Catalytic effect of alloxazinium and isoalloxazinium salts on oxidation of sulfides with hydrogen peroxide in micellar media

Article abstract of DOI:10.1135/cccc2009030

Three novel amphiphilic alloxazinium salts were prepared: 3-dodecyl-5-ethyl-7,8,10-trimethylisoalloxazinium perchlorate (1c), 1-dodecyl-5-ethyl-3-methylalloxazinium perchlorate (2b), and 3-dodecyl-5-ethyl-l-methylalloxazinium perchlorate (2c). Their catalytic activity in thioanisole (3) oxidation with hydrogen peroxide was investigated in micelles of sodium dodecylsulfate (SDS), hexadecyltrimethylammonium nitrate (CTANO₃) and Brij 35. Reaction rates were strongly dependent on the catalyst structure, on the type of micelles, and on pH value. Alloxazinium salts 2 were more effective catalysts than isoalloxazinium salts 1. Due to the contribution of micellar catalysis, the v_cat/v_o ratio of the catalyzed and non-catalyzed reaction rates was almost 80 with salt 2b solubilized in CTANO³ micelles. Nevertheless, the highest acceleration was observed with non-amphiphilic 5-ethyl-1,3-dimethylalloxazinium perchlorate (2a) in CTANO₃ micelles (v_cat/v_o = 134). In this case, salt 2a presumably acts as a phase-transfer catalyst bringing hydrogen peroxide from the aqueous phase into the micelle interior. Synthetic applicability of the investigated catalytic systems was verified on semipreparative scale.

Full text of DOI:10.1135/cccc2009030

Catalytic Effect of Alloxazinium and Isoalloxazinium Salts
973
CATALYTIC EFFECT OF ALLOXAZINIUM AND ISOALLOXAZINIUM
SALTS ON OXIDATION OF SULFIDES WITH HYDROGEN PEROXIDE
IN MICELLAR MEDIA
b
a3
Radek CIBULKAa1,*, Lenka BAXOVÁa2, Hana DVOŘÁKOVÁ , František HAMPL
,
Petra MÉNOVÁa4, Viktor MOJRa5, Baptiste PLANCQ and Serkan SAYIN
c
d
a
Department of Organic Chemistry, Institute of Chemical Technology, Prague,
Technická 5, 166 28 Prague 6, Czech Republic; e-mail: cibulkar@vscht.cz,
lenka.baxa@seznam.cz, hamplf@vscht.cz, petra.menova@seznam.cz, viktor.mojr@vscht.cz
Central Laboratories, Institute of Chemical Technology, Prague,
Technická 5, 166 28 Prague 6, Czech Republic; e-mail: hana.dvorakova@vscht.cz
Département de chimie, Pavillon Alexandre Vachon, Université Laval,
Québec, G1V 0A6, Canada; e-mail: baptiste.plancq.1@ulaval.ca
Department of Chemistry, Afyon Kocatepe University,
1
2
3
4
5
b
c
d
03200 Afyonkarahisar, Turkey; e-mail: pretty_ssayin@mynet.com
Received March 9, 2009
Accepted April 21, 2009
Publish ed on lin e Jun e 23, 2009
Dedicated to Dr. Alfred Bader on the occasion of his 85th birthday in recognition of his outstanding
contributions to the science of chemistry.
Th ree n ovel am ph iph ilic alloxazin ium salts were prepared: 3-dodecyl-5-eth yl-7,8,10-tri-
m eth ylisoalloxazin ium perch lorate (1c), 1-dodecyl-5-eth yl-3-m eth ylalloxazin ium perch lorate
(2b), an d 3-dodecyl-5-eth yl-1-m eth ylalloxazin ium perch lorate (2c). Th eir catalytic activity in
th ioan isole (3) oxidation with h ydrogen peroxide was in vestigated in m icelles of sodium
dodecylsulfate (SDS), h exadecyltrim eth ylam m on ium n itrate (CTANO3) an d Brij 35. Reaction
rates were stron gly depen den t on th e catalyst structure, on th e type of m icelles, an d on pH
value. Alloxazin ium salts 2 were m ore effective catalysts th an isoalloxazin ium salts 1. Due to
th e con tribution of m icellar catalysis, th e vcat/v0 ratio of th e catalyzed an d n on -catalyzed re-
action rates was alm ost 80 with salt 2b solubilized in CTANO3 m icelles. Neverth eless, th e
h igh est acceleration was observed with n on -am ph iph ilic 5-eth yl-1,3-dim eth ylalloxazin ium
perch lorate (2a) in CTANO3 m icelles (vcat/v0 = 134). In th is case, salt 2a presum ably acts as a
ph ase-tran sfer catalyst brin gin g h ydrogen peroxide from th e aqueous ph ase in to th e m icelle
in terior. Syn th etic applicability of th e in vestigated catalytic system s was verified on sem i-
preparative scale.
Keyw ord s: Sulfoxidation ; Micelles; Micellar catalysis; Organ ocatalysis; Green ch em istry;
Flavin s; Flavin ium salts; Alloxazin es; Isoalloxazin es; N-Alkylalloxan s; Oxidation s; Surfac-
tan ts; Ph ase tran sfer catalysis.
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 6, pp. 973–993
© 2009 Institute of Organic Chemistry and Biochemistry
doi:10.1135/cccc2009030

974
Cibulka et al.:
Flavin ium salts h ave been reported in m an y cases as efficien t catalysts for
oxidation of various substrates, usin g h ydrogen peroxide as a term in al oxi-
dan t1–8. Flavin ium salts are usually con sidered as flavoen zym e m im ics
wh ich can tran sform , e.g. sulfides to sulfoxides, tertiary am in es to N-oxides,
an d keton es to esters (Baeyer–Villiger oxidation ). In th e cited works, th e
term “flavin ium salts” h as been used for both isoalloxazin ium salts 1 substi-
tuted in position 10 (sim ilarly as n atural riboflavin ) an d isom eric alloxazin -
ium salts 2. Flavin ium salts as oxidation catalysts h ave several advan tages: (i)
th ey work un der m ild con dition s – th e reaction s usually proceed sm ooth ly
at room tem perature in m eth an ol1–7 or in m icellar system s8 in th e presen ce
of less th an 2 m ole % of th e catalyst (relative to th e substrate) with turn -
over n um ber (TON) up to 200 (ref.5a); (ii) th ey exh ibit h igh
ch em oselectivity in th e oxidation of sulfides producin g exclusively
sulfoxides with out overoxidation to sulfon es; (iii) th e oxidation s can be
perform ed even in aqueous solution s8; th is fact ran ks flavin ium salts
am on g th e en viron m en t-frien dly organ ocatalysts. In several in stan ces,
ch iral flavin ium salts affordin g sulfoxides6 or esters7 with m oderate en an -
tioselectivity h ave also been described.
Mech an ism of th e oxidation s with h ydrogen peroxide catalyzed by
flavin ium salts is depicted in Sch em e 1. Addition of h ydrogen peroxide to
flavin ium salts 1 or 2 affords flavin -4a-h ydroperoxides (1-OOH or 2-OOH,
SCHEME 1
Catalytic action of isoalloxazin ium 1 an d alloxazin ium 2 salts in oxidation s of substrates S
with h ydrogen peroxide
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 6, pp. 973–993

Catalytic Effect of Alloxazinium and Isoalloxazinium Salts
975
respectively) wh ich are th e agen ts oxidizin g th e substrate1,9. 5-Eth yliso-
alloxazin e-4a-h ydroperoxide (1-OOH) is an an alogue of n atural riboflavin -
4a-h ydroperoxide9 (R3 = R5 = H, R7 = R8 = CH3, R10 = ribityl) actin g as a ver-
satile oxidizin g cofactor in flavin m on ooxygen ases10. Isom eric alloxazin e-
4a-h ydroperoxide (2-OOH) h as n o n aturally occurrin g equivalen t.
Am on g th e above-m en tion ed oxidation s, sulfoxidation usually serves as
a m odel reaction for testin g th e flavoen zym e m im ics efficien cy1. It is
n ecessary to em ph asize th at in sulfoxidation s perform ed in th e presen ce
of flavin ium salts, a n on -catalyzed oxidation proceeds to som e exten t as
well1,2,5,6 (Sch em e 2). Th e n on -catalyzed reaction represen ts a con siderable
problem , especially in oxidation s catalyzed by ch iral flavin ium salts. Most
likely, it is respon sible for th e reported6 m oderate stereoselectivity in oxida-
tion s of sulfides to sulfoxides. Th e con tribution of th e n on -catalyzed oxida-
tion to th e overall process varies depen din g on th e structure of flavin ium
salt an d on reaction con dition s5, h owever, n o ration ale h as been given yet.
Th erefore, th e fin din g wh ich factors m ay favour catalyzed sulfoxidation
over th e n on -catalyzed reaction seem s to be a n ecessary prerequisite for th e
developm en t of stereoselective organ ocatalysts based on flavin ium salts.
SCHEME 2
Overall sch em e of sulfoxidation with h ydrogen peroxide in th e presen ce of flavin ium salts; vcat
an d v0 are rates of th e catalyzed an d n on -catalyzed reaction , respectively
Especially in th e sulfoxidation s perform ed in th e presen ce of isoallox-
azin ium salts 1 in h om ogen eous solution s, th e con tribution of th e n on -
catalyzed reaction is relatively h igh 5a,8. Neverth eless, th e n on -catalyzed
reaction can be successfully suppressed. Recen tly, we h ave reported8 on th e
positive effect of m icelles on th e vcat/v0 ratio of th e catalyzed an d n on -
catalyzed reaction rates in th e case of th ioan isole (3) oxidation catalyzed by
am ph iph ilic isoalloxazin ium salt 10-dodecyl-5-eth yl-3,7,8-trim eth yliso-
alloxazin ium perch lorate (1b) (Table I). Th e in creased vcat/v0 ratio in m icel-
les com pared to h om ogen eous solution s results from th e effect of m icellar
catalysis an d from th e suppression of th e n on -catalyzed process due to th e
separation of th e lipoph ilic substrate from th e aqueous ph ase con tain in g
h ydrogen peroxide. Th is effect stron gly depen ds on pH an d on th e n an o-
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 6, pp. 973–993

