An Iron(III) Catalyst with Unusually Broad Substrate Scope in Regioselective Alkylation of Diols and Polyols

Article abstract of DOI:10.1002/chem.201504477

In this study, [Fe(dibm)₃] (dibm=diisobutyrylmethane) is shown to have unusually broad scope as a catalyst for the selective monoalkylation of a diverse set of 1,2- and 1,3-diol-containing structures. The mechanism is proposed to proceed via a cyclic dioxolane-type intermediate, formed between the iron(III) species and two adjacent hydroxyl groups. This approach represents the first transition-metal catalysts that are able to replace stoichiometric amounts of organotin reagents in regioselective alkylation. The reactions generally lead to very high regioselectivities and high yields, on par with, or better than, previous methods used for regioselective alkylation.

Full text of DOI:10.1002/chem.201504477

DOI: 10.1002/chem.201504477
Full Paper
&
Homogeneous Catalysis
An Iron(III) Catalyst with Unusually Broad Substrate Scope in
Regioselective Alkylation of Diols and Polyols
Bo Ren,[a] Olof Ramstrçm,[b] Qiang Zhang,[a] Jiantao Ge,[a] and Hai Dong*[a]
Abstract: In this study, [Fe(dibm)3] (dibm=diisobutyrylme-
thane) is shown to have unusually broad scope as a catalyst
for the selective monoalkylation of a diverse set of 1,2- and
1,3-diol-containing structures. The mechanism is proposed
to proceed via a cyclic dioxolane-type intermediate, formed
between the iron(III) species and two adjacent hydroxyl
groups. This approach represents the first transition-metal
catalysts that are able to replace stoichiometric amounts of
organotin reagents in regioselective alkylation. The reactions
generally lead to very high regioselectivities and high yields,
on par with, or better than, previous methods used for re-
gioselective alkylation.
cessfully used for the selective alkylation of carbohydrates;[4b]
however, it only shows good selectivity for vicinal cis-diols. In
addition, although the use of catalytic amounts of organotin
reagents was recently shown to be efficient for regioselective
benzylation, the selectivities for vicinal trans-diols were poor,
unless stoichiometric amounts of the organotin species were
used.[3d] Furthermore, methods based on silver(I) or nickel(II)
salts are only applicable in particular cases.[5,6]
Introduction
Regioselective protection plays a very important role in organic
synthesis.[1] For example, regioselective protection is often the
most important part of synthetic strategies for carbohydrate
building blocks that are used as donors or acceptors in com-
plex oligosaccharide synthesis.[2] One of the most important
protection methods is alkylation, in which benzyl ethers (Bn)
are the most widely used alkylating groups in such syntheses
and are often installed by regioselective methods. Such alkyla-
tions have, for example, been achieved with organotin,[3] orga-
noboron,[4] silver salts,[5] and nickel-based complexes.[6] Of
these, organotin reagents have been most widely used in se-
lective alkylation due to the high regioselectivities obtained,
coupled with relatively straightforward manipulation.[3e,f] When
stoichiometric amounts of organotin reagents are used, good
selectivities can be obtained in the alkylation of diols and poly-
ols containing two adjacent hydroxyl groups, be these 1,2-
diols, 1,3-diols, cis-diols, or trans-diols. However, a major draw-
back of organotin reagents is their inherent potential toxicity.[7]
Accordingly, reducing the amount of organotin species or the
use of nontoxic alternatives to them are of benefit, provided
that they still result in good regioselectivities.[3a,d,4b,8] An orga-
noboron catalyst, a diarylborinic acid derivative (Taylor’s cata-
lyst), the best nontoxic catalyst found to date, has been suc-
Although several regioselective, transition-metal-catalyzed
acylation methods have been reported, especially the use of
copper catalysts,[9] no general alkylation protocols have been
identified to date. This has been addressed herein, and we
report that an iron(III)-based catalyst, [Fe(dibm)3] (dibm= diiso-
butyrylmethane), leads to high regioselectivities with very
broad substrate scope in the alkylation of diols and polyols
(Figure 1). The method developed can be used in place of stoi-
Figure 1. Iron(III)-catalyzed regioselective alkylation of diols and polyols.
chiometric amounts of organotin species, and generally results
in the same or better regioselectivities with similar or higher
yields. In addition, iron is not only of low cost and high abun-
dance, but is also nontoxic and generally environmentally
benign.[10]
[a] B. Ren, Q. Zhang, J. Ge, Prof. H. Dong
Key laboratory of Material Chemistry for
Energy Conversion and Storage, Ministry of Education
School of Chemistry and Chemical Engineering
Huazhong University of Science and Technology
Luoyu Road 1037, 430074 Wuhan (P.R. China)
Fax: (+86)27-87793242
E-mail: hdong@mail.hust.edu.cn
Results and Discussion
[b] Prof. O. Ramstrçm
Department of Chemistry, KTH-Royal Institute of Technology
Teknikringen 30, 10044 Stockholm (Sweden)
The origin of the resulting regioselectivities by organotin re-
agents has long been hypothesized to be controlled by com-
plex stannylene dimer or polymer structures.[3e,f] This concep-
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201504477.
