N-Substituent effects on the diethylzinc addition to benzaldehyde catalysed by bicyclic 1,4-amino alcohols

Article abstract of DOI:10.1016/j.tetasy.2009.01.021

Chiral enantiopure bicyclic 1,4-amino alcohols were synthesised by a new methodology that provided a common precursor, which was easily N-functionalised with a wide variety of substituents. The final compounds were used as chiral ligands in a model study of the enantioselective addition of diethyl zinc to benzaldehyde, aimed at understanding the influence of the N-substituent on both the rate and stereoselectivity of the reaction. This set of experiments also provided interesting insight into the non-catalysed addition that occurred by employing commercially available Et₂Zn solutions.

Full text of DOI:10.1016/j.tetasy.2009.01.021

Tetrahedron: Asymmetry 20 (2009) 340–350
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
N-Substituent effects on the diethylzinc addition to benzaldehyde catalysed
by bicyclic 1,4-aminoalcohols
*
Dina Scarpi , Ernesto G. Occhiato, Antonio Guarna
Dipartimento di Chimica Organica ‘U. Schiff’, Laboratory of Design, Synthesis and Study of Biologically Active Heterocycles (HeteroBioLab), Università degli Studi di
Firenze, Via della Lastruccia 13, I-50019 Sesto Fiorentino, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Chiral enantiopure bicyclic 1,4-aminoalcohols were synthesised by a new methodology that provided a
common precursor, which was easily N-functionalised with a wide variety of substituents. The final com-
pounds were used as chiral ligands in a model study of the enantioselective addition of diethyl zinc to
benzaldehyde, aimed at understanding the influence of the N-substituent on both the rate and stereose-
lectivity of the reaction. This set of experiments also provided interesting insight into the non-catalysed
addition that occurred by employing commercially available Et2Zn solutions.
Received 4 December 2008
Accepted 23 January 2009
Available online 23 February 2009
Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction
In particular, as we had already obtained data concerning the
influence of the steric hindrance of N-alkyls (i.e., the less hindered,
The asymmetric alkylation of aldehydes by diethylzinc has been
used as a benchmark for potential chiral ligands,1 since the first
reports by Oguni2 and Noyori.3 In particular, the transition state
model proposed by Noyori and Yamakawa in 19953b has been
largely used to rationalise the experimental data in almost every
subsequent work.
the more stereoselective), which were congruent with the transi-
tion state model6a elaborated on the basis of the literature,3b we
decided to also explore the steric and electronic effects of groups,
such as aryl, fluoroalkyl and sulfonyl that were capable of affecting
the ability of the N atom to coordinate to the Zn atom. Benzalde-
hyde was chosen as a model substrate for the organozinc addition.
To realise this, we had to elaborate a new and more general
synthesis of the bicyclic scaffold, that could be easily and readily
N-fuctionalised.
Moreover, in addition to the results obtained by employing
these new chiral ligands in the model reaction, we herein report
a set of experiments showing the presence of a non-catalysed addi-
tion occurring when using commercially available Et2Zn solutions
and these having a strong influence on the enantiomeric purity
of the final product at times.
Over the past few years, this application has been more deeply
studied with some interesting papers being published, reporting
observations concerning the structure–activity relationship.4 Gen-
erally,5 1,2-amino alcohols have proven to be the best ligands for
the diethylzinc addition to carbonyls, but a few examples of 1,3-
and 1,4-amino alcohols have also been investigated, among which
the chiral 1,4-amino alcohols reported by us, whose structure is
based on a 6,8-dioxa-3-aza-bicyclo[3.2.1]octane scaffold.6
The application of N-methyl amino alcohol 1 (Scheme 1) as a
chiral ligand6b in the addition of Et2Zn to aromatic aldehydes re-
sulted in the corresponding (S)-alcohols with an ee ranging from
92% to 98%. Analogously, the phenylacetylene Et2Zn-catalysed
addition to aromatic aldehydes afforded the corresponding (R)-
propargylic alcohols with an ee ranging from 68% to 70%. In both
cases, 1 was employed at 15% mol.
2. Results and discussion
The synthesis that we reported most recently6b of this class of
compounds was planned in order to obtain N-methyl amino alco-
hol 1 (Scheme 1) starting from commercially available reagents
from the chiral pool, that is, 3 and 4.
With this in mind, we considered a similar synthetic route,
where the starting N-methyl amine 4 was replaced by the corre-
sponding primary amine 5 (Scheme 2).
We started the synthesis by adding lactone 3 to a solution of a
pre-formed aluminum amide, obtained by mixing aminoacetalde-
hyde dimethyl acetal 5 and AlMe3. The reaction gave amidoalcohol
6, formed in 95% yield, which required no purification for the next
step. Selective O-benzylation7 was achieved by treatment of 6 with
In the context of gaining further mechanistic insights into the
Et2Zn addition to aromatic aldehydes, the results obtained so far
with this class of 1,4-amino alcohols prompted us to further inves-
tigate the relationship between the nature of the N-substituent and
the chiral induction ability of the corresponding chiral ligand.
* Corresponding author. Tel.: +39 055 4573502; fax: +39 055 4573531.
E-mail address: dina.scarpi@unifi.it (D. Scarpi).
0957-4166/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2009.01.021

D. Scarpi et al. / Tetrahedron: Asymmetry 20 (2009) 340–350
341
O
O
Me OMe
N
O
O
Me
OMe
OMe
O
H
N
N
O
+
OMe
Me
O
HO
O
O
2
1
3
OPg
4
Scheme 1. Retrosynthetic analysis of the N-methyl amino alcohol 1.
O
O
O
OMe
OMe
OMe
OMe
O
OMe
OMe
H
N
O
H
N
a
b
O
+
H2N
(95%)
O
(55%)
O
O
O
6
O
7
3
5
BnO
OH
O
O
OMe
OMe
O
c
O
O
d
O
N
N
O
7
OBn
(65%)
(14%)
MeO
O
O
8
BnO
9
e
(32%)
O
O
O
O
OMe
OMe
O
OMe
OMe
H
f
g
O
N
N
O
N
MeO2C
(65%)
(66%)
BnO
11
10
12
OBn
OBn
Scheme 2. (a) AlMe3 2.0 M in hexane, DCM, 0 °C?rt, 18 h; (b) KOH, cat. 18-C-6, BnBr, THF, rt, 700; (c) nBuLi 1.8 M in hexane, ClCO2Me, THF, ꢁ70 °C, 4 h; (d) SiO2ꢀH2SO4,
toluene, reflux, 100; (e) BH3ꢀSMe2, THF, reflux, 1 h; (f) ClCO2Me (1.3 equiv), Et3N (1.3 equiv), DCM, 0 °C?rt, 1 h; (g) TFA–DCM, 1:2, 0 °C?rt, 18 h.
KOH in anhydrous THF, in the presence of a catalytic amount of 18-
crown-6 ether, followed by the addition of benzyl bromide. Prod-
uct 7, obtained in 55% yield, was then protected as an N-carbome-
thoxy amide; the reaction required the deprotonation of the amide
by nBuLi at low temperature, followed by the slow addition of
methyl chloroformate. Compound 8 was obtained in a 65% yield
after purification. The cyclisation step was performed by treatment
of the N-carbomethoxy amide 8 under catalysis of sulfuric acid
adsorbed on silica gel in refluxing toluene. Cyclic compound 9
was obtained in a disappointing 14% yield after purification from
many by-products. This reaction usually affords the cyclic scaffold
with fairly good yields and in the case of the analogous N-methyl
system, the lactam was obtained in a 48% yield. Therefore, the
low yield was probably due to the presence of the N-carbomethoxy
moiety; once the hydrolysis of the acetals takes place, the molecule
is still too rigid and hindered to cyclise to compound 9. To increase
the flexibility of the molecule, we decided to reduce the amide
bond in 8 and cyclise the N-methyl amine 10 under literature con-
ditions.8 Unfortunately, the reduction by BH3ꢀSMe2 of N-carbo-
methoxyl amide 8 to N-carbomethoxyl amine 10 failed and in
more drastic conditions resulted only in the deprotection and
decomposition of the starting material. Therefore, to synthesise
compound 10, we had to first perform the amide reduction and
then the amide protection. The reduction of amide 7 by BH3ꢀSMe2
afforded compound 11 after 1h at reflux in 32% yield; N-carbo-
methoxyl protection proceeded smoothly under standard condi-
tions and compound 10 was finally obtained in a 65% yield. The
cyclisation step was then performed with an excess of trifluoroace-
tic acid and afforded compound 12 in 66% yield. The overall yield
starting from 3 and 5 was 7% after five steps.
A critical analysis of the experimental data suggested to us that
the amide bond could be better reduced in an earlier stage and that
this route did not represent a general method for obtaining the free
1,4-amino alcohol as we planned.
Therefore, we started the synthesis under conditions similar to
the literature conditions,8 and modified the scheme after the
reductive amination (Scheme 3).
Compound 13 was obtained in 85% yield by NaBH4 reduction of
commercially available lactone 3 at pH 4–7 in water.9 Amino alco-
hol 14 was then obtained by the reductive amination of 13 with
amine 5, NaBH3CN at pH 6,10 in 62% yield. N-Carbomethoxylation
was obtained by treatment of 14 with 2.6 equiv of methyl chloro-
formate followed by selective O-deprotection with catalytic K2CO3
in methanol of the N-,O-carbomethoxy compound (86% yield).11
The O-benzylation, performed on 15, afforded compound 10 which
was finally cyclised to product 12 in 62% yield. This alternative
route that eliminates any inconvenience due to the presence of
the amide functional group, allowed us to increase the yield in
intermediate 12 from 7% to 19% over five steps.
The first N-functionalised amino alcohol was directly obtained
from 12 by debenzylation with H2 over Pd/C (84% yield). All other
functionalisations required the N-deprotection that was obtained
by treatment of compound 12 with an excess of KOH in a refluxing
mixture of methanol and water in a 4:1 ratio for 48 h. Compound
16 was thus obtained in 94% yield and we did not observe any
decomposition or changes in the stereochemistry in the bicyclic
structure. Compound 16 was then used as the starting material
for all subsequent N-functionalisation.
