Selective oxidation of alcohols to carbonyl compounds mediated by fluorous-tagged TEMPO radicals

Article abstract of DOI:10.1002/adsc.200404343

Oxidation of primary, benzylic and secondary alcohols into their corresponding aldehydes and ketones with safe, inexpensive oxidants was achieved in good yields under mild conditions in the presence of catalytic amounts of 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO) radicals bearing perfluoroalkyl substituents. These "fluorous-tagged" TEMPOs were readily isolated from the reaction products by liquid-liquid or solid-phase extraction, considerably simplifying the purification step. Their recyclability was strongly influenced by the nature of the oxidizing system. The best results were obtained using either [bis(acetoxy)iodo]benzene (BAIB) or aqueous NaOCl as the primary oxidants. Fluorous TEMPO 10 could be reused up to six times in the BAIB oxidation of 1-octanol with only minor loss of catalytic activity.

Full text of DOI:10.1002/adsc.200404343

FULL PAPERS
Selective Oxidation of Alcohols to Carbonyl Compounds
Mediated by Fluorous-Tagged TEMPO Radicals
Orsolya Holczknecht, Marco Cavazzini, Silvio Quici, Ian Shepperson,
Gianluca Pozzi*
CNR-Istituto di Scienze e Tecnologie Molecolari (ISTM), Via Golgi 19, 20133 Milano, Italy
Fax: (þ39)-2-503-14159, e-mail: gianluca.pozzi@istm.cnr.it
Received: November 9, 2004; Accepted: February 4, 2005
Dedicated to Professor Fernando Montanari on the occasion of his 80th birthday
Abstract: Oxidation of primary, benzylic and secon- was strongly influenced by the nature of the oxidizing
dary alcohols into their corresponding aldehydes system. The best results were obtained using either
and ketones with safe, inexpensive oxidants was ach- [bis(acetoxy)iodo]benzene (BAIB) or aqueous
ieved in good yields under mild conditions in the pres- NaOCl as the primary oxidants. Fluorous TEMPO
ence of catalytic amounts of 2,2,6,6-tetramethylpiper- 10 could be reused up to six times in the BAIB oxida-
idine-1-oxyl (TEMPO) radicals bearing perfluoroalk- tion of 1-octanol with only minor loss of catalytic ac-
yl substituents. These “fluorous-tagged” TEMPOs tivity.
were readily isolated from the reaction products by
liquid-liquid or solid-phase extraction, considerably Keywords: alcohols; aldehydes; ketones; oxidation;
simplifying the purification step. Their recyclability TEMPO
Introduction
ondary alcohols are converted to carbonyl compounds
in high yields, even in large-scale operations. In addition,
The selective oxidation of primary and secondary alco- the oxidation of primary alcohols can be driven to give
hols into the corresponding carbonyl compounds is carboxylic acids by adding a phase-transfer catalyst to
one of the most important processes in organic synthesis the biphasic aqueous/organic system.[13] Inspired by
both at a laboratory and industrial scale.[1] Well-estab- Montanariꢀs work, various research groups have then
lished methods for this functional group transformation explored the use of a wealth of other organic and inor-
involve the use of stoichiometric amounts of either inor- ganic terminal oxidants {e.g., [bis(acetoxy)iodo]ben-
ganic oxidants [e.g., chromium(VI) salts][2] or organic zene (BAIB),[14] trichloroisocyanuric acid (TCCA),[15]
oxidants (e.g., activated DMSO).[3] However, these Oxone,[16] or iodine [17]} and anhydrous reaction condi-
methods hardly satisfy the current demand for non-pol- tions with the aim of further expanding the already
luting chemical processes of high atom efficiency, and wide applicability of the original procedure.[7] A differ-
new selective procedures which do not generate large ent approach, pioneered by Semmelhack and co-work-
amounts of by-products are highly desirable.[4,5] This ers,[11] consists of the use of oxygen as the primary oxi-
trend is illustrated by the ongoing development of cata- dant in conjunction with a TEMPO/metal catalyst com-
lytic systems based on stable nitroxyl radicals such as bination.[4] The obvious economic and environmental
2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO) in com- advantages associated with the use of oxygen account
bination with safe and easy to handle primary oxi- for the increasing interest in such catalytic systems.[18,19]
dants.[6,7,8] Early examples of TEMPO-catalyzed reac-
Whichever oxidant is used, separation of TEMPO
tions included the oxidation of secondary alcohols to ke- from the products remains an issue. Moreover, TEMPO
tones with m-chloroperbenzoic acid,[9] the oxidation of is quite expensive, so it is desirable to be able to separate
primary, secondary and benzylic alcohols in an electro- the catalyst after the oxidation reaction and to reuse it.
chemical process,[10] and the oxidation of allylic and ben- Tothis end, TEMPO has been immobilized ontobothin-
zylic alcohols to aldehydes by oxygen/CuCl in DMF.[11] organic and organic polymers,[20] affording heterogene-
In 1987 Montanari and co-workers introduced a far ous catalysts, which are readily separated from the reac-
more versatile and efficient catalytic procedure in which tion mixtures, but are usually far less versatile than the
buffered bleach acts as the terminal oxidant and bro- homogeneous TEMPO.[21] A recently emerged option
mide ion as a co-catalyst.[12] The oxidation reaction pro- is the immobilization of TEMPO onto soluble polymers,
ceeds under mild conditions and both primary and sec- which makes it possible to run the oxidation reaction un-
Adv. Synth. Catal. 2005, 347, 677–688
DOI: 10.1002/adsc.200404343
ꢁ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
677

Orsolya Holczknecht et al.
FULL PAPERS
der homogeneous or liquid-liquid biphasic conditions. synthesized following standard procedures. Compound
Precipitation of the catalyst is then induced by dilution 1 was prepared from the commercially available precur-
of the reaction mixture with an incompatible solvent. sors
4-hydroxy-2,2,6,6-tetramethylpiperidine-1-oxyl
Soluble polymer-supported TEMPO radicals have radical (5; 4-hydroxy-TEMPO) and pentadecafluorooc-
been prepared by ring-opening methathesis of norbor- tanoyl chloride (6) as described in the literature.[26] Rad-
nene derivatives.[22] Preliminary tests revealed that the ical 2 was obtained in 40% yield via esterification of 5
activity of these catalysts was consistently lower than with 3-(n-perfluorooctyl)propionyl chloride (7)[27] in
that of TEMPO in the oxidation of alcohols under Mon- dry Et2O in the presence of Et3N. The fluorous radicals
tanariꢀs conditions. Promising results have been ob- 3 and 4 were analogously prepared in 28% and 30%
tained in the oxidation of alcohols with various primary yields, respectively, from 4-amino-2,2,6,6-tetramethyl-
oxidants (including bleach) catalyzed by a TEMPO de- piperidine-1-oxyl (8; 4-amino-TEMPO) and the corre-
rivative tethered onto commercially available poly- sponding fluorous acyl chlorides.
(ethylene glycol) (PEG) of molecular weight of about
As expected, the light fluorous TEMPO derivatives
5000 Da.[23] The catalyst was easily recovered by selec- 1–4 are freely soluble in conventional organic sol-
tive precipitation from Et2O and its recyclability was vents, such as CH2Cl2 and MeOH, and their partition
studied using the oxidation of 1-octanol with the mild coefficients
P in perfluorocarbon/organic solvent
oxidant BAIB as a model reaction. An independent biphasic mixtures are<1 (P¼[Radical]Perfluorocarbon
/
study on the catalytic performance and recyclability of [Radical]Organic Solvent). As a higher fluorine content and
a series of PEG-supported TEMPO derivatives in the an increased number of perfluoroalkyl substituents RF
oxidation of alcohols under Montanariꢀs conditions
has been recently disclosed.[24] Low molecular weight
PEG supports afforded highly active catalytic systems,
but poor recovery yields by precipitation with Et2O.
On the other hand, TEMPO tethered onto PEGs with
a molecular weight>5000 Da were efficiently recov-
ered, but a pronounced reduction in catalytic activity
was observed, especially upon recycling. It appears
that coiling of the PEG polymeric backbone around
the TEMPO moiety might prevent the access of the sub-
strate to the active site of the catalyst.
Recyclable TEMPO catalysts can be designed without
recourse to polymeric supports with all their drawbacks.
Indeed, we recently developed a TEMPO derivative
bearing perfluoroalkyl substituents (a “fluorous-tag-
ged” TEMPO) which is a selective catalyst for the oxida-
tion of alcohols under mild, homogeneous conditions.[25]
The peculiar solubility properties ensured by the pres-
ence of the highly fluorinated domain allowed the easy
recovery of the catalyst by liquid-liquid extraction of
the reaction mixture. In this paper the preparation of a
series of fluorous-tagged TEMPO radicals and their ac-
tivity as catalysts in the oxidation of alcohols are fully ac-
counted. The influence of the linkage between the
TEMPO moiety and the fluorous domain on the selec-
tivity of the catalytic system, the use of various primary
oxidants and the recoverability of the different catalysts
are also discussed.
Figure 1. Light fluorous TEMPO radicals.
Results and Discussion
Synthesis of Fluorous-Tagged TEMPO Radicals
The simplest TEMPO derivatives 1–4 (Figure 1) bear-
ing just one fluorous ponytail and having a similar
weight percentage of fluorine (about 50%) were readily Scheme 1.
678
ꢁ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2005, 347, 677–688

