Supramolecular Assemblies as Promoters of Iodohydrin Formation

Article abstract of DOI:10.1002/ejoc.200300085

Finding alternative reaction media to replace polluting organic solvents is one aim of green chemistry. The ultimate green solvent, water, is the cheapest, non-toxic and most readily available reaction medium: three properties which make it an environmentally and economically attractive solvent. However, a fundamental problem in performing reactions in water is that many organic substrates are hydrophobic and not soluble in water. Several approaches are possible in solubilizing these compounds in aqueous media, one of which is carrying out reactions in aqueous solutions of surfactants at concentrations above their critical micellar concentration (cmc). Reactions of iodine with cyclohexene, 1-octene and styrene in water or in the presence of cationic surfactants do not give useful amounts of iodohydrins, but reactions in anionic surfactants give good yields. Iodohydrins are important functionalizable compounds and are readily prepared in the presence of sodium dodecyl sulfate (SDS) or sodium N-dodecanoyl sarcosinate (SANa). The critical conditions for these reactions were optimized with a rigorous statistical approach, the experimental design method. Use of these newly optimized reaction conditions gave high yields in short times for all of the alkenes examined. The use of anionic surfactants in water to form iodohydrins is a valid alternative to methods previously described. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2004.

Full text of DOI:10.1002/ejoc.200300085

FULL PAPER
Supramolecular Assemblies as Promoters of Iodohydrin Formation
Simona Cerritelli,[a,b] Marco Chiarini, Giorgio Cerichelli,* Marina Capone, and
[b]
[b]
[b]
Mario Marsili[
b]
Keywords: Synthesis design / Surfactants / Iodohydrins / Iodine / Green chemistry
Finding alternative reaction media to replace polluting or-
ganic solvents is one aim of green chemistry. The ultimate
green solvent, water, is the cheapest, non-toxic and most
readily available reaction medium: three properties which
make it an environmentally and economically attractive solv-
ent. However, a fundamental problem in performing reac-
factants do not give useful amounts of iodohydrins, but reac-
tions in anionic surfactants give good yields. Iodohydrins are
important functionalizable compounds and are readily pre-
pared in the presence of sodium dodecyl sulfate (SDS) or so-
dium N-dodecanoyl sarcosinate (SANa). The critical condi-
tions for these reactions were optimized with a rigorous stat-
tions in water is that many organic substrates are hydro- istical approach, the experimental design method. Use of
phobic and not soluble in water. Several approaches are pos-
sible in solubilizing these compounds in aqueous media, one
of which is carrying out reactions in aqueous solutions of sur-
these newly optimized reaction conditions gave high yields
in short times for all of the alkenes examined. The use of
anionic surfactants in water to form iodohydrins is a valid
factants at concentrations above their critical micellar con- alternative to methods previously described.
centration (cmc). Reactions of iodine with cyclohexene, 1-oc- ( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
tene and styrene in water or in the presence of cationic sur-
Germany, 2004)
Introduction
Finding reaction media to replace polluting organic sol-
vents is one of the principles of Green Chemistry. Water is
[
2]
Alkenes react readily across the double bond.[1] Electro- the cheapest, non-toxic and most readily available reaction
philic additions to unsaturated compounds, where the elec- medium, which makes it an environmental and economi-
trophile is a halogen molecule or derivative and the nucle- cally attractive solvent. However, a fundamental problem in
ophile is the alkene, are shown in the following scheme (Fig- performing reactions in water is that many organic sub-
ure 1).
strates are hydrophobic and sparingly soluble in water. Sev-
eral approaches are useful in solubilizing organic com-
pounds in aqueous media, one of which is carrying out re-
actions in the presence of amphiphiles, e.g. surfactants con-
taining both apolar hydrophobic and polar hydrophilic
groups on the same molecule. In water, this dual character
[
3]
leads to self-association or micellization. Because of this
self-association, micelles can control the course of organic
reactions because they provide different local environments
Figure 1. Scheme of halogenation of alkenes by iodine in water
˚
within a few Angströms. Therefore, unlike homogeneous
The main importance of these processes is that they gen-
erate saturated compounds with the potential to have differ-
ent vicinal substituents in a single-step reaction. For ex-
ample, newly formed α-halohydrins are attractive as they
can be used in further reactions because they are useful in-
termediates and are targets in organic syntheses.
solvents, micelles provide a variety of microenvironments
ranging from the hydrocarbon core to the polar interface
to bulk water, allowing hydrophobic and hydrophilic mol-
ecules to be solubilized in local domains. Micelles may af-
fect reaction rates and equilibria by bringing reactants to-
[4]
gether or keeping them apart, e.g. they can incorporate
both water-insoluble and ionic compounds at the micellar
[
[
a]
b]
Institute for Biomedical Engineering, Swiss Institute of surface. Therefore they can also promote regio-selectivity in
Technology and University Zurich,
Moussonstr. 18, 8044 Zürich, Switzerland
Dipartimento di Chimica, Ingegneria Chimica e Materiali,
Universit a` dell’Aquila,
[5]
the products by orienting reactants.
For both economic and environmental reasons recovery
of surfactants is crucial. Separation of products from crude
reaction mixtures containing surfactants can be difficult be-
cause these systems readily form emulsions during standard
Via Vetoio, 67010 Coppito LЈAquila, Italy
Fax: (internat.) ϩ39-0862-433753
E-mail: cerichel@univaq.it
Eur. J. Org. Chem. 2004, 623Ϫ630
DOI: 10.1002/ejoc.200300085
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
623

S. Cerritelli, M. Chiarini, G. Cerichelli, M. Capone, M. Marsili
FULL PAPER
extraction procedures. These problems have been faced and factant make this procedure environmentally friendly and
solved in different ways. As an example, Jaeger and co-wor- consistent with the principles of sustainable chemistry.[
kers[ have developed a general approach involving the syn- Therefore we undertook a rigorous study of the effects of
thesis of a chemically labile surfactant, which can be readily surfactants on the iodination reaction of alkenes.