Chem. Eur. J. 2016, 22, 2481 – 2486
2481
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
tion is likely to have held back the development of other re-
agents used in selective protection because it is difficult to
identify structures that can form analogous dimer or polymer
structures. In 2012, after comparisons of the product patterns
of three fundamentally different reactions—neighboring group
participation, acyl group migration, and organotin-mediated
protection—we could, however, propose that the regioselectiv-
ities achieved by organotin compounds were not related to
complex stannylene structures, but instead resulted from the
stereoelectronic effects of the substrates themselves.[3c] It was
thus predicted that analogous selectivities could be obtained if
the organotin reagents were replaced by reagents that could
form cyclic dioxolane-type intermediates with hydroxyl groups.
Two such examples are organosilicon reagents developed in
our laboratory,[8c] and organoboron reagents reported by the
group of Taylor.[8d] More recently, a method for selective benzy-
lation with catalytic amounts of organotin species was also de-
veloped by us.[3d] In this method, 0.1 equivalents of dibutyltin
oxide is able to form dibutylstannylene acetals with vicinal hy-
droxyl groups of the substrates. Potassium carbonate is used
to deprotonate hydroxyl groups of substrates and aids in recy-
cling of the catalyst.
tonitrile. When FeCl3 or FeBr3 were used as the catalyst, 3-O-
benzylated mannoside 2 was isolated in yields of 80–85%
(Table 1, entries 2 and 3). Poor or no selectivity was observed
when AlCl3 or CuCl2 were used as catalysts or when no catalyst
was used (Table 1, entries 4–6). These results indicated that
iron(III) could play a key role in selective alkylation, potentially
other than BnBr activation. However, when other cis-diols were
treated with FeCl3 or FeBr3, poor regioselectivities were ob-
served. We suspected that the identity of the ligand influenced
the ability of FeIII to serve as an alkylation catalyst and turned
our attention to two other catalysts, [Fe(acac)3] (acac=acetyla-
cetonate) and [Fe(dibm)3], which have been successfully used
in, for example, carbon–carbon bond formation.[13] Although
[Fe(acac)3] (0.2 equiv) gave a low rate of alkylation of 1 due to
the instability of the acac ligand at the reaction temperature
(formation of BnOAc was observed, likely as a result of the ace-
tate anion generated by the hydrolysis of acac), [Fe(dibm)3]
(0.2 equiv) provided 2 in 96% yield (Table 1, entries 7 and 8).
Decreasing the catalyst loading to 0.1 equivalents led to identi-
cal results (Table 1, entry 9), and the yields of 2 only slightly de-
creased to 91 and 88% when the amount of [Fe(dibm)3] cata-
lyst was decreased to 0.05 and 0.02 equivalents, respectively
(Table 1, entries 1 and 10). In the absence of K2CO3, the prod-
uct yield decreased dramatically (Table 1, entry 11); this indi-
cates the essentiality of this reagent. The product yield also de-
creased slightly with a reduced reaction temperature (Table 1,
entry 12). These results demonstrate that [Fe(dibm)3] is
a highly reactive catalyst for the regioselective alkylation of 1.
The scope of the method was further evaluated with other
substrates containing a cis-diol moiety and with other alkyla-
tion reagents (Table 2). In all tested substrates, the equatorial
hydroxyl group was selectively alkylated in yields of 75–98%.
For the alkylation of compounds 1, 4, 6, and 8 (Table 2, en-
tries 1–3), high yields were obtained under standard conditions
(0.05 equiv of [Fe(dibm)3]). However, the alkylation of com-
pounds 10, 12, 14, and 16 (Table 2, entries 4–6) required more
[Fe(dibm)3] (0.1 equiv) and a longer reaction time (24 h). The al-
kylation of unprotected glycosides required the use of a mixed
solvent system (MeCN/DMF, 9:1) due to their poor solubility in
MeCN alone (Table 2, entries 5 and 6). The alkylation of com-
pound 18 gave 95% yield of product 19, which indicated that
this method could be used for the synthesis of thioglycoside
building blocks (Table 2, entry 7). The method is also applicable
to different alkylating reagents, leading to highly regioselective
alkylation in high yields (Table 2, entries 1, 3, and 4). The
method was further applied to the alkylation of the primary
hydroxyl group of substrates containing 1,2- or 1,3-diols
(Table 3). In the alkylation of glycosides containing 1,3-diols
(20, 22, 24, and 26), and non-glycoside substrates containing
1,2-diols (28, 30, 32, 34, and 36), the use of standard condi-
tions gave products in which the terminal hydroxyl group was
alkylated in 70–98% yield (Table 3, entries 1–3). However, the
1,3-diols 38 and 40 (Table 3, entry 4) required more [Fe(dibm)3]
(0.1 equiv) and a longer reaction time (24 h) to reach comple-
tion.
Iron(III) reagents are widely used in organic synthesis, show-
ing typical Lewis acidic properties and characteristic reactions
with electron-rich entities.[10] Iron(III) is also known to form che-
lates with, for example, catechols, forming stable siderophore
complexes,[11] and other diols and polyols.[12] We therefore en-
visaged that iron(III) would form cyclic dioxolane-type struc-
tures with vicinal, nonaromatic hydroxyl groups, thereby selec-
tively activating oxygen and replacing dibutyltin oxide in alky-
lations. Methyl 4,6-O-benzylidene-a-d-mannopyranoside (1)
was chosen as the first substrate to verify this hypothesis
(Table 1). Thus, compound 1 was treated with BnBr (1.1 equiv)
in the presence of an iron catalyst and K2CO3 (1.5 equiv) in ace-
Table 1. Comparison of the results obtained by deviating from the stan-
dard conditions for the reaction.