Selective Oxidation of Alcohols to Carbonyl Compounds
FULL PAPERS
as depicted in Scheme 2. 1,3,5-Trichlorotriazine (14) was
treated with 4-hydroxy-TEMPO (5) in dry toluene at
room temperature, in the presence of powdered KOH
as a base, to give 15 in 45% yield. This compound had
been previously prepared in lower yields following a dif-
ferent procedure.[32] A two-fold excess of fluorous amine
´
17, prepared according to a modification of Rabaiꢀs
method,[33] was then reacted with 15 in boiling THF.
The ammonium salt 17 · HCl formed thereby was easily
separated by filtration of the cold reaction mixture and
neutralized with aqueous NaOH to give 17, which could
be recycled. The radical 10 was recovered in pure form
from the liquid phase in 65% yield after flash column
chromatography on silica. The non-fluorous triazine-
tethered TEMPO derivative 18 was synthesized like-
wise from 15 and an excess of di(n-octylamine).
Despite the similar fluorine content, the fluorous-tag-
ged radicals 9 and 10 showed different affinities for per-
fluorocarbons, as indicated by partition coefficient
measurements between perfluoro-1,3-dimethylcyclo-
hexane (PDMC) and standard organic solvents (Ta-
ble 1). Whereas 9 is only slightly preferentially soluble
in PDMC than in organic solvents, 10 partitions almost
completely in the fluorous phase. Fortunately, 10 is
also readily soluble in ethers such as tert-butyl methyl
ether (MTBE) and, at low concentrations, in CH2Cl2.
This residual affinity for organic solvents increases
with the temperature and greatly simplified the search
for convenient reaction conditions when 10 was tested
as a catalyst in the oxidation of alcohols. The high affin-
ity of 9 for AcOH is in agreement with the basic nature
of the nitrogen atom in the 4-position, which is efficient-
ly shielded from the electron-withdrawing perfluoroalk-
yl substituents thanks to the interposition of –(CH2)3–
spacers.[34]
Scheme 2.
are crucial to enhance the affinity for the fluorous phase,
the TEMPO radicals 9 (F¼59.9%) and 10 (F¼60.0%) Catalytic Activity
bearing two and four RF, respectively, were also synthe-
sized.
The ability of the fluorous-tagged TEMPO radicals
4-Amino-TEMPO (8) was reacted with an excess of 3- (1 mol % with respect to the substrate) to catalyze the
(n-perfluorooctyl)propyl iodide (11)[28] in boiling CH3
CN in the presence of K2CO3 as a base (Scheme 1). Al-
though efficient dialkylation of certain primary amines
has been achieved under similar conditions,[29] many pri-
mary amines react with 11 to give secondary amines as
the only products.[30] In the present case bis-N-alkylation
prevailed and the fluorous-tagged TEMPO 9 was ob-
tained as the major reaction product (yield¼50%). It
should be noted that also 4-amino-2,2,6,6-tetramethyl-
piperidine (12) was selectively bis-N-alkylated at the
primary nitrogen, affording 13 in 68% yield. However,
our attempts to convert 13 into 9 using standard condi-
tions for the oxidation of 2,2,6,6-tetramethylpiperidine
to TEMPO failed.[31]
Table 1. Partition coefficients P for fluorous-tagged TEMPO
radicals 9 and 10 between perfluoro-1,3-dimethylcyclohexane
(PDMC) and standard organic solvents.[a]
Organic Solvent
P [X]PDMC/[X]Org. Solv
.
9
10
Toluene
CH2Cl2
CH3CN
CH3OH
AcOH
3.1
1.4
14.0
7.0
38.0
11.3 (12.3)
>100 (>100)
20.5 (19.3)
0.1
14.5 (15.3)
[a]
Determined gravimetrically (Method A, Experimental
Section). Values determined by UV-Vis spectroscopy
(Method B, Experimental Section) in parentheses.
The nitroxyl radical 10 with four fluorous tags stem-
ming from a triazine core was conveniently synthesized
Adv. Synth. Catal. 2005, 347, 677–688
asc.wiley-vch.de
ꢁ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
679