2]
6]
hydrolyzed when the reaction is complete, which simplifies
Both cationic and anionic surfactants, at different con-
extraction of the products, although the surfactant is lost. centrations, but always above their critical micellar concen-
Alternatively, ionic surfactants can be recovered by cooling tration (cmc), were tested with the purpose of optimizing
the reaction mixture to below the Krafft point of the surfac- formation of the iodohydrins. Cationic surfactants were ce-
tant, leading to precipitation. Similarly non-ionic surfac- tyltrimethyl
ammonium
compounds:
CTAI3
ϩ
Ϫ
ϩ
Ϫ
tants can be recovered by heating the reaction mixture [C H N (CH ) I ], CTAIBr2 [C H N (CH ) IBr ],
16
33
3 3 3
16 33
3 3
2
above its cloud point. As a final example, surfactants can obtained by addition of cetyltrimethyl ammonium iodide
be recovered by ultrafiltration or by stripping products with (CTAI) to a solution of I or Br , respectively, and CTAI Br
2
2
2
ϩ
Ϫ
supercritical CO2.
[C H N (CH ) I Br ], obtained by addition of I to a
16 33 3 3 2 2
Electrophilic alkene additions have been extensively stud- solution of cetyltrimethyl ammonium bromide (CTABr).
ied and, particularly for bromination, the reaction mecha- Anionic surfactants were sodium dodecylsulfate (SDS,
[
7]
Ϫ
3
ϩ
nism in homogeneous solution is well understood. Rates C H OSO Na ), and sodium N-dodecanoyl sarcosinate
12
25
and products are extremely sensitive to the reaction con- [SANa, C H CON(CH )CH COONa].
11
23
3
2
ditions, which makes alkene halogenations useful probes of
micellar structure. There are several examples in the litera- or I3 as the iodine source with cyclohexene, 1-octene and
ture of kinetic and synthetic investigations on alkene bro- styrene. Reactions were carried out in the dark and easily
Reactions were carried out in aqueous solution using I2
Ϫ
mination aimed at analysis of the microenvironment at the followed visually until the end, monitoring the disappear-
micellarϪwater interface.[
8]
ance of the red iodine color. (The use of UV/Vis apparatus
Unlike syntheses of chloro- and bromohydrins,[9] forma- was ruled out because of the strong absorbance of turbid
tion of iodohydrins from alkenes and iodine in water is un- solutions.) Synthetic processes and yields were optimized
satisfactory owing to the reversibility of the addition of iod- using a mathematical chemometric method. A full-factorial
ine to a double bond, where the equilibrium is very unfavor- design for the four variables (temperature, concentration of
able[ (Figure 2).
10]
iodine, concentration of surfactant and solution pH) was
[
17]
generated with the Spectre program.
Reaction yields
were regressed on the data matrix and the importance of
each reaction parameter was calculated, in order to optim-
ize the reaction yields.
Figure 2. Scheme of iodination showing the ionic intermediates
formed during the reaction and reversibility of iodination of al-
kenes
Results and Discussion
Therefore, iodohydrins are typically obtained by reaction
[
11]
of epoxides with hydriodic acid, elementary iodine
or
؊
Iodinations in Water and Iodine (I or I3 ) Without
Surfactant
2
metal iodides or from halide ion exchange of chlorohydrins
or bromohydrins in acetone.[ Other reported procedures
12]
carry out iodination in the presence of scavengers (e.g. mer-
In order to establish the baseline reactivity of each sub-
curic oxide or silver oxide[ ), oxidizing agents of iodide strate, iodination was performed in water in the absence of
10]
[10]
ions (e.g. nitrous acid, oxygen, sodium nitrite, pyridinium surfactant. Under these conditions, only cyclohexene
[
13]
dichromate ), redox systems (e.g. periodic acid/sodium yielded a significant amount of product (Table 1). The
[14]
hydrogen sulfite systems ), or by using agents such as NIS trans-2-iodocyclohexanol, which was obtained in 60% yield
[
15]
(
N-iodosuccinimide ) in DME/H O (DME ϭ 1,2-dime- with respect to the iodine source, was recovered after three
2
thoxyethane) at Ϫ20 °C.
hours of reaction. For the other alkenes, the equilibrium
Because micelles can modify rates and equilibria of many was always toward reactants and undesired by-products,
organic reactions and there is extensive literature data on and only a very small amount of iodooctanol was isolated
brominations in micellar solutions,[
8,16]
aqueous surfactant after long reaction times (Table 1). The intense red color of
solutions have been used for syntheses of iodohydrins from residual iodine did not disappear during reaction. Neither
simple alkene molecules and iodine. The aim of this study the yields nor the reaction times for cyclohexene depended
Ϫ
was to develop a simple preparative procedure to obtain on the iodinating agent, and I or I3 reacted equally well.
2
iodohydrins. Although we did not have a homogeneous sys- Similar results were observed previously, where in aqueous
tem, because of a large excess of reactants over surfactant, solution there was no observable difference between the re-
Ϫ
[8b,18]
we considered the reaction mixture to be a normal micellar activities of Br and Br3
in alkene brominations.
2
system, using the pseudophase treatment.[ The absence of Therefore, it seems reasonable to conclude that reactions of
organic solvents and harsh reagents, the use of water as a bromine and iodine with alkenes follow similar mecha-
reaction medium and the possibility of recovering the sur- nisms.
19]
624
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 623Ϫ630

Supramolecular Assemblies as Promoters of Iodohydrin Formation
Table 1. Iodination in water
FULL PAPER
Alkene[a]
X[b]
Time (h)
Products
Yield (mol %)[c]
Ϫ
Cyclohexene
Cyclohexene
Styrene
Styrene
I
3
I
2
I
2
I
3
I
3
3
3
24
24
100
trans-2-iodocyclohexanol
trans-2-iodocyclohexanol
polymers
polymers
1-iodooctan-2-ol
60
60
Ϫ
Ϫ
5[e]
[
[
d]
d]
Ϫ
Ϫ
1-Octene
2-iodooctan-1-ol
[
a]
[b]
Ϫ
is always 0.03 . [c] With respect to [I Ϫ]. [d] 0.03
] or [I
2 3
Temperature 60 °C, 0.06  alkene. The molar concentration of iodine or I
3
[
e]

styrene. Sum of both regioisomers; ratio [1-iodooctan-2-ol]/[2-iodooctan-1-ol] 9:1.