976
Cibulka et al.:
TABLE I
Relative rates vcat/v0 of th ioan isole (3) oxidation with h ydrogen peroxide catalyzed by
flavin ium salts 1 (ref.8)
Reaction m edium
Catalyst
vcat/v0
pHa
Meth an ol–water (50:50 v/v)
An ion ic m icelles (SDSb)
1a
1b
1b
1b
6.6
24.7
38.9
16.6
2.4
4.4
2.4
3.4
c
Cation ic m icelles (CTANO3
)
Non -ion ic m icelles (Brij 35d)
a
b
c
Optim um pH value.
Sodium dodecylsulfate. Hexadecyltrim eth ylam m on ium n itrate.
d
Poly(eth ylen e glycol) dodecyl eth er; n ~ 23.
aggregate type: it is especially pron oun ced in cation ic m icelles (CTANO3)
wh ere th e vcat/v0 ratio is alm ost 40.
Bäckvall5a reported th at in con trast to isoalloxazin ium salt 1a, alloxazin -
ium salt 2a stron gly favours catalyzed oxidation of various sulfides over th e
n on -catalyzed reaction in h om ogen eous solution s. Th e observed vcat/v0 val-
ues were com parable with th ose obtain ed in reaction s catalyzed by th e
m ost efficien t am ph iph ilic isoalloxazin ium salt 1b in cation ic m icelles. Th is
fact in spired us to in vestigate sulfoxidation catalyzed by am ph iph ilic
alloxazin ium salts 2 in m icelles to com bin e both th e effect of th e catalyst
structure an d th e con tribution of m icellar catalysis. It is kn own 11 th at th e
reactivity of am ph iph ilic com poun ds in m icelles depen ds m arkedly on th e
relative position of th e polar h ead group, h ydroph obic alkyl ch ain , an d th e
reactive fun ction . Th erefore, we decided to prepare two isom eric am ph i-
ph ilic alloxazin ium salts 2b an d 2c. To exten d th e structure–activity study
of am ph iph ilic flavin ium salts, a n ew isoalloxazin ium salt 1c was in volved
in th is study as well.
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 6, pp. 973–993

Catalytic Effect of Alloxazinium and Isoalloxazinium Salts
977
RESULTS AND DISCUSSIONS
Preparation of Flavinium Catalysts
Preparation of isoalloxazin ium catalysts 1a an d 1b was described in our
previous paper8. Am ph iph ilic 3-dodecylisoalloxazin ium salt 1c was pre-
pared startin g from lum iflavin 4 (syn th esis of 4 was publish ed elsewh ere12),
wh ich was alkylated with dodecyl iodide in th e presen ce of potassium
carbon ate. 3-Dodecyl-7,8,10-trim eth ylisoalloxazin e (5) th us obtain ed was
con verted to th e fin al salt 1c by th e reaction with acetaldeh yde un der
reductive con dition s followed by oxidation of th e resultin g 5-eth yl-
1,5-dih ydro derivative with sodium n itrite in perch loric acid (Sch em e 3).
SCHEME 3
Preparation of isoalloxazin ium salt 1c
Alloxazin e 2a was prepared accordin g to th e procedure described by
Murah ash i5e. Syn th esis of lipoph ilic alloxazin ium salts 2b an d 2c is de-
picted in Sch em e 4. Ben zen e-1,2-diam in e prepared by h ydrogen ation of
1,2-din itroben zen e was con den sed with N-alkylalloxan s 6. Th e con den sa-
tion proceeds with excellen t regioselectivity yieldin g 3-alkylalloxazin es 7 as
th e sole products. Th eir alkylation with appropriate alkyl iodides afforded
1,3-dialkylated alloxazin es 8 wh ich were con verted to 5-eth ylalloxazin ium
salts 2b an d 2c usin g th e above-described m eth od (i.e. reaction with acet-
aldeh yde un der reductive con dition s followed by oxidation ). N-Meth yl-
(6a) an d N-dodecylalloxan (6b) were obtain ed by th e oxidation of corre-
spon din g N-alkylbarbituric acids with ch rom ium (VI) oxide usin g a m odi-
fied procedure described for th e preparation of un substituted alloxan 13
.
Th e described syn th esis of alloxazin es 8 gave better results th an a procedure
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 6, pp. 973–993

978
Cibulka et al.:
in volvin g con secutive alkylation of un substituted alloxazin e 9 wh ich leads
to a m ixture of alkylated alloxazin es (Sch em e 5).
SCHEME 4
Preparation of am ph iph ilic alloxazin ium salts 2b an d 2c
SCHEME 5
Alkylation of alloxazin e 9
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 6, pp. 973–993

Catalytic Effect of Alloxazinium and Isoalloxazinium Salts
979
Sulfoxidations Catalyzed by Amphiphilic Flavinium Salts
Catalytic activity of th e n ewly prepared am ph iph ilic flavin ium salts 1c, 2b
an d 2c in oxidation of sulfides with h ydrogen peroxide was in vestigated
un der th e con dition s used in our previous study8 perform ed with salt 1b.
Th ioan isole (3) was oxidized with 1.5 equivalen ts of h ydrogen peroxide in
an ion ic (SDS), cation ic (CTANO3) an d n on -ion ic (Brij 35) m icelles in th e
presen ce of 1.5 m ole % of flavin ium salt (relative to th e substrate) at 25 °C
(for details, see Experim en tal). Oxidation s in h om ogen eous solution s
(water/m eth an ol 50:50, v/v) catalyzed by h ydroph ilic flavin ium salts 1a
an d 2a were perform ed as a referen ce.
In all cases, th e efficien cy of th e catalysts was evaluated as th e vcat/v0 ratio
of th e catalyzed an d n on -catalyzed reaction rates. Th e rate vcat of th e cata-
lyzed reaction was calculated from th e observed reaction rate vobs an d th e
rate v0 of th e n on -catalyzed reaction (from Sch em e 2, vcat = vobs – v0). To
determ in e th e con tribution of blan k (v0), we m easured th e rate of th io-
an isole (3) oxidation with h ydrogen peroxide in th e absen ce of flavin ium
salt in th e correspon din g reaction m edia in in depen den t experim en ts. Th e
in itial rates of th e respective reaction s (up to 20% con version of th e sub-
strate) were used as th e reaction rates vobs an d v0. Th e obtain ed data allow a
com parison of th e reaction rate of th e catalyzed reaction (followin g at low
con version s th e first-order kin etics) an d th e n on -catalyzed reaction (follow-
in g th e secon d order kin etics). Despite th e fact th at th is approach is n ot
strictly correct it h as been widely used to evaluate th e catalytic activity of
flavin ium salts in m an y previous studies4,5a,5b,8. Th e obtain ed kin etic data
(vcat/v0) at various pH values in m icellar m edia as well as in h om ogen eous
solution s are sum m arized in Fig. 1.
Th e results obtain ed with am ph iph ilic alloxazin ium salts 2b an d 2c cor-
respon d qualitatively with th ose obtain ed with isoalloxazin ium derivative
1b (ref.8). Th e observed sim ilarity en sues from th e fact th at th e m ech an ism
of sulfoxidation catalyzed by both types of flavin ium salts is th e sam e
(Sch em e 1). Th e h igh est acceleration of th e oxidation catalyzed by salts 1
an d 2 was observed in cation ic m icelles. Th e efficien cy of both types of
flavin ium salts is pH-depen den t. All th e relative rate vcat/v0 versus pH pro-
files exh ibit a distin ct m axim um with th e on ly exception of th e reaction s
in SDS catalyzed by alloxazin ium derivatives 2 (ref.14). Th e explan ation of
th e “bell-sh aped” profile observed in all cases lies in an opposite effect of
proton on two steps of th e oxidation catalytic cycle (Sch em e 1). Wh ile th e
decom position of 4a-h ydroxyflavin s 1-OH or 2-OH to flavin ium salts 1 or 2
requires proton ation of th e h ydroxy group in pseudobases 1-OH or 2-OH,
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 6, pp. 973–993