We first tried on our substrate, the conditions reported by Chan
and Lam for N-arylations with arylboronic acids and cupric

342
D. Scarpi et al. / Tetrahedron: Asymmetry 20 (2009) 340–350
O
O
O
O
O
O
O
OMe
OMe
O
OMe
OMe
b
H
N
a
O
c
O
N
(85%)
(62%)
(86%)
OH
O
O
O
3
XO
OH
14
13
15 X = H
d
a: R = CO2Me
f: R = Ts
(67%)
10 X = Bn
b
: R = Ph
c: R = 4-OMe-C6H4
g
: R = CH2CH2F
h: R = CH2CF3
d
: R = 2-Me-C6H4
i
: R = Pr
e
e: R = Ms
l: R = Me
(62%)
O
O
O
O
R
R
m or n
(34-88%)
H
f
17b d
g (
17e f)
,
-
), h (
N
O
N
O
N
O
m
N
O
17g h
17i l
i (
, ), l ( , )
MeO2C
(94%)
BnO
17b-l
HO
1, 18b-i
BnO
BnO
12
16
(84%)
18a
Scheme 3. (a) NaBH4 (3 equiv), H2O pH 4–7, 0 °C, 3 h; (b) 5, NaBH3CN, MeOH pH 6, 0 °C?rt, 22 h; (c) (i) ClCO2Me (2.6 equiv), Et3N (2.6 equiv), DCM, 0 °C?rt, 18 h; (ii) K2CO3
(5% mol), MeOH, rt, 3.5 h; (d) NaH, TBAI, BnBr, THF, 0 °C?rt, 7 h; (e) TFA–DCM, 1:2, 0 °C?rt, 18 h; (f) KOH, MeOH–H2O, 4:1, reflux, 48 h; (g) ArB(OH)2, Cu(OAc)2, Et3N, DCM,
O2, rt, 5 h; (h) MsCl or TsCl, Et3N, DCM, rt, 28 h; (i) (i) 2-fluoroethyltosylate, NaHCO3, DMF, 80 °C, 19 h or (ii) 2,2,2-trilfuoroethyl trifluoromethanesulfonate, NaHCO3, abs.
EtOH, reflux, 17 h; (l) (i) HCHO 37%, NaBH3CN, pH 7, CH3CN, rt, 4 h or (ii) CH3CH2CHO, NaBH3CN, pH 6, MeOH, rt, 18 h; (m) 10% Pd/C, H2, EtOAc, rt, 18 h; (n) 20% Pd(OH)2/C,
cyclohexene, abs EtOH, reflux 4–7.5 h.
acetate,12 taking into account also the modifications recently re-
ported for similar scaffolds.10 Compound 16 was treated with
phenylboronic acid (2 equiv), cupric acetate (1 equiv) and anhy-
drous pyridine (2 equiv) in anhydrous dichloromethane as a sol-
vent in a vessel open to the air. After 45 h, the starting material
was consumed but this was due to a side reaction, that is, the cyclic
acetal opening. The use of triethylamine as a base resulted in 17b,
in 42% yield after purification. In the recent literature on the cop-
per-mediated C–N bond formation,13 the use of catalytic cupric
acetate was reported either at room temperature or at 40 °C, and
with oxygen as the vessel atmosphere. The target compound 17b
was obtained in 10–14% yield (cupric acetate was used in 10–
14% mol amount), but the starting material 16 could be recovered
in small amounts. Therefore, the subsequent N-arylations were
performed under an oxygen atmosphere at room temperature
and with a stoichiometric amount of cupric acetate. Since long
reaction times (i.e., 18–24 h) proved only to afford more by-prod-
ucts, without an increase in yield, the reactions were stopped after
5 h. In this way, target compound 17b was obtained in 50% yield;
these conditions represent a good compromise between a fair yield
and an acceptable loss of starting material. They were then applied
to N-arylation of 16 with 4-methoxyphenylboronic acid (43% yield
of 17c) and o-tolylboronic acid, that, as expected, afforded 17d in
only 10% yield for steric reasons.
The preparation of N-sulfonylamides involved standard reaction
conditions consisting of the addition of the suitable sulfonyl chlo-
ride (1.3 equiv) to a solution of 16 in DCM, in the presence of Et3N
as a base. Compounds N-mesyl 17e and N-tosyl 17f were obtained
with 71% and 81% yield, respectively.
The N-fluoroalkyl derivatives were prepared by the nucleophilic
substitution of 2-fluoroethyltosylate14 and 2,2,2-trifluoroethyl tri-
fluoromethanesulfonate15 by 16, under literature conditions,16
while compounds 17g and 17h, were obtained in 70% and 65%
yield, respectively.
To complete the synthesis of this series of N-functionalised 1,4-
amino alcohols O-debenzylation was performed. However, the
conditions reported above for 12 worked only for compounds
17e and 17f. In the case of 17b, for example, O-debenzylation by
hydrogenation over Pd/C in either CH3OH or EtOAc failed unex-
pectedly; we supposed that the tertiary amine was able to poison
the catalyst, even though removal of benzyl groups of N-aryl 3,4-
dibenzyloxypyrrolidines was reported to work under these condi-
tions, and in excellent yields.18 If the presence of the tertiary amine
was the problem, the use of a few drops of aqueous HCl could avoid
the catalyst poisoning;19 however, in our hands the only result was
O-debenzylation and contemporary N-dearylation (88% yield) of
17b. Therefore, we opted for the catalytic transfer hydrogenation
performed over Pd(OH)2/C in a refluxing mixture of cyclohexene
and ethanol:20 after 7 h, the desired product 18b was obtained in
a quantitative yield. These conditions were then applied to all com-
pounds 17, with the exception of sulfonamides 17e and 17f, where
the problem of the tertiary amine did not exist.
Compounds 1, 18b–d and 18g–i were obtained in yields ranging
from 56% to quantitative.
The thus obtained N-substituted 1,4-amino alcohols were final-
ly employed as chiral ligands in the addition of Et2Zn to benzalde-
hyde (Scheme 4). We applied the reaction conditions that worked
best with the N-methyl ligand 1,6b and the results are reported in
Table 1.
O
OH
Et
L*
(15% mol)
+
Et2Zn
+
Ph
H
Ph
OH
Ph
toluene, r.t.
19
20
21
Scheme 4. Catalysed addition of Et2Zn to benzaldehyde.
Finally, to prove the new synthetic strategy as general as we
planned, we decided to also synthesise the N-methyl derivative
1. The N-methyl compound 17l was obtained by reductive amina-
tion with 37% HCHO and NaBH3CN (44%);17 analogously, N-propyl
17i was obtained starting from 16 and propionaldehyde (53%).
Before starting to test all the chiral ligands synthesised in the
present work, we ran a ‘ligand-free’ control experiment (entry 1).
Surprisingly, racemic 20 was formed in 43% yield after 18 h, to-
gether with benzylic alcohol 21 (17%). It was clear that a compet-
itive non-catalysed reaction was also occurring and since it has

D. Scarpi et al. / Tetrahedron: Asymmetry 20 (2009) 340–350
343
Table 1
Distribution of products obtained by diethylzinc addition on benzaldehyde
*
Entry
L
R
Time (h)
19a (%)
20a (%)
21a (%)
(S)-eeb (%)
1c
None
None
None
1
1
1
18b
18c
18d
18d
18e
18f
18g
18g
18g
22
18
18
6
18
18
6
18
18
18
6
18
18
18
18
6
40
83
87
0
43
14
8
17
3
5
4
4
5
7
20
6
6
0
0
0
96
96
96
0
0
6
14
0
0
65
84
85
90
2
2d
3d
4c
Me
96
88
84
48
40
61
23
32
43
46
73
54
98
64
90
5d
8
6d
11
45
40
33
71
60
41
38
22
40
0
7c
C6H5
4-MeO-C6H4
2-Me-C6H4
8c
9c
10d
11c
12c
13c
14d
15d
166a
17c
18c
Ms
Ts
CH2CH2F
8
17
16
6
6
2
e
CH2CH3
CH2CF3
Pr
24
18
18
18h
18i
20
4
16
6
73
a
Determined by GC.
b
c
TM
Determined by GC using b DEX 120 column.
Et2Zn 1.0 M in hexanes, Batch S38116.
Et2Zn 1.0 M in hexanes, Batch S29515.
Tested at 10% mol.
d
e
been well established that dialkylzinc itself does not add to car-
bonyl groups, we supposed that the commercial solutions of
diethyl zinc used in our experiments contained trace amounts of
impurities responsible for the side-addition.21 Testing a different
batch of diethyl zinc resulted in a noteworthy decrease in the
amount of racemic 20 (‘only’ 14%, entry 2), that did not signifi-
cantly changed when stopping the reaction after 6 h (8%, entry
3). Furthermore, the amount of 21 decreased (3–5%, entries 2 and
3). The same set of three experiments was carried out in the pres-
ence of N-methyl 1 (entries 4–6) and resulted in the important
observation that the activity of the ligand was independent of
the diethyl zinc batch used: (S)-20 was always obtained with
96% ee, whereas the amount of 21 set to 4–5% and only the per-
centage of the remaining benzaldehyde was affected, increasing
from 0 (entry 4) to 8–11% (entries 5 and 6). Based on these data,
we could reasonably rely on the results obtained regardless of
the commercial organozinc reagent quality, at least with the best
ligand 1.
of 112.19°, 110.62° and 112.19°) and the N-methyl group was thus
oriented far from the coordination site, allowing an easier access of
the organozinc reagent for the ligand-metal complex formation
(the X-ray structure of 1 hydrochloride was also determined,22
and the ORTEP plot is depicted in Fig. 2).
In addition to the steric hindrance of the N-substituent, in
N-phenyl 18b (entry 7) the lone pair of the nitrogen atom conju-
gates with the
p system of the aromatic ring and, therefore, is less
inclined to complex zinc. MM2 calculations, performed on com-
pound 18b (Fig. 1), clearly showed the sp2 nitrogen hybridisation
(bond angles of 119.39°, 119.52° and 120.98°) and the resulting
conformation of the N-phenyl substituent partially screened the
coordination site, making the interaction of Et2Zn with the ligand
more difficult. For comparison, compound 1 was also analysed
(Fig. 1): the sp3 nitrogen hybridisation was evident (bond angles
Figure 2. X-ray structure of 1ꢀHCl.
Consequently, only the competitive non-catalysed reaction oc-
curred with ligand 18b. In fact, racemic 20 formed (48%) and most
of benzaldehyde 19 was left unreacted (45%). The presence of an
electron-donating p-methoxy group on the N-aryl substituent
(18c, entry 8) did not alter the outcome of the reaction; only the
b-reduction reaction increased (20% of 21), and racemic 20 and
starting material 19 were obtained in an equal 40% amount.