Ϫ
Ϫ
Iodinations in Micellar Systems
Thus in surfactant assemblies, the ratio Br /Br is affec-
3
Ϫ
ted, favoring Br3 . In homogeneous conditions this anion
is generally less reactive than Br by approximately one or-
Micellar effects on reaction rates have been analyzed with
pseudophase models, which separate the contributions of
reactant concentration at the interface and reactivity in that
region and include transfer equilibria of reactants between
2
[
10,21]
der of magnitude;
moreover in a cationic surfactant
Ϫ
Br3 interacts very strongly with the micellar head group.
For these reasons, alkene brominations in CTABr were
[
19]
water and micelles. In these models, micelles are regarded
as a separate phase from water, and the overall reaction rate
is the sum of the rates in each pseudophase. In the present
work, we do not attempt to analyze reaction rates quanti-
tatively, but instead try to gain a qualitative understanding
of the iodination reaction in these terms. All the alkenes
have extremely low water solubility, and we assume that re-
actions occur only at the micelleϪwater interface. We note
that the pseudophase kinetic treatments were developed lar-
gely for reactions of very dilute organic substrates and may
not be applicable to reactions in preparative conditions,
where treatment of transfer equilibria is complicated. How-
ever, the general concepts provide a useful guide for these
reactions even though they cannot be applied quantitat-
ively.
5
6
[8b,16]
slower by a factor of 10 Ϫ10 than in water.
We believe that, because of the greater polarizability of
iodine than bromine, similar or larger effects are involved
in alkene iodination in the presence of cationic surfactants,
resulting in a very low iodinating power, slow reactions and
formation of undesirable by-products. Iodide ions should
be concentrated at the cationic interface of the micelles, in-
terfering with the halogenation equilibrium. For these re-
asons we believe that cationic surfactants will not be useful
in syntheses of iodohydrins.
؊
Iodinations in Water, Iodine (I or I3 ) and Anionic
2
Surfactants
The effect of anionic surfactants on iodination was then
studied. We assumed that the negative charge of the micel-
lar surface would exclude the iodide anion from that region
؊
Iodinations in Water, Iodine (I or I ) and Cationic
2
3
Ϫ
Surfactants
and protect micellar-bound products from I , eliminating
the need for scavengers or oxidizing agents of the iodide.
Furthermore, a carbocationoid transition state should in-
teract favorably with the anionic head groups.
Both SDS and SANa affected reaction rates and prod-
ucts. Relative to halogenation in water, the anionic surfac-
tants increased yields and shortened reaction times for all
substrates, with strongly reduced by-product formation. All
results are shown in Table 2 and 3.
Additions of I or Br to a solution of CTABr or CTAI
2
2
form CTAI Br, CTAIBr or CTAI , depending on the rela-
2
2
3
tive initial conditions that were used in the iodinating reac-
tions. There was no productive reaction in the presence of
these cetyltrimethylammonium surfactants, even though re-
action times were at least 10 times longer than those for
reactions in water, and there were many side products (such
as diiodo compounds and oligomers). Moreover, with both
Cyclohexene gave similar yields of iodohydrins in the ab-
sence or presence of surfactants, but the reaction time was
approximately one-third as long with the micelles. Although
the local second-order rate constants in the micelles might
be expected to be lower than in water, owing to the lower
CTAIBr2 and CTAI Br bromo-iodinated products were
also recovered.
2
Owing to the lower polarity of the micellar surface com-
pared to the water, micelles reduce the rate of reactions that
involve polar intermediates and charge separations at the
[
8c]
[
8]
polarity at the micellar interface, both reactants, I and
interface. For bromination, we consider the equilibrium:
2
the alkene, are concentrated at the waterϪmicelle interface,
[
22Ϫ24]
where these ionic reactions take place,
thus overcom-
ing any kinetic inhibiting effect. Therefore, the overall reac-
tion rate is accelerated.
which is medium dependent and K decreases with
There are small differences in reaction times observed in
SDS and SANa, but no obvious trend. The reaction times
d
decreasing solvent polarity.[
8b,20]
Eur. J. Org. Chem. 2004, 623Ϫ630
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
625

S. Cerritelli, M. Chiarini, G. Cerichelli, M. Capone, M. Marsili
FULL PAPER
Table 2. Iodination in aqueous solutions of SDS
Alkene[a]
[S], ()[b]
X [c]
Time (h)
Products
Regioisomers ratio
Yield
(mol %)
[
d]
(
[a]/[b])
Ϫ
Cyclohexene
Cyclohexene
Styrene
0.025
0.05
0.1
I
I
I
I
3
2
2
3
1.5
0.5
24
trans-2-iodocyclohexanol
trans-2-iodocyclohexanol
2-iodo-1-phenylethan-1-ol
1-iodooctan-2-ol (a)
62
60
42
43
[
e]
Ϫ
1
-Octene
-Octene
0.025
16
2
2
2
-iodooctan-1-ol (b)
1-iodooctan-2-ol (a)
-iodooctan-1-ol (b)
1
0.05
I
2
7
57
2
[
a]
Temperature 60 °C, 0.06  alkene. [S] is surfactant concentration. [c] The molar concentration of iodine or I Ϫ
[b]
is always 0.03 . [d]
3
Ϫ
[e]
2 3
With respect to [I ] or [I ]. 0.03  styrene.