Orsolya Holczknecht et al.
FULL PAPERS
Table 2. Oxidation of 1-octanol to octanal with aqueous
Dismutation of TEMPO catalyzed by acetic acid issued
from the interaction between BAIB and the alcohol
then gives TEMPOH and the active oxammonium
salt.[14] In both cases the choice of the solvent is crucial.
Both TCCA and BAIB afford good results in terms of
reaction rates and selectivity for the carbonyl compound
when CH2Cl2 is used. In the case of BAIB reactions car-
ried out in other solvents, such as MTBE, are slug-
gish.[14,23] On the other hand, oxidation of alcohols with
TCCA/TEMPO to give carboxylic acids has been re-
ported in wet acetone.[15c] The behavior of the new cata-
lysts in the presence of these two primary oxidants was
thus investigated in CH2Cl2
When the fluorous-tagged TEMPO radicals (1 mol %
with respect to the substrate) were used in combination
with TCCA as the terminal oxidant (Table 3) under an-
hydrous conditions, reaction rates were similar to those
observed with aqueous NaOCl, but overoxidation of 1-
octanol to octanoic acid occurred to a larger extent.
Such a finding was not particularly surprising, since
NaOCl.[a]
Entry Catalyst t [min] Conversion [%] Selectivity [%]
1
2
3
4
1
2
3
4
9
10
10
10
15
15
10
15
15
10
10
30
98
96
>99
97
>99
>99
>99
95
98
>99
98
>99
>99
95
94
96
>99
5
6[b]
7[c]
8[d]
9
TEMPO 10
>99
[a]
Reaction conditions: Oxidant/S/C/KBr¼125/100/1/10. T¼
08C. Solvent¼CH2Cl2. Selectivity and conversions deter-
mined by GC (internal standard method).
T¼208C.
[b]
[c]
[d]
Solvent¼MTBE.
Solvent¼MTBE, bromide-free conditions.
oxidation of alcohols in combination with KBr the oxidation of aldehydes to acids promoted by
(10 mol %) and a slight excess of buffered bleach TCCA can occur also in the absence of nitroxyl radi-
(pH¼8.6) as the terminal oxidant,[12] was first evaluated cals.[15c] The addition of a base such as CH3COONa part-
using 1-octanol as a model substrate (Table 2). The ob- ly prevents this side-reaction (entries 1 and 3 vs. entries 2
served activity and selectivity of the fluorous-tagged cat- and 4),[15a] which had scarce relevance in the TEMPO-
alysts in the biphasic system CH2Cl2/H2O were generally catalyzed oxidation of 1-octanol (entry 8). The triazine
similar to those of TEMPO. Conversions higher than derivative 10 turned out to be the most selective among
95% were reached in 10–15 minutes, with almost com- the fluorous-tagged catalysts tested (entry 7), while rad-
plete selectivity for the aldehyde. However, a small ical 9 afforded large amounts of octanoic acid even in the
amount of octanoic acid was detected when 10 was em- presence of NaOAc. It should be noted that primary
ployed (entry 6). This cannot be avoided either by using amines are readily oxidized to nitriles by a combination
a different solvent such as tert-butyl methyl ether of TCCA and TEMPO.[35] Although the fate of 9 in the
(MTBE) (entry 7) or by working under bromide-free oxidation of 1-octanol with TCCA has not been investi-
conditions (entry 8). It should be noted that, although gated, we suppose that the presence of a tertiary amino
slightly decreased, the oxidation rate remained high group could facilitate degradation of the catalyst, thus
even in the absence of KBr and the same was observed enhancing the relative importance of the non-catalyzed
with all the new catalysts. This is quite interesting in TCCA oxidation of the intermediate aldehyde. The ni-
view of large scale operations, where bromide-free con- trogen atoms attached to the triazine ring in 10 are
ditions and the use of MTBE instead of CH2Cl2 would be
desirable.
Table 3. Oxidation of 1-octanol to octanal with TCCA.[a]
Besides aqueous NaOCl, [bis(acetoxy)iodo]benzene
(BAIB) and trichloroisocyanuric acid (TCCA) were
also tested as stoichiometric oxidants in the model reac-
tion. They were selected not only on the basis of the ex-
cellent results obtained with TEMPO, but also to evalu-
ate whether the insertion of fluorous tags could interfere
with different regeneration modes of the active oxidiz-
ing species. Indeed, TCCA and BAIB play different
roles in the oxidation of alcohols catalyzed by TEMPO.
TCCA, as well as OClꢀ and OClꢀ/Brꢀ, reacts with
TEMPO to form an N-oxammonium salt, which oxidiz-
es the alcohol to the corresponding carbonyl compound
giving the reduced form of TEMPO, the hydroxylamine
TEMPOH. The latter is then reoxidized by TCCA to re-
generate the oxoammonium cation.[15] BAIB is a mild
organic oxidant which is unable to oxidize TEMPO di-
rectly, but which regenerates TEMPO from TEMPOH.
Entry
Catalyst
Conversion [%]
Selectivity [%]
1
1
1
2
2
3
4
9
10
93
>99
95
>99
>99
>99
>99
>99
>99
74
97
78
93
86
90
57
98
95
2[b]
3
4[b]
5[b]
5[b]
6[b]
7[b]
8
TEMPO
[a]
TCCA¼trichloroisocyanuric acid. Reaction conditions:
oxidant/S/C¼105/100/1, T¼08C, solvent¼CH2Cl2. Selec-
tivity and conversions were determined by GC (internal
standard method) after a reaction time of 10 min.
In the presence of CH3COONa (3 mol equiv. with respect
to TCCA).
[b]
680
ꢁ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2005, 347, 677–688

Selective Oxidation of Alcohols to Carbonyl Compounds
FULL PAPERS
poor electron donors and are much less susceptible to
oxidative attack.[36]
The oxidation of 1-octanol with BAIB proceeded
slower than with TCCA or bleach independently of
the catalyst used, requiring at least two hours to go to
completion (Table 4) with a higher catalyst loading
(5 mol % with respect to the substrate). This is clearly
a consequence of the different reaction mechanism.
Overall reaction rates were generally similar to those
observed with TEMPO, but it should be noted that
rate profiles and selectivities (Figures 2 and 3) were
greatly influenced by the nature of the bond connecting
the fluorous domain with the rest of the molecule.
Initial reaction rates observed in the presence of the
light fluorous radicals 1 – 4 and with 9 were higher
than that observed with TEMPO (Figure 2). In particu-
lar, 3 and 4 featuring an amido bond gave TOFs at 30 mi-
nutes¼0.40 minꢀ1 and 0.55 minꢀ1, respectively, against
TOF at 30 minutes for TEMPO¼0.15 minꢀ1. Although
the oxidation catalyzed by 10 started quite slowly (TOF
Figure 3. Oxidation of 1-octanol to octanal with BAIB. Selec-
at 30 minutes¼0.07 minꢀ1), conversion after 2 hours tivity profiles for reactions mediated by fluorous-tagged
TEMPO radicals.
was close to that obtained with TEMPO (Table 4, en-
tries 7 and 9). It is worth noting that the non-fluorous tri-
azine-tethered TEMPO derivative 18 gave TOF at 30
minutes¼0.43 minꢀ1 and overall conversion and selec- entry 6) than in CH2Cl2 (entry 7). It has been shown that
tivity similar to that of 10 (Table 4, entries 7 and 8). alcohols are selectively oxidized with oxygen under FB
The induction time observed with 10 is thus related to conditions in the presence of catalytic amounts of TEM-
the heavy fluorous nature of this catalyst, with possible PO and fluorous copper(I) complexes at 908C.[38,39]
A
effects in terms of lipophobicity, self-aggregation and single homogeneous phase is formed at such a tempera-
formation of micro- and nanophases (“fluorous self-as- ture,[39] overcoming mass transfer limitations. In the
sembly”),[37] and not to its specific design. Finally, the ox- present case, the system remained biphasic at room tem-
idation of 1-octanol catalyzed by 10 proceeded more perature hampering the intimate contact between the
slowly under fluorous biphasic (FB) conditions (Table 4, fluorous catalyst and the organic species.
As shown in Figure 3, the selectivity for the aldehyde
in reactions catalyzed both by TEMPO and fluorous
radicals 1, 2, 9 and 10 was almost constant in time (>
97%). Radicals 3 and 4 initially afforded good aldehyde
selectivities (98% and 94% at 30 minutes for 3 and 4, re-
spectively), but after 30 minutes formation of octanoic
Table 4. Oxidation of 1-octanol to octanal with BAIB.[a]
Entry
Catalyst
Conversion [%]
Selectivity [%]
1
2
3
1
2
3
>99
>99
91
>99
>99
74
4
4
9
10
10
18
TEMPO
>99
>99
56
93
94
71
98
>99
>99
>99
>99
5
6[b]
7
8
9
93
[a]
BAIB ¼[bis(acetoxy)iodo]benzene. Reaction conditions:
oxidant/S/C¼110/100/5, T¼208C, solvent¼CH2Cl2. Se-
lectivity and conversions were determined by GC (internal
standard method) after a reaction time of 2 h.
Figure 2. Oxidation of 1-octanol to octanal with BAIB. Con-
version profiles for reactions mediated by fluorous-tagged
TEMPO radicals.
[b]
Fluorous biphasic (FB) conditions: CH2Cl2/PDMC.
Adv. Synth. Catal. 2005, 347, 677–688
asc.wiley-vch.de
ꢁ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
681

