Synthesis of Pyrene-4,5-dione on a 15 g Scale

Article abstract of DOI:10.1002/ejoc.201601198

A scalable and efficient method for the synthesis of pyrene-4,5-dione has been developed. The addition of N-methylimidazole (NMI, 5 mol-%) to a known oxidation reaction was shown to marginally improve the yield and dramatically improve the ease of the workup and thus the amount of product isolable in a day by using regular laboratory equipment.

Full text of DOI:10.1002/ejoc.201601198

A Journal of
Accepted Article
Title: Synthesis of Pyrene-4,5-dione on a 15 g Scale
Authors: Joshua C. Walsh; Kerry-Lynn M. Williams; Dominik Lungerich; Graham J.
Bodwell
This manuscript has been accepted after peer review and the authors have elected
to post their Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by using the Digital
Object Identifier (DOI) given below. The VoR will be published online in Early View as
soon as possible and may be different to this Accepted Article as a result of editing.
Readers should obtain the VoR from the journal website shown below when it is pu-
blished to ensure accuracy of information. The authors are responsible for the content
of this Accepted Article.
To be cited as: Eur. J. Org. Chem. 10.1002/ejoc.201601198
Link to VoR: https://doi.org/10.1002/ejoc.201601198
Supported by

European Journal of Organic Chemistry
10.1002/ejoc.201601198
COMMUNICATION
Synthesis of Pyrene-4,5-dione on a 15 g Scale
Joshua C. Walsh,[a] Kerry-Lynn M. Williams,[a] Dominik Lungerich,[b] and Graham J. Bodwell*[a]
Abstract: A scalable and efficient method for the synthesis of
Scheme 1. Harris’ synthesis of 2 and Bodwell’s modification.
pyrene-4,5-dione has been developed. The addition of 5 mol% N-
methylimidazole (NMI) to a known oxidation reaction was shown to
marginally improve the yield and dramatically improve the ease of
Harris’ procedure employs 2 g of pyrene (1) and delivers
roughly 1 g of 2. In our hands, this was found to be the practical
limit of the reaction. The major issue is the large amount of
intensely colored (essentially black) intractable material that is
produced in the reaction. Removal of this material by filtration is
problematic because it quickly clogs filter paper, even with the
addition of Celite. We previously reported that the use of paper
towels instead of filter paper was beneficial,[11] but the filtration is
still quite slow and does not completely remove the solid
material. Moreover, 4–6 washes are required to remove the
majority of the product from the solids and each wash brings
more solid material through into the filtrate. The subsequent
extraction is also challenging because the solid material that
made it through the filtration step pervades both layers and
adheres to the walls of the separatory funnel. With both layers
already being deeply colored, it is very difficult to visualize the
border between layers. All in all, the workup requires the use of
a rather large volume of solvent for a reaction of its scale. While
the reaction is certainly repeatable, it is bothersome from the
perspective of the experimentalist and does not lend itself at all
well to being performed on a scale beyond 2 g of pyrene.
the workup and thus the amount of product isolable in a day using
regular laboratory equipment.
Introduction
The use of pyrene as a key building block for the synthesis of
designed π molecular systems, molecular sensors and carbon-
based materials has increased dramatically over the past
decade.[1] Not surprisingly, this has been accompanied by a
number of new synthetic approaches to functionalized
pyrenes.[2] One of the most heavily exploited pyrene derivatives
over this time has been pyrene-4,5-dione (2). In 2016 alone, 2
has been used in such diverse areas as asymmetric synthesis,[3]
supramolecular chemistry,[4] functional organic materials,[5]
polycyclic heteroacenes,[6] and reactive intermediates.[7] The
surge in popularity of 2 can be traced back to Harris’ 2005
publication,[8,9] in which in situ-generated RuO4 was shown to
selectively oxidize pyrene’s K-region to afford 2 directly in
moderate yield on a ca. 1 g scale (Scheme 1). Although a larger
scale (9 g) synthesis of 2 was reported in 1998,[10] it does not
appear to have been utilized very often, possibly due to its
relative length (three steps vs. one). In 2011, our group reported
that replacing one of Harris’ co-solvents (acetonitrile) with THF
resulted in a significant reduction in the reaction time,[11] which
allows for a reaction, the very messy workup (see below) and
chromatography to be completed within a single day.