Therefore, in order to restore the capability of the nitrogen lone
pair for metal complexation, its conjugation with the
p system had
to be broken; this was possible by choosing an o-tolyl group for the
next N-aryl 18d ligand (entry 9), in which the presence of the
o-methyl could force the aromatic ring to assume
a skew
Figure 1. MM2 structures of ligands 1 and 18b.

344
D. Scarpi et al. / Tetrahedron: Asymmetry 20 (2009) 340–350
conformation. Actually, (S)-20 formed in a higher amount (61%)
and showed some optical activity (ee 6%); however, the o-tolyl
group must be too hindered for rapid complex formation and the
catalysed reaction proceeded too slowly to compete with the side
non-catalysed addition. Since the presence of the latter could obvi-
ously weigh more on the cases of very low enantioselectivities, we
ran a second experiment with the same ligand 18d, using the best
batch of Et2Zn in our hands and stopping the reaction after 6 h (en-
try 10); most benzaldehyde was recovered unreacted (71% of 19)
but alcohol (S)-20 was obtained with 14% ee. The result itself
was not good in terms of enantioselectivity, but proved the
hypothesis of the role of the o-tolyl substituent of the ligand.
Based on these results, we did not expect any enantioselectivity
for the reactions conducted with N-solfonylamides 18e and 18f
(entries 11 and 12), in which again the nitrogen lone pair was
not available for N–Zn interactions. This was in fact the case, as
we only obtained the racemic alcohol 20 (32% and 43%, respec-
tively) and recovered the unreacted 19 (60% and 41%, respectively).
So far the N-alkyl ligands seemed to best complex the organo-
zinc reagent. However, the literature1,5 has reported countless
examples of chiral amino alcohols able to afford, in shorter reaction
times (2–3 h) and with minimal ligand amounts (1–2% mol), the
same enantiopure products obtained by employing our 1,4-amino
alcohols in higher amounts (15–20% mol) and over longer reaction
times (18 h). The efficiency of a catalyst is usually explained in
terms of turn over and, according to the reaction mechanism and
to the transition state model proposed (Fig. 3),6a we thought that
its rate could be influenced by the strength of the N-Zn interaction
in complexes A and B.
highlighted by the comparison with the results obtained with N-
ethyl 22 (Fig. 4),6a tested at 10% mol over 24 h (entry 16): (S)-20
was obtained in 98% yield and 90% ee. Hence the replacement of
H with F lowered the reaction rate and the enantioselectivity, most
probably because of a weakening of the N–Zn interaction. In this
case, we could hardly ascribe this decrease in stereoselectivity to
the difference in steric hindrance of the two atoms. This hypothesis
seemed confirmed by the almost total loss of enantioselectivity (ee
2%, entry 17) in the reaction that used N-CH2CF3 18h as a chiral li-
gand, where the very poor nucleophilicity of the nitrogen, due to
the presence of three F atoms, prevented the formation of a suffi-
ciently strong N–Zn interaction in the metal-ligand complexes (A
and B, Fig. 3).23
Finally, when employing N-propyl 18i, the steric hindrance ef-
fect of the substituent caused, as expected, a decrease of the
enantioselectivity and (S)-alcohol formed in 73% ee (entry 18);
no significant change was obtained with the best Et2Zn batch (ee
77%; data not shown).
Thus N-methyl 1 proved to be the best ligand belonging to this
class of 1,4-amino alcohols. Therefore, assuming 1 as datum point,
all data in our hands seemed to highlight that N-EWG ligands were
less enantioselective because of the weakening of the N–Zn inter-
action in the TS. The effects of N-EDG ligands could only be verified
with the introduction of EDGs at the a-position with respect to the
nitrogen atom. The synthesis focussed on obtaining such struc-
tures, which required some less than trivial changes in the syn-
thetic route; the results will be reported in due course.
3. Conclusions
The modulation of the potency of this interaction could be
realised by choosing either electron-donating or electron-with-
drawing N-substituents. The N-fluoroalkyl 18g, in which the
electron-withdrawing fluorine caused a decrease of the nucleophi-
licity of the nitrogen, afforded (S)-20 in 46% yield and with 65% ee
(entry 13). The use of a second batch of Et2Zn resulted in an incre-
ment of the enantiomeric excess of (S)-20 to 84% (entry 14), show-
ing how deeply the non-catalysed reaction could mislead in the
evaluation of the efficiency of a ligand. Consistently, by reducing
the reaction time from 18 to 6 h (entry 15), the amount of starting
19 increased (from 22% to 40%) but, as already seen in the case of 1
(entries 5 and 6), the enantiomeric excess held steady. The effect of
the N-(2-fluoroethyl) group on the outcome of the reaction can be
Herein, we have reported on a new synthetic methodology
tuned to the 1,4-amino alcohol scaffold 16 which is easily N-func-
tionalisable with a wide variety of substituent (aryls, sulfonyls, al-
kyls and fluoroalkyls). The thus obtained chiral compounds 18b–i
were then used as chiral ligands in the addition of Et2Zn to benzal-
dehyde. The results of these experiments gave an insight into the
influence of the nature of the N-substituent on the stereoselectivity
of the addition. However, the presence of a non-catalysed addition
could not be neglected; in fact, depending on the diethyl zinc batch
used, different amounts of racemic alcohol 20 formed and there-
fore the evaluation of the efficiency of a chiral ligand can be mis-
leading at least when the catalysed reaction proceeds too slowly.
R
O
N
O
O
R
R
Et2Zn
Et2Zn
N
N
O
O
(S)-alcohol
O
PhCHO
Zn
Zn
O
HO
O
Et
O
Zn
H
Et
Et
Et
A
Ph
TS anti S
B
Figure 3. Model of transition state.
O
N
O
HO
22
Figure 4. Structure of N-ethyl amino alcohol 22.

D. Scarpi et al. / Tetrahedron: Asymmetry 20 (2009) 340–350
345
Thus, a ‘blank’ experiment had to be carried out for every commer-
cial batch purchased. The results in our hands showed that N-EWG
ligands caused a decrease in the enantioselectivity, probably be-
cause of the weakening of the N–Zn interaction in the TS. The effect
of the electron-donating substituent could only be evaluated by a-
functionalised 1,4-amino alcohols, whose synthesis and applica-
3.83 (dd, J 10.6, 2.2 Hz, 1H), 3.45 (dd, J 10.6, 7.3 Hz, 1H), 3.39–
3.22 (m, 2H), 3.34 (s, 3H), 3.32 (s, 3H), 1.58 (s, 3. H), 1.39 (s, 3H).
13C NMR d (ppm): 169.1 (s), 138.0 (s), 128.3 (d, 2C), 127.6 (d,
2C), 127.5 (d), 110.2 (s), 102.3 (d), 76.7 (d), 75.6 (d), 73.4 (t), 68.9
(t), 54.3 (q, 2C), 40.2 (t), 27.0 (q), 24.7 (q). MS m/z (%): 353 (M+,
1), 234 (7), 91 (48). Anal. Calcd for C18H27NO6: C, 61.17; H, 7.70;
N, 3.96. Found: C, 61.44; H, 7.63; N, 4.03.
tion will be reported in due course.
4.1.3. (4R,5R)-(5-Benzyloxymethyl-2,2-dimethyl[1,3]dioxolane-
4-carbonyl)-(2,2-dimethoxyethyl)carbamic acid methyl ester 8
A solution of amide 7 (995 mg, 2.81 mmol) in anhydrous THF
(10 mL) was cooled to ꢁ70 °C. A n-BuLi solution (1.6 M in hexane,
1.8 mL, 2.81 mmol) was added dropwise, maintaining the temper-
ature below ꢁ60 °C, and, after 30 min, a solution of methylchloro-
4. Experimental
4.1. General
Melting points are uncorrected. Chromatographic separations
were performed under pressure on silica gel by flash-column tech-
niques; Rf values refer to TLC carried out on 25-mm silica gel plates
(Merck F254), with the same eluent as indicated for the column
chromatography. 1H NMR (200 MHz) and 13C NMR (50.33 MHz)
spectra were recorded with a Varian XL 200 instrument in CDCl3
solution. Mass spectra were carried out by EI at 70 eV, unless
otherwise stated, on Shimadzu GC/MS QP5050 instruments. Micro-
analyses were carried out with a Perkin–Elmer 2400/2 elemental
analyser. Optical rotations were determined with a JASCO DIP-
370 instrument.
formate (217 lL, 2.81 mmol) in anhydrous THF (2 mL) was added.
After 4 h at ꢁ70 °C, the mixture was allowed to warm at rt and
quenched by addition of satd NH4Cl (10 mL). The product was ex-
tracted with EtOAc (3 ꢃ 10 mL) and the combined organic layers
were dried over Na2SO4. After filtration and evaporation of the sol-
vent, the crude was purified by flash chromatography (eluent:
EtOAc–petroleum ether, 1:3 + 0.5% Et3N; Rf = 0.29) and pure 8
was obtained as a colourless oil (746 mg, 65%). Compound 8:
½
a 2D6
ꢄ
¼ þ63:2 (c 1.0, CHCl3). 1H NMR d (ppm): 7.37–7.18 (m, 5H),
5.56 (d, J 7.0 Hz, 1H), 4.80 (qAB, J 5.9 Hz, 1H), 4.47 (s, 2H), 4.49–
4.42 (m, 1H), 3.85–3.62 (m, 2H), 3.73 (s, 3H), 3.57–3.43 (m, 2H),
3.30 (s, 3H), 3.26 (s, 3H), 1.60 (s, 3H), 1.40 (s, 3H). 13C NMR d
(ppm): 170.8 (s), 154.4 (s), 137.8 (s), 128.1 (d, 2C), 127.4 (d, 3C),
109.9 (s), 101.5 (d), 100.9 (d), 77.5 (d), 76.5 (d), 73.2 (t), 69.1 (t),
54.4 (q), 54.3 (q), 53.8 (q), 45.1 (t), 27.4 (q), 25.4 (q). MS m/z (%):
411 (M+, 0.3), 91 (90), 75 (100). Anal. Calcd for C20H29NO8: C,
58.38; H, 7.10; N, 3.40. Found: C, 58.17; H, 6.87; N, 3.26.
2,2,2-Trifluoroethyl trifluoromethanesulfonate15 and 2-fluoro-
ethyltosylate14 were synthesised by published procedures. The
1.0 M Et2Zn solution in hexanes was purchased from Aldrich (cat.
No. 296112, batches S29515 and S38116).