Orsolya Holczknecht et al.
FULL PAPERS
acid was observed and selectivities dropped. It appeared 11, 12 and 14). Longer reaction times failed to give com-
that the aldehyde initially formed was partly oxidized by plete conversion of those substrates (entry 14) due to the
the excess of primary oxidant still present in solution. progressive bleaching of the catalysts. Indeed, in the
ꢀ
ꢀ
The acidity of the NHCO group, which is enhanced presence of buffered NaOCl/KBr the stability of the ac-
by the presence of the electron-withdrawing perfluor- tive oxoammonium species at 208C is much lower than
oalkyl chain,[40] could thus influence the outcome of at 08C.[12] A number of techniques complementary to
the reaction, as previously found in the NaOCl-promot- the classic FB approach have been developed in the
ed oxidation of primary alcohol groups in carbohydrates last few years, providing new options for separating flu-
to carboxylic acids catalyzed by 4-acetamido-TEM- orous molecules from reaction mixtures.[42] Selective re-
PO.[41]
tention of fluorous-tagged molecules on fluorous re-
The substrate scope of the fluorous-tagged TEMPO verse phase silica gel (fluorous solid-phase extraction,
radicals was next evaluated in reactions promoted by F-SPE)[43] has found widespread application for quick
aqueous NaOCl or BAIB, the oxidants which gave the separations of mixtures involving fluorous reagents
best results in the model reaction.
and scavengers,[44] including light fluorous ones, and is
The light fluorous catalysts 1–4 behaved quite similar- now attracting increasing attention in the field of fluo-
ly in the presence of NaOCl, affording fast and selective rous catalysis.[45] The separation and recovery of the flu-
oxidation of primary, secondary and benzylic alcohols to orous-tagged TEMPO derivatives by F-SPE was investi-
their corresponding carbonyl derivatives. All reactions gated for the oxidation of cycloctanol with aqueous
were carried out at 08C in the presence of 1 mol % of NaOCl. After completion of the reaction the aqueous
catalyst and 1.25 mol. equivs. of terminal oxidant with layer was discarded and the organic solution was loaded
respect to the substrate. Under these conditions, most al- onto a commercially available fluorous reverse phase
cohols were quantitatively oxidized within 15 min, as silica gel cartridge. Cyclooctanone was eluted with a
found in the case of 1-octanol. Selected data for 1 and small amount of CH3CN, an organic solvent showing lit-
3 are reported in Table 5 (entries 1–8). The solubility tle affinity for fluorous compounds, and recovered after
of radical 10 in CH2Cl2 is strongly affected by the tem- evaporation of the solvent (Table 5, entries 4, 8 and 13).
perature, thus NaOCl oxidations were performed at Isolated yields of about 80% were obtained, which are
208C in order to dissolve the radical in the organic layer somewhat lower than GC yields. This is due to a partial
(Table 5, entries 9–14). Oxidation of primary, benzylic loss of the volatile product during the final evaporation
and cyclic secondary alcohols proceeded almost quanti- of CH3CN under reduced pressure. The light fluorous
tatively in only 10–15 minutes. However, as previously radical 1 and the heavy fluorous radical 10 were retained
demonstrated in the case of TEMPO,[12] reactions were selectively as an orange band at the top of the cartridge,
slowed down by increasing the temperature and this is whereas the light fluorous radical 3 moved down, al-
clearly shown by the relatively low conversions achieved though slower than cyclooctanone. Indeed, the reaction
when secondary alcohols were oxidized (Table 5, entries product was free from any fluorous residue as shown by
Table 5. Oxidation of alcohols to carbonyl compounds with aqueous NaOCl.[a]
Entry
Catalyst
Substrate
Product
t [min]
Conversion [%][b]
Selectivity [%]
1
2
3
4
5
6
7
8
1
1
1
1
3
3
3
3
4-Bromobenzyl alcohol
1-Undecanol
2-Octanol
Cyclooctanol
4-Bromobenzyl alcohol
1-Undecanol
2-Octanol
Cyclooctanol
4-Bromobenzyl alcohol
1-Undecanol
2-Undecanol
4-Bromobenzaldehyde
Undecanal
2-Octanone
Cyclooctanone
4-Bromobenzaldehyde
Undecanal
10
15
15
15
10
15
15
15
10
15
30
15
15
15
60
120
>99
>99
91
99 (80)
>99
>99
>99
>99 (77)
>99
96
95
92
99 (81)
77
92
98
98
98
>99
>99
98
>99
>99
97
2-Octanone
Cyclooctanone
4-Bromobenzaldehyde
Undecanal
2-Undecanone
2-Octanone
9
10[c]
10[c]
10[c]
10[c]
10[c]
10[c]
10
11
12
13
14
98
>99
>99
>99
>99
2-Octanol
Cyclooctanol
1-Phenylethanol
Cyclooctanone
Acetophenone
97
[a]
Reaction conditions: see Table 2.
Isolated yields in parentheses.
T¼208C.
[b]
[c]
682
ꢁ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2005, 347, 677–688

Selective Oxidation of Alcohols to Carbonyl Compounds
FULL PAPERS
the absence of signals in the 19F-NMR. By a simple sol- organic phase containing cyclooctanone with PDMC af-
vent switch to Et2O, it was possible to recover 1 (89% forded better results than F-SPE. In the first three runs,
of the starting material) and 10 (75%), although not 95%, 96% and 100% of the fluorous catalyst introduced
quantitatively. Recycling of these catalysts was attempt- was recovered, respectively. Only in the fourth run was
ed (Table 6), with moderate success in the case of 10 but an evident decrease of catalytic activity observed,
less satisfactorily with 1. The sudden drop in catalytic ac- matched by a loss of recovered catalyst (83%).
tivity observed with 1 can be accounted for by the cata-
The use of readily available, easy to handle and low
lyst undergoing degradation, since the orange color toxic trivalent iodine reagents such as BAIB is par-
characteristic of TEMPO derivatives disappeared dur- ticularly appealing for small-scale oxidation of alcohols
ing the second run. In the case of 10, the high fluorophi- because of the high yields and chemoselectivities ob-
licity of this compound allowed us to compare F-SPE to tained under mild reaction conditions.[46] Moreover,
another fluorous quick work-up method, namely liquid- most oxidation procedures developed for BAIB and re-
liquid extraction of the reaction mixture with a perfluor- lated compounds have been extended to polymer-sup-
ocarbon (F-LLE). As reported in Table 6, F-LLE of the ported hypervalent iodine reagents which can be recov-
ered by filtration and then regenerated.[47] In this con-
text, there is a strong interest for easily removable and
possibly recyclable TEMPO derivatives able to catalyze
BAIB oxidations. The oxidation of representative alco-
hols with BAIB in CH2Cl2 was thus investigated at 208C
Table 6. Recycling of F-tagged TEMPO: oxidation of cy-
clooctanol with NaOCl.[a]
Run Catalyst Method[b] Conversion [%] Selectivity [%]
in the presence of 5 mol % of fluorous catalyst and
1
2
1
2
3
1
2
3
4
1
F-SPE
F-SPE
99
60
99
84
63
99
90
83
64
>99
>99
>99
>99
>99
>99
>99
>99
>99
1.1 mol equivs. of terminal oxidant with respect to the
substrate (Table 7). On the basis of the preliminary ex-
periments with 1-octanol and the results obtained with
aqueous NaOCl, only 1 was tested among the light fluo-
rous radicals available.
The amount of fluorous radicals (1, 9, or 10) used in
BAIB-promoted oxidations was lower than that gener-
ally used in similar reactions catalyzed by TEMPO
(10 mol %).[14] Nevertheless, primary and benzylic alco-
hols were smoothly oxidized to the corresponding alde-
hydes in 2 hours (Table 7, entries 1, 2, 4–6, 9–12). The
oxidation of secondary alcohols (entries 3, 4, 7, 8, 13–
10[c]
10[c]
F-LLE
[a]
[b]
Reaction conditions: see Table 2. Reaction time¼15 min.
F-SPE¼fluorous solid phase extraction; F-LLE¼fluorous
liquid-liquid extraction.
T¼208C.
[c]
Table 7. Oxidation of alcohols to carbonyl compounds with BAIB.[a]
Entry
Catalyst
Substrate
Product
t [h]
Conversion[b] [%]
Selectivity [%]
1
2
3
1
1
1
1-Undecanol
4-Bromobenzyl alcohol
2-Octanol
Undecanal
4-Bromobenzaldehyde
2-Octanone
2
2
6
>99
>99
98
98
98
98
4
5
6
7
8
9
1
9
9
9
Cyclooctanol
1-Undecanol
4-Bromobenzyl alcohol
2-Octanol
Cyclooctanol
Cyclooctanone
Undecanal
4-Bromobenzaldehyde
2-Octanone
Cyclooctanone
Undecanal
4-Bromobenzaldehyde
Benzaldehyde
Cinnamaldehyde
2-Octanone
6
2
2
6
6
2
2
2
96
98
>99
>99
46[c]
>99
>99
>99
>99
>99
>99
>99
76
>99
>99
>99
>99
9
69[d]
10
10
10
10
10
10
10
10
1-Undecanol
>99 (80)
>99 (97)
>99
>99
92
91 (78)
83
93 (88)
10[e]
11
12
13
14[e]
15[e]
16[e]
4-Bromobenzyl alcohol
Benzyl alcohol
Cinnamyl alcohol
2-Octanol
Cyclooctanol
2-Undecanol
2
24
14
24
6
Cyclooctanone
2-Undecanone
Acetophenone
1-Phenylethanol
[a]
Reaction conditions: see Table 4.
Isolated yields in parentheses.
Conversion at 12 h¼51%
[b]
[c]
[d]
[e]
Conversion at 12 h¼73%
Using catalyst 10 recovered from the reaction summarized in the previous entry.
Adv. Synth. Catal. 2005, 347, 677–688
asc.wiley-vch.de
ꢁ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
683