Table 3. Iodination in aqueous solutions of SANa
Alkene[a]
[S], ()[b]
X[c]
Time (h)
Products
Regioisomers ratio
Yield
(mol %)
[
d]
(
[a]/[b])
Ϫ
Cyclohexene
Cyclohexene
Styrene
Styrene
0.025
0.05
0.15
0.05
0.025
I
I
I
I
I
3
2
2
2
3
(0.03)
(0.04)
(0.03)
(0.03)
1
0.5
24
24
24
trans-2-iodocyclohexanol
trans-2-iodocyclohexanol
2-iodo-1-phenylethan-1-ol
2-iodo-1-phenylethan-1-ol
1-iodooctan-2-ol (a)
60
50
47
22
51
[
[
e]
e]
Ϫ
1
-Octene
-Octene
(0.03)
3
3
2
-iodooctan-1-ol (b)
1-iodooctan-2-ol (a)
-iodooctan-1-ol (b)
1
0.05
I
2
(0.04)
24
45
2
[
a]
Temperature 60 °C and 0.06  alkene. [S] is surfactant concentration. [c] In parenthesis the molar concentration of iodine or I
[b]
Ϫ. [d]
3
Ϫ
[e]
2 3
With respect to [I ] or [I ]. 0.03  styrene.
shown in Table 2 and 3, demonstrate, as expected, that I is aqueous cetyltrimethyl ammonium surfactants, benzene
2
Ϫ
more reactive than I3 , while the yields of iodinated prod- NMR signals suggest that solute mobility at the aggregate
ucts are similar. This result fits the proposed exclusion of surface is not affected by the size of the aggregate.[ In a
26]
the anions from the negatively charged micellar surface.
number of micellar models, water was believed to penetrate
Styrene and 1-octene gave the desired products only when the micelle. Menger used space-filling models[28] to con-
the reaction was carried out in the presence of anionic sur- clude that there should be deep voids in the micelle, which
[29]
factants. When halogenations were carried out in water, the accommodate water molecules extensively. Fromherz,
polymerization of styrene took place much more rapidly Butcher and Lamb, and Dill and Flory postulated dif-
than halogen addition and the linear octene was recovered ferent structural organizations with partial exposure of
[
30]
[31]
mostly unchanged, together with a few oligomers.
alkyl groups to water. These micellar models are indepen-
With anionic surfactants, styrene formed 2-iodo-1-phenyl- dent of surfactant charge, and should be valid for cationic,
ethanol with yields of 47% (SANa) and 42% (SDS) for the anionic, zwitterionic and non-ionic micelles. Therefore, it is
two anionic surfactants studied. The behavior was similar reasonable to believe that styrene molecules are separately
for cyclization of 2-(ω-bromoalkoxy)phenoxide in dicationic located in clefts of the micellar pseudophase, so that styrene
[
25]
surfactants,
where intermolecular reactions were pre- polymerization is slowed down with respect to the halogen-
cluded by the positioning of the reactant in a well-struc- ation reaction. Such a picture does not contradict the well-
tured substrateϪsurfactant aggregate. As a first approxi- established dynamic nature of micelles. In the fast amphi-
mation, it is reasonable to assume that this effect is indepen- philic exchange of molecules between micelles and bulk
dent of micellar charge and depends only on the presence water, solubilization of substrates is a dynamic exchange
of large well-organized assemblies with the substrate at the equilibrium. Because of the very low alkene solubility in
surface. A recent study of benzene solubilization in aqueous water, this equilibrium is strongly shifted in favor of mi-
CTABr demonstrated that the aromatic molecule lies in celles, just as if a styrene droplet is diluted in an inert en-
clefts at the micellar surface replacing water molecules in vironment whose main effect is to inhibit polymerization.
[
26]
that surface. Similarly, a study on the chemical trapping Additionally, there is a concentration effect on the reaction
of water at the surfaces of cationic micelles showed that because both alkene and iodine are brought together by in-
cyclohexene lies at the surface and that its bromination teraction with the micellar surface, which accelerates forma-
takes place in a water-rich area of the surface,[ although tion of the cationic intermediate that rapidly generates the
evidence on micellar structure in water may not apply in the product. Thus intermolecular iodination is favored over
presence of apolar solutes.[ On the other hand, in some polymerization. Moreover, the reaction is regioselective and
27]
26]
626
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 623Ϫ630

Supramolecular Assemblies as Promoters of Iodohydrin Formation
FULL PAPER
only one of the two possible iodohydrins, 2-iodo-1-phenyle- tralization of the hydriodic acid formed during reaction. Fi-
than-1-ol, is formed, because the forming positive charge nally, two different concentrations of surfactant and iodine
is stabilized by the phenyl group in a heterolytic reaction. were examined, with concentrations above the cmc, which
However, 1-octene reacted in the presence of either anionic will also be significantly lowered by the very hydrophobic
surfactant to give a mixture of the two regioisomeric solutes. Higher [surfactant] forms more aggregates in solu-
iodohydrins. Because of the difference between phenyl and tion, increasing the waterϪmicellar interfacial reaction re-
alkyl groups in stabilizing carbenium-ion-like intermediates, gion.
regioselectivities differ for reactions in anionic micelles and
The independent variable, yk can be expressed as in
water. In water there is poor iodohydrin conversion, but a Equation (1). The first term represents the sum of the indi-
net prevalence, 9:1, of the Markovnikov product, 1-iodooc- vidual effects while the second term represents potential in-
tan-2-ol, while the anionic surfactants shifted the product teraction among the original variables.
ratio to 2:1 and 3:1, in SDS and SANa, respectively. Con-
sidering halohydrin formation, the prevalent product still
y
k
ϭ Σc
i
x
i
ϩ Σ
i
Σ
j
c
i
x
i
x
j
(1)
accords with the Markovnikov rule, but micelles clearly in-
crease the anti-Markovnikov iodohydrin. One difference be-
tween the two surfactants is that with SANa a greater ratio
of iodinating agent with respect to the surfactant was re-
quired, due to concomitant partial iodination of this am-
phiphile.