Results and Discussion
The current broad interest in dione 2 led us to investigate how
not only the yield, but also the ease of performing the K-region
oxidation of pyrene (1) might be improved. Initial work showed
that an increase or decrease in the temperature had a
detrimental effect on the yield. Performing the reaction in an ice
bath led to a marked decrease in the rate of consumption of
pyrene (1) and afforded 2 in just 17% yield after 2.5 h. On the
other hand, performing the reaction at 30 °C clearly resulted in
an increase in the amount of solid byproduct and the yield fell to
32%. In this regard, it was found that overly rapid addition of
NaIO4, which dissolves exothermically in water, has the same
effect.
The addition of nitrogen-containing heterocycles to
transition metal-catalyzed oxidations has been shown to have a
significant effect on the rate and outcome of certain reactions.
For example, Sharpless reported that pyridine and various
pyridine derivatives had a beneficial effect when added to the
methylrhenium trioxide-mediated epoxidation of stilbenes.[12] The
ligand was found to play three roles: (1) to speed up catalytic
turnover, (2) to prevent the hydrolysis of the epoxides that were
produced and, (3) above a threshold concentration, to increase
the catalyst lifetime in solution. These results contrasted the
effects observed when using nonaromatic tertiary amines, which
were found to strongly inhibit catalyst activity.
[a]
[b]
Joshua C. Walsh, Kerry-Lynn M. Williams, Prof. Dr. Graham J.
Bodwell
Department of Chemistry
Memorial University of Newfoundland
283 Prince Philip Drive, St. John’s, NL, Canada, A1B 3X7
E-mail: gbodwell@mun.ca
https://wordpress.com/posts/gbodwell.wordpress.com
Dominik Lungerich
Department of Chemistry and Pharmacy, Interdisciplinary Centre for
Molecular Materials (ICMM)
Friedrich-Alexander-Universität Erlangen-Nürnberg
Henkestrasse 42, 91054 Erlangen, Germany
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201601198
COMMUNICATION
For the study of how additives affect the conversion of 1 to
and then recovered somewhat as the loading was increased to a
2, loadings of 5 mol% were chosen for initial screening (Table 1). full equivalent (Table 2, Entries 7 and 8).
The addition of pyridine resulted in a drop in the yield of 2 to
37%, but there was an immediately noticeable improvement in
Table 2. Optimization of the loading of NMI in the K-region oxidation of
the “cleanliness” of the workup, which made it considerably
easier to perform (Table 1, Entry 1). Of course, defining and
assessing the level of cleanliness / messiness of a workup (a
very important issue for an experimentalist) is a subjective
matter. As such, the observed improvements in the workup in
this and subsequent experiments were categorized only as good
(A), moderate (B) or poor (C).
pyrene (1)[a] to afford dione 2
Entry[
NMI (mol%)[b]
Yield (%)[b]
1
2
3
4
5
6
7
8
1
5
44
52
45
39
30
12
17
21
10
15
20
25
50
100
Table 1. Additive screening in the K-region oxidation of pyrene (1)[a] to afford
dione 2.
Entry
Additive[b]
Yield (%)[c]
Cleaning
Effect[d]
[a] 2.0 g scale. [b] Isolated yield.
1
2
3
4
5
6
7
8
pyridine
37
43
26
43
36
45
45
52
B
B
C
C
C
B
A
A
With an experimentally more manageable procedure in
hand, attention was turned to increasing the scale of the reaction.
Upon moving to 5, 10 and 25 g of pyrene (1), the work-up
proved to be as straightforward as it was in the 2 g scale and the
yield remained steady at 51-52% (Table 3, Entries 2–4). Each
of these reactions has been repeated several times by different
experimentalists with consistent outcomes. The 25 g scale
reaction, which afforded 14.9 g of pyrene-4,5-dione (2), was the
largest scale at which the reaction could be performed
comfortably using the laboratory glassware that was available to
us. The reaction was performed once on a 100 g scale, but this
proved to be substantially more challenging from a practical
perspective due to the large volumes of solvent and amounts of
materials that had to be dealt with. In this regard, it is much
more convenient to perform four reactions at the 25 g scale than
one reaction at the 100 g scale.
