Photooxidation of benzyl alcohols with immobilized flavins

Article abstract of DOI:10.1002/adsc.200800576

Benzyl alcohols are oxidized cleanly and efficiently to the corresponding aldehydes under irradiation using flavin photocatalysts and aerial oxygen as the terminal oxidant in homogeneous aqueous solution. Turnover frequencies (TOF) of more than 800 h^-1 and turnover numbers (TON) of up to 68 were obtained. Several flavin photocatalysts with fluorinated or hydrophobic aliphatic chains were immobilized on solid supports like fluorous silica gel, reversed phase silica gel or entrapped in polyethylene pellets. The catalytic efficiency of the heterogeneous photocatalysts was studied for the oxidation of different benzyl alcohols in water and compared to the analogous homogeneous reactions. Removal of the heterogeneous photocatalyst stops the reaction conversion immediately, which shows that the immobilized flavin is the catalytically active species. The immobilized catalysts are stable, retain their reactivity if compared to the corresponding homogeneous systems and are easily removed from the reaction mixture and reused. TOF of up to 26 h^-1, TON of 280 and up to 3 reaction cycles without loss of activity are possible with the heterogeneous flavin photocatalysts.

Full text of DOI:10.1002/adsc.200800576

FULL PAPERS
DOI: 10.1002/adsc.200800576
Photooxidation of Benzyl Alcohols with Immobilized Flavins
Harald Schmaderer,a Petra Hilgers,a Robert Lechner,a and Burkhard Kçniga,*
a
Institute of Organic Chemistry, University of Regensburg, Universitꢀtsstr. 31, 93040 Regensburg, Germany
Fax : (+49)-941-943-1717; phone: (+49)-941-943-4575; e-mail: burkhard.koenig@chemie.uni-regensburg.de
Received: September 17, 2008; Revised: December 17, 2008; Published online: January 12, 2009
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/adsc.200800576.
Abstract: Benzyl alcohols are oxidized cleanly and heterogeneous photocatalyst stops the reaction con-
efficiently to the corresponding aldehydes under irra- version immediately, which shows that the immobi-
diation using flavin photocatalysts and aerial oxygen lized flavin is the catalytically active species. The im-
as the terminal oxidant in homogeneous aqueous so- mobilized catalysts are stable, retain their reactivity
lution. Turnover frequencies (TOF) of more than if compared to the corresponding homogeneous sys-
800 hÀ1 and turnover numbers (TON) of up to 68 tems and are easily removed from the reaction mix-
were obtained. Several flavin photocatalysts with flu- ture and reused. TOF of up to 26 hÀ1, TON of 280
orinated or hydrophobic aliphatic chains were immo- and up to 3 reaction cycles without loss of activity
bilized on solid supports like fluorous silica gel, re- are possible with the heterogeneous flavin photoca-
versed phase silica gel or entrapped in polyethylene talysts.
pellets. The catalytic efficiency of the heterogeneous
photocatalysts was studied for the oxidation of differ- Keywords: catalyst recycling; flavins; fluorous
ent benzyl alcohols in water and compared to the chemistry; immobilization; photooxidation; redox
analogous homogeneous reactions. Removal of the chemistry
Introduction
Although a typical catalyst loading in the photooxi-
dation reactions is about 1 mol%, it is advantageous
Flavin is a prosthetic group of flavoproteins and a ver- to immobilize the catalysts to facilitate its separation
satile electron carrier in biological systems. In nature, from the reaction mixture and a potential reuse. Fur-
flavins occur mostly in the form of flavin adenine di- thermore, for the synthesis of larger quantities of car-
nucleotide (FAD), flavin mononucleotide (FMN) or bonyl compounds, it would be desirable to accomplish
riboflavin co-factors.[1] The presence and structure of the catalytic oxidation of alcohols in a continuous re-
the surrounding protein, substitutions, and non-cova- action process which requires catalysts that are immo-
lent interactions significantly alters the redox proper- bilized on a suitable surface and are available for a
ties of these compounds. Furthermore, irradiation has number of reaction cycles.
a major effect on the reactivity of flavins by increas-
Only few examples of immobilized flavins have
been published. Bꢀckvall et al. described the catalytic
ing their redox power.[2]
Recently, flavins have received interest as flavoen- oxidation of sulfides to sulfoxides by H2O2 or NMM
zyme models[3] and for applications in chemical catal- to NMO in the osmium-catalyzed dihydroxylation
ysis.[4,5] Photooxidation of alcohols using catalytic with flavin derivatives immobilized in an ionic
amounts of flavin is particularly advantageous due to liquid.[4a,7] Rotello et al. reported on the development
the low toxicity of the reagent. The flavin-mediated of flavin derivatives appended on polystyrene copoly-
photooxidation of benzyl alcohols is an efficient pro- mers to study their redox properties.[3b] To the best of
cess using air as terminal oxidant; the reaction was re- our knowledge no use of heterogeneously immobi-
cently investigated and optimized for homogeneous lized flavins as catalysts in organic reactions has been
reaction in acetonitrile solution.[6] High power LEDs reported.
serve as a selective and efficient light source for the
Typical immobilization strategies for catalysts are
reaction, which makes applications to synthesis very covalent binding to a solid support, physical or elec-
easy.
trostatic adsorption,[8] and entrapment of catalysts in
porous materials. From these methods, electrostatic
Adv. Synth. Catal. 2009, 351, 163 – 174
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
163

Harald Schmaderer et al.
FULL PAPERS
and physical adsorption are the most simple to imple- cules with even higher fluorine content are required.
ment; however, both approaches are prone to catalyst Therefore, flavin 4 was alkylated with iodide 5 and
leaching. A special case of surface adsorption that potassium carbonate in dry dimethylformamide to
partly overcomes this drawback uses perfluorinated obtain flavin 6 in 62% yield. Flavin 6 exhibits a fluo-
alkyl chains as tags to impose fluorous properties on rine mass content of 57% (Scheme 1).
a given molecule. This allows separation of the per-
The synthesis and use of flavin 7 was recently re-
fluorinated compound from a complex reaction mix- ported.[6] 3-N-Methylation of the compound was ac-
ture by fluorous phase extraction or adsorption on complished in dry dimethylformamide with methyl
fluorous silica or perfluorinated polymers.[9] Fluorous iodide and caesium carbonate as base yielding 91% of
technology has been applied to catalyst recovery in flavin 8a. The use of a perfluorinated alkylation agent
transition metal[10] and organocatalysis.[11]
5 gave flavin 8b in 23% yield (Scheme 2).
We have investigated the immobilization of flavin
photocatalysts on unmodified, fluorous and reversed
phase silica gel, and the flavin entrapment in PE pel-
lets and glue. The catalytic activity of the heteroge-
ACHTUNGTRENNUNGneous photocatalysts was determined in benzyl alco-
hol photooxidations in aqueous solution and com-
pared to the analogous reactions in homogeneous so-
lution. We discuss and compare here the stability of
the different immobilized catalysts and their photoox-
idation efficacy.
Scheme 2. Preparation of N-methylated and fluorinated fla-
vins 8.
Results and Discussion
Synthesis
The synthesis of a hydrophobic flavin 13 started
from commercially available riboflavin. Riboflavin
The new flavin derivatives were synthesized according tetraacetate (10), which is easily available in large
to the Kuhn protocol.[12] Dinitro compound 1 was ob- amounts from riboflavin,[15] was methylated as de-
tained in an optimized route from commercially avail- scribed before yielding flavin 11[16] in 71% yield. The
able 4,5-dimethylnitroaniline.[13] Fluorinated amine 2 acetyl groups of the ribityl chain were cleaved by p-
was synthesized via azide substitution of iodide 5 and toluenesulfonic acid to give 3-methylriboflavin (12)[16]
subsequent reduction.[14]
(73%). Reaction with palmityl chloride gave 3-meth-
In a nucleophilic aromatic substitution, dinitro com- yltetrapalmityl riboflavin (13) as an orange soft solid
pound 1 was reacted with fluorinated amine 2. After in 52% yield (Scheme 3). This flavin shows high solu-
reduction of the remaining nitro group, the resulting bility in organic solvents and its hydrophobic proper-
amine was instantly used without isolation due to its ties are desired for immobilization on reversed phase
sensitivity to air. Cyclocondensation with alloxane silica gel.
monohydrate yielded flavin 4 in 38% over two steps.
In addition, some previously prepared flavin deriva-
The fluorine mass content of the compound reaches tives,[6] shown in Scheme 4, were tested for immobili-
47%. In order to enhance fluorine interactions, mole- zation and as catalysts.
Scheme 1. Synthesis of perfluorinated flavin 6.
164
asc.wiley-vch.de
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 163 – 174

Photooxidation of Benzyl Alcohols with Immobilized Flavins
FULL PAPERS
Scheme 3. Synthesis of riboflavin derivatives. Reaction con-
ditions: a) MeI, CsCO3, DMF, room temperature, 16 h; b) p-
TsOH, EtOH, D, 17 h; c) C16H31OCl, pyridine, CHCl3, room
temperature, 24 h
Scheme 5.
up within one hour, corresponding to a TOF of up to
10 hÀ1.
However, we realized that this reaction is even
more efficient in aqueous solution. Figure 1 shows a
standard screening set-up (left without irradiation;
right with irradiation). The photocatalytic activity of
different flavins was tested in solutions of benzyl alco-
hol in D2O (for detailed set-up see Experimental Sec-
tion).
With 10 mol% of riboflavin tetraacetate (10) as
photocatalyst the reaction was completed within five
minutes (Table 1, entry 1). Within one minute, it was
possible to convert 58% of the alcohol corresponding
to a TOF of 350 hÀ1 (entry 2) and lower catalyst load-
ings led to a TOF of more than 800 hÀ1 (entry 3). If
the reaction mixture is not stirred, the TOF drops by
a factor of 5, demonstrating the importance of effi-
cient mixing during the course of the reaction, espe-
cially for longer irradiation times (entry 4). Our previ-
ous investigations in acetonitrile solution showed that
the addition of thiourea dramatically accelerates this
photooxidation.[6] Therefore, we also tried to further
improve the rates in aqueous solution by the addition
Scheme 4. Thiourea-flavin derivatives investigated for cata-
lytic activity.
Photooxidations in Aqueous Homogeneous Solution
Recently, we studied the photooxidation of p-meth- of thiourea. However, in water the addition of thio-
oxybenzyl alcohol in the presence of various flavin urea decreases the reaction rates (entries 5 and 6).
A
ACHTUNGTRENNUNG
catalysts in acetonitrile solution.[6] In the catalytic Whereas in acetonitrile electronically activated sub-
cycle, oxidized flavin (Flox) is excited by irradiation strates, such as p-methoxybenzyl alcohol were re-
with light of suitable wavelength. As light source we quired for the conversion into the aldehyde,[6] the
used commercially available high power LEDs with scope of convertible substrates is significantly extend-
an emission maximum at 440 nm which matches the ed in water. Alteration of the redox potentials or a
longest wavelength absorption maximum of oxidized tighter catalyst–substrate interaction in the more
flavin. After irradiation, the flavin becomes strongly polar solvent may explain the observation. We were
oxidizing and benzyl alcohol is converted into the cor- able to convert unsubstituted benzyl alcohol (entry 8)
responding aldehyde. If oxygen is present in the reac- and electron-poor benzyl alcohols in moderate to
tion mixture, the reduced flavin species (Flred) is in- good yields (entries 9–11). As expected, due to the
stantly reoxidized, forming hydrogen peroxide as the electronic deactivation, the reaction rates drop com-
second reaction product.[4d,6,17,18] Overall, benzyl alco- pared to p-methoxybenzyl alcohol. An experiment
hol is oxidized in a light-driven catalytic cycle by with 2.5 mmol of p-methoxybenzyl alcohol and a cata-
aerial oxygen as terminal oxidant (Scheme 5).[17–20] In lyst loading of 1 mol% in 250 mL of water demon-
these experiments, we were able to almost completely strates the applicability on preparative laboratory
convert the alcohol starting material in a typical set- scale, as full conversion was obtained after 20 h in a
Adv. Synth. Catal. 2009, 351, 163 – 174
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
165

Harald Schmaderer et al.
FULL PAPERS
Figure 1. Standard screening setup (left: without irradiation; right: with irradiation).
Photooxidation with Silica Gel-Immobilized Flavins
Table 1. Catalytic photooxidations with flavins in homogene-
ous aqueous solution.[a]
After the optimization of the photooxidation condi-
tions in homogeneous solution, heterogeneous flavin
catalysts immobilized on silica gel were investigated.
Initially, flash silica gel particles with diameters of 35–
70 mm were used as solid support for immobilization
due to their large surface areas. Immobilization of the
catalysts was accomplished by soaking flash silica gel
with solutions of flavins in chloroform and evapora-
tion of the solvent, which gave homogeneous yellow
powders of flavin-coated silica gel (Figure 2).
Fluorinated flavins 4, 6 and 8b and non-fluorinated
flavins 7, 8a, 13, 14 and 15 were immobilized on silica
gel and fluorinated silica gel. The nature of the sup-
port has a dramatic effect on the stability of the im-
mobilized catalyst in aqueous solution. Fluorinated
flash silica gel turned out to be a more suitable sup-
port for all flavins compared to standard flash silica
gel. In water, considerable amounts of the chromo-
phores were washed off from not fluorinated flash
silica gel, giving yellow and fluorescing solutions. No
leaching of flavins was observed from the fluorinated
support in aqueous media and in toluene, even if fla-
Entry Flavin X
t
Aldehyde
TON TOF
[1/h]
[min] [%]
1
2
10
10
10
10
10
OMe
OMe
OMe
OMe
OMe
OMe
OMe
H
5
1
5
5
15
15
15
25
100
58
68
59
79
48
48
75
74
58
47
10
–
5.8
68
348
816
71
32
19
19
18
18
14
11
3[b]
4[c]
5[c,d]
5.9
7.9
4.8
4.8
7.5
7.4
5.8
4.7
6[c,d,e] 10
7[c,d]
8
11
10
10
10
10
9
10
11
COONa 25
COOMe 25
COOH 25
[a]
Conditions: V=1 mL, 2% DMSO-d6 in D2O, alcohol:
0.002M, 10 mol% of catalyst.
1 mol% of catalyst.
Without stirring.
Reaction in an NMR tube instead of a small glass vial.
Addition of thiourea (0.2 mM).
[b]
[c]
[d]
[e]
very clean reaction (see Supporting Information for vins without fluorous tags were applied. It was also
details). checked whether riboflavin tetraacetate (10) was
Earlier, Fukuzumi et al. presented a mechanism for washed off from the support in a reaction mixture
this reaction starting with an initial electron transfer containing p-methoxybenzyl alcohol. After stirring for
from the alcohol substrate to the oxidized flavin cata- 30 min under standard conditions, the characteristic
lyst, leading to the aldehyde and hydrogen peroxide absorption of flavin was not detectable in a UV/Vis
as products.[17,19] Later, this mechanism was supported spectrum, an indication of a flavin concentration in
by calculations[20] and we observed stoichiometric hy- solution smaller than c=10À7 mol/L (see Supporting
drogen peroxide formation in previous studies.[4d,6] 1- Information for details).
(4-Methoxyphenyl)-2,2-dimethylpropan-1-ol was used
The photocatalytic activity of immobilized flavins
as probe to discriminate between an electron transfer on silica supports (1% per mass) was tested in solu-
and a hydrogen abstraction mechanism for the photo- tions of benzyl alcohol in D2O (for detailed set-up see
oxidation of benzyl alcohols.[21] At the conditions of Experimental Section).
our experiments, we exclusively observe the electron
With riboflavin tetraacetate (10) as catalyst, the
transfer pathway (see Supporting Information for de- system retains its activity compared to homogeneous
tails).
solution and p-methoxybenzyl alcohol was completely
converted within one hour (Table 2, entries 1 and 2).
166
asc.wiley-vch.de
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 163 – 174