With styrene the phenyl group favors the formation of a
carbocationoid intermediate and the Markovnikov product,
but if 1-octene gives an intermediate iodonium-like species,
water could attach at either position. Previous studies
In order to calculate estimates of the coefficient c in
[
32]
Equation (1), the experimental design method,
was ad-
opted to perform a series of multivariate experiments. The
outstanding feature of such an array of experiments is that
all variables in Equation (1) are mutually orthogonal. Thus,
they allow computation of the single, unbiased ‘‘effect’’ of
every variable on the responses y . The obtained model
k
function can then be used to determine specific values for
each synthesis parameter (or interaction) that leads to
maximum yield and minimum reaction time. By using
SPECTRE software the following design matrix was ob-
tained as a result of a two-level full factorial design
showed that the linear chain of this hydrophobic alkene is
deeply located in the micellar aggregate.[8c] If octene is lo-
cated in the micelle, as shown in Figure 3), carbon atom C-
1
will be the more exposed to water, which partially offsets
(
Table 4). For each variable two levels were tested: 45 °C
and 65 °C; presence or absence of buffer to control pH;
.03 mol/L and 0.06 mol/L for [I ]; 0.01 mol/L and 0.1 mol/
the Markovnikov-predicted attack of water at C-2. In this
description we have assumed that the lifetime of a cationic
intermediate is too short to allow escape into water, which
increases preference for the Markovnikov product.
0
2
L for [SDS].
Table 4. Optimization of iodohydrin synthesis: experimental con-
ditions yields and times
Entry
T
Buffer
()
I
()
2
[SDS]
()
Yield[a]
(mol %)
Time[a]
(h)
(
°C)
1
2
3
4
5
6
7
8
9
45
65
45
65
45
65
45
65
45
65
45
65
45
65
45
65
0
0
0.06
0.06
0
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.03
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.06
0.01
0.01
0.01
0.01
0.1
0.1
0.1
46
58
57
58
73
78
65
72
46
45
39
15
79
60
59
32
4
0.5
12
1
Figure 3. Tentative orientations of anionic surfactants (head groups
and tails) and iodonium ion of octene in a micelle; conformations
are arbitrary and are drawn for purposes of illustration
2
0
0.3
12
0.8
18
5
20
22
8
0.06
0.06
0
0.1
0.01
0.01
0.01
0.01
0.1
0.1
0.1
An alternative explanation could involve radical species,
10
0
as frequently invoked for anti-Markovnikov regiochemistry, 11
but we excluded it by carrying out all the experiments in the 12
0.06
0.06
0
0
0.06
0.06
1
1
1
1
3
4
5
6
absence of light, which could have generated iodine atoms.
20
21
20
Optimization of Iodohydrin Syntheses
0.1
A rigorous statistical approach was used to optimize the
syntheses. First, the yield (y) was assumed to be a function
of four reaction parameters: temperature (T), buffer con-
[
a]
Yields and times were the y
k
responses used to estimate sensi-
tivities (see text).
centration [B], iodine concentration [I ] and surfactant con-
After running all 16 experiments the obtained yields and
2
centration [SDS]. The temperature could not be increased reaction times (Table 4) were introduced into the modeling
above 65 °C, owing to the volatility and low solubility in function of SPECTRE and a refined mathematical model
water of iodine present. Phosphate buffer (0.06 , pH 7.0) was obtained showing the effects of the four system vari-
was added to examine the effect on the products of neu- ables and of two important interactions. These effects are
Eur. J. Org. Chem. 2004, 623Ϫ630
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
627

S. Cerritelli, M. Chiarini, G. Cerichelli, M. Capone, M. Marsili
FULL PAPER
given as ‘‘sensitivity’’ in Table 5. The sensitivity value is a
direct measure of how much yield and time change when
the related variable changes by half of the difference in the
two levels of the design matrix. For example, in the design
matrix [SDS] varies from 0.01 to 0.1 mol/L and half of this
change will cause a change in yield (y1) of 9.94 units in the
yield scale. The importance of the reaction parameters can
thus be immediately read off the sensitivities. Therefore
from Table 5 [SDS] appears to be the most important factor
in increasing the yield, whereas it leaves reaction time unaf-
fected. In contrast, the presence of buffer and the concen-
tration of I showed the strongest effect on lowering the
2
reaction time.
Figure 4. Iodination reaction yields for various alkenes; three dif-
ferent alkenes were reacted with iodine either aqueous conditions
(black); in 0.05  SDS (light gray); or by using the optimized con-
ditions derived from the systematic approach described in the text
Table 5. Effects of reaction parameters on iodohydrin formation
shown as sensitivity
(
dark gray); the respective yields, determined as total product after
Sensitivity
2
extraction with Et O with respect to the iodine concentration, are
Variable
y1: yield
y2: time
shown
Temperature
Buffer
Ϫ2.9693
Ϫ5.6802
Ϫ8.5203
9.9405
Ϫ6.1916
Ϫ5.2930
Ϫ1.7687
3.2920
6.5452
gave no iodinated product with only side products reco-
[
[
I
2
]
SDS]
0.10334 vered, the result of a polymerization reaction that is signifi-
1.7686
0.83910
Temperature ϫ [I
Buffer ϫ [I
2
]
cantly faster than the productive reactions. This is probably
due to the hydrophobicity of this aromatic alkene, which
2
]
is insoluble in water so that, when iodine is added, ionic
The obtained function was then further processed by the polymerization occurs in the styrene phase. In contrast, ad-
software system to calculate the specific values for each par- dition of a surfactant allows compartmentalization of styr-
[
25]
ameter that will deliver maximum yield in the shortest reac- ene into micelles, resulting in formation of a single halog-
tion time. Such optimized conditions are shown in Table 6, enated product, 2-iodo-1-phenylethan-1-ol, in good yield
and yields are almost quantitative for cyclohexene and 1- (67%). The polymerization was not completely suppressed
octene, but not for styrene.