2.2’-bipyridine
1,10-phenanthroline
diethylene triamine
PPh3
DBU
DavePhos
N-methylimidazole
[a] 2.0 g scale. [b] 5 mol%. [c] Isolated yield. [d] A = good (much cleaner
workup), B = moderate (somewhat cleaner workup), C = poor (unchanged
workup).
The use of the bidentate ligand 2,2’-bipyridine also
resulted in a cleaner workup and the yield (43%) approached
that of our modified procedure (Table 1, Entry 2).[11] Upon
moving to 1,10-phenanthroline, the yield dropped to 26% and
the workup was as problematic as before (Table 1, Entry 3). It
may be that this ligand also undergoes K-region oxidation.[13]
A
reaction employing diethylene triamine (a nonaromatic,
tridentate tertiary amine) also had virtually no effect on the yield
or the workup (Table 1, Entry 4), whereas the use of
triphenylphosphine resulted in a lower yield of 2 (36%) after an
unimproved workup (Table 1, Entry 5). With the addition of the
Table 3. Scale-up of the K-region oxidation of pyrene (1) to afford dione 2
with 5 mol% NMI as an additive.
Entry
Pyrene (1) (g)
Yield (%)[a]
1
2
3
4
5
2
5
10
25
100
52
52
51
52
41
amidine base 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU,
a
relatively strong organic base) came the return of the noticeably
improved workup (Table 1, Entry 6). Upon using DavePhos, a
popular Buchwald-Hartwig ligand, the yield (45%) held steady
and there was a pronounced improvement in the cleanliness of
the workup to the point that it could be described as
straightforward (Table 1, Entry 7). N-Methylimidazole (NMI) had
an equally good cleaning effect on the workup and the yield
(52%) improved slightly. All of these reactions were run for 2.5 h
and, according to TLC analysis, there did not appear to be any
pronounced differences in the rate of consumption of pyrene (1).
Since the best result was obtained with NMI, which also
has the benefit of being inexpensive, it was selected for the
investigation of how the loading affects the yield of 2 (Table 2).
The cleaning effect of NMI was consistently good over the range
of 1–100 mol%, but the yield was quite variable. The yield of 2
was greatest at the initially chosen loading of 5 mol% (Table 2,
Entry 2), but still reasonably good at 1 mol% and 10 mol%
(Table 2, Entries 1 and 3). Curiously, the yield fell off sharply as
the loading was increased to 25 mol% (Table 2, Entries 4–6)
[a] Isolated yield.
Two pyrene derivatives, 2-tert-butylpyrene (3) and 2,7-di-
tert-butylpyrene (5) were also subjected to the new reaction
conditions and it was found that the oxidation smoothly
converted them to the respective diones 4 and 6 in slightly lower
yield on a 25 g scale (Scheme 2). Smaller scale reactions also
proceeded smoothly. The times required for the reactions of 3
and 5 to go to completion (2.0 and 1.5 h, respectively) were
shorter than for pyrene (1) (2.5 h),[14] which may be a reflection
of increasing electron density in the pyrene system with each
additional tert-butyl substituent. It may also that the tert-butyl
groups add steric bulk, which disfavors aggregation and/or
improves dissolution of starting materials. Paying attention to
the reaction time is important because, for all three substrates,
prolonged reaction resulted in the formation of several
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201601198
COMMUNICATION
byproducts (TLC analysis), a messier workup and diminishing
yields of the respective diones.
Experimental Section
For general experimental information, please see ref. 11.[11]
Pyrene-4-5-dione (2). To a solution of pyrene (25.0 g, 124 mmol) in
CH2Cl2 (500 mL) and THF (500 mL) were added RuCl3·3H2O (3.23 g,
12.4 mmol), N-methylimidazole (0.501 g, 6.18 mmol) and H2O (625 mL).