Entry
Conditions
Yield[a] [%]
1
2
3
4
5
6
7
8
optimized conditions[b]
FeCl3 (0.2 equiv)
FeBr3 (0.2 equiv)
AlCl3 (0.2 equiv)
CuCl2 (0.2 equiv)
without catalyst
[Fe(acac)3] (0.2 equiv)
[Fe(dibm)3] (0.2 equiv)
[Fe(dibm)3] (0.1 equiv)
[Fe(dibm)3] (0.02 equiv)
without K2CO3
91
85
80
[c]
–
–
–
–
[c]
[c]
[d]
96
96
88
9
10
11
12
[d]
–
reaction at 608C
80
[a] Yield of product isolated. [b] Reactant (50 mg), benzyl bromide (BnBr;
1.1 equiv), [Fe(dibm)3] (0.05 equiv), K2CO3 (1.5 equiv), MeCN (1 mL), 808C,
8 h. [c] Poor or no selectivity (3-OBn/2-OBn ca. 1/1). [d] Low conversion
(5–10% of 2).
The substrates containing 1,2-trans-diols were next investi-
gated. When stoichiometric amounts of organotin reagents are
Chem. Eur. J. 2016, 22, 2481 – 2486
www.chemeurj.org
2482
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Table 2. [Fe(dibm)3]-catalyzed regioselective alkylation of substrates con-
taining cis-diol.[a]
Table 3. [Fe(dibm)3]-catalyzed regioselective alkylation of substrates con-
taining a primary hydroxyl group.[a]
Entry
Reactant
Product
Yield [%]
Entry
1
Reactant
Product
Yield [%]
3a: 96, 3b: 98
R=PMB, BrBn
3c: 98, 3d: 97
R=allyl, CNBr
21: 80
23: 98
1
2
3
89
5: 91
27: 96
2
7a: 91, 7b: 75
R=Bn, allyl
9a: 87, 9b: 80
R=Bn, BrBn
29: 94
31: 75
33: 70[b]
35: 93
37: 80
3
11 a: 89, 11 b: 78
R=Bn, PMB
4[b]
39: 80
41: 88
4[c]
5[c]
86
15: 83
17: 90
[a] Reaction conditions: reactant (50 mg), BnBr (1.1 equiv), [Fe(dibm)3]
(0.05 equiv), K2CO3 (1.5 equiv), MeCN (1 mL), 808C, 8–12 h. Yields are
given for products isolated. [b] Mixed with 12% O2-benzylated product.
[c] BnBr (1.5 equiv), [Fe(dibm)3] (0.1 equiv), 24 h.
6[b,c]
7
95
products 47 and 49. Good selectivity was found for benzyla-
tion of 64 in comparison with poor selectivity for benzylation
of 62. Alkylation of thioglycosides 50 and 52 gave even higher
yields (Table 4, entries 1 and 2). Carbohydrate polyols without
an axial hydroxyl group displayed the lowest reactivities, and
0.2 equivalents of [Fe(dibm)3] had to be used in these cases
(Table 4, entries 3 and 4). As observed with the cis-diols, the al-
kylation of unprotected glycosides required the use of a mixed
solvent system (MeCN/DMF. 9:1; Table 4, entry 4). The alkyla-
tion of methyl a-glucopyranosides 54 and 58 gave O2-alkylat-
ed products 55 and 59 in 78 and 75% yield, respectively. In
contrast, methyl b-glucopyranosides 56 and 60, which differed
only in the nature of the O6 substituent (TBS in 56 and OH in
60), resulted in different selectivities. Good regioselectivity for
O2 was obtained for compound 60, whereas substrate 56 dis-
played poor selectivity. In all cases, better regioselectivities and
yields were obtained by using catalytic amounts of
[Fe(dibm)3], rather than stoichiometric amounts of organotin
reagents required (Table S1 in the Supporting Information).
In analogy to other FeIII complexes, cyclic dioxolane/dioxane-
type intermediates can be proposed to play key roles in the
catalytic mechanism (Figure 2). Following the formation of the
cyclic intermediate a between the substrate and [Fe(dibm)3],
intermediate a would further react with the alkylating reagent,
leading to intermediate c in which one position is alkylated
and the other is occupied by the Fe species. The process may
potentially also proceed via ionic species b. For comparison,
a recent calculation on dialkylstannylene acetals does not sup-
port the formation of an ionic species from the cyclic dioxo-
lane-type dialkylstannylene acetals.[14] Intermediate c would
subsequently undergo ligand exchange with a second sub-
[a] Reaction conditions: reactant (50 mg), RBr or 4-methoxybenzyl chlo-
ride (PMBCl; 1.1 equiv), [Fe(dibm)3] (0.05 equiv), K2CO3 (1.5 equiv), MeCN
(1 mL), 808C, 8–12 h. Yields are given of products isolated. [b] BnBr
(1.5 equiv), [Fe(dibm)3] (0.1 equiv), 24 h. For 10, 80 and 90% yields of 11 a
were isolated with 0.05 and 0.2 equivalents of [Fe(dibm)3], respectively.