and some polymers were still recovered at the end of the re-
action
Finally a large increase in yield was obtained in micellar
and optimized conditions for the linear 1-octene, which is
almost completely unreactive in water, but was almost com-
Conclusion
Improvements in yields and shorter reaction times, rela- pletely converted into a mixture of the two regioisomeric
tive to results in water, were obtained for all of the sub- iodohydrins in the presence of anionic surfactants (Fig-
strates when reactions were carried out in the presence of ure 4) but with a decrease in the relative amount of the
anionic surfactants and this was further improved when op- Markovnikov product. The presence of micellar aggregates
timized conditions were used (Table 2 and 6; Figure 4). clearly influences the regioselectivity of the reaction. Only
Cyclohexene gave very high yields under the optimized con- one of the possible iodohydrins is formed from styrene, with
ditions: trans-2-iodocyclohexanol was obtained almost regioselectivity predicted by the Markovnikov rule, with
quantitatively but with no improvement in the reaction polymerization partially suppressed by the anionic micelles.
time.
However, the anti-Markovnikov product of octene increases
An interesting result was obtained for the only aromatic in anionic micelles owing to orientation of 1-octene in mi-
system tested. After reacting for 24 hours in water, styrene celles. This is an interesting effect of micellar aggregates on
Table 6. Iodination in aqueous solution of SDS after optimization
Alkene[a]
Time
h)
Products
Regioisomers ratio
([a]/[b])
Yield
(mol %)
[
b]
(
Cyclohexene
0.5
8
7
trans-2-iodocyclohexanol
2-iodo-1-phenylethan-1-ol
1-iodooctan-2-ol (a)
95
67
95
[
c]
[d]
Styrene
-Octene
1
2
2
-iodooctan-1-ol (b)
0.03  phosphate buffer and 0.06  alkene. [b] With respect to [I
Styrene was added in small amounts over 8 hours.
[
a]
]. [c] 0.03  styrene. [d]
2
Temperature 60 °C, 0.15  SDS, 0.03  I
2
628
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 623Ϫ630

Supramolecular Assemblies as Promoters of Iodohydrin Formation
the regiochemistry of the reaction, but it does not seem to (1.2 mmol), and the alkenes (2.4 mmol) were stirred, with a mag-
FULL PAPER
netic stirrer, in water (10 mL) at 60 °C until the red color of the
iodine disappeared, but not longer than 100 h. Products were reco-
vered by simple extraction with diethyl ether (5 times), drying the
change the overall reaction mechanism. Different solvation
powers of micellar pseudophase and bulk water favor for-
mation of a reaction intermediate that is more iodonium-
ion like in micelles, while water stabilizes the more carboc-
ation-like intermediate. In the micellar pseudophase, attack
4
combined organic layers with MgSO , and removing the solvent
in vacuo.
of water is favored on one of the two reactive carbon sites General Iodination Reactions with Surfactant: Reactions were per-
formed as described above except that the iodine source and the
in the intermediate, resulting in a mixture of products.
alkene were added to the aqueous surfactant at concentrations
above the cmc.
Even if the system is not homogeneous it behaves like a
‘normal’’ micellar system; in fact we were able to qualitat-
‘
ively rationalize results in general terms of pseudophase CTAI3 and CTAIBr2 were formed by adding I2 and Br2
models.
(0.03 mmol) respectively, to aqueous CTAI (0.05 mmol); CTABrI
(0.03 mmol) to CTABr (0.05 mmol) in
water. These intense orange solutions were stirred at 60 °C before
2
We report a new, easy and environmentally friendly was prepared by adding I
2
method of synthesizing iodohydrins, important building
blocks for organic synthesis, from alkenes and iodine in
aqueous micellized anionic surfactants. We demonstrated
that in the described conditions and in the presence of
anionic micellar aggregates the equilibrium reaction with
iodine is shifted towards the desired products, avoiding the
use of scavengers or harsh oxidant. Cationic micelles are
not useful in formation of iodohydrins as they probably in-
duce the undesired back reaction with iodide ion.
adding the substrates.
The products were isolated by continuous liquidϪliquid extraction.
Surfactant Recovery: At the end of the reaction we stripped the
product from the waterϪsurfactant system for 6 h with
a
liquidϪliquid extractor using diethyl ether as the solvent. The reco-
vered waterϪsurfactant system could then be reused for new reac-
tion cycles.
Cetyltrimethylammonium Iodide (CTAI) Synthesis: Potassium iod-
ide was added to a solution of CTABr (1 g) in water (50 mL). The
precipitate was recovered by filtration and purified by recrystalliza-
Experimental Section
tion from acetone to yield CTAI as a white solid (1.1 g, 95% yield).
1
H NMR (CDCl
), 3.1Ϫ3.7 (m, 11 H, CH
): δ ϭ 14.0 (CH ), 22.5, 23.1, 25.9, 29.1, 29.19, 29.24,
9.3, 29.5, 31.7 (CH ), 53.6 (NCH ), 66.9 (NCH ) ppm.
3
): δ ϭ 0.81 (t, J ϭ 6.0 Hz, 3 H, CH
3
), 1.0Ϫ1.5
General Remarks: Reactions were carried out in aqueous solution
with deionized, doubly distilled water. Commercial alkenes (Ald-
rich, Fluka) of the highest purity were used without further purifi-
cation. CTABr was purified according to the method of Duynstee
13
(
m, 28 H, CH
2
2
N and CH N) ppm.
3
C
NMR (CDCl
2
3
3
2
3
2
[
33]
SDS and SANa (Fluka) were used without trans-2-Iodocyclohexanol: NMR spectra conform to ref.[34] 1H
and Grunwald.
further purification. 1H NMR (200.132 MHz) and C NMR
13
3
NMR (CDCl ): δ ϭ 3.58Ϫ3.78 (m, 1 H, CHI); 3.85Ϫ4.15 (m, 1
1
3
(50.323 MHz) spectroscopy was performed with an FT Bruker AC H, CHOH) ppm. C NMR (CDCl ): δ ϭ 23.6, 27.2, 33.0, 37.8,
3
200E NMR spectrometer. Peaks are reported relative to internal 42.4, 75.2 ppm.