4.1.1. (4R,5R)-5-Hydroxymethyl-2,2-dimethyl[1,3]dioxolane-4-
carboxylic acid (2,2-dimethoxyethyl)amide 6
A solution of 5 (2.27 mL, 20.9 mmol) in anhydrous DCM (77 mL)
was cooled at 0 °C; an AlMe3 solution (2.0 M in hexane, 10.5 mL,
20.9 mmol) was added dropwise and, after 30 min, lactone 3
(3.0 g, 19.0 mmol) was added. After 1 h the ice bath was removed
and the resulting solution was stirred for 19 h at room tempera-
ture. After cooling at 0 °C, concd HCl was added dropwise until a
slurry formed; water (50 mL) and brine (25 mL) were then added
and additional concd HCl until the slurry disappeared (pH ꢂ 2).
After separation of the phases, the aqueous one was extracted with
DCM (2 ꢃ 25 mL); the combined organic layers were dried over
Na2SO4. After filtration and evaporation of the solvent, crude 6
was obtained (4.78 g, 95%) and this was used in the next step with-
4.1.4. (1R,5R,7R)-7-Benzyloxymethyl-2-oxo-6,8-dioxa-3-aza-
bicyclo[3.2.1]octane-3-carboxylic acid methyl ester (9)
Prepared as reported in the literature6b starting from 8 (721 mg,
1.75 mmol) and obtaining pure 9 (74 mg, 14%) as a white solid
after FCC purification (eluent: EtOAc–petroleum ether, 2:3;
Rf = 0.33). Compound 9: mp 76–77 °C. ½a D23
¼ ꢁ27:5 (c 1.0, CHCl3).
ꢄ
1H NMR d (ppm): 7.37–7.25 (m, 5H), 5.77 (d, J 2.6 Hz, 1H), 4.69 (d, J
4.8 Hz, 1H), 4.59–4.45 (m, 2H), 4.27 (qAB, J 4.4 Hz, 1H), 3.82 (s, 3H),
3.75–3.53 (m, 4H). 13C NMR d (ppm): 165.8 (s), 153.3 (s), 137.3 (s),
128.3 (d, 2C), 127.7 (d, 3C), 98.3 (d), 78.5 (d), 77.6 (d), 73.7 (t), 67.4
(t), 54.1 (q), 51.7 (t). MS m/z (%): 307 (M+, 1), 201 (M+ꢁOBn, 14),
172 (32), 91 (100). Anal. Calcd for C15H17NO6: C, 58.63; H, 5.58;
N, 4.56. Found: C, 58.57; H, 5.55; N, 4.28.
out further purification. Compound 6: ½a D23
¼ þ3:5 (c 0.71, CHCl3).
ꢄ
1H NMR d (ppm): 7.02–6.90 (br s, 1H), 4.64–4.49 (m, 2H), 4.38 (t, J
5.1 Hz, 1H), 3.85–3.77 (m, 1H), 3.64–3.52 (m, 2H), 3.40 (s, 6H),
3.38–3.25 (m, 1H), 1.55 (s, 3H), 1.39 (s, 3H). 13C NMR d (ppm):
170.6 (s), 110.0 (s), 102.3 (d), 77.6 (d), 76.8 (d), 61.5 (t), 54.5 (q),
54.4 (q), 40.4 (t), 29.6 (q), 24.7 (q). MS m/z (%): 263 (M+, 5), 232
(27), 174 (100).
4.1.5. (4S,5R)-(5-Benzyloxymethyl-2,2-dimethyl[1,3]dioxolan-
4-ylmethyl)-(2,2-dimethoxyethyl)amine 11
Amidoalcohol 7 (1.0 g, 2.82 mmol) was dissolved into anhy-
drous THF (30 mL), and BH3ꢀSMe2 (10 M, 560
added dropwise. The resulting solution was refluxed for 1 h; after
cooling, abs EtOH (1.1 mL), 3 M NaOH (840 L) and water (56
lL, 5.64 mmol) was
4.1.2. (4R,5R)-5-Benzyloxymethyl-2,2-dimethyl[1,3]dioxolane-
4-carboxylic acid (2,2-dimethoxyethyl)amide 7
l
mL) were added dropwise and the resulting suspension was re-
fluxed for 1 h. After this time, the mixture was cooled at room tem-
perature and the product was extracted with EtOAc (2 ꢃ 75 mL);
the combined organic layers were dried over Na2SO4. After filtra-
tion and evaporation of the solvent, the crude was purified by flash
chromatography (eluent: DCM–MeOH, 30:1 + 0.5% Et3N; Rf = 0.40)
and pure 11 was obtained as a colourless oil (305 g, 32%). Com-
Freshly powdered KOH (917 mg, 16.3 mmol) was added to a
solution of amido alcohol 6 (3.91 g, 14.8 mmol) in anhydrous
THF (30 mL), followed by 18-C-6 ether (157 mg, 4% mol) and ben-
zyl bromide (1.94 mL, 16.3 mmol). The suspension was left to stir
at room temperature for 70 min. After this time, DCM (50 mL),
water (30 mL) and 1 M HCl (10 mL) were added; the phases were
separated and the organic one was washed with H2O (2 ꢃ 40 mL)
and dried over Na2SO4. After filtration and evaporation of the sol-
vent, the crude was purified by flash chromatography (eluent:
EtOAc–petroleum ether, 1:1 + 0.5% Et3N; Rf = 0.31) and pure 7
was obtained as a colourless oil (2.86 g, 55%). Compound 7:
pound 11: ½a 2D5
ꢄ
¼ ꢁ10:2 (c 0.76, CHCl3). 1H NMR d (ppm): 7.44–
7.28 (m, 5H), 4.55 (qAB, J 12.1 Hz, 2H), 4.44 (t, J 5.9 Hz, 1H),
4.38–4.27 (m, 2H), 3.57–3.45 (m, 2H), 3.37 (s, 3H), 3.36 (s, 3H),
2.78–2.72 (m, 4H), 2.06–1.87 (br s, 1H), 1.43 (s, 3H), 1.35 (s, 3H).
13C NMR d (ppm): 137.7 (s), 128.3 (d, 2C), 127.7 (d, 2C), 127.6
(d), 108.3 (s), 103.7 (d), 76.5 (d), 75.7 (d), 73.5 (t), 68.6 (t), 54.0
(q), 51.4 (t), 49.2 (t), 28.1 (q), 25.5 (q). MS m/z (%): 324 (M+ꢁCH3,
½
a 2D4
ꢄ
¼ þ12:9 (c 1.0, CHCl3). 1H NMR d (ppm): 7.46–7.22 (m, 5H),
6.81–6.67 (br s, 1H), 4.62–4.48 (m, 4H), 4.26 (t, J 5.5 Hz, 1H),

346
D. Scarpi et al. / Tetrahedron: Asymmetry 20 (2009) 340–350
2), 264 (27), 91 (100). Anal. Calcd for C18H29NO5ꢀH2O: C, 61.52; H,
was purified by flash chromatography, eluting first with EtOAc
8.70; N, 3.99. Found: C, 61.72; H, 8.75; N, 4.08.
and then with DCM–MeOH, 20:1 (Rf = 0.50), affording pure 14
(1.62 g, 62%) as a pale yellow oil. Compound 14: ½a D26
¼ ꢁ7:2 (c
ꢄ
4.1.6. (4S,5R)-(5-Benzyloxymethyl-2,2-dimethyl[1,3]dioxolan-4-
ylmethyl)-(2,2-dimethoxyethyl)carbamic acid methyl ester 10
From 11: A solution of 11 (1.2 g, 3.5 mmol) in anhydrous DCM
0.50, CHCl3). 1H NMR d (ppm): 4.43 (t, J 5.5 Hz, 1H), 4.34–4.27
(m, 2H), 3.80–3.65 (m, 2H), 3.38 (s, 6H), 3.01–2.80 (m, 2H), 2.75
(d, J 5.5 Hz, 2H), 1.43 (s, 3H), 1.34 (s, 3H). 13C NMR d (ppm):
107.9 (s), 103.2 (d), 77.44 (d), 75.8 (d), 60.4 (t), 54.3 (q, 2C), 50.8
(t), 48.6 (t), 27.4 (q), 24.9 (q). MS m/z (%): 249 (M+, 0.3), 174
(60 mL) was cooled at 0 °C; Et3N (630
lL, 4.6 mmol, 1.3 equiv)
and methyl chloroformate (350 L, 4.6 mmol, 1.3 equiv) were then
l
(M+ꢁCH(OCH3)2, 34), 118 (74). Anal. Calcd for C11H23NO5ꢀ 4H2O:
1
added dropwise. The mixture was then allowed to warm to room
temperature and was stirred for 1 h. The solution was then washed
with water (40 mL), brine (40 mL) and dried over Na2SO4. After fil-
tration and evaporation of the solvent, the residue was purified by
FCC (eluent: EtOAc–petroleum ether, 1:2; Rf = 0.35) to afford pure
10 as a colourless oil (904 mg, 65%).
C, 52.06; H, 9.33; N, 5.52. Found: C, 52.11; H, 9.64; N, 5.75.
4.1.9. (4S,5R)-(2,2-Dimethoxyethyl)-(5-hydroxymethyl-2,2-
dimethyl[1,3]dioxolan-4-ylmethyl)carbamic acid methyl
ester 15
From 15: A solution of 15 (1.72 g, 5.6 mmol) in anhydrous THF
(56 mL) was cooled at 0 °C; NaH (60% suspension in mineral oil,
248 mg, 6.2 mmol) and tetrabutylammonium iodide (10% mol)
were added, and the suspension was stirred under nitrogen for
A solution of 14 (1.62 g, 6.5 mmol) in anhydrous DCM (65 mL)
was cooled at 0 °C; Et3N (1.2 mL, 8.5 mmol, 1.3 equiv) and methyl
chloroformate (657 lL, 8.5 mmol, 1.3 equiv) were added dropwise.
A further 1.3 equiv of Et3N and methyl chloroformate was added
after 20 min, the ice bath was removed and the resulting solution
was stirred at room temperature until 14 was consumed (TLC; elu-
ent: EtOAc–petroleum ether, 2:1). The solution was then washed
with water (2 ꢃ 40 mL), brine (60 mL) and dried over Na2SO4. After
filtration and evaporation of the solvent, the residue was taken up
into methanol (15 mL) and anhydrous K2CO3 (5% mol) was added.