[c] MeCN/DMF (9/1, 1 mL). For 16, 60% yield of 17 was isolated with only
MeCN as the solvent.
used, substrates containing trans-diols will usually exhibit poor
selectivity if the substituents adjacent to the diol are either
both equatorial or both axial. In contrast, good regioselectivity
for the hydroxyl group adjacent to the equatorial substituent
is observed if one of the adjacent substituents is equatorial
and the other one is axial.[3e] Neither the organoboron-cata-
lyzed method[4] nor the organotin-catalyzed method[3a,d] can be
applied to the selective alkylation of substrates containing
a trans-diol moiety. As discussed above, the alkylation of sub-
strates containing cis-diols and terminal diols by using
[Fe(dibm)3] revealed, for most cases, higher selectivities and
higher catalyst activity than the organotin method. To explore
the effects of [Fe(dibm)3] on the alkylation of 1,2-trans-diols,
4,6-O-benzylidene-protected glucopyranosides 42 and 44 and
4,6-O-benzylidene-protected galactopyranosides 46 and 48, in
which the 2,3-hydroxyl groups are free, and 2,6-position pro-
tected glucopyranoside 62 and mannoside 64, in which the
3,4-hydroxyl groups are free, were tested (Table 4, entries 1, 2,
and 5). Gratifyingly, the O2-alkylated products 43 and 45 were
isolated in 74 and 89% yield from 42 and 44, respectively,
whereas slightly lower yields were observed for O3-alkylated
Chem. Eur. J. 2016, 22, 2481 – 2486
www.chemeurj.org
2483
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Table 4. [Fe(dibm)3]-catalyzed regioselective alkylation of substrates con-
taining trans-diols.[a]
Table 5. Competitive benzylation of hydroxyl groups.[a]
Entry
Reactant
Product
Yield
Entry Reactant
Product
Yield
[%]
[%]
43: 74
45: 89
53: 92
90[b]
1
1
47: 66[b]
49: 70[b]
51: 89
2
3
91
86
2
71a/71b
(33/30)
4
5
55: 78
66 and 70
85
57: 50[d]
3[c]
[a] Reaction conditions: reactant (50 mg), BnBr (1.1 equiv), [Fe(dibm)3]
(0.05 equiv), K2CO3 (1.5 equiv), MeCN (1 mL), 8–12 h. Yields are given for
products isolated. [b] 31/31a (3:1).
59: 75
61: 70
4[c,e]
uct 31 was obtained from diol 30 in 90% yield, whereas no al-
kylation product was obtained from alcohol 25. We further en-
visaged that only cyclic intermediates with five- and six-mem-
bered rings would be stable and lead to alkylation. According-
ly, [Fe(dibm)3] would only be able to catalyze the alkylation of
1,2- and 1,3-diols. To test this hypothesis, we compared the al-
kylation of ethylene glycol (66; a 1,2-diol), 1,3-propanediol (68;
a 1,3-diol), and 1,4-butanediol (70; a 1,4-diol) with [Fe(dibm)3]
as the catalyst (Table 5, entries 2–5). The monoalkylated prod-
ucts 67 (obtained from 66) and 69 (obtained from 68) were
isolated in 91 and 86% yield, respectively. In contrast, a mixture
of mono- and dialkylated products, 71a and 71b (1/0.9), was
obtained from 70, which indicated that [Fe(dibm)3] was ineffec-
tive in the regioselective alkylation of this 1,4-diol. Similarly,
when diols 66 and 70 were coreacted with 1.1 equivalents of
BnBr in a competing alkylation, monoalkylated product 67 was
isolated in 85% yield from diol 66, whereas only trace amounts
of alkylation products were obtained from 70 (Figure S1 in the
Supporting Information). All of these results support that
a cyclic intermediate plays a key role in the catalytic mecha-
nism of [Fe(dibm)3].
63: 55[f]
65: 87
5
[a] Reaction conditions: reactant (50 mg), BnBr (1.1 equiv), [Fe(dibm)3]
(0.05 equiv), K2CO3 (1.5 equiv), MeCN (1 mL), 808C, 8–12 h. Yields are
given for products isolated. TBS=tert-butyldimethylsilyl. [b] O2-alkylated
products were also obtained, 29% for 46, 20% for 48. [c] BnBr (1.5 equiv),
[Fe(dibm)3] (0.2 equiv), 24 h. [d] O3-alkylated products were also obtained
in 40% yield. [e] MeCN/DMF (9/1, 1 mL). [f] 3-OBn/4-OBn in a ratio of ap-
proximately 55/45.
Conclusion
[Fe(dibm)3] catalysis represents an efficient and broadly appli-
cable method for the regioselective alkylation of 1,2- and 1,3-
diols and polyols containing 1,2- and 1,3-diols. This transition-
metal catalyst could be used as a universal replacement for
stoichiometric amounts of organotin reagents in the regiose-
lective alkylation of diols (both cis and trans) and polyols. The
products were obtained in the same or better regioselectivity
and in similar or higher yields than those of existing methods.