980
Cibulka et al.:
an d, con sequen tly, it is facilitated at lower pH values, th e form ation of
flavin -4a-h ydroperoxides 1-OOH or 2-OOH from salts 1 or 2 proceeds with
a loss of proton th us bein g disfavoured un der th e sam e con dition s15. Th e
optim um pH value depen ds also on th e aggregate type. In an ion ic m icelles,
th e m axim um relative rates vcat/v0 occur at h igh er pH values wh ile in
cation ic m icelles th ey are sh ifted to lower pH values.
FIG. 1
pH depen den ce of th e vcat/v0 ratio of in itial rates of catalyzed an d n on -catalyzed oxidation of 3
with h ydrogen peroxide in m icelles of SDS (a), CTANO3 (b), Brij 35 (c) an d in h om ogen eous
solution s (m eth an ol/water 50:50, v/v) (d). Isoalloxazinium salts 1a (ᮀ), 1b (᭺), 1c (᭝) and
alloxazinium salts 2a (᭿), 2b (᭹), 2c (ꢀ) were used as catalysts. Con dition s: c(3) = 1.5 × 10–2
m ol l–1, c(H2O2) = 2.25 × 10–2 m ol l–1, c(1 or 2) = 2.3 × 10–4 m ol l–1, c(buffer) = 1.0 × 10–1 m ol l–1
c(surfactan t) = 1.0 × 10–1 m ol l–1; t = 25 °C. Th e lin es h elp th e eyes to follow th e data poin ts
,
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 6, pp. 973–993

Catalytic Effect of Alloxazinium and Isoalloxazinium Salts
981
Neverth eless, th ere are im portan t quan titative differen ces in th e activity
of isoalloxazin ium 1 an d alloxazin ium 2 derivatives. As expected, alloxazin -
ium salts 2 are m ore efficien t catalysts th an isoalloxazin ium salts 1 regard-
less of th e type of th e reaction system 14. Wh ile th e relative rate vcat/v0 of th e
th ioan isole (3) oxidation in th e presen ce of 1.5 m ole % isoalloxazin ium salt
1b was approxim ately 40 un der optim um con dition s8, th e sam e reaction
catalyzed by th e sam e m olar am oun t of alloxazin ium salt 2b was acceler-
ated by th e factor of 80 in com parison with th e n on -catalyzed oxidation .
In com parison with isoalloxazin ium salts 1, th e optim um pH values for
alloxazin ium catalysts 2 are greater by ca. 3 un its. We suppose th at th is
sh ift is a con sequen ce of th e arom aticity of pyrazin ium groupin g in allox-
azin ium salts 2. In con trast to oxidation s catalyzed by flavin ium salts 1, in
an alogous catalytic cycle in volvin g flavin ium salts 2 th e tran sform ation of
pseudobases 2-OH to flavin ium salts 2 (Sch em e 1) sh ould be accom pan ied
by th e gain of delocalization en ergy wh ich m igh t be a drivin g force for th is
step5a. Th is assum ption is supported by th e obtain ed apparen t pKa′ values
describin g th e equilibrium between flavin ium salt an d th e correspon din g
pseudobase (Sch em e 6). Apparen t pKa′ con stan ts are derived as follows:
form ally, th e form ation of 4a-h ydroxyflavin (pseudobase) represen ts th e
addition of water to flavin ium salt with th e sim ultan eous release of proton .
High er apparen t pKa′ values of alloxazin ium salts 2 in com parison with
th ose obtain ed for isoalloxazin ium salts 1 (by 3–6 un its, depen din g on th e
m edium – Table II) th us in dicate easier elim in ation of water from pseudo-
base 2-OH.
SCHEME 6
Th e equilibrium between flavin ium salt an d pseudobase for isoalloxazin ium salts (1) (a) an d
alloxazin ium salts (b)
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 6, pp. 973–993

982
Cibulka et al.:
TABLE II
Apparen t pKa′ values of flavin ium salt–pseudobase equilibrium for selected derivatives
pKa
′
Flavin ium salt
H2O
SDS
CTANO3
Brij 35
1a
1b
2a
2b
2.8
–
1.8
3.9
8.9
9.6
0.6
0.1
5.6
3.5
2.8
1.4
6.4
5.2
7.9
–
Com parin g th e catalytic efficacy of regioisom ers in both isoalloxazin e
an d alloxazin e series (1b vs 1c an d 2b vs 2c), we observed th at th e isom ers
1b an d 2b with lipoph ilic alkyl ch ain attach ed to th e “upper” site of th e
flavin ium skeleton (i.e. in th e position s 1 or 10) were m ore efficien t th an
th eir isom ers with lipoph ilic alkyl ch ain in th e position 3. We assum e th at
th e relative position of th e polar h ead group an d th e h ydroph obic alkyl
ch ain in am ph iph ilic salts 1b an d 2b favours th eir orien tation at m icellar
surface with th e position 4a directed at aqueous ph ase; th is orien tation
facilitates th e attack by h ydrogen peroxide.
Sulfoxidations Catalyzed by Non-lipophilic Flavinium Salts 1a and 2a
in Micellar Media
In th e n ext part of th is study we h ave tested th e catalytic activity of th e
n on -am ph iph ilic salts 1a an d 2a in th e oxidation of th ioan isole (3) in
m icellar m edia to establish th e role of lipoph ilic alkyl ch ain in 1b, 1c an d
2b, 2c. As we h ad expected, th e in troduction of lipoph ilic alkyl ch ain in to
position 10 of th e isoalloxazin ium skeleton (salt 1b) in creased its catalytic
activity in com parison with salt 1a (Figs 1a–1c an d 2a); h owever, th is in -
crease was on ly m oderate. Surprisin gly, th e in troduction of lipoph ilic alkyl
ch ain in to position 3 of th e isoalloxazin ium skeleton (salt 1c) even de-
creased th e catalytic activity as com pared with n on -am ph iph ilic
h om ologue 1a. Th is un expected effect of th e h ydroph obic alkyl ch ain on
catalytic activity of flavin s in m icellar system s is especially pron oun ced in
th e case of alloxazin ium salts 2. Despite th e fact th at th e catalytic activity
of am ph iph ilic salt 2b in CTANO3 (Fig. 1b) was really h igh , even h igh er
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 6, pp. 973–993

Catalytic Effect of Alloxazinium and Isoalloxazinium Salts
983
th an th e activity of th e m ost efficien t flavin reported so far4–8, it was over-
com e by th at of th e n on -am ph iph ilic h om ologue 2a. We observed relative
rate vcat/v0 = 134 at optim um 14 pH (Fig. 2b).
Th ese in terestin g results lead us to con sider th e fun ction in g of flavin ium
salts on th e m icellar ph ase – bulk aqueous ph ase in terface. Lipoph ilic salts
are tigh tly boun d in m icelles by h ydroph obic in teraction s. Con sequen tly,
reaction s of flavin ium subun it with agen ts com in g from th e aqueous ph ase
(proton , h ydrogen peroxide) proceed at th e in terface on ly. On th e oth er
h an d, n on -am ph iph ilic salts can be easily extracted in to th e aqueous ph ase
an d m ay act sim ilarly as ph ase-tran sfer catalysts16. For sulfoxidation s in
m icelles catalyzed by n on -am ph iph ilic flavin ium salts, we propose th e fol-
lowin g m ech an ism (Fig. 3): first, relatively h ydroph ilic flavin ium salt Fl+
is extracted in to th e aqueous ph ase an d reacts with h ydrogen peroxide
form in g flavin -4a-h ydroperoxide Fl-OOH. Th is n eutral species is less h ydro-
ph ilic an d, th erefore, it can be easily extracted back in to th e m icellar
pseudoph ase wh ere it reacts with th e lipoph ilic substrate. Pseudobase
Fl-OH resultin g from th e oxidation elim in ates water with th e form ation of
flavin ium salt Fl+.
Th e above-m en tion ed h ypoth esis is supported by th e values of bin din g
con stan ts17 KS determ in ed for flavin in ium salts 2 an d th eir correspon din g
pseudobases 2-OH to m icelles of CTANO3 (Table III). Th e bin din g con stan t
KS of salt 2a is approxim ately 10 tim es lower th an th at of pseudobase
FIG. 2
pH depen den ce of th e vcat/v0 ratio of in itial rates of catalyzed an d n on -catalyzed oxidation of 3
with h ydrogen peroxide in m icelles of SDS (ᮀ), CTANO3 (᭺) an d Brij 35 (᭝). Flavin ium salts
1a (a) and 2a (b) were used as catalysts. Conditions: c(3) = 1.5 × 10–2 m ol l–1, c(H2O2) = 2.25 × 10–2
m ol l–1, c(1a or 2a) = 2.3 × 10–4 m ol l–1, c(buffer) = 1.0 × 10–1 m ol l–1, c(surfactan t) = 1.0 × 10–1
m ol l–1; t = 25 °C. Th e lin es h elp th e eyes to follow th e data poin ts
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 6, pp. 973–993