The resulting solution was stirred at rt for 3.5 h; the solvent was
then removed by evaporation and the crude was directly purified
by FCC (eluent: DCM–MeOH, 30:1; Rf = 0.11) to afford pure 15 as
30 min. Benzyl bromide (686
lL, 5.8 mmol) was then added and,
after 30 min, the ice bath was removed and the suspension was
stirred at room temperature for further 7 h. The reaction was then
quenched by addition of satd NH4Cl (28 mL) and water (56 mL);
the product was extracted with EtOAc (3 ꢃ 70 mL) and the com-
bined organic phases were washed with brine (56 mL) and dried
over Na2SO4. After filtration and evaporation of the solvent, the
crude was purified by FCC (eluent EtOAc–petroleum ether, 1:2;
Rf = 0.39), affording pure 10 as a yellowish oil (1.5 g, 67%). Com-
pound 10: ½a 2D3
ꢄ
¼ ꢁ54:4 (c 0.87, CHCl3). 1H NMR d (ppm): 7.34
a colourless oil (1.72 g, 86%). Compound 15: ½a D23
¼ ꢁ58:7 (c 0.93,
ꢄ
(br s, 5H), 4.63–4.49 (m, 2H), 4.41–4.33 (m, 2H), 3.79–3.49 (m,
5H), 3.70 (s, 3H), 3.38 (s, 3H), 3.35 (s, 3H), 3.31–3.12 (m, 2H),
1.46 (s, 3H), 1.32 (s, 3H). 13C NMR d (ppm) [two rotamers]: 156.7
(s), 137.8 (s), 128.3 (d, 2C), 127.7 (d, 3C), 108.8 (s), 103.7 and
103.5 (d), 76.3 (d), 75.7 and 75.5 (d), 73.5 (t), 68.4 (t), 54.9 (q),
54.4 and 54.0 (q), 52.7 (q), 49.9 and 49.7 (t), 48.4 and 47.9 (t),
28.0 (q), 25.5 (q). MS m/z (%): 382 (M+-CH3, 4), 307 (4), 218 (14),
91 (74), 75 (100). Anal. Calcd for C20H31NO7: C, 60.44; H, 7.86; N,
3.52. Found: C, 60.18; H, 7.91; N, 3.49.
CHCl3). 1H NMR d (ppm): 4.52–4.25 (m, 2H), 4.22–4.14 (m, 1H),
3.67 (s, 3H), 3.62–3.52 (m, 4H), 3.34 (s, 6H), 3.31–3.23 (m, 2H),
1.43 (s, 3H), 1.29 (s, 3H). 13C NMR d (ppm) [two rotamers]: 156.6
(s), 108.3 (s), 103.4 (d), 77.1 (d), 76.2 (d), 60.9 (t), 54.8 (q), 54.3
and 54.0 (q), 52.7 (q), 49.8 (t), 48.3 and 47.9 (t), 27.9 (q), 25.2
(q). MS m/z (%): 292 (M+ꢁCH3, 2), 186 (5), 144 (8), 75 (100). Anal.
1
Calcd for C13H25NO7ꢀ 4H2O: C, 50.07; H, 8.24; N, 4.49. Found: C,
50.06; H, 8.65; N, 4.50.
4.1.10. (1S,5S,7R)-7-Benzyloxymethyl-6,8-dioxa-3-azabicyclo
[3.2.1]octane-3-carboxylic acid methyl ester 12
4.1.7. (2R,3R)-2,2-Dimethyltetrahydrofuro[3,4-d][1,3]dioxol-4-
ol 13
A solution of 10 (1.5 g, 3.8 mmol) in DCM (7.6 mL) was cooled at
0 °C and TFA (3.8 mL) was dropwise added. After 10 min, the ice
bath was removed and the solution was stirred at rt overnight.
The reaction was stopped by cautious addition of satd NaHCO3
(15 mL) and Na2CO3 (s), under vigorous stirring, until pH was 7–
8. The phases were then separated, the aqueous layer was ex-
tracted with DCM (2 ꢃ 8 mL). The combined organic layers were
washed with water (15 mL), brine (15 mL) and dried over Na2SO4.
After filtration and evaporation of the solvent, the crude was puri-
fied by FCC (eluent: EtOAc–petroleum ether, 1:3; Rf = 0.26), afford-
ing pure 12 (736 mg, 66%) as a pale yellow oil. Compound 12:
A solution of 3 (1.95 g, 12.3 mmol) in water (100 mL) was
cooled in an ice bath, and NaBH4 (350 mg, 3 equiv) was added por-
tion-wise over 2 h under stirring. The pH was monitored continu-
ously and maintained between 4 and 7 by the addition of H2SO4
(3 N). When the reduction was complete (TLC; eluent: Et2O–petro-
leum ether, 1:1), water was evaporated and CH3OH was added to
the residue and evaporated (3 ꢃ 40 mL). Finally, CHCl3 (100 mL)
was added to the residue and the solution was dried over Na2SO4.
Filtration through a pad of Celite–silica gel and evaporation of the
solvent provided crude 13 (1.68 g, 85%) as a colourless oil. This was
used in the next reaction without further purification. Compound
13: 1H NMR d (ppm): 5.42 (s, 1H), 4.84 (dd, J 6.2, 3.7 Hz, 1H),
4.58 (d, J 5.86 Hz, 1H), 4.12–3.99 (m, 2H), 2.64 (br s, 1H), 1.47 (s,
3H), 1.32 (s, 3H).
½
a 2D6
ꢄ
¼ ꢁ46:9 (c 1.25, CHCl3). 1H NMR d (ppm) [two rotamers]:
7.42–7.27 (m, 5H), 5.53 (s, 1H, minor rotamer), 5.49 (s, 1H; major
rotamer), 4.66–4.52 (m, 2H), 4.48–4.34 (m, 1H), 4.25–4.15 (m, 1H),
3.97–3.83 (m, 1H), 3.82–3.71 (m, 2H), 3.70 (s, 3H, major rotamer),
3.67 (s, 3H, minor rotamer), 3.59–3.51 (m, 1H), 3.33–3.23 (m, 1H),
3.11–2.99 (m, 1H). 13C NMR d (ppm) [two rotamers]: 137.7 (s),
128.4 (d, 2C), 127.8 (d, 2C), 127.5 (d), 98.7 and 98.4 (d), 76.2 and
75.8 (d), 73.8 (t), 72.7 and 72.3 (d), 68.3 (t), 52.8 (q), 48.3 (t),
43.4 and 43.1 (t). MS m/z (%): 293 (M+, 5), 202 (M+ꢁBn, 6), 91
(100). Anal. Calcd for C13H17NO3ꢀH2O: C, 60.19; H, 6.62; N, 4.68.
Found: C, 60.23; H, 6.30; N, 4.76.
4.1.8. (4S,5R)-{5-[(2,2-Dimethoxyethylamino)methyl]-2,2-dimethyl
[1,3]dioxolan-4-yl}methanol 14
A solution of 13 (1.68 g, 10.5 mmol) and 5 (1.26 mL, 11.6 mmol)
in CH3OH (53 mL) was cooled at 0 °C; NaBH3CN (726 mg, 11.6
mmol) was added, and the pH was adjusted to 6 with glacial
CH3CO2H. The ice bath was removed after 15 min and the solution
was stirred for 22 h at rt. After evaporation of the solvent, the res-
idue was treated with satd Na2CO3 (25 mL) and the product was
extracted with CHCl3 (3 ꢃ 20 mL). The combined organic layers
were washed with water (20 mL), brine (20 mL) and dried over
Na2SO4. After filtration and evaporation of the solvent, crude 14
4.1.11. (1S,5S,7R)-7-Benzyloxymethyl-6,8-dioxa-3-azabicyclo
[3.2.1]octane 16
Powdered KOH (1.4 g, 25 mmol) was added to a solution of 12
(736 mg, 2.5 mmol) in MeOH (32 mL) and water (8 mL). The

D. Scarpi et al. / Tetrahedron: Asymmetry 20 (2009) 340–350
347
resulting solution was refluxed for 48 h and a second portion of
KOH (1.4 g, 25 mmol) was added after 24 h. After cooling at rt,
the solvent was removed in vacuo, the residue was taken up into
water (15 mL) and the product was extracted with CHCl3 (3 ꢃ 15
mL). After filtration and evaporation of the solvent, amino alcohol
16 was obtained as a pale yellow oil (553 mg, 94%). Compound 16:
13C NMR d (ppm): 149.0 (s), 137.8 (s), 132.8 (s), 131.4 (d), 128.3
(d, 2C), 127.7 (d, 2C), 127.6 (d), 126.7 (d), 123.8 (d), 120.1 (d),
99.9 (d), 77.1 (d), 74.0 (d), 73.5 (t), 68.2 (t), 55.5 (t), 49.6 (t), 18.2
(q). MS m/z (%): 325 (M+, 7), 234 (M+ꢁBn, 94), 118 (100), 91 (93).
Anal. Calcd for C20H23NO3: C, 73.82; H, 7.12; N, 4.30. Found: C,
73.95; H, 7.46; N, 4.26.
½
a 2D4
ꢄ
¼ ꢁ52:0 (c 1.03, CHCl3). 1H NMR d (ppm): 7.40–7.30 (m, 5H),
5.40 (s, 1H), 4.61 (qAB, J 12.1 Hz, 2H), 4.24–4.10 (m, 2H), 3.99–3.84
(m, 2H), 3.24–3.12 (m, 1H), 2.88 (qAB, J 13.2 Hz, 2H), 2.66 (d, J 13.2
Hz, 1H). 13C NMR d (ppm): 137.3 (s), 128.2 (d, 2C), 127.6 (d), 127.5
(d, 2C), 100.7 (d), 77.0 (d), 74.5 (d), 73.6 (t), 67.5 (t), 49.0 (t), 45.2
(t). MS m/z (%): 144 (M+ꢁBn, 38), 91 (100). Anal. Calcd for
C13H17NO3ꢀH2O: C, 64.71; H, 7.38; N, 5.80. Found: C, 64.24; H,
7.37; N, 5.90.
4.2.4. (1S,5S,7R)-7-Benzyloxymethyl-3-methanesulfonyl-6,8-
dioxa-3-azabicyclo[3.2.1]octane 17e
At first, Et3N (62
lL, 0.44 mmol) and CH3SO2Cl (50 mg, 0.44
mmol) were added to a solution of 16 (80 mg, 0.34 mmol) in anhy-
drous DCM (7 mL), stirred at room temperature and under nitrogen
atmosphere. The reaction was monitored by TLC (eluent: DCM–
MeOH, 20:1) and, after 28 h, DCM was added (8 mL) and the organ-
ic phase was washed with 1 M HCl (10 mL), satd NaHCO3 (10 mL),
water (2 ꢃ 10 mL), brine (10 mL) and finally dried over Na2SO4.