NaIO4 (119 g, 555 mmol) was added in small portions over 20 min. The
resulting slurry was stirred at room temperature for 2.5 h and the organic
solvents were removed under reduced pressure. CH2Cl2 (500 mL) and
H2O (500 mL) were added to dissolve the solids and the layers were
separated. The aqueous phase was extracted with CH2Cl2 (4 × 250 mL)
and the combined organic layers were washed with H2O (3 × 500 mL),
dried over anhydrous Na2SO4 and concentrated under reduced pressure
to afford a dark orange solid. Column chromatography (10 cm width  15
cm height, CH2Cl2) gave pyrene-4,5-dione (2) as bright orange crystals
(14.9 g, 52%) Rf (CH2Cl2) = 0.32: mp: >300 °C dec, Lit.[8] 299–302 °C; 1
H
NMR (CDCl3, 300 MHz) δ 8.46 (dd, J = 7.4, 1.3 Hz, 2H), 8.15 (dd, J = 8.0,
1.3 Hz, 2H), 7.83 (s, 2H), 7.74 (t, J = 7.7 Hz, 2H) ppm; 13C NMR (CDCl3,
75 MHz) δ 180.4, 135.7, 132.0, 130.2, 130.1, 128.4, 128.0, 127.3 ppm.
HRMS [APCI(+)] calcd for C16H8O2 232.0524, found: 232.0523.
Alternative workup using precipitation: Instead of evaporating the
post-extraction solution (combined organic layers) to dryness, it was
concentrated under reduced pressure to a volume of ca. 250 mL and
then diluted with methanol (ca. 500 mL). The resulting suspension was
then cooled to 0 °C for 1 h and the product was collected by suction
filtration to yield 2 as a brown powder (40–45%).
Scheme 2. K-Region oxidations of 1, 3 and 5 on a 25 g scale.
The new procedure is advantageous not only in terms of
convenience and scale, but also from a sustainability viewpoint.
Using the new procedure described here,
a
single
experimentalist can convert 25.0 g of pyrene (1) into 14.9 g of
dione 2 in a period of 6 h (includes the reaction, workup and
chromatography), during which ca. 6 L of solvent is used. The
large majority of this solvent is dichloromethane for extraction
2-tert-Butylpyrene-4,5-dione (4). To a solution of 2-tert-butylpyrene
(25.0 g, 97.0 mmol) in CH2Cl2 (500 mL) and THF (500 mL) were added
RuCl3⋅3H2O (2.59 g, 9.70 mmol), N-methylimidazole (0.398 g, 4.85
mmol) and H2O (625 mL). NaIO4 (93.5 g, 437 mmol) was added in small
portions over 20 min. The resulting slurry was stirred at room
temperature for 2 h and the organic solvents were removed under
reduced pressure. CH2Cl2 (500 mL) and H2O (500 mL) were added to
dissolve the solids and the layers were separated. The aqueous phase
was extracted with CH2Cl2 (4 × 250 mL) and the combined organic
extracts were washed with H2O (3 × 500 mL), dried over anhydrous
Na2SO4 and concentrated under reduced pressure to afford a dark
orange solid. Column chromatography (10 cm width  25 cm height,
CH2Cl2) gave 2-tert-butylpyrene-4,5-dione (4) as a bright orange solid
(13.4 g, 48%): Rf (CH2Cl2) = 0.64: mp: >250 °C dec. ; 1H NMR (CDCl3,
300 MHz) 8.57 (d, J = 2.1 Hz, 1H), 8.44 (dd, J = 7.4, 1.3 Hz, 1H), 8.14
(dd, J = 8.0, 1.3 Hz, 1H), 8.14 (d, J = 2.1 Hz, 1H), 7.81 (s, 2H), 7.70 (t, J
= 7.7 Hz, 1H), 1.49 (s, 9H) ppm; 13C NMR (CDCl3, 75 MHz) 180.85,
180.81, 151.7, 135.6, 132.3, 131.9, 131.8, 130.2, 130.1, 130.0, 128.6,
128.5, 127.61, 127.55, 127.1, 126.6, 35.3, 31.2 ppm. MS [APCI(+)] m/z
(%): 311 ([M+Na]+, 100); HRMS [APCI(+)] calcd for C20H16O2 288.1153,
found: 288.1150.
and chromatography, which is largely recoverable.[15]
By
comparison, using our previous procedure, the same
experimentalist was able to convert 2.0 g of pyrene (1) into 1.1 g
of dione 2 in a period of 7 h, during which ca. 4 L of solvent was
used.