Orsolya Holczknecht et al.
FULL PAPERS
16) proceeded more slowly, the reaction rates depending Conclusion
on the steric hindrance of the substrate analogously to
what was found with TEMPO.[15] In addition, the oxida- Fluorous-tagged TEMPO derivatives synthesized from
tion of secondary alcohols was found to be quite sensi- easily available precursors were used as catalysts for
tive to the nature of the fluorous catalyst employed: the oxidation of a variety of alcohols under mild condi-
nearly quantitative conversions of 2-octanol and cyclo- tions, affording results very similar to those obtained
octanol were achieved in 6 hours in the presence of the with TEMPO. Reactions were performed in conven-
light fluorous TEMPO 1 (entries 3, 4), whilst the oxida- tional organic solvents under homogeneous or liquid-
tion of the same alcohols proceeded sluggishly in the liquid aqueous-organic conditions, and fluorous separa-
presence of the 4-amino-substituted TEMPO 9 (entries tion techniques were applied to isolate the fluorous-tag-
7, 8). After an initial induction time, the heavy fluorous ged radicals from the organic products. Promising re-
radical 10 gave slow but steady reactions (entries 13 and sults were obtained using popular primary oxidants
14). It is worth noting that 10 isolated in a few minutes by such as aqueous NaOCl and, in particular, BAIB. With
F-LLE of the reaction mixture containing 2-octanone the latter, the heavy fluorous radical 10 could be reused
(entry 13) could be successfully reused (entry 14). This up to six times in the oxidation of 1-octanol showing only
recovery/recycling process was sequentially applied (en- minor loss of catalytic activity. The new fluorous-tagged
tries 14–16), allowing the oxidation of other secondary radicals thus share the advantages usually associated
alcohols to the corresponding ketones in good yields. with the use of polymer-supported TEMPO derivatives
The quick separation of the catalyst also simplified the (simplified work-up procedures, quick recovery, recy-
recovery of the products, which were isolated pure after clability) without some of their common limitations
column chromatography using a short pad of silica gel. (lack of versatility, mass transfer limitations, poor acces-
Recovery/recycling of radicals 1 and 9 after F-SPE of sibility to the active site). Results obtained in the present
the reaction mixture was also possible, but less efficient, work are also intended to provide further guidelines for
as shown in the case of the oxidation of 1-octanol (Ta- the rational design of fluorous-tagged stable nitroxyl
ble 8).
radicals and for the development of all-fluorous catalyt-
Thanks to the milder oxidizing environment, the serv- ic systems. In this context, the combination of fluorous-
ice lifetime of radical 1 in BAIB-promoted reactions is tagged TEMPO derivatives with a recyclable fluorous
longer than that of the same radical in the presence of version of BAIB[48] and the use of molecular oxygen as
aqueous NaOCl. However, degradation of 1 (and 9) still the primary oxidant are worth investigating.
occurs, thus limiting the number of recycles. This can be
improved by choosing a different fluorous-tagged TEM-
PO radical such as 10, which was recovered by F-LLE
Experimental Section
(98% of the introduced amount for each recovery) and
reused five times with a minor loss of catalytic activity
observed in the 6th run (Table 8).
General Remarks
Solvents were purified by standard methods and dried if neces-
sary, except perfluoro-(1,3-dimethyl)cyclohexane (Apollo Sci-
entific Ltd., UK) that was used as received. 4-Hydroxy-TEM-
PO (5) pentadecafluorooctanoyl chloride (6), 4-amino-TEM-
PO (8) were purchased from Sigma-Aldrich and used as re-
ceived. 3-(n-Perfluorooctyl)propionyl chloride (7),[27] 3-(n-per-
fluorooctyl)propyl iodide (11),[28] and the fluorous-tagged
TEMPO 1 were prepared as described in the literature.[26] Flu-
orous amine 17,[33,34] and triazine-tethered TEMPO 15 are
known compounds.[32] The original synthetic procedures have
been conveniently modified as described below. Reactions
were monitored by TLC on silica gel 60 F254. Column chroma-
tography was carried out on silica gel SI 60 (Merck, Germany),
0.063–0.200 mm (normal) or 0.040–0.063 mm (flash). Fluo-
rous solid phase extraction (F-SPE) on fluorous reverse phase
silica gel was carried out on FluoroFlashꢂ SPE cartridges, 5 g,
10 mL (Fluorous Technologies Inc., USA). Melting points (un-
corrected) were determined with a capillary melting point ap-
paratus Bꢃchi SMP-20. 1H NMR (300 MHz), 13C NMR
(75.4 MHz) and 19F NMR (282 MHz) spectra were recorded
in CDCl3 on a Bruker AC 300 spectrometer with tetramethyl-
silane (d¼0), CDCl3 (d¼77) and CFCl3 (d¼0) as internal
standards, respectively. TEMPO derivatives are paramagnetic
compounds giving NMR spectra with only broad signals that
Table 8. Recycling of F-tagged TEMPO: oxidation of 1-octa-
nol with BAIB.[a]
Run Catalyst Method[b] Conversion [%] Selectivity [%]
1
1
F-SPE
>99
>99
81
>99
>99
91
21
40
1
1
9
F-SPE
F-LLE
>99
>99
72
93
92
93
87
91
98
98
55
>99
>99
>99
>99
98
10
85
98
[a]
[b]
Reaction conditions: see Table 4. Reaction time¼2 h.
F-SPE¼fluorous solid phase extraction; F-LLE¼fluorous
liquid-liquid extraction.
684
ꢁ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2005, 347, 677–688