The operational simplicity of this new method, the avoidance
of organotin reagents, and the more universal applicability
Figure 2. Proposed catalytic mechanism.
strate, directly or via dibm, leading to the regeneration of the
cyclic intermediate a.
Support for the cyclic dioxolane-type intermediate mecha-
nism was found when diol 30 and its corresponding monohy-
dric alcohol 25 were coreacted with 1.1 equivalents of BnBr in
a competitive reaction (Table 5, entry 1). Monoalkylated prod-
Chem. Eur. J. 2016, 22, 2481 – 2486
www.chemeurj.org
2484
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
69.9, 69.0, 63.3, 55.4, 55.0 ppm; HRMS (ESI-TOF): m/z calcd for
C22H26O7Na [M+Na]+: 425.1576; found: 425.1564.
compared with other reagents renders this method attractive
for the preparation of selectively alkylated diol and polyol de-
rivatives. The alkylation patterns obtained by using catalytic
amounts of [Fe(dibm)3] were the same as those obtained with
methods that used stoichiometric amounts of organotin re-
agents. Accordingly, it is possible to predict the results of the
alkylation substrates through searching the extensive literature
on organotin-based alkylation. The results also support the
principle of organotin-based protection that we proposed in
2012.[3c] In other words, the regioselectivities achieved with or-
ganotin reagents are not related to the structure of the stanny-
lene complex, but to stereoelectronic effects of the substrates
themselves, and analogous selectivities should be obtained
when the organotin species is replaced by any reagent that
can form cyclic intermediates with two hydroxyl groups.
Methyl 4,6-O-benzylidene-3-O-4-bromobenzyl-a-d-manno-
pyranoside (3b)
1H NMR (400 MHz, CDCl3): d=7.55–7.32 (m, 7H; PhCH, Br-PhCH2),
7.21 (d, J=8.3 Hz, 2H; Br-PhCH2), 5.59 (s, 1H; PhCH), 4.82–4.70 (m,
2H; Br-PhCH2, H-1), 4.66 (d, J=12.2 Hz, 1H; Br-PhCH2), 4.27 (dd,
J1 =4.1 Hz, J2 =9.6 Hz, 1H; H-4), 4.08 (dd, J1 =7.1 Hz, J2 =14.3 Hz,
1H; H-6a), 4.01 (s, 1H; H-2), 3.94–3.72 (m, 3H; H-3, H-5, H-6b), 3.37
(s, 3H; OMe), 2.73 ppm (s, 1H; 2-OH); 13C NMR (100 MHz, CDCl3):
d=137.6, 137.1, 131.7, 129.6, 129.1, 128.4, 126.2, 121.9, 101.8,
101.2, 78.9, 75.7, 72.3, 70.0, 69.0, 63.3, 55.1 ppm; HRMS (ESI-TOF):
m/z calcd for C21H23O6BrNa [M+Na]+: 473.0576; found: 473.0552.
Methyl 3-O-allyl-4,6-O-benzylidene-a-d-mannopyranoside
(3c)
1H NMR (400 MHz, CDCl3): d=7.49 (d, J=7.7 Hz, 2H; PhCH), 7.44–
7.30 (m, 3H; PhCH), 5.83–5.61(m, 1H; CH2CH=CH2), 5.59 (s, 1H;
PhCH), 5.37–5.25 (m, 2H; CH2CH=CH2), 5.19 (d, J=10.4 Hz, 1H;
CH2CH=CH2), 4.78 (s, 1H; H-1), 4.39–4.23 (m, 2H; CH2CH=CH2, H-4),
4.20 (dd, J1 =5.9 Hz, J2 =12.8 Hz,1H; CH2CH=CH2), 4.12–4.00 (m,
2H; H-2, H-6a), 3.91–3.75 (m, 3H; H-3, H-5, H-6b), 3.39 (s, 3H; OMe),
2.77 ppm (s, 1H; 2-OH); 13C NMR (100 MHz, CDCl3): d=137.7, 134.6,
129.0, 128.3, 126.2, 117.6, 101.7, 101.2, 78.9, 75.3, 72.0, 70.0, 69.0,
63.3, 55.1 ppm; HRMS (ESI-TOF): m/z calcd for C17H22O6Na [M+Na]+
: 345.1314; found: 345.1296.