In the case of pyrene-4,5-dione (2), column chromatography
can be avoided through the use of precipitation. The brown solid
obtained in this fashion is of lower purity than the bright orange
solid obtained from column chromatography (TLC, 1H NMR
analysis), but is of sufficient quality for use in reactions such as
reductive O-alkylation.[11] The precipitation method was not as
effective for the isolation of 4 and 6.
At this point in time, the role played by the NMI (and other
additives) in cleaning up the reaction and the reason(s) why it is
most effective at 5 mol% loading are unclear. No meaningful
correlations between any structural or physical property of the
additives (N-hybridization, denticity, pKa of the conjugate acid)
and the yield of the reaction leading to 2 or the cleaning effect
could be identified. More extensive studies will be required to
shed light on these issues.
2,7-Di-tert-butylpyrene-4,5-dione (6). To a solution of 2,7-di-tert-
butylpyrene (25.0 g, 81.2 mmol) in CH2Cl2 (500 mL) and THF (500 mL)
were added RuCl3⋅3H2O (2.17 g, 8.11 mmol), N-methylimidazole (0.333 g,
4.06 mmol) and H2O (625 mL). NaIO4 (78.3 g, 365 mmol) was added in
small portions over 20 min. The resulting slurry was stirred at room
temperature for 1.5 h and the organic solvents were removed under
reduced pressure. CH2Cl2 (500 mL) and H2O (500 mL) were added to
dissolve the solids and the layers were separated. The aqueous phase
was extracted with CH2Cl2 (4 × 250 mL) and the combined organic layers
were washed with H2O (3 × 500 mL), dried over anhydrous Na2SO4 and
concentrated under reduced pressure to afford a dark orange solid.
Column chromatography (10 cm width  25 cm height, CH2Cl2) gave 2,7-
bis(tert-butyl)pyrene-4,5-dione (6) as a bright orange solid (12.1 g, 44%):
Rf (CH2Cl2) = 0.48: mp: >250 °C dec, Lit.[8] 241–244 °C; 1H NMR (CDCl3,
300 MHz) 8.55 (d, J = 2.1 Hz, 2H), 8.12 (d, J = 2.1 Hz, 2H), 7.80 (s, 2H),
Conclusions
The addition of 5 mol% NMI to the ruthenium-mediated K-region
oxidation of pyrene (1) has a pronounced effect on the ease with
which the reaction workup can be performed. This allows it to
be performed comfortably on up to 25 g of pyrene (1) using
standard laboratory equipment. Using this procedure, a single
worker can easily generate ca. 15 g of dione 2 in one day.
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201601198
COMMUNICATION
1.49 (s, 18H) ppm; 13C NMR (CDCl3, 75 MHz) 181.1, 151.1, 131.9, 131.8,
129.8, 128.4, 127.3, 126.5, 35.2, 31.2 ppm. MS [APCI(+)] m/z (%): 367
([M+Na]+, 100); HRMS [APCI(+)] calcd for C24H24O2 344.1777, found:
344.1776.
X. Sun, A. Atxabal, M. Melle-Franco, L. E. Hueso, A. Mateo-Alonso,
Phys. Chem. Chem. Phys. 2016, 18, 11616-11619. (d) P. K. Sahoo, C.
Giri, T. S. Haldar, R. Puttreddy, K. Rissanen, P. Mal, Eur. J. Org. Chem.
2016, 1283-1291.
[7]
L. E. Longobardi, L. Liu, S. Grimme, D. W. Stephan, J. Am. Chem. Soc.
2016, 138, 2500-2503.
[8]
[9]
J. Hu, D. Zhang, F. W. Harris, J. Org. Chem. 2005, 70, 707-708.
A SciFinder search (limited to preparation, reactant or reagent of the
exact structure) afforded 113 references for 2, 80 of which were
published after reference 8.
Acknowledgements
Financial support of this work (G. J. B.) from the Natural
Sciences and Engineering Research Council (NSERC) of
Canada is gratefully acknowledged. Prof. J.-P. Lumb (McGill
University) is thanked for valuable discussions.