After filtration and evaporation of the solvent, the crude was puri-
fied by FCC (eluent: DCM–MeOH, 60:1; Rf = 0.65), affording pure
17e as a white solid (86 mg, 81%). Compound 17e: mp 90–91 °C.
4.2. General procedure for arylation with arylboronic acids
To a solution of amino alcohol 16 (118 mg, 0.5 mmol) in anhy-
drous DCM (3.2 mL) were added the desired aryl boronic acid (1.0
mmol), anhydrous Cu(AcO)2 (0.5 mmol) and Et3N (1.0 mmol). The
resulting blue suspension was stirred under an oxygen atmosphere
(balloon) at room temperature for 5 h. The dark green suspension
was then diluted with DCM (12 mL) and washed with 10% aq
NH4OH (3 ꢃ 10 mL). The aqueous phases were extracted once with
DCM (10 mL) and the combined organic layers were washed with
water (2 ꢃ 10 mL), brine (15 mL) and dried over Na2SO4. After fil-
tration and evaporation of the solvent, the crude was straightly
purified by FCC (eluent: EtOAc–petroleum ether, 1:7 + 0.5% Et3N)
to afford pure N-aryl compounds 17b–d.
½
a 2D6
ꢄ
¼ ꢁ47:4 (c 0.74, CHCl3). 1H NMR d (ppm): 7.36–7.28 (m,
5H), 5.59 (s, 1H), 4.59 (s, 2H), 4.51–4.45 (m, 1H), 4.30–4.21 (m,
1H), 4.03–3.83 (m, 2H), 3.69 (d, J 11.9 Hz, 1H), 3.52 (d, J 11.4 Hz,
1H), 3.16 (dd, J 12.1, 2.9 Hz, 1H), 2.92 (d, J 11.4 Hz, 1H), 2.74 (s,
3H). 13C NMR d (ppm): 137.7 (s), 128.4 (d, 2C), 127.9 (d, 2C),
127.8 (d), 98.3 (d), 76.3 (d), 73.8 (t), 72.5 (d), 68.0 (t), 49.1 (t),
44.7 (t), 34.5 (q). MS m/z (%): 313 (M+, 1), 234 (M+ꢁMs, 8), 128
(8), 91 (100). Anal. Calcd for C14H19NO5S: C, 53.66; H, 6.11; N,
4.47. Found: C, 53.95; H, 6.46; N, 4.86.
4.2.5. (1S,5S,7R)-7-Benzyloxymethyl-3-(toluene-4-sulfonyl)-
6,8-dioxa-3-azabicyclo[3.2.1]octane 17f
4.2.1. (1S,5S,7R)-7-Benzyloxymethyl-3-phenyl-6,8-dioxa-3-
azabicyclo[3.2.1]octane 17b
Prepared as reported for 17e. Starting from 16 (89 mg, 0.38
mmol) and TsCl (95 mg, 0.50 mmol) and obtaining, after purifica-
tion by FCC (eluent: EtOAc–petroleum ether, 1:1; Rf = 0.26), pure
17f (100 mg, 71%) as a white solid. Compound 17f: mp 145–146
White solid (50%). Rf = 0.38. mp 64–65 °C. ½a D23
¼ ꢁ59:1 (c 0.75,
ꢄ
CHCl3). 1H NMR d (ppm): 7.38–7.30 (m, 5H), 7.28–7.22 (m, 2H),
6.82 (t, J 7.0 Hz, 1H), 6.69–6.64 (m, 2H), 5.65 (s, 1H), 4.59–4.55
(m, 1H), 4.53 (qAB, J 12.1 Hz, 2H), 4.31–4.22 (m, 1H), 3.91–3.83
(m, 1H), 3.75–3.66 (m, 1H), 3.54–3.47 (m, 2H), 3.15 (dd, J 12.5,
2.9 Hz, 1H), 2.94 (dd, J 11.7, 1.5 Hz, 1H). 13C NMR d (ppm): 149.1
(s), 137.8 (s), 129.2 (d, 3C), 128.4 (d, 2C), 127.8 (d, 2C), 118.3 (d),
112.5 (d, 2C), 99.1 (d), 75.7 (d), 73.6 (t), 73.1 (d), 68.4 (t), 51.2
(t), 46.0 (t). MS m/z (%): 311 (M+, 7), 220 (M+ꢁBn, 46), 105 (100),
91 (71). Anal. Calcd for C19H21NO3: C, 73.29; H, 6.80; N, 4.50.
Found: C, 73.19; H, 6.98; N, 4.69.
°C. ½a 2D6
ꢄ
¼ þ4:6 (c 0.85, CHCl3). 1H NMR d (ppm): 7.63 (d, J 8.4
Hz, 2H), 7.38–7.27 (m, 7H), 5.51 (s, 1H), 4.60 (s, 2H), 4.41–4.38
(m, 1H), 4.27–4.18 (m, 1H), 4.08–3.89 (m, 2H), 3.68 (d, J 12.1 Hz,
1H), 3.57 (d, J 11.0 Hz, 1H), 2.79 (dd, J 11.7, 2.2 Hz, 1H), 2.58 (d, J
11.4 Hz, 1H), 2.44 (s, 3H). 13C NMR d (ppm): 143.8 (s), 137.7 (s),
132.1 (s), 129.6 (d, 2C), 128.3 (d, 2C), 127.7 (d, 2C), 127.6 (d),
127.4 (d, 2C), 98.3 (d), 76.2 (d), 73.7 (t), 72.4 (d), 68.1 (t), 49.2
(t), 44.8 (t), 21.6 (q). MS m/z (%): 389 (M+, 0.4), 298 (M+ꢁBn, 12),
234 (22), 128 (21), 91 (100). Anal. Calcd for C20H23NO5S: C,
61.68; H, 5.95; N, 3.60. Found: C, 61.46; H, 5.91; N, 3.50.
4.2.2. (1S,5S,7R)-7-Benzyloxymethyl-3-(4-methoxyphenyl)-6,8-
dioxa-3-azabicyclo[3.2.1]octane 17c
Pale pink solid (43%). Rf = 0.34. mp 91–92 °C. ½a D23
ꢄ
¼ ꢁ61:5 (c
4.2.6. (1S,5S,7R)-7-Benzyloxymethyl-3-(2-fluoroethyl)-6,8-
dioxa-3-azabicyclo[3.2.1]octane 17g
0.92, CHCl3). 1H NMR d (ppm): 7.39–7.31 (m, 5H), 6.92–6.82 (m,
2H), 6.68–6.60 (m, 2H), 5.65 (s, 1H), 4.57 (qAB, J 12.1 Hz, 2H),
4.58–4.53 (m, 1H), 4.35–4.25 (m, 1H), 3.98–3.90 (m, 1H), 3.83–
3.75 (m, 1H), 3.79 (s, 3H), 3.44 (d, J 11.4 Hz, 2H), 3.12 (dd, J 12.1,
2.6 Hz, 1H), 2.90 (dd, J 11.4, 1.1 Hz, 1H). 13C NMR d (ppm): 152.5
(s), 143.3 (s), 137.7 (s), 128.2 (d, 2C), 127.7 (d, 2C), 127.6 (d),
114.5 (d, 2C), 114.0 (d, 2C), 99.1 (d), 75.8 (d), 73.5 (t), 73.1 (d),
68.2 (t), 55.6 (q), 52.0 (t), 47.0 (t). MS m/z (%): 341 (M+, 41), 250
(M+ꢁBn, 64), 135 (100), 120 (32), 91 (39). Anal. Calcd for
C20H23NO4: C, 70.36; H, 6.79; N, 4.10. Found: C, 70.55; H, 6.73; N,
4.14.
In a 5-mL flask equipped with a reflux condenser and magnetic
stirrer were placed amino alcohol 16 (124 mg, 0.53 mmol), 2-flu-
oroethyltosylate (127 mg, 0.58 mmol), NaHCO3 (89 mg, 1.06
mmol) and dry DMF (1.6 mL). The mixture was heated at 80 °C
for 19 h and, after cooling, aqueous NH4OH 10% (16 mL) was
added and the product was extracted with Et2O (4 ꢃ 10 mL).
The combined organic phases were washed with water (3 ꢃ 10
mL) and dried over Na2SO4. After filtration and evaporation of
the solvent, the thus obtained crude was purified by FCC (eluent:
EtOAc–petroleum ether, 1:6 + 0.5% Et3N; Rf = 0.29) to afford pure
17g as
a
colourless oil (104 mg, 70%). Compound 17g:
4.2.3. (1S,5S,7R)-7-Benzyloxymethyl-3-o-tolyl-6,8-dioxa-3-azabic-
½
a 2D3
ꢄ
¼ ꢁ68:4 (c 1.38, CHCl3). 1H NMR d (ppm): 7.37–7.27 (m,
yclo[3.2.1]octane 17d
5H), 5.45 (s, 1H), 4.58 (qAB, J 11.7 Hz, 2H), 4.56–4.50 (m, 1H),
4.36–4.28 (m, 2H), 4.27–4.18 (m, 1H), 4.12–4.04 (m, 1H), 3.98–
3.90 (m, 1H), 2.86–2.80 (m, 2H), 2.72–2.50 (m, 3H), 2.39 (d, J
11.0 Hz, 1H). 13C NMR d (ppm): 137.9 (s), 128.2 (d, 2C), 127.7
(d, 2C), 127.5 (d), 99.6 (d), 81.7 (dt, JCF 167.9 Hz), 76.6 (d), 73.7
(d), 73.4 (t), 68.0 (t), 56.7 (dt, JCF 19.8 Hz), 56.7 (t), 52.0 (t). MS
Yellow oil (10%). Rf = 0.48. ½a D26
ꢄ
¼ ꢁ123:7 (c 0.5, CHCl3). 1H NMR
d (ppm): 7.33–7.02 (m, 9H), 5.56 (d, J 1.5 Hz, 1H), 4.57 (qAB, J 11.7
Hz, 2H), 4.49–4.45 (m, 1H), 4.40–4.31 (m, 1H), 4.22–4.14 (m, 1H),
4.01–3.94 (m, 1H), 3.58 (dd, J 12.1, 2.2 Hz, 1H), 3.15 (d, J 11.4 Hz,
1H), 2.88 (d, J 12.5 Hz, 1H), 2.68 (d, J 11.7 Hz, 1H), 2.27 (s, 3H).