Experimental Section
General
All commercially available starting materials and solvents were of
reagent grade and used without further purification. Chemical re-
actions were monitored by TLC on precoated silica gel 60
(0.25 mm thickness) plates. High-resolution mass spectra were ob-
tained by ESI and Q-TOF detection. Flash column chromatography
was performed on silica gel 60 (SDS 0.040–0.063 mm). 1H and
13C NMR spectra were recorded on Bruker Avance 400 or Bruker
DMX 500 spectrometers at 298 K in CDCl3, with the residual signals
from CHCl3 (1H: d=7.25 ppm; 13C: d=77.2 ppm) as an internal
Methyl 4,6-O-benzylidene-3-O-2-cyanobenzyl-a-d-mannopyr-
anoside (3d)
1
standard. H NMR assignments were made by first-order analysis of
the spectra and supported by standard H–1H correlation spectros-
1
1H NMR (400 MHz, CDCl3): d=7.61 (d, J=7.6 Hz, 1H; PhCH), 7.56–
7.46 (m, 4H; CN-PhCH2), 7.41–7.33 (m, 4H; PhCH), 5.61 (s, 1H;
PhCH), 4.98 (d, J=12.4 Hz, 1H; CN-PhCH2), 4.92 (d, J=12.4 Hz, 1H;
CN-PhCH2), 4.78 (s, 1H; H-1), 4.29 (dd, J1 =4.3 Hz, J2 =9.8 Hz, 1H;
H-4), 4.23–4.13 (m, 2H; H-2, H-6a), 3.97–3.77 (m, 3H; H-3, H-5, H-
6b), 3.39 (s, 3H; OMe), 2.90 ppm (s, 1H; 2-OH); 13C NMR (100 MHz,
CDCl3): d=141.6, 137.6, 133.1, 132.9, 129.1, 128.3, 126.2, 117.8,
111.7, 101.7, 101.3, 78.8, 76.6, 71.0, 69.8, 69.0, 63.4, 55.1 ppm;
HRMS (ESI-TOF): m/z calcd for C22H23O6NNa [M+Na]+: 420.1423;
found: 420.1400.
copy (COSY).
General method for regioselective alkylation of diols and
polyols
Diol and polyol reactants (50 mg) were allowed to react with alky-
lation reagent (RBr and PMBCl, 1.1–1.5 equiv) in dry acetonitrile
(1 mL) or a mixture of solvents (MeCN/DMF: 9/1) at 808C for 8–
24 h, in the presence of [Fe(dibm)3] (0.05–0.2 equiv) and K2CO3
(1.5 equiv). The reaction mixture was directly purified by flash
column chromatography (hexanes/EtOAc=3:1 to 1:1) to afford the
pure selectively protected derivatives.
Methyl 3-O-allyl-6-(tert-butyldimethylsilyloxy)-a-d-manno-
pyranoside (7b)
The procedure for the preparation of catalyst [Fe(dibm)3] has been
reported in ref. [13].
1H NMR (400 MHz, CDCl3): d=6.10–5.88 (m, 1H; CH2CH=CH2), 5.32
(dd, J1 =17.2, J2 =1.5 Hz, 1H; CH2CH=CH2), 5.22 (dd, J1 =10.4, J2 =
1.1 Hz, 1H; CH2CH=CH2), 4.74 (s, 1H; H-1), 4.23–4.11 (m, 2H;
CH2CH=CH2), 3.97 (s, 1H; H-2), 3.93–3.79 (m, 3H; H-4, H-6a, H-6b),
3.67–3.52 (m, 2H; H-3, H-5), 3.37 (s, 3H; OMe), 3.06 (s, 1H; 4-OH),
2.42 (s, 1H; 2-OH), 0.90 (s, 9H; Si(CH3)3), 0.09 ppm (s, 6H; Si(CH3)2);
13C NMR (100 MHz, CDCl3): d=134.6, 118.0, 100.5, 79.0, 71.0, 70.8,
69.4, 67.9, 65.0, 55.0, 26.0, 18.4, À5.4 ppm; HRMS (ESI-TOF) m/z
calcd for C16H32O6SiNa [M+Na]+: 371.1866; found: 371.1851.
Spectroscopic data of known products were in accordance with
those reported in the literature.
Methyl 4,6-O-benzylidene-3-O-4-methoxybenzyl-a-d-manno-
pyranoside (3a)
1H NMR (400 MHz, CDCl3): d=7.62–7.18 (m, 7H; PhCH, MeO-
PhCH2), 6.85 (d, J=8.6 Hz, 2H; MeO-PhCH2), 5.60 (s, 1H; PhCH),
4.77 (d, J=11.4 Hz, 1H; MeO-PhCH2), 4.72(s, 1H; H-1), 4.62 (d, J=
11.4 Hz, 1H; MeO-PhCH2), 4.27 (dd, J1 =3.9 Hz, J2 =9.6 Hz, 1H; H-4),
4.07 (dd, J1 =7.1 Hz, J2 =14.3 Hz, 1H; H-6a), 3.96 (s, 1H; H-2), 3.91–
3.75 (m, 6H; H-3, H-5, H-6b, MeO-PhCH2), 3.36 (s, 3H; OMe),
2.83 ppm (s, 1H; 2-OH); 13C NMR (100 MHz, CDCl3): d=159.5, 137.7,
130.2, 129.7, 129.0, 128.3, 126.2, 113.9, 101.7, 101.2, 78.9, 75.4, 72.8,
Methyl 6-(tert-butyldimethylsilyloxy)-3-O-4-methoxybenzyl-
b-d-galactopyranoside (11 b)
1H NMR (400 MHz, CDCl3): d=7.31 (d, J=8.5 Hz, 2H; MeO-PhCH2),
6.89 (d, J=8.6 Hz, 2H; MeO-PhCH2), 4.67 (s, 2H; MeO-PhCH2), 4.16
(d, J=7.7 Hz, 1H; H-1), 3.99 (d, J=3.2 Hz, 1H; H-4), 3.91 (dd, J1 =
10.2, J2 =6.5 Hz, 1H; H-6a), 3.86- 3.72 (m, 5H; H-6b, MeO-PhCH2, H-
Chem. Eur. J. 2016, 22, 2481 – 2486
www.chemeurj.org
2485
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Y. W. Huang, C. C. Lee, K. L. Chang, S. C. Hung, Nature 2007, 446, 896–
899.