[10] E. R. R. Young, R. L. Funk, J. Org. Chem. 1998, 63, 9995-9996.
[11] G. Venkataramana, P. Dongare, L. N. Dawe, D. W. Thompson, Y. Zhao,
G. J. Bodwell, Org. Lett. 2011, 13, 2240-2243.
[12] J. Rudolph, K. L. Reddy, J. P. Chiang, K. B. Sharpless, J. Am. Chem.
Soc. 1997, 119, 6189-6190.
Keywords: pyrene • K-region • oxidation • ruthenium • N-
[13] Y. Jiang, C. Chen, Synlett, 2010, 1679-1681.
[14] In all of the reactions, the starting material is never completely
consumed (TLC analysis). Once the starting material has been
methylimidazole
reduced to just
terminated.
a faint fluorescent spot, the reaction should be
[1]
[2]
T. M. Figueira-Duarte, K. Müllen, Chem. Rev. 2011, 111, 7260-7314.
(a) J. M. Casas-Solvas, J. D. Howgego, A. P. Davis, Org. Biomol.
Chem. 2014, 12, 212-232. (b) J. M. Casas-Solvas, T. J. Mooibroek, S.
Sandramurthy, J. D. Howgego, A. P. Davis, Synlett 2014, 25, 2591-
2594.
[15] The dichloromethane used for extraction and chromatography is
collected from rotary evaporation with a pink (from chromatography) to
purple (from extraction) coloring due to the presence of iodine. Shaking
the recovered dichloromethane with aqueous sodium thiosulfate
solution in a separatory funnel affords colorless solvent, which after
drying over anhydrous Na2SO4, yields dichloromethane suitable for use
in a subsequent experiment. The dichloromethane that was recovered
during this work was used only for chromatography of K-region
oxidation reactions. The use of sodium thiosulfate washes during the
workup prior to chromatography did remove most of the iodine, but the
dichloromethane in the receiving flask of the rotary evaporator was still
pink during both the extraction stage and the chromatography. As such,
it proved to be more expedient to perform a sodium thiosulfate wash on
the recovered solvent.
[3]
[4]
Y. Jiang, J.-H. Fu, T.-Z. Li, F. Sha, X.-Y. Wu, Org. Biomol. Chem. 2016,
14, 6435-6441.
L. M. Salonen, D. D. Medina, E. Carbo-Argibay, M. G. Goesten, L.
Mafra, N. Guldris, J. M. Rotter, D. G. Stroppa, C. Rodriguez-Abreu,
Chem. Commun. 2016, 52, 7986-7989.
[5]
[6]
(a) M. Khadem, J. C. Walsh, G. J. Bodwell, Y. Zhao, Org. Lett. 2016,
18, 2403-2406. (b) Y. Liu, Q. Bai, J. Li, S. Zhang, C. Zhang, F. Lu, B.
Yang, P. Lu, RSC Adv. 2016, 6, 17239-17245. (c) P.-Y. Gu, J. Zhang,
G. Long, Z. Wang, Q. Zhang, J. Mater. Chem. C 2016, 4, 3809–3814.
(a) N. R. Arroyo, M. F. Rozas, P. Vazquez, G. P. Romanelli, M. V.
Mirifico, Synthesis 2016, 48, 1344-1352. (b) S. Fusco, C. Maglione, A.
Velardo, V. Piccialli, R. Liguori, A. Peluso, A. Rubino, R. Centore, Eur. J.
Org. Chem. 2016, 1772-1780. (c) A. B. Marco, C. Gozalvez, M. Olano,
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201601198
COMMUNICATION
Entry for the Table of Contents
COMMUNICATION
Pyrene Oxidation*
Joshua C. Walsh, Kerry-Lynn M.
Williams, Dominik Lungerich, Graham J.
Bodwell*
Page No. – Page No.
Text for Table of Contents
Synthesis of Pyrene-4,5-dione on a 15
g Scale
The addition of 5 mol% NMI to the ruthenium-mediated K-region oxidation of pyrene has a dramatic
effect on the workup. The previously very messy and laborious workup that limited the scale of the
reaction has thus become much easier to perform, which enables the reaction to be performed
comfortably on up to 25 g of pyrene.
This article is protected by copyright. All rights reserved