3
CDCl at δ ϭ 7.26 ppm for proton and δ ϭ 77.36 ppm for carbon.
1
-Iodooctan-2-ol: NMR spectra conform to ref.[34] 1H NMR
Product distribution and yields were determined by gas chromato-
graphic analyses, using a calibrated internal standard on an HP
(CDCl
CH ), 2.05 (s, 1 H), 3.07Ϫ3.55 (m, 3 H, CH
C NMR (CDCl ): δ ϭ 14.1, 16.6, 22.6, 25.6, 29.1, 31.7, 36.6,
71.1 ppm.
3
): δ ϭ 0.89 (t, J ϭ 6.5 Hz, 3 H, CH
3
), 1.05Ϫ1.6 (m, 10 H,
2
2
I and CHOH) ppm.
1
3
5
890 gas chromatography system equipped with a flame-ionization
detector (FID), with a fused-silica wide-bore column (30 ϫ
.53 mm), and poly(5%diphenyl-95% dimethylsiloxane) (CP ϭ 8)
3
0
1
2
-Iodooctan-1-ol: H NMR (CDCl
3
): δ ϭ 0.89 (t, J ϭ 6.5 Hz, 3 H,
or poly(dimethylsiloxane) (CP ϭ 5) as the stationary phase. Yields
were determined on the total recovered product with respect to
total iodine and expressed as mol%.
CH
CH
3
), 1.05Ϫ1.6 (m, 10 H, CH
OH), 4.18Ϫ4.39 (m, 1 H, CHI) ppm. C NMR (CDCl
2
), 2.17 (s, 1 H), 3.95Ϫ4.03 (q, 2 H,
1
3
2
3
): δ ϭ
14.1, 16.6, 22.6, 25.6, 29.1, 31.7, 33.6, 68.6 ppm.
Standard iodohydrins were prepared from the corresponding al-
2-Iodo-1-phenylethanol: NMR spectra conform to ref.[35] 1H NMR
kenes using periodic acid and sodium bisulfite.[14]
(
CDCl ): δ ϭ 2.39 (s, 1 H, OH), 3.25Ϫ3.48 (m, 2 H, CH I), 4.70
3
2
1
3
All reactions involved non-homogeneous mixtures, owing to the
very low solubility of both reactants in water and their higher con-
centrations with respect to surfactants. Reactions were protected
from light and followed visually, monitoring the disappearance of
the intense iodine red color by removing the aluminum foil shield
for a few seconds.
(m, 1 H, CHOH), 7.30 (m, 5 H, Csp2 H) ppm. C NMR (CDCl
δ ϭ 15.1, 73.9, 125.7, 128.3, 128.6, 141.2 ppm.
3
):
Acknowledgments
We wish to thank Dr. Jason C. Schense for help in editing the
manuscript, and Prof. Clifford A. Bunton for his advice and helpful
discussions. We also want to thank CNR Progetto Finalizzato Beni
Culturali, COFIN ’99 and INCA Green Chemistry Project, for fin-
ancial support.
In the presence of I
2
, a higher surfactant concentration was chosen
Ϫ
(0.05 ) than that used in the presence of I
3
(0.025 ). This is
mainly due to the different solubility of these halogenating agents,
the trihalide being much more soluble in water than the molecu-
lar iodine.
[
1] [1a]
F. A. Carey, R. J. Sundberg, Advanced Organic Chemistry,
3rd ed., Plenum Press, New York, 1990. [ J. March, Advanced
Organic Chemistry, 4th ed., VCH Publishers, New York, 1992.
1b]
General Iodination Reactions without Surfactant: Molecular iodine
Ϫ
3
(1.2 mmol), or I
, formed in situ by adding potassium iodide
Eur. J. Org. Chem. 2004, 623Ϫ630
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
629

S. Cerritelli, M. Chiarini, G. Cerichelli, M. Capone, M. Marsili
FULL PAPER
[
1c]
[16]
J. Fuhrhop, G. Penzlin, Organic Synthesis 2nd ed., VCH
Publishers, New York, 1993.
M. T. Bianchi, G. Cerichelli, G. Mancini, F. Marinelli, Tetra-
hedron Lett. 1984, 25, 5205Ϫ5208.
[17]
[
2] [2a]
OECD (Organization for Economic Cooperation and De-
velopment) Workshop on Sustainable Chemistry, 1998. [ P.
T Anastas, J. C. Warner, Green Chemistry: Theory and Practice,
Oxford University Press, Oxford, 1998. [ P. T Anastas, T. C.
Williamson, D. Hjeresen, J. Breen, Promoting Green Chemistry
Initiatives, in Environm. Sci. Technol., 1999, March 1.
M. Marsili, E. Marengo, H. Saller, Anal. Chim. Acta 1988,
210, 33Ϫ50.
2b]
[
18]
Ϫ
Ϫ
is 16 Ϫ1 (G.
As the equilibrium constant Br
2
ϩ Br ϭ Br
3
2c]
Jones, S. Baeckström, J. Am. Chem. Soc. 1934, 56, 1517Ϫ1523;
D. B. Scaife, H. J. V. Tyrrell, J. Chem. Soc. 1958, 386Ϫ392) it
is reasonable that the difference in reactivity observed between
the two brominating agents is due only to the different alkenes
and not to different brominating abilities of the two species. It
is reasonable to apply the same consideration to iodine-triiod-
ide.
[
3] [3a]
G. S. Hartley, Micellization, Solubilization and Microemul-
sions, vol. 1, (Ed.: K. L. Mittal), Plenum Press, New York,
[
3b]
1977.
J. H. Fendler, E. J. Fendler, Catalysis in Micellar and
[3c]
Macromolecular Systems, Academic Press, New York, 1975.
[
[
19] [19a]
J. H. Fendler, Membrane Mimetic Chemistry, Wiley Intersci-
ence, New York, 1982.