2), 3.54 (s, 3H; OMe), 3.46–3.35 (m, 2H; H-3, H-5), 2.53 (s, 1H; 4-
OH), 2.46 (s, 1H; 2-OH), 0.89 (s, 9H; Si(CH3)3), 0.08 ppm (s, 6H;
Si(CH3)2); 13C NMR (100 MHz, CDCl3): d=159.6, 130.0, 129.7, 114.2,
104.0, 80.5, 74.8, 71.8, 71.1, 66.1, 62.3, 57.0, 55.4, 26.0, 18.4,
À5.3 ppm; HRMS (ESI-TOF): m/z calcd for C21H36O7SiNa [M+Na]+:
451.2128; found: 451.2122.
[2] a) Q. Zhang, E. R. van Rijssel, M. T. C. Walvoort, H. S. Overkleeft, G. A. van
der Marel, J. D. C. Codee, Angew. Chem. Int. Ed. 2015, 54, 7670–7673;
b) S.-L. Tang, N. L. B. Pohl, Org. Lett. 2015, 17, 2642–2645; c) S. Visansiri-
kul, J. P. Yasomanee, P. Pornsuriyasak, M. N. Kamat, N. M. Podvalnyy, C. P.
Gobble, M. Thompson, S. A. Kolodziej, A. V. Demchenko, Org. Lett. 2015,
17, 2382–2384; d) D. Schmidt, F. Schuhmacher, A. Geissner, P. H. See-
berger, F. Pfrengle, Chem. Eur. J. 2015, 21, 5709–5713; e) S. J. Danishef-
sky, Y. K. Shue, M. N. Chang, C. H. Wong, Acc. Chem. Res. 2015, 48, 643–
652; f) D. Y. Zhu, K. N. Baryal, S. Adhikari, J. L. Zhu, J. Am. Chem. Soc.
2014, 136, 3172–3175.
[3] a) M. Giordano, A. Iadonisi, J. Org. Chem. 2014, 79, 213–222; b) Y. X.
Zhou, J. Y. Li, Y. J. Zhan, Z. C. Pei, H. Dong, Tetrahedron 2013, 69, 2693–
2700; c) H. Dong, Y. X. Zhou, X. L. Pan, F. C. Cui, W. Liu, J. Y. Liu, O. Ram-
strçm, J. Org. Chem. 2012, 77, 1457–1467; d) H. F. Xu, Y. C. Lu, Y. X.
Zhou, B. Ren, Y. X. Pei, H. Dong, Z. C. Pei, Adv. Synth. Catal. 2014, 356,
1735–1740; e) T. B. Grindley, Adv. Carbohydr. Chem. Biochem. 1998, 53,
17–142; f) S. David, A. Thieffry, A. Forchioni, Tetrahedron Lett. 1981, 22,
2647–2650.
[4] a) D. Lee, C. L. Williamson, L. N. Chan, M. S. Taylor, J. Am. Chem. Soc.
2012, 134, 8260–8267; b) L. N. Chan, M. S. Taylor, Org. Lett. 2011, 13,
3090–3093.
[5] a) B. Ren, M. Y. Wang, J. Y. Liu, J. T. Ge, H. Dong, ChemCatChem 2015, 7,
761–765; b) S. Malik, V. A. Dixit, P. V. Bharatam, K. P. R. Kartha, Carbohydr.
Res. 2010, 345, 559–564; c) A. Bouzide, G. Sauve, Tetrahedron Lett.
1997, 38, 5945–5948.
[6] U. B. Gangadharmath, A. V. Demchenko, Synlett 2004, 2191–2193.
[7] a) M. M. Whalen, B. G. Loganathan, K. Kannan, Environ. Res. 1999, 81,
108–116; b) S. M. Jenkins, K. Ehman, S. Barone, Jr., Dev. Brain Res. 2004,
151, 1–12.
[8] a) B. Ren, M. Rahm, X. L. Zhang, Y. X. Zhou, H. Dong, J. Org. Chem. 2014,
79, 8134–8142; b) Y. X. Zhou, M. Rahm, B. Wu, X. L. Zhang, B. Ren, H.