348
D. Scarpi et al. / Tetrahedron: Asymmetry 20 (2009) 340–350
m/z (%): 281 (M+, 0.3), 190 (M+ꢁBn, 70), 91 (100). Anal. Calcd for
(113 mg, 44%). Compound 17l: ½a D26
ꢄ
¼ ꢁ73:1 (c 1.50, CHCl3). 1H
1
C15H20FNO3ꢀ 4H2O: C, 63.03; H, 7.23; N, 4.90. Found: C, 63.28; H,
NMR d (ppm): 7.30–7.18 (m, 5H), 5.35 (s, 1H), 4.50 (qAB, J 12.1
Hz, 2H), 4.23–4.19 (m, 1H), 4.17–4.08 (m, 1H), 4.03–3.95 (m, 1H),
3.88–3.80 (m, 1H), 2.71–2.57 (m, 2H), 2.36 (dd, J 11.7, 2.2 Hz,
1H), 2.08 (s, 3H). 13C NMR d (ppm): 137.9 (s), 128.1 (d, 2C), 127.6
(d, 2C), 127.4 (d), 99.6 (d), 76.7 (d), 73.7 (d), 73.3 (t), 68.1 (t),
58.3 (t), 53.9 (t), 44.4 (q). MS m/z (%): 249 (M+, 1), 158 (M+ꢁBn,
89), 91 (100). Anal. Calcd for C14H19NO3: C, 67.45; H, 7.68; N,
5.62. Found: C, 67.32; H, 7.46; N, 5.86.
7.09; N, 4.87.
4.2.7. (1S,5S,7S)-7-Benzyloxymethyl-3-(2,2,2-trifluoroethyl)-
6,8-dioxa-3-azabicyclo[3.2.1]octane 17h
In a 5-mL flask equipped with a reflux condenser and magnetic
stirrer were placed amino alcohol 16 (136 mg, 0.58 mmol), 2,2,2-
trifluoroethyl trifluoromethanesulfonate (134 mg, 0.58 mmol),
NaHCO3 (97 mg, 1.16 mmol) and absolute ethanol (1.7 mL). The
mixture was heated at reflux for 17 h and, after cooling, the solvent
was removed under vacuum. Aqueous NH4OH 10% (10 mL) was
added to the residue and the product was extracted with DCM
(4 ꢃ 5 mL). The combined organic phases were washed with brine
(10 mL) and dried over Na2SO4. After filtration and evaporation of
the solvent, the thus obtained crude was purified by FCC (eluent:
EtOAc–petroleum ether, 1:7 + 0.5% Et3N; Rf = 0.42), affording pure
4.3. General procedures for O-debenzylation of substrates 12 and
17b–l
Method A: The substrate (1 mmol) was dissolved into EtOAc (10
mL), under a nitrogen atmosphere, and 10% Pd/C (0.1 mmol Pd)
was added. The mixture was flushed first with nitrogen then with
hydrogen and finally left under a hydrogen atmosphere (balloon)
at room temperature for 18 h. The catalyst was removed by filtra-
tion through a Celite pad, washed with EtOAc and the solvent was
removed under vacuum. The crude was purified by FCC with the
specified eluent.
Method B: In a 50-mL flask equipped with a reflux condenser
and magnetic stirrer the substrate (1 mmol) was dissolved into
absolute ethanol (14 mL), and cyclohexene (14 mL) and 20%
Pd(OH)2/C (0.68 mmol Pd) were added. The mixture was refluxed
under nitrogen atmosphere for 4–7.5 h (TLC) and, after cooling,
the catalyst was removed by filtration through a Celite pad,
washed with methanol and the solvent was removed under vac-
uum. The crude was purified by FCC with the specified eluent.
17h as
a colourless oil (120 mg, 65%). Compound 17h:
½
a 2D4
ꢄ
¼ ꢁ59:1 (c 0.99, CHCl3). 1H NMR d (ppm): 7.36–7.28 (m,
5H), 5.45 (s, 1H), 4.57 (qAB, J 12.1 Hz, 2H), 4.34–4.30 (m, 1H),
4.27–4.18 (m, 1H), 4.09–4.01 (m, 1H), 3.95–3.87 (m, 1H), 2.97–
2.83 (m, 5H), 2.64 (d, J 10.6 Hz, 1H). 13C NMR d (ppm): 137.8 (s),
128.2 (d, 2C), 127.8 (d, 2C), 127.6 (d), 125.3 (qs, JCF 280.8 Hz),
99.4 (d), 76.5 (d), 73.6 (d), 73.5 (t), 68.0 (t), 57.1 (qt, JCF 30.5 Hz),
56.5 (t), 51.9 (t). MS m/z (%): 317 (M+, 2), 226 (M+ꢁBn, 77), 166
(34), 91 (100). Anal. Calcd for C15H18F3NO3: C, 56.78; H, 5.72; N,
4.41. Found: C, 56.44; H, 5.55; N, 4.62.
4.2.8. (1S,5S,7S)-7-Benzyloxymethyl-3-propyl-6,8-dioxa-3-azabi-
cyclo[3.2.1]octane 17i
Amino alcohol 16 (166 mg, 0.71 mmol) was dissolved into
4.3.1. (1S,5S,7R)-7-Hydroxymethyl-6,8-dioxa-3-azabicyclo-
[3.2.1]octane-3-carboxylic acid methyl ester 18a
CH3OH (3.6 mL) and propionaldehyde (56
lL, 0.78 mmol) was
added dropwise followed by NaBH3CN (49 mg, 0.78 mmol) and
few drops of glacial acetic acid (pH ꢂ 6). The mixture was left to
stir at room temperature for 18 h. After evaporation of the solvent,
NaHCO3 satd (5 mL) was added (pH ꢂ 8) to the residue and the
product was extracted with CHCl3 (3 ꢃ 5 mL); the combined organ-
ic phases were washed with water (8 mL) and dried over Na2SO4.
After filtration and evaporation of the solvent, the thus obtained
crude was purified by FCC (eluent: EtOAc–petroleum ether, 1:7;
Rf = 0.56), affording pure 17i as a thick colourless oil (104 mg, 53%).
Prepared according to Method A, starting from 12 (148 mg, 0.5
mmol) and obtaining 18a (86 mg, 84%). Compound 18a: EtOAc;
Rf = 0.34. Colourless oil. ½a D26
ꢄ
¼ ꢁ44:2 (c 0.51, CHCl3). 1H NMR d
(ppm): 5.52 (br s, 1H), 4.40 (br s, 1H), 4.18–4.09 (m, 1H), 3.95–
3.86 (m, 2H), 3.73 (s, 3H), 3.79–3.64 (m, 2H), 3.38–3.22 (m, 1H),
3.08 (d, J 13.2 Hz, 1H). 13C NMR d (ppm) [two rotamers]: 98.7
and 98.4 (d), 77.9 (d), 72.3 and 71.9 (d), 61.2 (t), 52.9 (q), 48.3
(t), 43.4 and 43.2 (t). MS m/z (%): 203 (M+, 8), 126 (18). Anal. Calcd
for C8H13NO5: C, 47.29; H, 6.45; N, 6.89. Found: C, 47.56; H, 7.11; N,
6.62.
Compound 17i: ½a D24
ꢄ
¼ ꢁ71:7 (c 1.0, CHCl3). 1H NMR d (ppm):
7.36–7.27 (m, 5H), 5.44 (s, 1H), 4.58 (qAB, J 12.1 Hz, 2H), 4.34–
4.29 (m, 1H), 4.25–4.17 (m, 1H), 4.12–4.05 (m, 1H), 3.98–3.90
(m, 1H), 2.82–2.72 (m, 2H), 2.43 (dd, J 11.7, 2.2 Hz, 1H), 2.29–
2.09 (m, 3H), 1.41–1.25 (m, 2H), 0.84 (t, J 7.3 Hz, 3H). 13C NMR d
(ppm): 138.0 (s), 128.2 (d, 2C), 127.7 (d, 2C), 127.5 (d), 99.9 (d),
76.7 (d), 74.0 (d), 73.5 (t), 68.3 (t), 59.3 (t), 56.7 (t), 52.1 (t), 19.7
(t), 11.9 (q). MS m/z (%): 187 (M+ꢁBn, 14), 158 (100), 142 (43).
Anal. Calcd for C16H23NO3: C, 69.29; H, 8.36; N, 5.05. Found: C,
69.32; H, 8.46; N, 4.86.
4.3.2. (1S,5S,7R)-(3-Phenyl-6,8-dioxa-3-azabicyclo[3.2.1]oct-7-yl)
methanol 18b
Prepared according to Method B, starting from 17b (78 mg, 0.25
mmol) and obtaining 18b (55 mg, quantitative). Compound 18b:
EtOAc–n-hexane, 2:3 + 0.5% Et3N ; Rf = 0.12. White solid. mp 102–
103 °C. ½a 2D6
ꢄ
¼ ꢁ33:1 (c 0.82, CHCl3). 1H NMR d (ppm): 7.36–7.24
(m, 2H), 6.90–6.78 (m, 3H), 5.68 (s, 1H), 4.58–4.52 (m, 1H), 4.19
(qAB, J 5.5 Hz, 1H), 4.06–3.85 (m, 2H), 3.63–3.51 (m, 2H), 3.21 (dd,
J 12.1, 2.9 Hz, 1H), 2.98 (dd, J 11.7, 1.5 Hz, 1H). 13C NMR d (ppm):
149.0 (s), 129.2 (d, 2C), 119.3 (d), 113.3 (d, 2C), 99.0 (d), 77.9 (d),
73.0 (d), 60.8 (t), 51.6 (t), 47.0 (t). MS m/z (%): 221 (M+, 41), 144
(M+ꢁPh,11), 105 (100), 77 (29). Anal. Calcd for C12H15NO3: C,
65.14; H, 6.83; N, 6.33. Found: C, 65.46; H, 7.18; N, 6.26.