C. A. Bunton, F. Nome, F. H. Quina, L. S. Romsted, Acc.
Chem. Res. 1991, 24, 357Ϫ364. [
noy, J. Am. Chem. Soc. 1967, 89, 4698.
19b]
F. M. Menger, C. E. Port-
[
[
4]
5]
[19c]
C. A. Bunton, G. Savelli, Adv. Phys. Org. Chem. 1986, 22,
L. S. Romsted, in
2
13Ϫ309.
Surfactants in Solution (Eds.: K. L. Mittal, B. Lindman), Ple-
num Press, New York, 1984; vol. 2, p. 1015.
M.-F. Ruasse, J. Aubard, B. Galland, A. Adenier, J. Phys.
Chem. 1986, 90, 4382Ϫ4388.
D. Acharya, M. N. Das, J. Org. Chem. 1969, 34, 2828Ϫ2831.
R. Bacaloglu, C. A. Bunton, G. Cerichelli, F. Ortega, J. Phys.
Chem. 1989, 93, 1490Ϫ1497.
P. Di Profio, R. Germani, G. Savelli, G. Cerichelli, M. Chiarini,
G. Mancini, C. A. Bunton, N. D. Gillitt, Langmuir 1998, 14,
2662Ϫ2669.
M Chiarini, in Competition of nucleophiles on the surface of
the micellar aggregates and synthetic application, PhD thesis,
LЈAquila, Italia, 1999.
G. Savelli, R. Germani, L. Brinchi, in Reactions and Synthesis
in Surfactant Systems (Ed.: J. Texter), Surfactant Science series
n°100, Marcel Dekker, New York, 2001.
20]
[
[
[
[
6] [6a]
[21]
[22]
D. A. Jaeger, M. R. Frey, J. Org. Chem. 1982, 47, 311Ϫ315.
6b]
[
D. A. Jaeger, J. Jamrozik, T. G. Golich, M. W. Clennan, J.
Mohebalian, J. Am. Chem. Soc. 1989, 111, 3001Ϫ3006. [ D.
A. Jaeger, Y. M. Sayed, A. K. Dutta, Tetrahedron Lett. 1990,
6c]
[
[
[
23]
24]
25]
3
1, 449Ϫ450.
7] [7a]
G. H Shimdt, D. G. Garratt, The Chemistry of Doubly
Bonded Functional Groups, John Wiley & Sons, New York,
[
7b]
1
977.
ditions to Unsaturated Systems, 2nd ed., Elsevier, New York,
982.
P. B. D. De La Mare, R. Bolton, Electrophilic Ad-
1
G. Cerichelli, L. Luchetti, G. Mancini, G. Savelli, Langmuir
1999, 15, 2631Ϫ2634.
G. Cerichelli, G. Mancini, Langmuir 2000, 16, 182Ϫ187.
G. Mancini, C. Schiavo, G. Cerichelli, Langmuir 1996, 12,
3567Ϫ3573.
8] [8a]
R. B. Lennox, R. McClelland, J. Am. Chem. Soc. 1986,
[
8b]
[26]
108, 3771Ϫ3781.
G. Cerichelli, C. Grande, L. Luchetti, G.
[
27]
Mancini, C. A. Bunton, J. Org. Chem. 1987, 52, 5167Ϫ5171,
[
8c]
and ref. therein.
G. Cerichelli, C. Grande, L. Luchetti, G.
[
28]
Mancini, J. Org. Chem. 1991, 56, 3025Ϫ3030, and ref. therein.
F. M. Menger, Acc. Chem. Res. 1979, 12, 111Ϫ117.
P. Fromherz, Chem. Phys. Lett. 1981, 77, 460Ϫ466.
J. A. Butcher, G. W. Lamb, J. Am. Chem. Soc. 1984, 106, 1217.
9] [9a]
[29]
[30]
J. R. Atkinson, R. P. Bell, J. Chem. Soc. 1963, 3260Ϫ3269.
R. P. Bell, M. Pring, J. Chem. Soc., (B) 1966, 1119Ϫ1126.
[
9b]
[
10]
11]
12]
13]
14]
15]
[31] [31a]
J. W. Cornforth, D. T. Green, J. Chem. Soc., (C) 1970,
46Ϫ849 and ref. therein.
H. Sharghi, K. Niknamb, M. Pooyana, Tetrahedron 2001, 57,
057Ϫ6064.
G. Sumrell, B. M. Wiman, R. G. Howell, M. C. Harvey, Can.
J. Chem. 1964, 42, 2710Ϫ2712.
R. Antonioletti, M. D’Auria, A. De Mico, G. Piancatelli, A.
Scettri, Tetrahedron 1983, 39, 1765Ϫ1768.
H. Ohta, Y. Sakata, T. Takeuchi, Y. Ishii, Chem. Lett. 1990,
K. A. Dill, P. J. Flory, Proc. Natl. Acad. Sci. (U. S. A.)
31b]
8
1980, 77, 3115. [
Sci. (U. S. A.) 1981, 78, 676Ϫ680.
K. A. Dill, P. J. Flory, Proc. Natl. Acad.
[
[
[
[
[
[
[
[
[
32]
33]
34]
35]
6
William G. Hunter, J. Stuart Hunter, in Statistics for Exper-
imenters, John Wiley & Sons, New York 1978.
E. F. J. Duynstee, E. Grunwald, J. Am. Chem. Soc. 1959, 81,
4540Ϫ4542.
H. Masuda, K. Takase, M. Nishio, A. Hasegawa, Y. Nishi-
yama, Y. Ishii, J. Org. Chem. 1994, 59, 5550Ϫ5555.
H. Kotsuki, T. Shimanouchi, R. Ohshima, S. Fujiwara, Tetra-
hedron 1998, 54, 2709Ϫ2722.
7
73Ϫ736.
M. Smietana, V. Gouverneur, C. Mioskowski, Tetrahedron
Lett. 2000, 41, 193Ϫ195.
Received February 10, 2003
630
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 623Ϫ630