Dong, J. Org. Chem. 2013, 78, 11618–11622; c) Y. X. Zhou, O. Ramstrçm,
H. Dong, Chem. Commun. 2012, 48, 5370–5372; d) D. Lee, M. S. Taylor,
J. Am. Chem. Soc. 2011, 133, 3724–3727.
Methyl 6-O-benzyl-3-O-p-bromobenzyl-b-d-galactopyrano-
side (27)
1H NMR (400 MHz, CDCl3): d=7.45 (d, J=8.3 Hz, 2H; PhCH2), 7.40–
7.20 (m, 7H; PhCH2, Br-PhCH2), 4.72–4.60 (m, 2H; PhCH2), 4.58 (s,
2H; Br-PhCH2), 4.15 (d, J=7.8 Hz, 1H; H-1), 4.02 (d, 1H; H-4), 3.85–
3.68 (m, 3H; H-2, H-6a, H-6b), 3.58 (d, J=6.0 Hz, 1H; H-5), 3.53 (s,
3H; OMe), 3.38 (d, J=6.2 Hz, 1H), 2.81 (s, 1H; 2-OH), 2.63 ppm (s,
1H; 4-OH); 13C NMR (100 MHz, CDCl3): d=138.0, 136.9, 131.7,
129.6, 128.5, 127.9, 121.9, 104.1, 80.5, 73.8, 73.5, 71.2, 71.1, 69.2,
66.6, 57.1 ppm; HRMS (ESI-TOF): m/z calcd for C21H25O6BrNa
[M+Na]+: 475.0732; found: 475.0696.
Methyl 2-O-benzyl-6-(tert-butyldimethylsilyloxy)-a-d-gluco-
pyranoside (55)
1H NMR (400 MHz, CDCl3): d=7.41–7.27 (m, 5H; PhCH2), 4.75–4.64
(m, 2H; PhCH2), 4.61 (d, J=3.5 Hz, 1H; H-1), 3.93 (t, J=9.1 Hz, 1H;
H-3), 3.88–3.76 (m, 2H; H-6a, H-6b), 3.59 (dd, J1 =4.7 Hz, J2 =9.5 Hz,
1H; H-5), 3.56–3.49 (m, 1H; H-4), 3.39–3.31 (m, 4H; H-2, OMe), 3.00
(s, 1H; 4-OH), 2.60 (s, 1H; 3-OH), 0.89 (s, 9H; Si(CH3)3), 0.08 ppm (s,
6H; Si(CH3)2); 13C NMR (100 MHz, CDCl3): d=138.1, 128.7, 128.3,
97.7, 79.3, 77.5, 77.2, 76.8, 73.1, 72.4, 70.3, 64.2, 55.3, 26.0, 18.5,
À5.3 ppm; HRMS (ESI-TOF): m/z calcd for C20H34O6SiNa [M+Na]+:
421.2022; found: 421.2004.
[9] a) I. H. Chen, K. G. M. Kou, D. N. Le, C. M. Rathbun, V. M. Dong, Chem.
Eur. J. 2014, 20, 5013–5018; b) Y. Matsumura, T. Maki, S. Murakami, O.
Onomura, J. Am. Chem. Soc. 2003, 125, 2052–2053; c) C. L. Allen, S.
Miller, Org. Lett. 2013, 15, 6178–6181; d) E. V. Evtushenko, J. Carbohydr.
Chem. 2010, 29, 369–378; e) E. V. Evtushenko, Carbohydr. Res. 2012,
359, 111 –119; f) E. V. Evtushenko, J. Carbohydr. Chem. 2015, 34, 41–54.
[10] a) I. Bauer, H. J. Knçlker, Chem. Rev. 2015, 115, 3170–3387; b) B. D.
Sherry, A. Fürstner, Acc. Chem. Res. 2008, 41, 1500–1511; c) Y. Bour-
dreux, A. LemØtais, D. Urban, J.-M. Beau, Chem. Commun. 2011, 47,
2146–2148.
Acknowledgements
This study was supported by the National Nature Science
Foundation of China (no. 21272083) and the Chutian Project
sponsored by Hubei Province. We are also grateful to the staff
in the Analytical and Test Center of HUST for support with the
NMR instruments.
[11] M. Sandy, A. Butler, Chem. Rev. 2009, 109, 4580–4595.
[12] a) S. A. Cantalupo, S. R. Fiedler, M. P. Shores, A. L. Rheingold, L. H. Doer-
rer, Angew. Chem. Int. Ed. 2012, 51, 1000–1005; Angew. Chem. 2012,
124, 1024–1029; b) H. Tang, Y. Ye, T. Li, J. Zhou, G. Chen, Analyst 2003,
128, 974–979.
Keywords: alkylation · carbohydrates · green chemistry · iron ·
regioselectivity
[13] J. C. L. Lo, J. H. Gui, Y. K. Yabe, C. M. Pan, P. S. Baran, Nature 2014, 516,
343–348.
[14] S. M. Lu, R. J. Boyd, T. B. Grindley, J. Org. Chem. 2015, 80, 2989–3002.
[1] a) B. Wu, J. Ge, B. Ren, Z. Pei, H. Dong, Tetrahedron 2015, 71, 4023–
4030; b) L. Wang, M. Y. Dong, T. L. Lowary, J. Org. Chem. 2015, 80, 2767–
2780; c) B. Ren, H. Dong, O. Ramstrçm, Chem. Asian J. 2014, 9, 1298–
1304; d) K. L. M. Hoang, X. W. Liu, Nat. Commun. 2014, 5, 5051; e) H.
Dong, M. Rahm, N. Thota, L. Deng, T. Brinck, O. Ramstrçm, Org. Biomol.
Chem. 2013, 11, 648–653; f) C. C. Wang, J. C. Lee, S. Y. Luo, S. S. Kulkarni,
Received: November 9, 2015
Published online on January 20, 2016
Chem. Eur. J. 2016, 22, 2481 – 2486
www.chemeurj.org
2486
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim