, A. Trunschke, D. Müller, V. Quaschning, E. Kemnitz, H.
Lieske, J. Mol. Catal. A 207 (2004) 51.
[1] A. Corma, M.J. Climent, H. Garcia, J. Primo, Appl. Catal. 49 (1989)
109.
[2] K. Gaare, D. Akporiaye, J. Mol. Catal. A 109 (1996) 177.
[3] K. Smith, Z. Zhenhua, P.K.G. Hodgson, J. Mol. Catal. A 134 (1998)
121.
[4] D. Rohan, C. Canaff, E. Fromentin, M. Guisnet, J. Catal. 177 (1998)
296.
[5] E.G. Derouane, C.J. Dillon, D. Bethell, S.B. Derouane-Abd Hamid,
J. Catal. 187 (1999) 209.
[6] U. Freese, F. Heinrich, F. Roessner, Catal. Today 49 (1999) 237.
[7] T.W. Bastock, J.H. Clark, P. Landon, K. Martin, J. Chem. Res., Synop.
(1994) 104.
[8] B.M. Choudary, M. Sateesh, M.L. Kantam, K.V.R. Prasad, Appl.
Catal. A 171 (1998) 155.
[29] Y. Izumi, M. Ono, M. Kitagawa, M. Yoshida, K. Urabe, Microporous
Mater. 5 (1995) 255.

278
J. Deutsch et al. / Journal of Catalysis 231 (2005) 269–278
[30] F. Effenberger, E. Sohn, G. Epple, Chem. Ber. 116 (1983)
1195.
[31] L. Kolditz, in: Anorganische Chemie, vol. 1, VEB Deutscher Verlag
der Wissenschaften, Berlin, 1980, p. 178.
[33] F.E. Ray, W.A. Moomaw, J. Am. Chem. Soc. 55 (1933) 3833.
[34] E.G. Derouane, G. Crehan, C.J. Dillon, D. Bethell, H. He, S.B.
Derouane-Abd Hamid, J. Catal. 194 (2000) 410.
[35] J. Kaur, K. Griffin, B. Harrison, I.V. Kozhevnikov, J. Catal. 208
(2002) 448.
[32] C.R. Noller, R. Adams, J. Am. Chem. Soc. 46 (1924) 1889.