Photooxidation of Benzyl Alcohols with Immobilized Flavins
FULL PAPERS
Figure 2. Top : Normal light irradiation; bottom: UV blue light irradiation. Left: solid flavin 10, middle: non-modified silica
gel, right: immobilized catalyst.
However, in some reactions, oxidation to minor conditions the catalytic activity of 3-N-alkylated and
amounts of p-methoxybenzoic acid occurs. Shorter re- 3-N-H flavins is comparable (Table 1, entries 5 and 7),
action times of 30 or 15 min, respectively, suppress indicating that this difference has to be attributed to
the undesired acid by-product (entries 3 and 4).
the immobilization. Neither the redox potentials[4d]
Oxygen is the terminal oxidant of this reaction. nor the shape of HOMO and LUMO[22] of the flavin
Therefore, it was investigated whether the photooxi- chromophore are significantly influenced by 3-N-alky-
dation is accelerated under an oxygen atmosphere lation. A different orientation of the flavin chromo-
(entries 5–7). However, the conversion drops by a phore on the support surface may be envisaged to
factor of two compared to the standard experiments cause the differences in reactivity.
in aerial environment (entries 2–4). The adverse
Flavins 7 and 15a showed considerably higher
effect of high oxygen concentrations may be due to TOFs of 4.7 and 4.9 hÀ1, respectively, compared to all
the quenching of the flavin excited state by oxygen.
other tested derivatives, exceeding riboflavin tetraace-
Hydrogen peroxide, the second product of the reac- tate (10) as catalyst (3.6 hÀ1 under comparable condi-
tion, is not able to oxidize p-methoxybenzyl alco- tions).
hol.[4d] A standard reaction mixture with 10 equiva-
In general, the TOF of the photoreaction with im-
lents of hydrogen peroxide instead of a flavin catalyst mobilized flavins drops compared to experiments in
does not show any conversion after stirring for 30 min homogeneous solution by a factor of 8–20, while
(see Supporting Information).
TON and aldehyde yields remain comparable
Like in homogeneous solution (Table 1), the addi- (Table 3, entries 1–6).
tion of thiourea decelerates the conversion (entries 8
Recycling of the immobilized photocatalysts was
and 9),[6] but a larger scope of possible substrates is demonstrated by placing a reaction mixture into a sy-
observed. Electronically not activated benzyl alcohols ringe with a filter. After each 15 min of irradiation
were oxidized with the catalysts on fluorinated silica and stirring, the reaction mixture was removed and
1
gel with reasonable TOFs and good overall conver- the conversion was monitored by H NMR. To the re-
sions (entries 10–13).
maining immobilized catalyst in the syringe, a fresh
A surprising correlation arises from the analysis of benzyl alcohol solution was added and the procedure
the catalytic activity of different immobilized flavins was repeated (see Supporting Information for de-
(entries 1, 16–24). All 3-N-substituted flavins show a tails). The activity of the immobilized photocatalyst
reduced catalytic activity in the photooxidation of p- remained almost unchanged for three cycles with high
methoxybenzyl alcohol in aqueous solution when im- TOFs of 10 hÀ1 and then dropped by 30% in the next
mobilized on silica support. In homogeneous reaction two cycles (Table 4, entries 1–5). These results con-
Adv. Synth. Catal. 2009, 351, 163 – 174
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
167

Harald Schmaderer et al.
FULL PAPERS
Table 2. Catalytic photooxidations with flavins immobilized on silica gel.[a]
Entry
Flavin
X
t [h]
Aldehyde [%]
Acid [%]
TON
TOF [1/h]
1
2
3
4
10
10
10
10
10
10
10
10
10
10
10
10
10
10
4
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OMe
Me
2
1
71
83
78
65
46
33
29
52
55
41
44
36
24
<3
61
93
46
10
54
10
97
77
59
18
2.8
<3
29
17
3
3
0
0
0
8
12
9
0
0
0
0
9
4
3
0
1
0
3
8
4
3
–
–
7.1
8.3
7.8
6.5
4.6
3.3
2.9
5.2
5.5
4.1
4.4
3.6
2.4
<0.3
6.1
9.3
4.6
1.0
5.4
5.0
9.7
7.7
5.9
1.8
280
–
3.6
8.3
16
0.5
0.25
1
0.5
0.25
2
2
2
2
2
2
2
2
2
2
2
2
2
26
5[b]
6[b]
7[b]
8[c]
9[d]
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25[e]
26[f]
4.6
6.6
12
2.6
2.8
2.1
2.2
1.8
1.2
<0.15
3.1
4.7
2.3
0.5
2.7
2.5
4.9
3.9
3.0
0.9
18
H
COOMe
COOH
COONa
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OMe
OMe
7
8a
8b
11
14
15a
15b
15c
15d
13
6
2
2
2
2
15.5
4
–
[a]
Conditions: V=1 mL; 1% per mass flavin on fluorinated silica gel; 10 mol% of catalyst; alcohol: 0.002M.
Oxygen atmosphere.
Addition of thiourea (0.002M).
Addition of thiourea (0.001M).
Reversed phase silica gel instead of fluorinated, pure alcohole solution.
Immobilized on fluorinated glass instead of fluorous silica gel, V=7 mL, alcohol: 0.01M, without stirring
[b]
[c]
[d]
[e]
[f]
Table 3. Comparison of homogeneous, silica gel and polyethylene based catalysis.[a]
Entry
X
System
t [min]
Aldehyde [%]
TON
TOF [1/h]
1
2
3
4
5
6
OMe
OMe
H
in solution
on silica gel
in solution
on silica gel
in solution
on silica gel
in solution
PE-pellet
1
58
78
75
44
58
36
59
65
5.8
7.8
7.5
4.4
5.8
3.6
–
348
16
18
2.2
14
1.8
–
30
25
120
25
120
5
H
COOMe
COOMe
OMe
OMe
7[b]
8[b,c]
240
–
–
[a]
Conditions: V=1 mL, D2O, tetraacetyl riboflavin (10) as catalyst, alcohol: 0.002M.
Without stirring.
Alcohol: 0.01M.
[b]
[c]
firm that the immobilized flavins are catalytically 30 min did not result in further reaction conversion
active and not chromophore molecules that are leach- (see Supporting Information). The photooxidation
ing from the support. As an additional experiment, proceeds only in the presence of the heterogeneous
we removed the immobilized catalyst from a reaction photocatalyst.
mixture that was irradiated for 30 min (yield: 72% of
To show the applicability of the photooxidation to
aldehyde). Continued irradiation of the same solution the preparative laboratory scale we used p-methoxy-
without the catalyst under identical conditions for benzyl alcohol as substrate and solvent. 3-Methylte-
168
asc.wiley-vch.de
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 163 – 174

Photooxidation of Benzyl Alcohols with Immobilized Flavins
FULL PAPERS
Table 4. Recycling experiments with immobilized catalysts.[a]
Entry
Alcohol [M]
Immobilization
t [min]
Aldehyde [%]
Run
TON
TOF [1/h]
1[b]
2[b]
3[b]
4[b]
5[b]
6
0.002
0.002
0.002
0.002
0.002
0.01
on SiO2
on SiO2
on SiO2
on SiO2
on SiO2
PE
15
15
15
15
15
240
240
22
25
24
15
8
1
2
3
4
5
1
2
2.2
2.5
2.4
1.5
0.8
–
8.8
10
9.6
6.0
3.2
–
46
29
7
0.01
PE
–
–
[a]
Conditions: V=1 mL, D2O, tetraacetylriboflavin (10) as catalyst, p-methoxybenzyl alcohol as substrate.
10 mol% of catalyst.
[b]
trapalmitylriboflavin (13), which is a very low melting Flavin Immobilization by Entrapment in PE Pellets
oily solid and completely insoluble in water was im- or Glues
mobilized on reversed phase silica gel, comprising
C18-alkyl chains and was used as catalyst. Preparation In some solvents, the catalysts were washed off from
of the supported catalyst was carried out in the same the silica gel support. Therefore, an entrapment of fla-
way as for fluorinated silica gel. The photooxidation vins in polymer pellets and in simple glue was tried.
in neat p-methoxybenzyl alcohol allows an overall As water is the best solvent for the photocatalytic oxi-
conversion of 2.8% to the aldehyde with a TOF of dation completely insoluble polyethylene (PE) was
18 hÀ1 and high TON of 280 (Table 2, entry 25).
selected as material for the entrapment. The prepara-
An attempt to create an active photocatalyst by im- tion of flavin containing PE pellets is simple: The
mobilization of flavin 6 on commercially available flavin chromophore and commercially available PE
fluorinated glass[23] failed due to the small amount of powder are mixed in a ratio of 1:100 by weight. This
deposited photoactive compound. No significant con- mixture is compressed to 125 MPa at 808C yielding
version was observed with this catalyst even after 4 h yellow, fluorescing pellets (Figure 3).
of irradiation (Table 2, entry 26).
To determine the catalytic activity, the flavin-PE
pellets were placed at the top of 1 mL of an aqueous
Figure 3. Top : Normal light, bottom: UV light. Left: solid flavin 10, middle: PE powder, right: immobilized catalyst + 1 cent
coin for comparison.
Adv. Synth. Catal. 2009, 351, 163 – 174
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
169

Harald Schmaderer et al.
FULL PAPERS
Table 5. Catalytic photooxidations with entrapped flavins.[a]
Conclusions
Entry Flavin Alcohol
[mM]
t
Immobilization
Aldehyde
[%]
The photooxidation of benzyl alcohols to the corre-
sponding aldehydes with flavin catalysts using oxygen
as the terminal oxidizing agent proceeds very rapidly
and efficiently in aqueous solution. A TOF of more
than 800 hÀ1 and a TON of up to 68,800 were ob-
served. It was possible to oxidize benzyl alcohols
which are electronically not activated for this reac-
tion.
To create heterogeneous photocatalysts, several
new flavin derivatives were prepared and immobilized
on solid supports (fluorous silica gel, reversed phase
silica gel, PE pellets). The immobilized photocatalysts
retained their catalytic activity for benzyl alcohol oxi-
dation. In comparison to the analogous homogeneous
[h] method
1
2
3
4
5
6
7
10
11
8a
7
–
13
10
10
10
10
10
10
2
4
4
4
4
4
1
PE
PE
PE
PE
65
55
14
n.s.c.[b]
n.s.c.[b]
29
pure PE
coated on glass
2
12 superglue
n.s.c.[b]
[a]
[b]
Conditions: V=1 mL, D2O.
n.s.c.=no significant conversion.[24]
p-methoxybenzyl alcohol solution (0.01m in D2O) in a reactions, the reaction rates decrease by a factor of 8–
small glass reaction vessel, which was irradiated from 20 for immobilization on silica gel and by a factor of
the bottom by an LED (440 nm) without stirring. The 50 for the catalyst entrapment in polyethylene pellets.
1
reaction conversion was monitored by H NMR.
The highest catalytic activity (TOF 26 hÀ1), best sta-
Riboflavin tetraacetates 10 and 11 were found to bility (TON 280) and up to 3 reaction cycles without
be the most active catalysts if immobilized in PE pel- loss of performance are possible with flavins 7, 10 and
lets. A conversion of the benzyl alcohol to the alde- 13 immobilized on fluorous silica gel and reversed
hyde in 55–65% yield was observed after 4 h (Table 5, phase silica gel, respectively.
entries 1 and 2). In absolute numbers, 5.5–6.5 mmol
In summary, we have shown that heterogeneous
were converted in such a reaction. However, com- photocatalysts can be derived from functionalized fla-
pared to the reaction in homogeneous aqueous solu- vins by physical adsorption on a fluorinated silica gel
tion which reaches under identical conditions the support or by mechanical entrapment. The immobi-
same conversion after 5 min (Table 1, entry 4), the ef- lized chromophores are the catalytically active spe-
ficiency is drastically reduced by a factor of 50. In ad- cies, as the photooxidation reaction stops immediately
dition, one pellet contains about 4.6 mmol of flavin, when the heterogeneous catalyst is removed. The het-
which means that the catalyst loading is 45%. Obvi- erogeneous catalysts significantly facilitate product
ously, not all flavin molecules are accessible for the isolation and catalyst recycling. Flavin-catalyzed pho-
photocatalytic reaction as they are trapped in the tooxidations are suited for preparative laboratory
bulk of the pellet leading to a lower effective loading. scale conversions. The introduced heterogeneous cata-
Other entrapped flavins 7 and 8a (entries 3 and 4) lysts facilitate the use of flavin photooxidation cata-
showed only low or no significant photooxidation ac- lysts in synthesis and pave the way to applications in
tivity. As expected, no conversion was obtained if a energy efficient transformations in continuous flow
PE pellet without flavin was used. Recycling of flavin reactions.
catalyst pellets is limited; the activity drops by 50%
in the second run (Table 4, entries 6 and 7).
In order to obtain stable immobilized flavin layers,
Experimental Section
we entrapped riboflavin tetraacetate (10) by mixing a
chloroform solution with superglue[25] and evaporating
the mixture. This yielded a stable polymer layer con-
taining the chromophore in a glass, which unfortu-
Dinitrobenzene 1,[13] fluorinated amine 2,[14] 10-(2-meth-
oxyethyl)-7,8-dimethylbenzo[g]pteridine-2,4(3H,10H)-dione
(7)[6] and riboflavin tetraacetate (10)[15] were prepared by
G
ACHTUNGTRENNUNG
nately showed no conversion when irradiated in the known methods. All other chemicals were purchased from
commercial suppliers, checked by 1H NMR spectrometry
presence of substrate solution for 12 h (entry 7).
and then used as received. Before use, solvents were dis-
Simple evaporation of a chloroform solution of tetra-
tilled. Dry N,N-dimethylformamide was purchased from
palmitylriboflavin (13) without adding glue yielded a
stable chromophore layer in a glass vial. This layer is
stable in aqueous solution, but is destroyed by stir-
ring. Therefore, the photooxidation was investigated
Fluka. Fluorinated silica gel with a particle size of 35–70 mm
and reversed phase silica gel (40–63 mm) were purchased
from Fluka. Thin layer chromatography was carried out on
silica gel 60 F254 aluminium sheets (Merck) or on precoated
by irradiation of the flavin layer with a solution of p-
plastic sheets Polygram SIL G/UV254 (Macherey–Nagel,
methoxybenzyl alcohol that was not stirred. Within
1 h, 29% of the substrate was converted into the cor-
responding aldehyde (entry 6).
Dꢂren, Germany), with detection under 254 nm or 333 nm
UV light or by naked eye (flavins are intensively yellow-col-
oured). Flash column chromatography was carried out on
170
asc.wiley-vch.de
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 163 – 174

Photooxidation of Benzyl Alcohols with Immobilized Flavins
FULL PAPERS
silica gel 35–70 mm, 60 ꢃ from Acros. NMR spectra were re-
corded with a Bruker spectrometer equipped with a robotic
sampler at 300 MHz (1H NMR) or 75 MHz (13C NMR). Tet-
ramethylsilane (TMS) was used as an external standard.
Electrospray ionization (ES-MS) mass spectra were record-
ed on a ThermoQuest Finnigan TSQ 7000 spectrometer.
High resolution mass spectrometry (HR-MS) was performed
with a ThermoQuest Finnigan MAT 95 spectrometer. Melt-
ing points were measured on a Bꢂchi SMP-20 apparatus and
are uncorrected. IR spectra were measured on a Biorad
Spectrometer Excalibur FTS 3000.
added. After refluxing for five more days, the solvent was
evaporated yielding a yellow-orange solid. Flash column
chromatography (Rf =0.25; petrol ether:chloroform, 4:1)
gave an orange solid; yield: 294 mg (480 mmol, 24%); mp
858C; IR (ATR): n=3348, 1633, 1578, 1505, 1333, 1242,
1
1193, 1146, 1007 cmÀ1; H NMR (300 MHz, CDCl3): d=2.20
(s, 3H, Ar-CH3), 2.30 (s, 3H, Ar-CH3), 2.42–2.59 (m, 2H,
CH2C8F17), 3.65–3.72 (m, 2H, NH-CH2), 6.61 (s, 1H, Ar-H),
7.96 (s, 1H, Ar-H), 8.02 (tr, J=5.49 Hz, 1H, NH); 13C NMR
(75 MHz, CDCl3): d=18.7 (Ar-CH3), 20.9 (Ar-CH3), 30.8
(CH2-C8F17), 35.1 (N-CH2), 113.5, 125.5, 127.0, 130.7, 143.1,
147.7 (6ꢄAr-C); 19F NMR (75 MHz, CDCl3): d=À126.6 (m,
2 F, CF2), À123.8 (m, 2 F, CF2), À123.2 (m, 2 F, CF2), À122.4
(m, 4 F, 2ꢄCF2), À122.1 (m, 2 F, CF2), À114.3 (quin, J=
15.3 Hz, 2 F, CH2CF2), À81.2 (tr, J=9.82 Hz, 3 F, CF3); ES-
MS: m/z (%)=613.2 (MH+, 100); HR-MS-EI: m/z=
612.0714 [M]+C, calcd. for C18H13F17N2O2: 612.0706 [delta
À1.38 ppm].
General Procedure for the Immobilization of Flavins
on Fluorous Silica Gel
Flavins (2–5 mg) were dissolved in chloroform (20–30 mL)
in a flask (50 mL) and the appropriate amount of silica gel
(200–500 mg) was added. The solvent was slowly evaporat-
ed. Drying under vacuum gave flavin catalysts on fluorous
silica with a loading of 1% per mass as yellow powders. In
the case of tetraacetylriboflavin (10) which was mainly used
in our experiments, this represents 18 mmol flavin/g silica
gel.
10-(1H,1H,2H,2H-Perfluorodecyl)flavin (4)
Nitro compound 3 was dissolved in methanol (30 mL) and
chloroform (5 mL). Palladium on activated charcoal (28 mg)
was added and the mixture was hydrogenated with 12 bar at
room temperature for 16 h. After filtration, alloxane mono-
hydrate (365 mg, 2.57 mmol) and boric acid (700 mg,
11.3 mmol) were added and the mixture was stirred at room
temperature for three days in the dark. After evaporation of
the solvents, chloroform (300 mL) was added and the organ-
ic phase was washed with water and brine. The organic
phase was dried over magnesium sulfate and evaporated.
The crude product was purified by flash column chromatog-
General Procedure for the Immobilization of Flavins
in Polyethylene Pellets
Flavin (2.5 mg) and polyethylene powder (247.5 mg) were
mixed and brought into a press. A pressure of 125 MPa was
adjusted and the pressing tool was heated to 80–1008C until
the pressure decreased. The apparatus was allowed to cool
down to room temperature and the yellow pellets (1–2 mm
thick, 1.3 cm diameter) were removed. To remove traces of
flavin that was not completely incorporated, the pellets
were washed with water three times.
raphy
(Rf =0.3;
chloroform:ethyl
acetate:methanol,
20:10:3); yield: 118 mg (171 mmol, 38%); yellow solid; mp
3108C (decomp.); IR (ATR): n=3182, 3069, 1724, 1672,
1
1581, 1538, 1509, 1196, 1141, 824 cmÀ1; H NMR (300 MHz,
CDCl3): d=2.47 (s, 3H, Ar-CH3), 2.58 (s, 3H, Ar-CH3),
2.68–2.85 (m, 2H, CH2C8F17), 4.99 (tr, J=7.55 Hz, 2H, N-
CH2), 7.42 (s, 1H, Ar-H), 8.10 (s, 1H, Ar-H), 8.43 (br s, 1H,
NH); 13C NMR: not measured due to low solubility;
19F NMR (75 MHz, CDCl3): d=À126.6 (m, 2 F, CF2),
À123.4 (m, 2 F, CF2), À123.1 (m, 2 F, CF2), À122.3 (m, 4 F,
2ꢄCF2), À122.0 (m, 2 F, CF2), À113.9 (tr, J=13.2 Hz, 2 F,
CH2CF2), À81.2 (tr, J=9.82 Hz, 3 F, CF3); ES-MS: m/z
(%)=689.2 (MH+, 100). HR-MS-EI: m/z=688.0767 [M]+C,
calcd. for C22H13F17N4O2: 688.0767 [delta 0.00 ppm].
General Procedure for Testing of the Photocatalytic
Activity of Flavins
The photocatalytic activity of flavins was tested in solutions
of different benzyl alcohols (0.002M) in D2O (V=1 mL).
Flavins (10 mol%) were added as DMSO-d6 stock solution
(0.01M) for the experiments in homogeneous solution or as
immobilized flavins on silica gel (1% per mass) for hetero-
geneous experiments. The reaction mixture was stirred and
irradiated at room temperature with an array of six LEDs
(440 nm, 5 W) in small glass vials (see Figure 1). The prog-
1
ress of the reaction was monitored by H NMR. Heteroge-
3,10-Bis-(1H,1H,2H,2H-perfluorodecyl)flavin (6)
neous catalysts were removed by filtration before the mea-
surement. The 1H NMR resonance signals of the alcohol
and the aldehyde are well separated and were assigned un-
ambiguously. The clean conversion allows the quantitative
monitoring of the reaction progress by integration of the ar-
omatic resonance signals (see also Supporting Information).
Flavin 4 (68 mg, 100 mmol) was dissolved in dry dimethyl-
formamide (15 mL) and subsequently, potassium carbonate
(72 mg, 521 mmol) and fluorinated iodide
5 (1.30 g,
2.26 mmol) were added and the mixture was stirred for 24 h
at room temperature in the dark. The suspension was dilut-
ed with chloroform and washed with water (5ꢄ100 mL) and
brine (2ꢄ100 mL). The organic phase was dried over mag-
nesium sulfate and the solvents were evaporated. The crude
product was purified by flash column chromatography (Rf =
0.2; dichloromethane:methanol, 100:1); yield: 71 mg (62
mmol, 62%); yellow solid; mp 1558C; IR (ATR): n=1664,
N-(1H,1H,2H,2H-Perfluorodecyl)-4,5-dimethyl-2-
nitroaniline (3)
Dinitro-o-xylene 1 (392 mg, 2.0 mmol) was dissolved in etha-
nol (50 mL). Subsequently, triethylamine (416 mL, 3.0 mmol)
and the fluorinated amine 2 (1.31 g, 2.83 mmol) in ethanol
(3 mL) were added and the mixture was refluxed for one
day. Another portion of amine 2 (300 mg, 648 mmol) was
1585, 1546, 1196, 1144, 656 cmÀ1
;
1H NMR (300 MHz,
CDCl3): d=2.45 (s, 3H, Ar-CH3), 2.46–2.83 (m, 7H, Ar-CH3
+ CH2C8F17), 4.43 (tr, J=7.14 Hz, 2H, N-CH2), 4.96 (tr, J=
Adv. Synth. Catal. 2009, 351, 163 – 174
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
171

Harald Schmaderer et al.
FULL PAPERS
7.00 Hz, 2H, N-CH2), 7.41 (s, 1H, Ar-H), 8.07 (s, 1H, Ar-
H); 13C NMR: not measured due to low solubility; 19F NMR
(75 MHz, CDCl3): d=À126.6 (m, 4 F, 2ꢄCF2), À124.0 (m, 2
F, CF2), À123.5 (m, 2 F, CF2), À123.2 (m, 4 F, 2ꢄCF2),
À122.3 to À122.1 (m, 12 F, 6ꢄCF2), À114.8 (tr, J=12.9 Hz,
2 F, CH2CF2), À113.9 (tr, J=13.5 Hz, 2 F, CH2CF2), À81.3
to À81.2 (m, 6 F, CF3); ES-MS: m/z (%)=1135.2 (MH+,
100); HR-MS-EI: m/z=1134.0722, [M]+C, calcd. for
C32H16F34N4O2: 1134.0730 [delta 0.73 ppm].
2ꢄCF2), À122.2 (m, 2 F, CF2), À114.8 (quin, J=16.6 Hz, 2 F,
CH2CF2), À81.2 (tr, J=10.1 Hz, 3 F, CF3); ES-MS: m/z
(%)=747.3 (MH+, 100), 769.3 (MNa+, 16); HR-MS-EI:
m/z=746.1178 [M]+C, calcd. for C25H19F17N4O3: 746.1186
[delta 1.03 ppm].
3-Methylriboflavin Tetraacetate (11)[16,26]
Riboflavin tetraacetate 10 (1.63 g, 3.0 mmol) was dissolved
in dry dimethylformamide (20 mL) and subsequently, caesi-
um carbonate (1.47 g, 4.50 mmol) and methyl iodide
(1.8 mL, 29.0 mmol) were added. After stirring for 16 h at
room temperature in the dark, water (5 mL) was added and
the solvents were evaporated. The crude product was dis-
solved in chloroform (250 mL) and washed with water (2ꢄ
100 mL) and brine. The organic phase was dried over mag-
nesium sulfate and the solvents were evaporated. Purifica-
tion was done by flash column chromatography (Rf =0.15;
chloroform:methanol, 50:1); yield: 1.19 g (2.13 mmol, 71%);
orange solid; mp 1838C; IR (ATR): n=2920, 1737, 1709,
10-Methoxyethyl-3-methylflavin (8a)[16]
Flavin 7 (306 mg, 1.02 mmol) was dissolved in dry dimethyl-
formamide (40 mL) and subsequently, caesium carbonate
(488 mg, 1.50 mmol) and methyl iodide (1.37 g, 9.64 mmol)
were added and the mixture was stirred for 18 h at room
temperature in the dark. The suspension was diluted with
chloroform and washed with water (3ꢄ100 mL) and brine.
The organic phase was dried over magnesium sulfate and
the solvents were evaporated. The crude product was puri-
fied by flash column chromatography (Rf =0.2; chloroform:-
methanol, 100:1); yield: 293 mg (930 mmol, 91%); yellow
solid; mp 2558C (decomp.). IR (ATR): n=2921, 1703, 1651,
1659, 1532, 1373, 1209, 1034 cmÀ1
;
1H NMR (300 MHz,
CDCl3): d=1.70 (s, 3H, Ac-CH3), 2.05 (s, 3H, Ac-CH3),
2.20 (s, 3H, Ac-CH3), 2.27 (s, 3H, Ac-CH3), 2.40 (s, 3H, Ar-
CH3), 2.52 (s, 3H, Ar-CH3), 3.45 (s, 3H, N-CH3), 4.22 (dd,
J=12.35 Hz, J=5.76 Hz, 1H, CH), 4.40 (dd, J=12.21 Hz,
J=2.61 Hz, 1H, CH), 4.59–5.26 (m, 2H, CH), 5.35–5.45 (m,
2H, CH), 5.62–5.65 (m, 1H, CH), 7.51 (s, 1H, Ar-H), 7.97
(s, 1H, Ar-H); 13C NMR (75 MHz, CDCl3): d=19.5 (Ar-
CH3), 20.4 (Ac-CH3), 20.8 (Ac-CH3), 20.9 (Ac-CH3), 21.1
(Ac-CH3), 21.5 (Ar-CH3), 28.7 (N-CH3), 44.6 (CH2), 61.9
(CH2), 69.0 (CH), 69.4 (CH), 70.4 (CH), 115.4 (C-9), 131.2
(C-9a), 132.9 (C-6), 134.7 (C-5a), 135.6 (C-4a), 136.7 (C-7),
147.6 (C-8), 149.1 (C-10a), 155.3 (C-2), 160.0 (C-4), 169.7
(CO), 169.9 (CO), 170.4 (CO), 170.7 (CO); ES-MS: m/z
(%)=559.2 (MH+, 100); HR-MS-EI: m/z=558.1962 [M]+C,
calcd. for C26H30N4O10: 558.1962 [delta À0.01 ppm].
1582, 1540,1451, 1231, 1110, 1014, 970 cmÀ1
;
1H NMR
(300 MHz, CDCl3): d=2.41 (s, 3H, Ar-CH3), 2.51 (s, 3H,
Ar-CH3), 3.26 (s, 3H, O-CH3), 3.49 (s, 3H, N-CH3), 3.88 (tr,
J=5.21 Hz, 2H, O-CH2), 4.87 (tr, J=5.08 Hz, 2H, N-CH2),
7.62 (s, 1H, Ar-H), 7.99 (s, 1H, Ar-H); 13C NMR (75 MHz,
CDCl3): d=19.6 (Ar-CH3), 21.6 (Ar-CH3), 45.4 (N-CH2),
59.3 (O-CH3), 69.6 (O-CH2), 116.7 (C-9), 132.2 (C-9a), 132.3
(C-6), 135.0 (C-5a), 135.5 (C-4a), 136.7 (C-7), 147.6 (C-8),
148.7 (C-10a), 156.0 (C-2), 160.2 (C-4); ES-MS: m/z (%)=
315.0 (MH+, 100); HR-MS-EI: m/z=314.1374 [M]+C, calcd.
for C16H18N4O3: 314.1379 [delta 1.56 ppm].
3-(1H,1H,2H,2H-Perfluorodecyl)-10-methoxyethyl-
flavin (8b)
Flavin 7 (109 mg, 363 mmol) was dissolved in dry dimethyl-
formamide (15 mL) and potassium carbonate (251 mg,
1.81 mmol) was added. After stirring for 20 min, fluorinated
iodide 5 (625 mg, 1.09 mmol) in dry dimethylformamide
(5 mL) was added. After one day, another portion of iodide
5 (417 mg, 726 mmol) was added and the mixture was stirred
for three days at room temperature in the dark. The mixture
was diluted with chloroform and washed with water (3ꢄ
100 mL) and brine (100 mL). The organic phase was dried
over magnesium sulfate and the solvents were evaporated.
The crude brown solid was purified by flash column chroma-
tography (Rf =0.2; chloroform:methanol, 20:1); yield: 63 mg
(84 mmol, 23%); yellow solid; mp 1758C; IR (ATR): n=
2952, 1708, 1663, 1585, 1552, 1197, 1144, 1111, 1003, 958,
3-Methylriboflavin (12)[16,27]
3-Methylriboflavin tetraacetate 11 (280 mg, 501 mmol) was
dissolved in ethanol (50 mL) and p-toluenesulfonic acid
(98 mg, 569 mmol) was added. After refluxing for 17 h, an-
other portion of p-toluenesulfonic acid (49 mg, 285 mmol)
was added and the mixture was refluxed for 3 h. After cool-
ing to room temperature, the solution was stored in the re-
frigerator overnight and a yellow solid precipitated that was
filtered off and dried; yield: 142 mg (364 mmol, 73%);
yellow solid; mp 2758C (decomp.); IR (ATR): n=3230,
1716, 1617, 1579, 1532, 1235 cmÀ1
;
1H NMR (300 MHz,
DMSO-d6): d=2.38 (s, 3H, Ar-CH3), 2.47 (s, 3H, Ar-CH3),
3.27 (s, 3H, N-CH3), 3.44–3.48 (m, 1H, OH), 3.64 (br s, 3H,
3ꢄOH), 4.23–4.26 (m, 1H, CH), 4.51 (tr, J=5.63 Hz, 1H,
CH), 4.58–4.62 (m, 1H, CH), 4.77 (d, J=5.49 Hz, 1H,
CH),4.88–5.00 (m, 2H, CH), 5.13 (d, J=4.67 Hz, 1H, CH),
7.89 (s, 1H, Ar-H), 7.91 (s, 1H, Ar-H); 13C NMR (75 MHz,
DMSO-d6): d=18.8 (Ar-CH3), 20.8 (Ar-CH3), 28.0 (N-CH3),
47.1 (CH2), 63.4 (CH2), 68.8 (CH), 72.8 (CH), 73.6 (CH),
117.4 (C-9), 130.7 (C-6), 132.0 (C-9a), 134.2 (C-5a), 135.7
(C-4a), 135.9 (C-7), 146.2 (C-8), 149.3 (C-10a), 155.0 (C-2),
159.7 (C-4); ES-MS: m/z (%)=391.0 (MH+, 100); HR-MS-
LSI: m/z= 391.1621 [M + H]+, calcd. for C18H23N4O6:
391.1618 [delta À0.87 ppm].
1
719, 657 cmÀ1. H NMR (300 MHz, CDCl3): d=2.44 (s, 3H,
Ar-CH3), 2.49–2.67 (m, 5H, Ar-CH3 + 3-N-CH2), 3.29 (s,
3H, O-CH3), 3.91 (tr, J=5.08 Hz, 2H, 10-N-CH2), 4.45 (tr,
J=7.55 Hz, 2H, CH2C8F17), 4.89 (tr, J=5.21 Hz, 2H, O-
CH2), 7.67 (s, 1H, Ar-H), 8.03 (s, 1H, Ar-H); 13C NMR
(75 MHz, CDCl3): d=19.6 (Ar-CH3), 21.7 (Ar-CH3), 29.8
(CH2C8F17), 34.3 (N-CH2), 45.6 (N-CH2), 59.4 (O-CH3), 69.6
(O-CH2), 116.8 (C-9), 132.4 (C-9a, C-6, C-5a), 135.3 (C-4a),
137.0 (C-7), 148.1 (C-8), 148.9 (C-10a), 155.1 (C-2), 159.9
(C-4); 19F NMR (75 MHz, CDCl3): d=À126.6 (m, 2 F, CF2),
À124.0 (m, 2 F, CF2), À123.2 (m, 2 F, CF2), À122.4 (m, 4 F,
172
asc.wiley-vch.de
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 163 – 174

Photooxidation of Benzyl Alcohols with Immobilized Flavins
FULL PAPERS
Goodman, V. M. Rotello, M. L. Waters, Angew. Chem.
2004, 116, 742–745; Angew. Chem. Int. Ed. 2004, 43,
724–727 ; e) G. Cooke, Angew. Chem. 2003, 115, 5008–
5018; Angew. Chem. Int. Ed. 2003, 42, 4860–4870; f) F.
Guo, B. H. Chang, C. J. Rizzo, Bioorg. Med. Chem.
Lett. 2002, 12, 151–154; g) C. Behrens, M. Ober, T.
Carell, Eur. J. Org. Chem. 2002, 3281–3289; h) J. Bute-
nandt, R. Epple, E.-U. Wallenborn, A. P. M. Eker, V.
Gramlich, T. Carell, Chem. Eur. J. 2000, 6, 62–72;
i) V. M. Rotello, Curr. Opin. Chem. Biol. 1999, 3, 747–
751; j) R. Deans, V. M. Rotello, J. Org. Chem. 1997, 62,
4528–4529; k) E. Breinlinger, A. Niemz, V. M. Rotello,
J. Am. Chem. Soc. 1995, 117, 5379–5380.
3-Methyltetrapalmitylriboflavin (13)
3-Methylriboflavin 12 (280 mg, 501 mmol) was suspended in
a mixture of dry chloroform (15 mL) and pyridine (15 mL).
A solution of palmityl chloride (1.55 mL, 5.13 mmol) in dry
chloroform (5 mL) was added dropwise within 1 h at 08C.
Afterwards, the mixture was stirred for 24 h at room temper-
ature. After addition of water (5 mL), the suspension was
heated to 608C for 1 h and the solvents were evaporated af-
terwards. The crude product was dissolved in chloroform
(30 mL) and was washed with sodium hydrogen carbonate
solution (2ꢄ100 mL) and brine (2ꢄ100 mL). The organic
phase was dried over magnesium sulfate and the solvents
were evaporated. The orange solid was purified by flash
column chromatography (Rf =0.2; petrol ether:ethyl acetate,
2:1); yield: 173 mg (129 mmol, 52%); orange solid; mp 50–
538C; IR (ATR): n=2917, 2850, 1741, 1664, 1584, 1545,
[4] a) A. A. Lindꢅn, M. Johansson, N. Hermanns, J.-E.
Bꢀckvall, J. Org. Chem. 2006, 71, 3849–3853; b) A. A.
Lindꢅn, N. Hermanns, S. Ott, L. Krꢂger, J.-E. Bꢀckvall,
Chem. Eur. J. 2005, 11, 112–119; c) Y. Imada, H. Iida,
S.-I. Murahashi, T. Naota, Angew. Chem. 2005, 117,
1732–1734; Angew. Chem. Int. Ed. 2005, 44, 1704–
1706; d) R. Cibulka, R. Vasold, B. Kçnig, Chem. Eur. J.
2004, 10, 6223–6231; e) Y. Imada, H. Iida, S. Ono, S.-I.
Murahashi, J. Am. Chem. Soc. 2003, 125, 2868–2869;
f) M. J. H. Moonen, M. W. Fraaije, I. M. C. M. Rietjens,
C. Laane, W. J. H. van Berkel, Adv. Synth. Catal. 2002,
344, 1023–1035; g) S.-I. Murahashi, S. Ono, Y. Imada,
Angew. Chem. 2002, 114, 2472–2474; Angew. Chem.
Int. Ed. 2002, 41, 2366–2368; h) S.-I. Murahashi, T.
Oda, Y. Masui, J. Am. Chem. Soc. 1989, 111, 5002–
5003; i) S. Shinkai, Y.-I. Ishikawa, O. Manabe, Chem.
Lett. 1982, 11, 809–812.
1
1466, 1151, 721 cmÀ1; H NMR (300 MHz, CDCl3): d=0.86–
2.53 (m, 130H, 2ꢄAr-CH3 + 4ꢄC15H31), 3.48 (s, 3H, N-
CH3), 4.17–4.23 (m, 1H, CH), 4.43–4.48 (m, 1H, CH), 4.92
(br s, 2H, CH), 5.39–5.49 (m, 2H, CH), 5.67–5.69 (m, 1H,
CH), 7.54 (s, 1H, Ar-H), 8.02 (s, 1H, Ar-H); 13C NMR
(75 MHz, CDCl3): d=14.2 (CH3), 19.5 (Ar-CH3), 21.5 (Ar-
CH3), 28.8 (N-CH3), 44.6 (CH2), 61.9 (CH2), 69.1 (CH), 70.4
(2ꢄCH), 115.6 (C-9), 131.4 (C-9a), 133.0 (C-6), 134.7 (C-
5a), 135.7 (C-4a), 136.5 (C-7), 147.4 (C-8), 149.2 (C-10a),
155.3 (C-2), 160.0 (C-4), 172.5 (CO), 172.6 (CO), 173.1
(CO), 173.5 (CO), C15H31 (signals not assigned); ES-MS:
m/z (%)=1344.4 (MH+, 100).
[5] J. Piera, J.-E. Bꢀckvall, Angew. Chem. 2008, 120, 3558–
3576; Angew. Chem. Int. Ed. 2008, 47, 3506–3523.
[6] J. Svoboda, H. Schmaderer, B. Kçnig, Chem. Eur. J.
2008, 14, 1854–1865.
Acknowledgements
This work was supported by the Deutsche Forschungsgemein-
schaft through a graduate fellowship to H.S. (GRK 640) and
the program “Eigene Stelle” to P.H.
[7] A. Closson, M. Johansson, J.-E. Bꢀckvall, Chem.
Commun. 2004, 1494–1495.
[8] Leading reviews: a) D. E. DeVos, B. F. Sels, P. A.
Jacobs, Adv. Catal. 2002, 46, 1–87; b) J. A. Gladysz,
Pure Appl. Chem. 2001, 73, 1319–1324; c) G. Oehme,
in: Comprehensive Asymmetric Catalysis I-III, (Eds.:
E. N. Jacobsen, A. Pfaltz, H. Yamamoto), Springer-
Verlag, Berlin, Germany, 1999; Vol. 3, pp 1377–1386.
[9] Leading reviews: a) J. A. Gladysz, D. P. Curran, Tetra-
hedron 2002, 58, 3823–3825; b) C. C. Tzschucke, C.
Markert, W. Bannwarth, S. Roller, A. Hebel, R. Haag,
Angew. Chem. 2002, 114, 4136–4173; Angew. Chem.
Int. Ed. 2002, 41, 3964–4000; c) D. P. Curran, in: Hand-
book of Fluorous Chemistry. (Eds.: J. A. Gladysz, D. P.
Curran, I. T. Horvath), Wiley-VCH, Weinheim, 2004,
pp 101–127.
[10] Leading examples: a) L. V. Dinh, J. A. Gladysz, Angew.
Chem. 2005, 117, 4164–4167; Angew. Chem. Int. Ed.
2005, 44, 4095–4097; b) M. Matsugi, D. P. Curran, J.
Org. Chem. 2005, 70, 1636–1642; c) M. Contel, C.
Izuel, M. Laguna, P. R. Villuendas, P. J. Alonso, R. H.
Fisch, Chem. Eur. J. 2003, 9, 4168–4178; d) C. C.
Tzschucke, C. Markert, H. Glatz, W. Bannwarth,
Angew. Chem. 2002, 114, 4678–4681; Angew. Chem.
Int. Ed. 2002, 41, 4500–4503.
References
[1] a) P. Hemmerich, Chem. Org. Nat. Prod. 1976, 33, 451–
525; b) S. Gishla, V. Massey, Eur. J. Biochem. 1989,
181, 1–17.
[2] a) S. O. Mansoorabadi, C. J. Thibodeaux, H. Liu, J.
Org. Chem. 2007, 72, 6329–6342; b) V. Massey, Bio-
chem. Soc. Trans. 2000, 28, 283–296; c) Chemistry and
Biochemistry of Flavoenzymes, (Ed.: F. Mꢂller), CRC,
Boca Raton, 1991; d) B. J. Fritz, S. Kasai, K. Matsui,
Photochem. Photobiol. 1987, 45, 113–117; e) A. Bowd,
P. Byrom, J. B. Hudson, J. H. Turnbull, Photochem.
Photobiol. 1968, 8, 1–10; f) B. Kçnig, M. Pelka, H.
Zieg, T. Ritter, H. Bouas-Laurent, R. Bonneau, J.-P.
Desvergne, J. Am. Chem. Soc. 1999, 121, 1681–1687.
[3] a) B. J. Jordan, G. Cooke, J. F. Garety, M. A. Pollier, N.
Kryvokhyzha, A. Bayir, G. Rabani, V. M. Rotello,
Chem. Commun. 2007, 1248–1250; b) J. B. Carroll, B. J.
Jordan, H. Xu, B. Erdogan, L. Lee, L. Cheng, C. Tiern-
an, G. Cooke, V. M. Rotello, Org. Lett. 2005, 7, 2551–
2554; c) M. Gray, A. J. Goodmann, J. B. Carroll, K.
Bardon, M. Markey, G. Cooke, V. M. Rotello, Org.
Lett. 2004, 6, 385–388; d) S. M. Butterfield, C. M.
[11] Leading examples: a) L. Zu, J. Wang, H. Li, W. Wang,
W. Org. Lett. 2006, 8, 3077–3079; b) Z. Dalicsek, F.
Adv. Synth. Catal. 2009, 351, 163 – 174
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
173

Harald Schmaderer et al.
FULL PAPERS
Pollreisz, ꢆ. Gçmçry, T. Soꢇs, Org. Lett. 2005, 7, 3243–
3246.
[21] a) E. Baciocchi, S. Belvedere, M. Bietti, O. Lanzalunga,
Eur. J. Org. Chem. 1998, 299–302; b) M. Bietti, E. Ba-
ciocchi, S. Steenken, J. Phys. Chem. A 1998, 102, 7337–
7342; c) M. Fabbrini, C. Galli, P. Gentili, J. Mol. Cat.
B: 2002, 18, 169–171; d) P. Baiocco, A. M. Barreca, M.
Fabbrini, C. Galli, P. Gentili, Org. Biomol. Chem. 2003,
1, 191–197.
[22] Spartan program package; method B3LYP 6–31**.
[23] The fluorinated glass plates were a generous gift from
Fluorous Technologies Inc., Pittsburgh, USA.
[24] Under these experimental conditions with long reaction
times, a slow background reaction was detectable
which leads to conversions of 2–3% per hour.
[25] Commercially available cyanacrylate based superglue
from Henkel Loctite, Munich, Germany.
[26] a) S. Shinkai, T. Yamashita, Y. Kusano, O. Manabe, J.
Org. Chem. 1980, 45, 4947–4952; b) M. Insinska-Rak,
E. Sikorska, J. L. Bourdelande, I. V. Khmelinskii, W.
Prukała, K. Dobek, J. Karolczak, I. F. Machado, L. F. V.
Ferreira, E. Dulewicz, A. Komasa, D. R. Worrall, M.
Kubicki, M. Sikorski, J. Photochem. Photobiol. A:
Chem. 2007, 186, 14–23.
[12] R. Kuhn, F. Weygand, Ber. dtsch. chem. Ges. 1935, 68,
1282–1288, and references cited therein.
[13] a) A. Monge, J. A. Palop, A. Lꢇpez de Cerꢈin, V. Sena-
dor, F. J. Martinez-Crespo, Y. Sainz, S. Narro, E.
Garcia, C. de Miguel, M. Gonzꢈlez, E. Hamilton, A. J.
Barker, E. D. Clarke, D. T. Greenhow, J. Med. Chem.
1995, 38, 1786–1792; b) T. Sugaya, K. Nobuyuki, A. Sa-
kaguchi, S. Tomioka, Synthesis 1995, 1257–1262;
c) R. R. Holmes, R. P. Bayer, J. Am. Chem. Soc. 1960,
82, 3454–3456.
[14] a) F. Szçnyi, A. Cambon, J. Fluorine Chem. 1989, 42,
59–68; b) F. Szçnyi, F. Guennouni, A. Cambon, J. Flu-
orine Chem. 1991, 55, 85–92.
[15] D. B. McCormick, J. Heterocycl. Chem. 1970, 7, 447–
450.
[16] D. B. McCormick, J. Heterocycl. Chem. 1974, 11, 969–
974.
[17] S. Fukuzumi, T. Goto, K. Ishikawa, T. Tanaka, Chem.
Lett. 1988, 11, 1923–1926.
[18] T. C. Bruice, Acc. Chem. Res. 1980, 13, 256–262.
[19] S. Fukuzumi, K. Tanii, T. Tanaka, J. Chem. Soc. Chem.
Commun. 1989, 816–818.
[27] E. L. Loechler, T. C. Hollocher, J. Am. Chem. Soc.
1980, 102, 7312–7321.
[20] W. Tong, H. Ye, H. Zhu, V. T. DꢉSouza, J. Mol. Struct.
1995, 333, 19–27.
174
asc.wiley-vch.de
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 163 – 174