Selective Oxidation of Alcohols to Carbonyl Compounds
FULL PAPERS
are not reported. ElectroSpray ionization (ESI) mass spec-
trometry: Bruker APEX II ICR-FTMS mass spectrometer
(source: nano ESI at 458C). FT-IR spectra were recorded on
a Perkin Elmer 1725X instrument. GC analyses were per-
formed on an Agilent 6850 instrument (column: HP-1 100% di-
methylpolysiloxane 30 mꢁ320 mmꢁ0.25 mm); carrier¼He
(constant flow); mode¼split (split ratio¼80:1); injector T¼
2508C; detector (FID) T¼2808C. The products of the oxida-
tion reactions were determined by comparison with the com-
mercially available carbonyl compounds and carboxylic acids.
Elemental analyses: Departmental Service of Microanalysis
(University of Milano).
the aqueous layer extracted three times with Et2O. The com-
bined organic layer was washed with water, brine and dried
over Na2SO4. The crude amide was purified bycolumn chroma-
tography (silica gel, hexane/Et2O 4/1) to give the title com-
pound as a light orange solid; yield: 0.16 g (28%); mp 99–
ꢀ
1028C. IR (KBr, selected data): n¼3309 (N H), 1651
(C O), 1367 (N O) cmꢀ1; anal. calcd. for C20H22F17N2O2
(645.37): C 37.22, H 3.44, N 4.34; found: C 37.21, H 3.38, N 4.13.
¼
ꢀ
(C8F17C3H6)2N-TEMPO (9)
In a Schlenk tube, a mixture of 8 (0.17 g, 1 mmol), K2CO3
(0.43 g, 3 mmol) and 3-(n-perfluorooctyl)propyl iodide (11;
2.21 g, 3.75 mmol) in dry CH3CN (10 mL) was stirred under ni-
trogen for 60 h at 808C. The suspension was cooled to room
temperature and diluted with Et2O (30 mL). After filtration
of the solid, the solvents were evaporated, and the product
was purified by column chromatography (silica gel, hexane/
Et2O, 9/1 to recover the excess of 11, then hexane/Et2O, 7/3) af-
fording the title compound as an orange solid; yield: 0.54 g
(50%); mp 65.5–66.58C. IR (KBr, selected data): n¼2981
C8F17C2H4COO-TEMPO (2)
3-(n-Perfluorooctyl)propionyl chloride (7; 1.02 g, 2.00 mmol)
was dissolved in dry Et2O and added dropwise to a solution
of 4-hydroxy-TEMPO (5; 0.31 g, 1.83 mmol) and Et3N
(0.31 mL, 2.23 mmol) in Et2O (10 mL) at 08C. The reaction
mixture was stirred at 08C for one hour then allowed to
warm to room temperature and then stirred overnight. Water
(10 mL) was added to the reaction and the organic phase sep-
arated, the aqueous phase was extracted twice with Et2O. The
organic phase was washed with water and brine and dried over
Na2SO4. The solvent was evaporated to leave the crude product
which was purified by column chromatography (silica gel, hex-
ane/AcOEt, 4/1) to give the title compound as a pale orange
solid; yield: 0.47 g (40%); mp 79–818C. IR (KBr, selected
(N H), 1372 (N O) cmꢀ1; anal. calcd. for C31H29F34N2O
(1091.52): C 34.11, H 2.68, N 2.57; found: C 33.96, H 2.53, N
2.81; HR-MS (ESI): m/z¼1092.18611 ([MþH]þ); calcd. for
[C31H30F34N2O]þ: 1092.18142.
ꢀ
ꢀ
Fluorous Amine (13)
data): n¼1732 (C O), 1372 (N O) cmꢀ1; anal. calcd. for
C20H21F17NO3 (646.36): C 37.16, H 3.27, N 2.17; found:
C 37.06, H 3.25, N 1.78.
¼
ꢀ
4-Amino-2,2,6,6-tetramethylpiperidine (12; 0.16 g, 1 mmol),
K2CO3 (0.43 g, 3 mmol) and dry CH3CN (6 mL) were placed
in a flame-dried Schlenk tube under nitrogen. The mixture
wasstirredfor 15 minatroomtemperature, then3-(n-perfluoro-
octyl)propyl iodide (11; 1.76 g, 3 mmol) was added and the re-
action was stirred for 48 h at 608C. The suspension was cooled
to room temperature and diluted with Et2O (30 mL). After fil-
tration of the solid, the solvents were evaporated, and the prod-
uct was purified by column chromatography (silica gel, MTBE)
affording the title compound as a white solid; yield: 0.73 g
(68%); mp 33–348C. 1H NMR (300 MHz, CDCl3): d¼1.01 (t,
J¼12.2 Hz, 2H), 1.12 (s, 6H), 1.18 (s, 6H), 1.51–1.59 (m, 2H),
1.61–1.75 (m, 4H), 1.98–2.25 (m, 4H), 2.51 (t, J¼6.7 Hz,
4H), 2.89–3.04 (m, 1H); 13C NMR (75.4, CDCl3): d¼19.8,
C7F15CONH-TEMPO (3)
Pentadecafluorooctanoyl chloride (6; 0.25 mL, 0.99 mmol)was
dissolved in dry Et2O (3 mL) and added dropwise to solution of
4-amino-TEMPO (8; 0.16 g, 0.92 mmol) and triethylamine
(0.16 mL, 1.12 mmol) in dry Et2O (3 mL). The reaction mix-
ture was stirred for an hour at 08C and then allowed to warm
to room temperature and was left stirring overnight. The reac-
tion was quenched with water (5 mL), the organic layer sepa-
rated and the aqueous layer extracted three times with Et2O.
The combined organic layer was washed with water, brine
and dried over Na2SO4. The crude amide was purified by col-
umn chromatography (silica gel, hexane/Et2O, 4/1), followed
by recrystallization from hexane and Et2O to give the title com-
pound as a light orange solid; yield: 0.14 g (28%); mp 94–978C.
2
28.9, (t, JC,F ¼21.5 Hz), 35.6, 41.4, 48.9, 51.3, 51.7, 105.0–
122.8 (m, RF); 19F NMR (282 MHz, CDCl3): d¼ ꢀ126.7 (br s,
4F), ꢀ124.3 (br s, 4F), ꢀ123.3 (br s, 4F), ꢀ122.8 to ꢀ121.7
3
3
(m, 12F), ꢀ114.7 (t, JF,F ¼14 Hz, 4F), ꢀ81.2 (t, JF,F
¼
10.5 Hz, 6 F); anal. calcd. for C31H30F34N2 (1076.53): C 34.59,
H 2.81, N 2.60; found: C 34.71, H 2.66, N 2.95.
ꢀ
¼
IR (KBr, selected data): n¼3330 (N H), 1695 (C O), 1368
(N O) cmꢀ1; anal. calcd. for C17H18F15N2O2 (567.31): C 35.99,
ꢀ
H 3.20, N 4.94; found: C 36.08, H 3.35, N 5.10.
Triazine-O-TEMPO (15)
A solution of 4-hydroxy-TEMPO (5; 0.74 g, 4.24 mmol) in dry
toluene (10 mL) was added dropwise at room temperature to a
suspension of powdered KOH (0.28 g, 5 mmol) and freshly
crystallized 1,3,5-trichlorotriazine (14; 0.92 g, 5 mmol) in dry
toluene (5 mL). The mixture was stirred for 48 h, then diluted
with CH2Cl2 (20 mL) and filtered. The liquid layer was washed
with water, brine and dried over sodium sulfate. The crude
product was purified by column chromatography (silica gel,
CH2Cl2/MeOH, 98/2) to give the title compound as an orange
C8F17C2H4CONH-TEMPO (4)
3-(n-Perfluorooctyl)propionyl chloride (7; 0.51 g, 1.00 mmol)
was dissolved in dry Et2O (3 mL) and added dropwise to solu-
tion of 8 (0.16 g, 0.92, mmol) and triethylamine (0.16 mL,
1.12 mmol) in dry Et2O (3 mL). The reaction mixture was stir-
red for an hour at 08C and then allowed to warm to room tem-
perature and was left stirring overnight. The reaction was
quenched with water (5 mL), the organic layer separated and
Adv. Synth. Catal. 2005, 347, 677–688
asc.wiley-vch.de
ꢁ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
685

Orsolya Holczknecht et al.
FULL PAPERS
solid; yield: 0.61 g (45%). Physical data were in agreement with
those reported in the literature.[32]
cal balance. The partition coefficient P was determined as the
ratio between the weight of the fluorous phase residue and the
weight of the organic phase residue.
Method B: A 10-mL vial equipped with a magnetic stirrer
was charged with the fluorous-tagged TEMPO radical
(8 mmol), perfluoro-(1,3-dimethyl)cyclohexane (2 mL) and
the organic solvent (2 mL). The mixture was thermostatted
at 258C and vigorously stirred for 4 hours. Samples (1 mL) tak-
en out of each phase were diluted 200 times with Et2O and their
UV/Vis absorbance at 226 nm was measured. The concentra-
tion of the radical in each phase was determined by comparison
with a calibration curve (concentration vs. absorbance) previ-
ously constructed with known solutions. Partition coefficients
P were in good agreement with those obtained with Method
A, as shown in Table 1 for radical 10.
(C8F17C3H6)2NH (17)
In a flame-dried Schlenk tube, a mixture of perfluoroalkyl io-
dide 11 (4.70 g, 8 mmol), K2CO3 (1.10 g, 8 mmol) and benzyla-
mine (0.42 mL, 3.85 mmol) in dry CH3CN (20 mL) was stirred
under nitrogen for 30 h at 808C. The suspension was cooled to
room temperature and diluted with Et2O (40 mL). After filtra-
tion of the solid, the solvents were evaporated, and the product
was purified by column chromatography (silica gel, hexane to
recover the excess of 11, then hexane/Et2O 9/1), affording
the tertiary amine 16 as a pale yellow oil; yield: 3.58 g (91%).
The benzyl group was cleaved by hydrogenolysis as described
by Gladysz and co-workers to give the title compound as a
white solid; yield: 3.21 g (98%).[34] Physical data of 16 and 17
were in agreement with those reported in the literature.
Typical Oxidation Procedures
Reactions carried out with fluorous-tagged radicals 1 (NaOCl)
and 10 (BAIB) are also illustrative of F-SPE and F-LLE work-
up procedures, respectively.
[(C8F17C3H6)2N]2-triazine-O-TEMPO (10)
In a flame-dried Schlenk tube, fluorous amine 17 (1.24 g,
1.32 mmol) and triazine-O-TEMPO 15 (96 mg, 0.30 mmol)
were dissolved in dry THF (15 mL). The solution was stirred
under nitrogen for 6 h at 858C, during which a white precipitate
(17·HCl) was formed. Thesuspension was cooledto room tem-
perature, diluted with Et2O (20 mL) and filtered on a Bꢃchner
funnel. The solid was washed Et2O (10 mL) and the combined
liquid layers were evaporated to give a thick orange oil. Flash
column chromatography (silica gel, petroleum ether/MTBE,
75/25) afforded the pure title compound as an orange solid;
yield: 415 mg (65%); mp 73–778C. IR (KBr, selected data):
Oxidation of Cycloctanol using Aqueous NaOCl
A 50-mL two-necked, round-bottom flask equipped with a
magnetic stirrer and a dropping funnel was charged with cyclo-
octanol (0.64 g, 5 mmol), radical 1 (28 mg, 0.05 mmol) and CH2
Cl2 (12.5 mL). To the stirred solution cooled to 08C, a 0.50 M
aqueous solution of KBr (1 mL, 0.50 mmol) was added, fol-
lowed by dropwise addition of a 0.35 M aqueous solution of
NaOCl (17.8 mL, 6.25 mmol) buffered at pH 8.6 by NaHCO3.
The reaction mixture was vigorously stirred for a further 15 mi-
nutes. The organic phase was separated and the aqueous phase
was extracted with CH2Cl2 (2ꢁ4 mL). The combined organic
layers were washed with water (3ꢁ5 mL) and dried over
MgSO4. The solvent was evaporated under vacuum to leave a
residue which was loaded onto a fluorous reverse phase silica
gel cartridge and eluted with CH3CN (6 mL) to isolate cyclo-
ctanone (0.50 g, 80%). The cartridge was then washed with
Et2O (10 mL) to remove the fluorous-tagged catalyst 1
(25 mg, 89%) which was used in a subsequent run (Table 6).
n¼1573, 1525 (C N), 1366 (N O) cmꢀ1; anal. calcd. for
C56H41F68N6O2 (2121.86): C 31.70, H 1.95, N 3.96; found:
C 32.12, H 1.99, N 3.98; HR-MS (ESI): m/z¼2144.20114
([MþNa]þ); calcd. for [(C56H41F68N6O2)Na]þ: 2144.21027.
¼
ꢀ
[(C8H17)2N]2-triazine-O-TEMPO (18)
In a flame-dried Schlenk tube, triazine-O-TEMPO 15 (0.16 g,
0.50 mmol) and an excess of dioctylamine (0.56 g, 2.32 mmol)
were dissolved in dry THF (3 mL). The solution was stirred un-
der nitrogen for 5 h at 758C, then cooled to room temperature
and diluted with Et2O (20 mL). The solution was washed with
water, brine and dried over MgSO4. Flash column chromatog-
raphy (silica gel, CH2Cl2) afforded the title compound as a red-
orange, thick oil; yield: 204 mg, (65%). IR (KBr, selected data):
Oxidation of 4-Bromobenzyl Alcohol using BAIB
A water-cooled jacketed vial fitted with a stirring bar was
charged with a solution of 4-bromobenzyl alcohol (187 mg,
1 mmol) and n-decane (71 mg, 0.5 mmol, internal standard
for GC) in CH2Cl2 (2 mL). To the stirred solution thermostat-
ted at 208C, radical 10 (106 mg, 0.05 mmol) was added, fol-
lowed by BAIB (0.71 g, 1.1 mmol). After 2 hours the reaction
was complete as shown by GC analysis. The solution was ex-
tracted with PDMC [perfluoro-(1,3-dimethyl)cyclohexane,
3ꢁ2 mL]. The combined fluorous extracts containing 10
were evaporated to dryness under vacuum to recover the fluo-
rous-tagged catalyst 10 (104 mg, 98%) which could be reused
without further treatments (Table 8). Pure 4-bromobenzalde-
hyde (180 mg, 97%) was obtained after elution of the organic
layer through a short silica pad (light petroleum ether/
CH2Cl2, 1/1).
n¼1568, 1517 (C N), 1361 (N O) cmꢀ1; anal. calcd. for
C44H85N6O2 (730.18): C 72.38, H 11.73, N 11.51; found: C
72.51, H 11.62, N 11.69.
¼
ꢀ
Determination of Partition Coefficients P
Method A: A 10-mL vial equipped with a magnetic stirrer was
charged with the fluorous-tagged TEMPO radical (25 mmol),
perfluoro-(1,3-dimethyl)cyclohexane (2 mL) and the organic
solvent (2 mL). The mixture was thermostatted at 258C and
vigorously stirred for 4 h. A 1-mL sample was taken out of
each phase, evaporated to dryness and weighed on an analyti-
686
ꢁ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2005, 347, 677–688

Selective Oxidation of Alcohols to Carbonyl Compounds
FULL PAPERS
[16] C. Bolm, A. S. Magnus, J. P. Hildebrand, Org. Lett. 2000,
2, 1173–1175.
Oxidation of 1-Octanol using TCCA.
A stoppered vial fitted with a stirring bar was charged with a
solution of 1-octanol (130 mg, 1 mmol) and n-decane (71 mg,
0.5 mmol, internal standard for GC) in CH2Cl2 (1 mL). To
the stirred solution cooled to 08C in an ice-bath, CH3COONa
(258 mg, 3.15 mmol) was added, followed by a solution of rad-
ical 4 (6.5 mg, 0,01 mmol) in CH2Cl2 (1 mL). After 5 minutes
TCCA (244 mg, 1.05 mmol) was introduced and the reaction
mixture was stirred for 10 minutes. The mixture was diluted
with Et2O (10 mL), filtered on a PTFE septum and the clear
liquid layer was analyzed by GC to determine conversion
and selectivity of the reaction.
[17] R. A. Miller, R. S. Hoerrner, Org. Lett. 2003, 5, 285–287.
[18] F. Minisci, F. Recupero, A. Cecchetto, C. Gambarotti, C.
Punta, R. Faletti, R. Paganelli, G. F. Pedulli, Eur. J. Org.
Chem. 2004, 109–119, and references cited therein.
[19] P. Gamez, I. W. C. E. Arends, R. A. Sheldon, J. Reedijk,
Adv. Synth. Catal. 2004, 346, 805–811.
[20] a) T. Fey, H. Fischer, S. Bachmann, K. Albert, C. Bolm, J.
Org. Chem. 2001, 66, 8154–8159, and references cited
therein; b) A. Dijksman, I. W. C. E. Arends, R. A. Shel-
don, Chem. Commun. 2000, 271–272.
[21] For a discussion of this issue and the development of an
efficient heterogeneous TEMPO catalyst, see: R. Ciri-
minna, C. Bolm, T. Fey, M. Pagliaro, Adv. Synth. Catal.
2002, 344, 159–163.
Acknowledgements
[22] C. Tanyeli, A. Gꢃmꢃs¸, Tetrahedron Lett. 2003, 44, 1639–
1642.
[23] G. Pozzi, M. Cavazzini, S. Quici, M.Benaglia, G. Del-
lꢀAnna, Org. Lett. 2004, 6, 441–443.
We gratefully acknowledge the European Union (Contract No.
HPRN-CT-2000-00002, Development of Fluorous Phase tech-
nology for Oxidation processes) for financial support.
[24] P. Ferreira, W. Hayes, E. Phillips, D. Rippon, S. C. Tsang,
Green Chem. 2004, 6, 310–312.
[25] G. Pozzi, M. Cavazzini, O. Holczknecht, S. Quici, I. Shep-
person, Tetrahedron Lett. 2004, 45, 4249–4251.
[26] J. Zhou, S. Chen, H. Duan, M. Jiang, Y. Zhang, Lang-
muir 2001, 17, 5685–5687.
[27] B. Hungerhoff, H. Sonnenschein, F. Theil, J. Org. Chem.
2002, 67, 1781–1785.
[28] J.-M. Vincent, A. Rabion, V. K. Yachandra, R. H. Fish,
References and Notes
[1] M. Hudlicky, Oxidations in Organic Synthesis, American
Chemical Society, Washington, DC, 1990, pp. 114–149.
[2] F. A. Luzzio, Org. React. 1998, 53, 1–221, and references
cited therein.
[3] T. T. Tidwell, Synthesis 1990, 857–870, and references
cited therein.
[4] R. A. Sheldon, I. W. C. E. Arends, G.-J. ten Brink, A.
Dicksmann, Acc. Chem. Res. 2002, 35, 774–781, and ref-
erences cited therein.
[5] R. A. Sheldon, I. W. C. E. Arends, in: Modern Oxidation
Methods, (Ed.: J.-E. Bꢄckvall), Wiley-VCH, Weinheim,
2004, pp. 83–114.
[6] A. E. J. De Nooy, A. C. Besemer, H. van Bekkum, Syn-
thesis 1996, 1153–1174.
[7] W. Adam, C. R. Saha-Mçller, P. A. Ganeshpure, Chem.
Rev. 2001, 101, 3499–3548.
[8] R. A. Sheldon, I. W. C. E. Arends, Adv. Synth. Catal.
2004, 346, 1051–1071.
[9] J. A. Cella, J. A. Kelley, E. F. Kenehan, J. Org. Chem.
Can. J. Chem. 2001, 79, 888–895.
´
`
[29] F. De Campo, D. Lastecoueres, J.-M. Vincent, J.-B. Verl-
hac, J. Org. Chem. 1999, 64, 4969–4971.
[30] I. Shepperson, S. Quici, G. Pozzi, M. Nicoletti, D. OꢀHa-
gan, Eur. J. Org. Chem. 2004, 4545–4551.
[31] N. Merbouh, J. M. Bobbitt, C. Brueckner, Org. Prep.
Proced. Int. 2004, 36, 3–31.
[32] M. J. Adam, L. D. Hall, Carbohydr. Res. 1985, 144, 23–
31.
´
´
´
´
[33] Z. Szlavik, G. Tarkanyi, A. Gçmçry, G. Tarczay, J. Rabai,
J. Fluorine Chem. 2001, 108, 7–14.
[34] C. Rocaboy, W. Bauer, J. A. Gladysz, Eur. J. Org. Chem.
2000, 14, 2621–2628.
[35] F.-E. Chen, Y.-Y. Kuang, H.-F. Dai, L. Lu, M. Huo, Syn-
thesis 2003, 2629–2631.
[36] P. L. Anelli, L. Lunazzi, F. Montanari, S. Quici, J. Org.
Chem. 1984, 49, 4197–4203.
1975, 40, 1860–1862.
[10] M. F. Semmelhack, C. S. Chou, D. A. Cortes, J. Am.
Chem. Soc. 1983, 105, 4492–4494.
[11] M. F. Semmelhack, C. R. Schmid, D. A. Cortes, C. S.
´
´
[37] For an example of the importance of “fluorous self-as-
sembly” in catalysis, see: E. de Wolf, A. L. Spek,
B. W. M. Kuipers, A. P. Philipse, J. D. Meeldijk, P. H. H.
Bomans, P. M. Frederik, B.-J. Deelman, G. van Koten,
Tetrahedron 2002, 58, 3911–3922.
Chou, J. Am. Chem. Soc. 1984, 106, 3374–3376.
[12] P. L. Anelli, C. Biffi, F. Montanari, S. Quici, J. Org.
Chem. 1987, 52, 2559–2562.
[13] P. L. Anelli, S. Banfi, F. Montanari, S. Quici, J. Org.
Chem. 1989, 54, 2970–2972.
[38] B. Betzemeier, M. Cavazzini, S. Quici, P. Knochel, Tetra-
hedron Lett. 2000, 41, 4343–4346.
[14] A. De Mico, R. Margarita, L. Parlanti, A. Vescovi, G.
Piancatelli, J. Org. Chem. 1997, 62, 6974–6977.
[39] M. Contel, C. Izuel, M. Laguna, P. R. Villuendas, P. J.
Alonso, R. H. Fish, Chem. Eur. J. 2003, 9, 4168–4178.
[40] A. Granados, R. H. de Rossi, J. Am. Chem. Soc. 1995,
117, 3690–3696.
[41] P. L. Bragd, A. C. Besemer, H. van Bekkum, J. Mol. Cat-
al. A 2001, 170, 35–42.
[15] a) C.-J. Jenny, B. Lohri, M. Schlageter, (Hoffmann-La
Roche), Eur. Pat. Appl. EP 775 684, 1997; US Patent
5,821,374, 1998; Chem. Abstr. 1997, 127, P65794x; b) L.
De Luca, G. Giacomelli, A. Porcheddu, Org. Lett.
2001, 3, 3041–3043; c) L. De Luca, G. Giacomelli, S. Ma-
sala, A. Porcheddu, J. Org. Chem. 2003, 68, 4999–5001.
Adv. Synth. Catal. 2005, 347, 677–688
asc.wiley-vch.de
ꢁ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
687

Orsolya Holczknecht et al.
FULL PAPERS
[42] Handbook of Fluorous Chemistry, (Eds.: J. A. Gladysz,
[46] H. Tohma, T. Maegawa, S. Takizawa, Y. Kita, Adv. Synth.
Catal. 2002, 344 , 328–337, and references cited therein.
[47] Y. Tashino, H. Togo, Synlett 2004, 2010–2012, and refer-
ences cited therein.
´
D. P. Curran, I. Horvath), Wiley-VCH, Weinheim, 2004.
[43] D. P. Curran, Synlett 2001, 1488–1496.
[44] W. Zhang, Chem. Rev. 2004, 104, 2531–2556, and refer-
ences cited therein.
[48] C. Rocaboy, J. A. Gladysz, Chem. Eur. J. 2003, 9, 88–95.
[45] a) M. Cavazzini, S. Quici, G. Pozzi, Tetrahedron 2002, 58,
3943–3949; b) B. Croxtall, E. G. Hope, A. M. Stuart,
Chem. Commun. 2003, 2430–2431.
688
ꢁ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2005, 347, 677–688