984
Cibulka et al.:
2a-OH. It is reason able to assum e th at th e bin din g con stan t value of th e
h ydroperoxide 2a-OOH sh ould be sim ilar to th at of pseudobase 2a-OH. Th e
bin din g con stan ts values of lipoph ilic salt 2b an d its pseudobase 2b-OH
in dicate th at both th ese species are relatively tigth ly boun d to m icelle.
FIG. 3
Proposed m echanism of sulfoxidation catalyzed by non-am phiphilic flavins in m icellar system s
TABLE III
Bin din g con stan t KS of alloxazin ium salts 2 an d pseudobase 2-OH to m icelles of CTANO3
Com poun d
KS, m ol–1
l
2a
13
156
196
750
2a-OH
2b
2b-OH
Semipreparative Sulfoxidations Catalyzed by Flavinium Salts
To dem on strate th e applicability of our catalytic system s based on m icellar
solution s of flavin ium salts, we perform ed sulfoxidation of th ioan isole (3)
on sem ipreparative scale (100 m g of th e substrate) usin g th e m ost efficien t
catalysts both from isoalloxazin e an d alloxazin e series (salts 1b an d 2a, re-
spectively). Oxidation s were carried out in cation ic an d in an ion ic m icelles
at optim al pH an d at 25 °C usin g sm all excess of h ydrogen peroxide. Reac-
tion tim e was gen erally 30 m in ; if th e catalyzed reaction was n ot com plete,
th e tim e was exten ded un til com plete con version was ach ieved (TLC). Opti-
m ized isolation procedure (see Experim en tal) afforded pure product free of
surfactan t.
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 6, pp. 973–993

Catalytic Effect of Alloxazinium and Isoalloxazinium Salts
985
Th e results sum m arized in Table IV clearly sh ow th at all catalytic system s
em ployin g salt 2a gave full con version with in about 30 m in (Table IV, en -
tries 3, 4 an d 8). Moreover, th e reaction catalyzed with 2a can be success-
fully perform ed even with substan tially decreased am oun t of th e catalyst
(Table IV, en tries 4 an d 5). In th is case, turn over n um ber (TON) reach ed an
im posin g value of 840. Sim ilarly as in kin etic experim en ts, isoalloxazin ium
salt 1b was less efficien t th an alloxazin ium catalyst 2a. To ach ieve com plete
con version , sulfoxidation s catalyzed with 1b require lon ger tim e both in
SDS an d in CTANO3 m icelles (Table IV, en tries 2 an d 7). Blan ks (en tries 1
an d 6) afforded on ly m in or am oun ts of sulfoxide after 30 m in .
TABLE IV
Preparative sulfoxidation of 3 with h ydrogen peroxide catalyzed with flavin ium saltsa
Catalyst
loadin g
m ole%
Reaction
tim eb
m in
Yieldc
%
En try
Medium
pH
Catalyst
1
2
3
4
5
6
7
8
SDS
4.4
4.4
7.5
7.5
7.5
2.4
2.4
7.5
n on e
1b
–
30
90
9
97
97
96
96d
4
SDS
1.5
1.5
0.3
0.1
–
SDS
2a
30
SDS
2a
30
SDS
2a
40
CTANO3
CTANO3
CTANO3
n on e
1b
30
1.5
1.5
120
30
94
97
2a
a
100 m g of 3 was oxidized with 1.5 equivalen ts of h ydrogen peroxide in th e presen ce of th e
catalyst (0.1–1.5 m ole %) in 4 m l of buffered solution of surfactan t (c(SDS) = 0.1 m ol l–1
c(CTANO3) = 0.02 m ol l–1). Reaction tim e was 30 m in or un til th e con version was com -
plete with th e exception of blan k experim en ts. Isolated yield. TON = 840; calculated from
isolated yield.
,
b
c
d
CONCLUSION
Th e results of our curren t study clearly dem on strate th at alloxazin ium
salts 2 are m ore efficien t catalysts of th e sulfides oxidation with h ydrogen
peroxide th an isoalloxazin ium salts 1. In com parison with h om ogen eous
catalytic system s operatin g with com m on “h ydroph ilic” flavin ium salts1–7
,
m icellar system s m ade of cation ic surfactan ts an d con tain in g am ph iph ilic
alloxazin ium salt 2b with th e h ydroph obic alkyl ch ain in position 1 are
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 6, pp. 973–993

986
Cibulka et al.:
m ore efficien t. Th e obtain ed value of th e relative rate vcat/v0 (about 80)
gives eviden ce of a sign ifican t con tribution of m icellar catalysis. However,
to our great surprise, replacin g am ph iph ilic alloxazin ium salt 2b in cation ic
m icelles by n on -am ph iph ilic salt 2a lead to furth er in crease in th e catalytic
activity. We h ave con cluded th at th e species in volved in th e catalytic cycle
(i.e. salt 2a an d h ydroperoxide 2a-OOH) h ave th eir h ydroph ilic an d lipo-
ph ilic properties well balan ced an d th erefore salt 2a m ay act like a ph ase-
tran sfer catalyst brin gin g h ydrogen peroxide from th e aqueous ph ase in to
th e m icelle in terior. We assum e th at th e con cept of ph ase-tran sfer catalysis
with covalen tly boun d peroxide in sulfoxidation s catalyzed by flavin ium
salts can be applied even to real two-ph ase system s. In vestigation of th is
possibility will be th e subject of our n ext study.
EXPERIMENTAL
Gen eral
Tem perature data are un corrected. NMR spectra were recorded on a Varian Mercury Plus 300
(299.97 MHz for 1H an d 75.44 MHz for 13C) an d a Bruker Avan ce DRX 500 spectrom eter
(500.13 MHz for 1H an d 125.77 MHz for 13C). Ch em ical sh ifts are given in ppm relative to
tetram eth ylsilan e (δ 0.0) used as an in tern al stan dard, couplin g con stan ts (J) are given in Hz.
TLC an alyses were carried out on a DC Alufolien Kieselgel 60 F254 (Merck). Preparative col-
um n ch rom atograph y separation s were perform ed on a silica gel Kieselgel 60 0.040–0.063 m m
(Merck). Elem en tal an alyses (C, H, N) were perform ed on a Perkin –Elm er 240 an alyser; deter-
m in ation of Cl was m ade on an X-ray fluorescen ce spectrom eter ARL 9400 XP+. UV-Vis spec-
tra were recorded on a Varian Cary 50 spectroph otom eter. HPLC an alyses were perform ed
on an Ecom HPLC System (colum n : Separon SGX C18, 150 × 4.6 m m ; 0.5 m l m in –1) usin g
UV spectroph otom etric detection .
Ch em icals
Startin g m aterials an d reagen ts used for th e preparation of th e catalysts were obtain ed from
Aldrich , Lan caster an d Fluka, an d were used with out purification . Solven ts an d substrates
used for kin etic experim en ts (an alytical grade) were purch ased from Fluka an d Pen ta. Th e
con cen tration of h ydrogen peroxide was determ in ed by m an gan om etric titration . Prepara-
tion an d ch aracterization of isoalloxazin ium catalysts 1a an d 1b was described in our previ-
ous paper8. Alloxazin ium salt 2a was prepared accordin g to th e procedure described by
Murah ash i5e. Syn th esis of lum iflavin 4 (used for syn th esis of 1c) is publish ed elsewh ere12
.
N-Substituted barbituric acids (used for th e syn th esis of N-alkylalloxan s 6) were prepared
from N-substituted ureas18 by th e reaction with m alon ic acid (in th e case of m eth yl
h om ologue) or with dieth yl m alon ate (in th e case of dodecyl derivative). N-Meth ylurea for
th e syn th esis of N-m eth ylbarbituric acid was obtain ed from Sigm a–Aldrich . N-Dodecylurea
was prepared from th e correspon din g N-dodecylam in e an d sodium cyan ate accordin g to th e
described procedure19
.
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 6, pp. 973–993

Catalytic Effect of Alloxazinium and Isoalloxazinium Salts
987
3-Dodecyl-5-ethyl-10-methylisoalloxazinium perchlorate (1c). Acetaldeh yde (0.7 m l, 12.5 m m ol)
an d palladium on carbon (10%, 21 m g) were added to a suspen sion of 5 (50 m g, 0.118 m m ol)
in acetic acid (14 m l) an d water (1.4 m l). Th e resultin g m ixture was stirred for 3 days in an
autoclave un der H2 (0.6 MPa) at room tem perature. Th en th e catalyst was filtered off, acetic
acid was evaporated un der reduced pressure, an d th e rem ain in g green solid was dried in
vacuo. Th e residue was suspen ded in 2 M perch loric acid (1.5 m l), an d sodium perch lorate
(0.321 g, 2.29 m m ol) an d sodium n itrite (0.119 g, 1.72 m m ol) were added. Meth an ol (7.5 m l)
was added to th is system in order to obtain a h om ogen eous solution . Th e m ixture was th en
stirred at 10 °C for 2 h . Meth an ol was evaporated un der reduced pressure, th e precipitated
solid was filtered off, wash ed with 2 M perch loric acid an d dried in vacuo. Yield 41 m g
(63%). Violet crystallin e solid, m .p. 104–107 °C. 1H NMR (500 MHz, CD3CN): 0.91 t, 3 H,
J = 6.6 (alkyl-CH3); 1.31 m , 16 H ((CH2)8); 1.66 m , 2 H (NCH2CH2CH2); 1.80 t, 3 H, J = 6.5
(N+CH2CH3); 2.20 m , 2 H (NCH2CH2); 2.59 s, 3 H (7-CH3Ar); 2.65 s, 3 H (8-CH3Ar); 4.01 t,
2 H, J = 7.3 (NCH2); 4.17 s, 3 H (NCH3); 4.99 bs, 1 H (N+CHH); 6.06 bs, 1 H (N+CHH);
7.90 s, 1 H (9-HAr); 8.23 s, 1 H (6-HAr). 13C NMR (125 MHz, CD3CN): 14.4 (alkyl-CH3); 15.5
(N+CH2CH3); 20.3 (7-CH3Ar); 21.4 (8-CH3Ar); 23.3 (alkyl-CH2); 27.3 (alkyl-CH2); 27.4
(alkyl-CH2); 27.7 (alkyl-CH2); 29.9 (alkyl-CH2); 30.21 (alkyl-CH2); 30.26 (alkyl-CH2); 30.32
(alkyl-CH2); 30.4 (alkyl-CH2); 32.1 (NCH3); 32.4 (alkyl-CH2); 42.3 (NCH2); 53.2 (N+CH2);
118.8 (CHAr); 121.7 (CHAr); 127.6 (CAr); 131.3 (CAr); 136.7 (CAr); 143.0 (CH3-CAr); 150.2
(CAr); 153.4 (CH3-CAr); 153.9 (CO); 156.9 (CO). For C27H41ClN4O6·3H2O (525.05) calculated:
53.41% C, 7.80% H, 5.84% Cl, 9.23% N; foun d: 53.36% C, 7.39% H, 5.96% Cl, 9.14% N.
5-Ethyl-1,3-dimethylalloxazinium perchlorate (2a). Salt 2a was prepared accordin g to th e
publish ed procedure5e. Yield 84%. Oran ge solid, m .p. 246–249 °C. 1H NMR (500 MHz,
CD3CN): 1.84 t, 3 H (N+CH2CH3); 3.54 s, 3 H (3-NCH3); 3.85 s, 3 H (1-NCH3); 5.00 bs, 1 H
(N+CHH); 6.10 bs, 1 H (N+CHH); 8.27–8.38 m , 3 H (7,8,9-HAr); 8.55 d, 1 H, J = 8.9 (6-HAr).
13C NMR (125 MHz, CD3CN): 15.6 (N+CH2CH3); 30.7 (3-NCH3); 31.6 (1-NCH3); 52.9
(N+CH2); 120.7 (CHAr); 122.7 (CAr); 131.1 (CHAr); 137.1 (CHAr); 137.8 (CHAr); 148.0 (CAr);
148.8 (CAr); 149.8 (CO); 156.4 (CO). For C14H15ClN4O6·H2O (388.77) calculated: 43.25% C,
4.41% H, 9.12% Cl, 14.41% N; foun d: 43.52% C, 4.19% H, 9.27% Cl, 14.34% N.
1-Dodecyl-5-ethyl-3-methylalloxazinium perchlorate (2b ). Salt 2b was prepared from
alloxazin e 8a (50 m g, 0.126 m m ol) accordin g to th e procedure described for 1c. Yield 34 m g
(51%). Brown solid, m .p. 114–118 °C. 1H NMR (500 MHz, CD3CN): 0.86 t, 3 H, J = 6.5
(alkyl-CH3); 1.26 m , 16 H ((CH2)8); 1.44 m , 2 H (NCH2CH2CH2); 1.73 m , 2 H (NCH2CH2);
1.84 t, 2 H, J = 7.0 (N+CH2CH3); 3.50 s, 3 H (NCH3); 4.46 t, 2 H, J = 7.5 (NCH2); 5.13 bs, 1 H
(N+CHH); 6.21 bs, 1 H (N+CHH); 8.23–8.31 m , 3 H (7,8,9-HAr); 8.55 d, 1 H, J = 8.9 (6-HAr).
13C NMR (125 MHz, CD3CN): 14.4 (alkyl-CH3); 15.6 (N+CH2CH3); 23.4 (alkyl-CH2); 27.2
(alkyl-CH2); 27.7 (alkyl-CH2); 29.9 (alkyl-CH2); 30.1 (alkyl-CH2); 30.2 (alkyl-CH2); 30.28
(alkyl-CH2); 30.33 (alkyl-CH2); 30.4 (alkyl-CH2); 30.7 (alkyl-CH2); 32.6 (N-CH3); 45.1
(NCH2); 52.9 (N+CH2); 120.7 (CHAr); 122.7 (CAr); 131.1 (CHAr); 137.1 (CHAr); 137.7 (CHAr);
148.0 (CAr); 148.4 (CAr); 149.5 (CO); 156.4 (CO). For C25H37ClN4O6·H2O (543.07) calculated:
55.29% C, 7.24% H, 6.53% Cl, 10.32% N; foun d: 54.90% C, 7.26% H, 6.40% Cl, 10.30% N.
3-Dodecyl-5-ethyl-1-methylalloxazinium perchlorate (2c). Salt 2c was prepared from
alloxazin e 8b (70 m g, 0.176 m m ol) accordin g to th e procedure described for 1c. Yield 50 m g
(76%). Brown solid, m .p. 143–147 °C. 1H NMR (500 MHz, CD3CN): 0.86 t, 3 H, J = 6.6
(alkyl-CH3); 1.25 m , 16 H ((CH2)8); 1.39 m , 2 H (NCH2CH2CH2); 1.67 m , 2 H (NCH2CH2);
1.76 t, 3 H, J = 7.0 (N+CH2CH3); 3.78 s, 3 H (NCH3); 4.07 m , 2 H (NCH2); 5.13 bs, 1 H
(N+CHH); 6.21 bs, 1 H (N+CHH); 8.22–8.32 m , 3 H (7,8,9-HAr); 8.54 d, 1 H, J = 8.9 (6-HAr).
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 6, pp. 973–993

988
Cibulka et al.:
13C NMR (125 MHz, CD3CN): 14.4 (alkyl-CH3); 15.6 (N+CH2CH3); 23.4 (alkyl-CH2); 27.5
(alkyl-CH2); 27.7 (alkyl-CH2); 29.9 (alkyl-CH2); 30.0 (alkyl-CH2); 30.2 (alkyl-CH2); 30.27
(alkyl-CH2); 30.34 (alkyl-CH2); 30.4 (alkyl-CH2); 31.5 (N-CH3); 32.6 (alkyl-CH2); 44.7
(NCH2); 52.9 (N+CH2); 120.6 (CHAr); 122.7 (CAr); 130.8 (CHAr); 136.7 (CHAr); 137.6 (CHAr);
147.5 (CAr); 148.5 (CAr); 149.4 (CO); 156.0 (CO). For C25H37ClN4O6·H2O (543.07) calculated:
55.29% C, 7.24% H, 6.53% Cl, 10.32% N; foun d: 55.17% C, 7.38% H, 6.34% Cl, 10.19% N.
3-Dodecyl-7,8,10-trimethylisoalloxazine (5). A m ixture of lum iflavin 4 (0.800 g, 1.88 m m ol),
potassium carbon ate (2.160 g, 15.63 m m ol), an d dodecyl iodide (2.770 g, 9.35 m m ol) in dry
DMF (180 m l) was stirred at room tem perature for 45 h . Th e salts were filtered off an d th e
solven t was rem oved un der reduced pressure. Th e crude product was purified by colum n
ch rom atograph y (ch loroform /m eth an ol 100:5) an d by crystallization from eth an ol/ water
(2:1). Yield 1.050 g (79%). Yellow solid, m .p. 195–196 °C. 1H NMR (500 MHz, CDCl3):
0.87 t, 3 H, J = 6.9 (alkyl-CH3); 1.24 m , 16 H ((CH2)8); 1.34 m , 2 H (NCH2CH2CH2); 1.71 m ,
2 H (NCH2CH2); 2.44 s, 3 H (7-CH3Ar); 2.54 s, 3 H (8-CH3Ar); 4.09 m , 2 H (NCH2); 4.10 s,
3 H (NCH3); 7.40 s, 1 H (9-HAr); 8.04 s, 1 H (6-HAr). 13C NMR (125 MHz, CDCl3): 14.3
(alkyl-CH3); 19.7 (7-CH3Ar); 21.7 (8-CH3Ar); 22.9 (alkyl-CH2); 27.2 (alkyl-CH2); 28.0
(alkyl-CH2); 29.0 (NCH3); 29.48 (alkyl-CH2); 29.55 (alkyl-CH2); 29.60 (alkyl-CH2); 29.75
(alkyl-CH2); 29.83 (alkyl-CH2); 29.86 (alkyl-CH2); 32.1 (alkyl-CH2); 42.3 (NCH2); 115.5
(CHAr); 131.9 (CAr); 132.8 (CHAr); 134.8 (CAr); 136.0 (CAr); 136.9 (8-CH3-CAr); 148.0
(7-CH3-CAr); 149.2 (CAr); 155.8 (CO); 159.9 (CO). For C25H36N4O2 (424.59) calculated:
70.72% C, 8.55% H, 13.20% N; foun d: 70.28% C, 8.92% H, 13.27% N.
N-Methylalloxan hydrate (6a). Ch rom ium oxide (26.3 g, 264 m m ol) was dissolved in a m ix-
ture of acetic acid (100 m l) an d water (12 m l). Reaction m ixture was cooled with ice an d
N-m eth ylbarbituric acid (24.4 g, 172 m m ol) was added in sm all portion s keepin g th e tem -
perature below 50 °C. After th e addition was com plete, th e reaction m ixture was stirred at
45–50 °C for 30 m in . Th e reaction m ixture was allowed to stay at 4 °C for 1 week an d th en
it was tem pered to room tem perature. Th e solid was filtered off, wash ed with cold acetic
acid (3 × 20 m l) an d dieth yl eth er (25 m l) an d dried in vacuo. Yield 22.3 g (74%). Wh ite
crystals, m .p. 134–138 °C (ref.20 150–151 °C). 1H NMR (300 MHz, DMSO-d6): 3.04 s, 3 H
(CH3); 7.62 m , 2 H (C(OH)2); 11.43 m , 1 H (NH). 13C NMR (75 MHz, DMSO-d6 + 4% D2O):
28.2 (CH3); 86.0 (C(OH)2); 166.4 (CO); 168.7 (CO); 169.6 (CO). For C5H6N2O5 (174.11) cal-
culated: 34.49% C, 3.47% H, 16.09% N; foun d: 34.27% C, 3.50% H, 15.86% N.
N-Dodecylalloxan hydrate (6b ). Ch rom ium oxide (1.1 g, 10.88 m m ol) was dissolved in
a m ixture of acetic acid (20 m l) an d water (2 m l). Th e reaction m ixture was cooled with ice
an d N-dodecylbarbituric acid (2.93 g, 9.87 m m ol) was added in sm all portion s keepin g th e
tem perature below 50 °C. After th e addition of all th e substrate, th e reaction m ixture was
stirred at 45–50 °C for 30 m in . Reaction m ixture was diluted with water (125 m l) an d ex-
tracted with eth yl acetate (3 × 150 m l). Organ ic layer was wash ed with water an d dried over
m agn esium sulfate. After th e evaporation of th e solven t in vacuo, th e crude product was pu-
rified by colum n ch rom atograph y (eth yl acetate/h exan e 1:1). Yield 0.94 g (29%). Wh ite
pasty com poun d. 1H NMR (300 MHz, DMSO-d6): 0.83 t, 3 H, J = 7.0 (alkyl-CH3); 1.22 m ,
18 H ((CH2)9); 1.45 m , 2 H (NCH2CH2); 3.62 t, 2 H, J = 7.6 (NCH2); 7.61 m , 2 H (C(OH)2);
11.40 m , 1 H (NH). For C16H26N2O4·H2O (328.41) calculated: 58.52% C, 8.59% H, 8.53% N;
foun d: 58.27% C, 8.54% H, 8.66% N.
3-Methylalloxazine (7a). A solution of 1,2-din itroben zen e (2.000 g, 11.90 m m ol) in acetic
acid (100 m l) was h ydrogen ated in th e presen ce of palladium on carbon (10%, 0.150 g) at
0.1 MPa at room tem perature. Total reaction tim e was 3 h . Th e reaction m ixture was filtered
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 6, pp. 973–993

Catalytic Effect of Alloxazinium and Isoalloxazinium Salts
989
th rough Celite, an d a h ot solution (70 °C) of boric acid (0.819 g, 13.24 m m ol) an d
N-m eth ylalloxan m on oh ydrate (2.071 g, 11.89 m m ol) in acetic acid (75 m l) was added im -
m ediately. Th e m ixture was stirred at room tem perature for 20 h . Th e precipitate form ed
was filtered off, wash ed with acetic acid an d dieth yl eth er, an d dried in vacuo. Yield 1.750 g
(65%). Cream crystallin e solid, m .p. 285 °C. 1H NMR (300 MHz, DMSO-d6): 3.29 s, 3 H
(CH3); 7.77 m , 1 H (7-HAr); 7.92 m , 2 H (6,8-HAr); 8.18 d, 1 H, J = 8.2 (6-HAr); 12.23 s, 1 H
(NH). 13C NMR (125 MHz, DMSO-d6): 29.1 (NCH3); 127.9 (CHAr); 129.1 (CHAr); 129.7 (CAr);
130.7 (CHAr); 133.7 (CHAr); 139.9 (CAr); 143.5 (CAr); 145.2 (CAr); 150.3 (CO); 160.0 (CO). For
C
11H8N4O2 (228.21) calculated: 57.89% C, 3.53% H, 24.55% N; foun d: 58.12% C, 3.41% H,
24.55% N.
3-Dodecylalloxazine (7b). A solution of 1,2-din itroben zen e (0.248 g, 1.48 m m ol) in acetic
acid (20 m l) was h ydrogen ated in th e presen ce of palladium on carbon (10%, 20 m g) at
0.1 MPa at room tem perature. Total reaction tim e was 3 h . Th e reaction m ixture was filtered
th rough Celite, an d a h ot solution (70 °C) of boric acid (0.102 g, 1.65 m m ol) an d N-dodecyl-
alloxan (0.485 g, 1.48 m m ol) in acetic acid (14 m l) was added im m ediately. Th e m ixture was
stirred at room tem perature for 20 h . Th e precipitate form ed was filtered off, wash ed with
acetic acid an d dieth yl eth er an d dried in vacuo. Yield 0.487 g (86%). Yellow solid, m .p.
279–282 °C. 1H NMR (300 MHz, DMSO-d6): 0.83 t, 3 H, J = 6.8 (alkyl-CH3); 1.21 m , 18 H
((CH2)9); 1.61 m , 2 H (NCH2CH2); 3.90 m , 2 H (NCH2); 7.77 m , 1 H (7-HAr); 7.93 m , 2 H
(6,8-HAr); 8.19 d, 1 H, J = 8.2 (6-HAr); 12.20 s, 1 H (NH). 13C NMR (125 MHz, DMSO-d6):
14.1 (alkyl-CH3); 22.6 (alkyl-CH2); 26.8 (alkyl-CH2); 27.4 (alkyl-CH2); 29.21 (alkyl-CH2);
29.26 (alkyl-CH2); 29.32 (alkyl-CH2); 29.5 (alkyl-CH2); 29.60 (alkyl-CH2); 29.63 (alkyl-CH2);
31.7 (alkyl-CH2); 42.4 (NCH2); 127.9 (CHAr); 129.2 (CHAr); 129.6 (CAr); 130.7 (CHAr); 133.8
(CHAr); 139.8 (CAr); 143.6 (CAr); 145.4 (CAr); 150.4 (CO); 160.0 (CO). For C22H30N4O2
(382.51) calculated: 69.08% C, 7.91% H, 14.65% N; foun d: 69.38% C, 7.74% H, 14.48% N.
1-Dodecyl-3-methylalloxazine (8a). A m ixture of alloxazin e 7a (0.800 g, 3.51 m m ol), potas-
sium carbon ate (2.422 g, 17.53 m m ol), an d dodecyl iodide (2.077 g, 7.01 m m ol) in dry DMF
(40 m l) was stirred in in ert atm osph ere (N2) at room tem perature for 22 h . Th e salts were fil-
tered off an d wash ed with ch loroform . Solven ts from th e filtrate were rem oved un der re-
duced pressure. Dich lorom eth an e (70 m l) an d water (20 m l) were added to th e residue an d
after separation of th e organ ic layer, th e aqueous ph ase was wash ed with dich lorom eth an e
(7 × 20 m l). Organ ic ph ase was dried over an h ydrous m agn esium sulfate. After th e evapora-
tion of th e solven t in vacuo, th e crude product was purified by colum n ch rom atograph y
(ch loroform /m eth an ol 100:3). Yield 1.373 g (99%). Yellow solid, m .p. 152 °C. 1H NMR (500
MHz, CDCl3): 0.87 t, 3 H, J = 6.5 (alkyl-CH3); 1.26 m , 16 H ((CH2)8); 1.44 m , 2 H
(NCH2CH2CH2); 1.82 m , 2 H (NCH2CH2); 3.60 s, 3 H (NCH3); 4.44 t, 2 H, J = 7.6 (NCH2);
7.75 t, 1 H, J = 8.0 (7-HAr); 7.89 t, 1 H, J = 8.1 (8-HAr); 8.02 d, 1 H, J = 8.4 (6-HAr); 8.34 d,
1 H, J = 8.4 (9-HAr). 13C NMR (125 MHz, CDCl3): 14.1 (alkyl-CH3); 22.6 (alkyl-CH2); 26.8
(alkyl-CH2); 27.4 (alkyl-CH2); 29.1 (NCH3); 29.21 (alkyl-CH2); 29.30 (alkyl-CH2); 29.33
(alkyl-CH2); 29.47 (alkyl-CH2); 29.57 (alkyl-CH2); 29.61 (alkyl-CH2); 31.9 (alkyl-CH2); 42.8
(NCH2); 127.9 (CHAr); 129.0 (CHAr); 129.6 (CAr); 130.8 (CHAr); 133.7 (CHAr); 139.9 (CAr);
143.4 (CAr); 145.0 (CAr); 150.3 (CO); 159.9 (CO). For C23H32N4O2 (396.54) calculated:
69.67% C, 8.13% H, 14.13% N; foun d: 69.94% C, 7.90% H, 13.84% N.
3-Dodecyl-1-methylalloxazine (8b). A m ixture of alloxazin e 7b (0.410 g, 1.07 m m ol), potas-
sium carbon ate (0.741 g, 5.36 m m ol), an d m eth yl iodide (0.761 g, 5.36 m m ol) in dry DMF
(15 m l) was stirred in in ert (N2) atm osph ere at room tem perature for 22 h . Th e salts were fil-
tered off an d wash ed with ch loroform . Solven ts from th e filtrate were rem oved un der re-
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 6, pp. 973–993

990
Cibulka et al.:
duced pressure. Dich lorom eth an e (40 m l) an d water (15 m l) were added to th e residue an d
after separation of th e organ ic layer, aqueous ph ase was wash ed with dich lorom eth an e (7 ×
15 m l). Th e organ ic ph ase was dried over an h ydrous m agn esium sulfate. After th e evapora-
tion of th e solven t in vacuo, th e crude product was purified by colum n ch rom atograph y
(ch loroform /m eth an ol 100:3). Yield 0.210 g (49%). Yellow solid, m .p. 99–102 °C. 1H NMR
(500 MHz, CDCl3): 0.87 t, 3 H, J = 6.6 (alkyl-CH3); 1.24 m , 16 H ((CH2)8); 1.37 m , 2 H
(NCH2CH2CH2); 1.73 m , 2 H (NCH2CH2); 3.82 s, 3 H (NCH3); 4.17 t, 2 H, J = 7.5 (NCH2);
7.75 t, 1 H, J = 8.1 (7-HAr); 7.89 t, 1 H, J = 8.2 (8-HAr); 8.02 d, 1 H, J = 8.4 (6-HAr); 8.34 d, 1
H, J = 8.3 (9-HAr). 13C NMR (125 MHz, CDCl3): 14.1 (alkyl-CH3); 22.6 (alkyl-CH2); 26.8
(alkyl-CH2); 27.4 (alkyl-CH2); 28.9 (NCH3); 29.21 (alkyl-CH2); 29.29 (alkyl-CH2); 29.34
(alkyl-CH2); 29.47 (alkyl-CH2); 29.57 (alkyl-CH2); 29.61 (alkyl-CH2); 31.8 (alkyl-CH2); 42.5
(NCH2); 127.9 (CHAr); 129.0 (CHAr); 129.6 (CAr); 130.8 (CHAr); 133.7 (CHAr); 139.9 (CAr);
143.4 (CAr); 145.0 (CAr); 150.3 (CO); 159.9 (CO). For C23H32N4O2 (396.54) calculated:
69.67% C, 8.13% H, 14.13% N; foun d: 69.61% C, 8.01% H, 13.99% N.
Kin etic Measurem en ts
Th e oxidation of th ioan isole (3) (both in m icellar an d in h om ogen eous solution s) was per-
form ed in vials th erm ostatted to 25.0 ± 0.5 °C. Th e reaction m ixtures were prepared by m ix-
in g aliquot portion s of stock solution s of reactan ts an d oth er additives: th ioan isole (3) (c =
6.00 × 10–1 m ol l–1 in m eth an ol), flavin ium salt 1 or 2 (c = 9.0 × 10–3 m ol l–1 in m eth an ol,
fresh ly prepared; it was n ot added in blan k experim en ts), surfactan t (c = 1.00 × 10–1 m ol l–1
in water), in tern al stan dard (1,3-dim eth oxyben zen e, c = 3.0 × 10–2 m ol l–1 in 0.1 M solution
of surfactan t) an d ph osph ate buffer (c = 5.00 × 10–1 m ol l–1). Th e prepared m ixtures were
th erm ostatted for 20 m in before th e experim en t was started. Th e reaction s were in itiated by
th e addition of 100 µl of h ydrogen peroxide solution (c = 4.50 × 10–1 m ol l–1 in water) to
th e prepared m ixture. Th e resultin g con cen tration s of reactan ts an d auxiliaries in th e reac-
tion m ixtures were as follows: c(3) = 1.5 × 10–2 m ol l–1, c(H2O2) = 2.25 × 10–2 m ol l–1
c(flavin ium salt) = 2.3 × 10–4 m ol l–1, c(1,3-dim eth oxyben zen e) = 7.5 × 10–3 m ol l–1
,
,
c(buffer) = 1.0 × 10–1 m ol l–1, c(surfactan t) = 1.0 × 10–1 m ol l–1
.
Th e con cen tration of m eth yl ph en yl sulfoxide was m on itored by HPLC usin g 1,3-di-
m eth oxyben zen e as an in tern al stan dard. Sam ple preparation : 100 µl of th e reaction m ixture
was diluted with 100 µl of aceton itrile.
Preparative Sulfoxidation s in SDS Micelles. Gen eral Procedure
Aqueous solution of SDS (c = 5.00 × 10–1 m ol l–1; 800 µl) was m ixed with deion ized water
(1 200 µl) an d ph osph ate buffer (c = 5.00 × 10–1 m ol l–1; 2 000 µl). Th ioan isole (95 µl,
0.805 m m ol) an d flavin ium salt (0.0120 or 0.0024 or 0.0008 m m ol) were added. Th e reac-
tion m ixture was em ulgated by son ication an d th erm ostatted to 25.0 ± 0.5 °C for 20 m in .
Th e reaction was in itiated by th e addition of 124 µl of h ydrogen peroxide (30% aqueous
solution ; 1.2 m m ol) to th e prepared m ixture. After 30 m in or after com plete con version was
ach ieved (m on itored by TLC usin g toluen e as a m obile ph ase), sodium perch lorate (4 g) an d
dieth yl eth er (10 m l) were added. Th e resultin g suspen sion was allowed to stan d for 15 m in
an d th en filtered. Th e filtrate was separated in a separatory fun n el an d th e aqueous ph ase was
extracted with dieth yl eth er (10 × 6 m l). Th e com bin ed eth ereal extracts were dried over an -
h ydrous sodium sulfate an d evaporated to dryn ess. Th e crude product was an alyzed by 1H
NMR.
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 6, pp. 973–993

Catalytic Effect of Alloxazinium and Isoalloxazinium Salts
991
Preparative Sulfoxidation s in CTANO3 Micelles. Gen eral Procedure
Aqueous solution of CTANO3 (c = 5.00 × 10–1 m ol l–1; 160 µl) was m ixed with deion ized
water (1 840 µl) an d ph osph ate buffer (c = 5.00 × 10–1 m ol l–1; 2 000 µl). Th ioan isole (95 µl,
0.805 m m ol) an d flavin ium salt (0.012 m m ol) were added. Th e reaction m ixture was
em ulgated by son ication an d th erm ostatted to 25.0 ± 0.5 °C for 20 m in . Th e reaction was
in itiated by th e addition of 124 µl of h ydrogen peroxide (30% aqueous solution ; 1.2 m m ol)
to th e prepared m ixture. After 30 m in or after com plete con version was ach ieved (m on i-
tored by TLC usin g toluen e as a m obile ph ase), sodium perch lorate (4 g) an d dieth yl eth er
(10 m l) were added. Th e resultin g suspen sion was allowed to stan d for 15 m in an d th en fil-
tered. Th e filtrate was separated in a separatory fun n el an d th e aqueous ph ase was extracted
with dieth yl eth er (10 × 6 m l). Th e com bin ed eth ereal extracts were dried over an h ydrous
sodium sulfate an d evaporated to dryn ess. Th e crude product was an alyzed by 1H NMR.
Study of Flavin ium Salt–Pseudobase Equilibrium
Th e apparen t pKa′ values of 1 an d 2 were calculated usin g Origin 6.1 software21 (Boltzm an n
fun ction ) by n on -lin ear regression an alysis of absorban ce versus pH data at th e m axim um
absorption wavelen gth s of 1 an d 1-OH or 2 an d 2-OH. Th e pKa′s given in Table II are th e
m ean s of th e obtain ed values. Eith er ph osph ate buffer (for pH > 1.6) or HNO3 (for pH < 1.6)
was used to adjust th e pH values of th e system s. Acidity fun ction (H0) values22 were used in -
stead of pH values for HNO3 solution s.
Bin din g Con stan ts of Flavin ium Salts an d Th eir Pseudobases in Cation ic Micelles
Th e values of th e bin din g con stan t KS of 2 an d 2-OH to m icelles of CTANO3 were m easured
spectroph otom etrically17 from th e variation of absorban ce Aobs at 450 n m (in th e case of
salt 2) or 296 n m (in th e case of pseudobase 2-OH) with surfactan t con cen tration csurf at pH
value by 2 un its lower (for KS of flavin ium form ) or by 2 un its h igh er (for KS of pseudobase)
th an correspon din g pKa′ in th e appropriate m icellar system . Th e bin din g con stan t was
calculated usin g Eq. (1) (see n ote in ref.17
)
(Aobs – Aw)/[Dn] = AMKS – AobsKS
(1)
wh ere Aobs is th e observed absorban ce, AW an d AM are th e correspon din g absorban ces of 2
in water an d in m icellar pseudoph ase an d [Dn] is th e con cen tration of m icellized surfactan t
([Dn] = csurf – cm c; cm c stan ds for critical m icelle con cen tration ). Th e values of AW were
m easured in in depen den t experim en ts. Th e bin din g con stan t was obtain ed as a slope of th e
lin ear plot (Aobs – Aw)/[Dn] versus Aobs. Ph osph ate buffer was used to adjust pH values of th e
system s.
The authors wish to thank the Czech Science Foundation (Grant No. 203/07/1246) for the
financial support.
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 6, pp. 973–993

992
Cibulka et al.:
REFERENCES AND NOTES
1. a) Bäckvall J.-E. in: Modern Oxidation Methods (J.-E. Bäckvall, Ed.), pp. 193–222.
Wiley-VCH, Weinheim 2004; b) Gelalcha F. G.: Chem. Rev. 2007, 107, 3338; c) Yasushi I.,
Takeshi N.: Chem. Record 2007, 7, 354.
2. Murahashi S.-I., Oda T., Masui Y.: J. Am. Chem. Soc. 1989, 111, 5002.
3. Mazzini C., Lebreton J., Furstoss R.: J. Org. Chem. 1996, 61, 618.
4. Bergstad K., Bäckvall J.-E.: J. Org. Chem. 1998, 63, 6650.
5. a) Minidis A. B. E., Bäckvall J.-E.: Chem. Eur. J. 2001, 7, 297; b) Lindén A. A., Krüger L.,
Bäckvall J.-E.: J. Org. Chem. 2003, 68, 5890; c) Lindén A. A., Hermanns N., Ott S.,
Krüger L., Bäckvall J.-E.: Chem. Eur. J. 2005, 11, 112; d) Lindén A. A., Johansson M.,
Hermanns N., Bäckvall J.-E.: J. Org. Chem. 2006, 71, 3849; e) Imada Y., Iida H., Ono S.,
Masui Y., Murahashi S.-I.: Chem. Asian J. 2006, 136.
6. a) Shinkai S., Yamaguchi T., Kawase A., Kitamura A., Manabe O.: Chem. Commun. 1987,
1506; b) Shinkai S., Yamaguchi T., Manabe O., Toda F.: Chem. Commun. 1988, 1399;
c) Murahashi S.-I.: Angew. Chem., Int. Ed. Engl. 1995, 34, 2443.
7. Murahashi S.-I., Ono S., Imada Y.: Angew. Chem. Int. Ed. 2002, 41, 2366.
8. Baxová L., Cibulka R., Hampl F.: J. Mol. Catal. A 2007, 277, 53.
9. a) Kemal C., Chan T. W., Bruice T. C.: Proc. Natl. Acad. Sci. U.S.A. 1977, 74, 405;
b) Bruice T. C.: Acc. Chem. Res. 1980, 13, 256; c) Miller A.: Tetrahedron Lett. 1982, 23,
753.
10. a) Palfey B., Massey V. in: Comprehensive Biological Catalysis (M. Sinnott, Ed.), Vol. 3,
pp. 83–154. Academic Press, London 1998; b) Moonen M. J. H., Fraaije M. W., Rietjens
I. M. C. M., Laane C., van Berkel W. J. H.: Adv. Synth. Catal. 2002, 10, 1023.
11. a) Ogino K., Yoshida T., Yamamoto H., Tagaki W.: Chem. Lett. 1992, 1197; b) Menger
F. M., Keiper J. S., Mbadugha B. N. A., Caran K. L., Romsted L. S.: Langmuir 2000, 16,
9095; c) Kotoučová H., Cibulka R., Hampl F., Liška F.: J. Mol. Catal. A 2001, 174, 59;
d) Cibulka R., Hampl F., Kotoučová H., Mazáč J., Liška F.: Collect. Czech. Chem. Commun.
2000, 65, 227; e) Kivala M., Cibulka R. Hampl F.: Collect. Czech. Chem. Commun. 2006,
71, 1642.
12. Yoneda F., Sakuma Y., Ichiba M., Shinomura K.: J. Am. Chem. Soc. 1976, 98, 830.
13. Holmgren A. V., Wenner W.: Org. Synth. 1952, 32, 6.
14. Unfortunately, we were not able to achieve the optimum pH value in SDS micelles in
the case of amphiphilic salts 2 since no suitable buffer is available for sulfoxidation
experiments at pH > 8. All the anions in inorganic buffers operating above pH 8 react
with hydrogen peroxide under formation of corresponding peroxoanions (peroxo-
phosphates and peroxoborates). The so-called biological buffers containing tertiary
amino group cannot be used either due to formation of the corresponding N-oxides.
15. Merényi G., Lind J.: J. Am. Chem. Soc. 1991, 113, 3146.
16. Dehmlow E. V., Dehmlow S. S.: Phase Transfer Catalysis, 3rd ed. Wiley-VCH, Weinheim
1993.
17. Micellar binding of any organic substance is generally described by a binding constant
KS = [Sm]/[Sw][Dn], where [Sm] and [Sw] are the concentrations of the substance in
micellar and aqueous phase, respectively, and [Dn] is the concentration of micellized
surfactant, see: Sepulveda L., Lissi E., Quina F.: Adv. Colloid Interface Sci. 1986, 25, 1.
18. Stein A., Gregor H. P., Spoerri P. E.: J. Am. Chem. Soc. 1956, 78, 6185.
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 6, pp. 973–993

Catalytic Effect of Alloxazinium and Isoalloxazinium Salts
993
19. Ohno A., Kunitomo J., Kawai Y., Kawamoto T., Tomishima M., Yoneda F.: J. Org. Chem.
1996, 61, 9344.
20. Li W.-S., Zhang N., Sayre L. M.: Tetrahedron 2001, 57, 4507.
21. Origin 6.1. OriginLab Corporation, Northampton 2000.
22. a) Milyaeva N. M.: Zh. Neorg. Khim. 1958, 3, 2011; b) Dawber J. G.: J. Chem. Soc. A 1968,
1532.
Collect. Czech. Chem. Commun. 2009, Vol. 74, No. 6, pp. 973–993