4.2.9. (1S,5S,7S)-7-Benzyloxymethyl-3-methyl-6,8-dioxa-3-azabi
cyclo[3.2.1]octane 17l
Amino alcohol 16 (240 mg, 1.02 mmol) was dissolved into
CH3CN (14 mL), and HCHO (36.5%, 387 lL, 5.10 mmol) was added
dropwise followed by NaBH3CN (83 mg, 1.33 mmol). A white pre-
cipitate was formed; after 10 min, a few drops of glacial AcOH were
added (pH ꢂ 7) and more precipitate formed. The suspension was
left under stirring at room temperature for 4 h. After evaporation
of the solvent, water (10 mL) and Na2CO3 (s) were added
(pH ꢂ 8) to the residue, and the product was extracted with CHCl3
(4 ꢃ 12 mL) and the combined organic phases were dried over
Na2SO4. After filtration and evaporation of the solvent, the thus ob-
tained crude was purified by FCC (eluent: EtOAc–n-hexane,
1:6 + 0.5% Et3N; Rf = 0.30), affording pure 17l as a colourless oil
4.3.3. (1S,5S,7R)-(3-(4-Methoxyphenyl)-6,8-dioxa-3-azabicyclo
[3.2.1]oct-7-yl)methanol 18c
Prepared according to Method B, starting from 17c (86 mg, 0.25
mmol) and obtaining 18c (55 mg, 88%). Compound 18c: DCM–
MeOH, 30:1 + 0.5% Et3N; Rf = 0.31. Pale pink solid. mp 123–124
°C. ½a 2D4
ꢄ
¼ ꢁ41:8 (c 0.58, CHCl3). 1H NMR d (ppm): 6.87–6.75 (m,
4H), 5.64 (s, 1H), 4.52–4.47 (m, 1H), 4.04–3.93 (m, 2H), 3.75 (s,
3H), 3.53 (s, 1H), 3.44 (d, J 12.1 Hz, 2H), 3.15 (dd, J 12.1, 2.9 Hz,
1H), 2.92 (dd, J 11.4, 0.7 Hz, 1H). 13C NMR d (ppm): 153.4 (s),

D. Scarpi et al. / Tetrahedron: Asymmetry 20 (2009) 340–350
349
143.0 (s), 115.5 (d, 2C), 114.6 (d, 2C), 99.0 (d), 78.1 (d), 73.4 (d),
60.4 (t), 55.7 (q), 52.9 (t), 48.5 (t). MS m/z (%): 252 (M++1, 10),
251 (M+, 61), 135 (100), 120 (37). Anal. Calcd for C13H17NO4ꢀH2O:
C, 60.69; H, 6.92; N, 5.44. Found: C, 60.71; H, 6.70; N, 5.67.
½
a 2D4
ꢄ
¼ ꢁ67:8 (c 0.12, CHCl3). 1H NMR d (ppm): 5.50 (s, 1H), 4.36–
4.30 (m, 1H), 4.18–4.10 (m, 1H), 4.10–4.03 (m, 2H), 3.78–3.68 (br
s, 1H), 3.10–2.90 (m, 5H), 2.68 (d, J 11.0 Hz, 1H). 13C NMR d
(ppm): 125.0 (qs, JCF 280.2 Hz), 99.0 (d), 78.6 (d), 73.7 (d), 60.0
(t), 57.3 (qt, JCF 31.1 Hz), 56.5 (t), 52.7 (t). MS m/z (%): 227 (M+,
1
4.3.4. (1S,5S,7R)-(3-o-Tolyl-6,8-dioxa-3-azabicyclo[3.2.1]oct-7-yl)
methanol 18d
9), 182 (56), 112 (100). Anal. Calcd for C8H12F3NO3ꢀ 4H2O: C,
41.47; H, 5.44; N, 6.05. Found: C, 41.17; H, 5.17; N, 6.02.
Prepared according to Method B, starting from 17d (30 mg,
92
l
mol) and obtaining 18d (14 mg, 65%). Compound 18d: DCM–
4.3.9. (1S,5S,7R)-(3-Propyl-6,8-dioxa-3-azabicyclo[3.2.1]oct-7-yl)
methanol 18i
MeOH, 30:1 + 0.5% Et3N; Rf = 0.18. Pale yellow oil. ½a D26
¼ ꢁ90:2
ꢄ
(c 1.2, CHCl3). 1H NMR d (ppm): 4H), 5.57 (d, J 1.8 Hz, 1H), 4.49–
4.46 (m, 1H), 4.28–4.03 (m, 3H), 3.66 (dd, J 11.7, 2.2 Hz, 1H),
3.24 (d, J 11.7 Hz, 1H), 3.01 (d, J 12.1 Hz, 1H), 2.69 (d, J 12.1 Hz,
1H), 2.34 (s, 3H). 13C NMR d (ppm): 148.4 (s), 132.6 (s), 131.7 (d),
126.8 (d), 124.3 (d), 120.0 (d), 99.6 (d), 78.8 (d), 74.0 (d), 60.4 (t),
55.9 (t), 50.0 (t), 18.3 (q). MS m/z (%): 235 (M+, 30), 118 (100). Anal.
Calcd for C13H17NO3: C, 66.36; H, 7.28; N, 5.95. Found: C, 66.48; H,
7.50; N, 5.82.
Prepared according to Method B, starting from 17i (88 mg, 0.32
mmol) and obtaining 18i (50 mg, 84%). Compound 18i: DCM–
MeOH, 40:1; Rf = 0.36. Thick colourless oil. ½a D24
¼ ꢁ68:5 (c 0.20,
ꢄ
CHCl3). 1H NMR d (ppm): 7.12 (br s, 1H), 5.52 (s, 1H), 4.37–4.31
(m, 1H), 4.24–4.08 (m, 2H), 3.85 (d, J 11.8 Hz, 1H), 3.08–2.98 (m,
1H), 2.62 (dd, J 11.9, 2.3 Hz, 1H), 2.41–2.30 (m, 3H), 1.49 (qt, J
8.0 Hz, 2H), 0.89 (t, J 7.4 Hz, 3H). 13C NMR d (ppm): 99.0 (d),
78.3 (d), 74.7 (d), 59.1 (t, 2C), 56.0 (t), 53.4 (t), 19.2 (t), 11.7 (q).
MS m/z (%): 187 (M+, 18), 158 (100), 142 (45). Anal. Calcd for
C9H17NO3: C, 57.73; H, 9.15; N, 7.48. Found: C, 57.52; H, 8.71; N,
7.05.
4.3.5. (1S,5S,7S)-(3-Methanesulfonyl-6,8-dioxa-3-azabicyclo
[3.2.1]oct-7-yl)methanol 18e
Prepared according to Method A, starting from 17e (67 mg, 0.21
mmol) and obtaining 18e (44 mg, 94%). Compound 18e: EtOAc;
4.3.10. (1S,5S,7R)-(3-Methyl-6,8-dioxa-3-azabicyclo[3.2.1]oct-7-
yl) methanol 1
Rf = 0.36. White solid. mp 179–180 °C. ½a D26
¼ ꢁ40:4 (c 0.70,
ꢄ
CH3OH). 1H NMR (CD3OD) d (ppm): 5.61 (s), 4.53–4.49 (m, 1H),
4.18–4.02 (m, 2H), 3.95–3.83 (m, 1H), 3.65 (d, J 12.1 Hz, 1H),
3.42 (d, J 11.4 Hz, 1H), 3.23 (dd, J 12.1, 2.6 Hz, 1H), 2.98 (dd, J
11.0, 0.7 Hz, 1H), 2.88 (s, 3H). 13C NMR (CD3OD) d (ppm): 99.7
(d), 80.0 (d), 73.4 (d), 61.0 (t), 50.3 (t), 45.7 (t), 34.1 (q). MS m/z
(%): 222 (M+ꢁ1, 2), 144 (M+ꢁMs, 100). Anal. Calcd for C7H13NO5S:
C, 37.66; H, 5.87; N, 6.27. Found: C, 38.38; H, 5.82; N, 6.09.
Prepared according to Method B, starting from 17l (61 mg, 0.24
mmol) and obtaining 1 (32 mg, 84%). Compound 1: DCM–MeOH,
30:1; Rf = 0.28. Colourless oil. ½a D24
ꢄ
¼ ꢁ92:4 (c 0.95, CHCl3). [lit.1b
½
a 2D2
ꢄ
¼ ꢁ91:2 (c 0.89, CHCl3)]. 1H NMR d (ppm): 5.51 (s, 1H),
4.39–4.26 (m, 1H), 4.19–4.08 (m, 2H), 3.84 (d, J 11.7 Hz, 1H),
3.07–2.88 (m, 2H), 2.67–2.55 (m, 1H), 2.39 (d, J 11.0 Hz, 1H),
2.29 (s, 3H), 1.77–1.46 (br s, 1H). Anal. Calcd for C7H13NO3: C,
52.82; H, 8.23; N, 8.80. Found: C, 52.95; H, 8.46; N, 8.86.
4.3.6. (1S,5S,7R)-(3-(Toluene-4-sulfonyl)-6,8-dioxa-3-azabicyclo
[3.2.1]oct-7-yl)methanol 18f
4.4. General procedure for addition reaction of diethylzinc to
aldehydes
Prepared according to Method A, starting from 17f (66 mg, 0.17
mmol) and obtaining 18f (29 mg, 56%). Compound 18f: DCM–
MeOH, 30:1 + 0.5% Et3N; Rf = 0.34. Colourless oil. ½a D26
¼ ꢁ7:1 (c
ꢄ
Diethylzinc (1.0 M solution in hexanes, 1.5 mmol) was added to
a solution of ligand 18 (0.15 mmol) in anhydrous toluene (1.5 mL).
After 30 min, the aldehyde (1 mmol) was added dropwise and the
resulting mixture was stirred at room temperature for 18 h. The
reaction was quenched by 1 M aqueous HCl solution (5 mL) and
the product was extracted three times with EtOAc. The combined
organic phases were dried over Na2SO4. After filtration and evapo-
ration of the solvent, the crude alcohol was obtained and directly
analysed by GC to determine product composition and ee.
1.0, CHCl3). 1H NMR d (ppm): 7.60 (d, J 8.1 Hz, 2H), 7.32 (d, J 8.4
Hz, 2H), 5.51 (s, 1H), 4.41 (s, 1H), 4.26–4.12 (m, 2H), 4.03–3.90
(m, 1H), 3.60 (qAB, J 12.1 Hz, 2H), 2.76 (dd, J 12.1, 2.6 Hz, 1H),
2.54 (d, J 11.0 Hz, 1H), 2.42 (s, 3H). 13C NMR d (ppm): 144.1 (s),
131.8 (s), 129.8 (d, 2C), 127.5 (d, 2C), 98.3 (d), 78.2 (d), 72.0 (d),
60.7 (t), 49.2 (t), 44.8 (t), 21.6 (q). MS m/z (%): 299 (M+, 1), 184
(18), 155 (29), 144 (M+ꢁTs, 100). Anal. Calcd for C13H17NO5S: C,
52.16; H, 5.72; N, 4.68. Found: C, 52.32; H, 6.01; N, 4.63.
4.3.7. (1S,5S,7R)-(3-(2-Fluoroethyl)-6,8-dioxa-3-azabicyclo[3.2.1]
oct-7-yl)methanol 18g
Acknowledgements
Prepared according to Method B, starting from 17g (104 mg,
0.37 mmol) and obtaining 18g (61 mg, 86%). Compound 18g:
Mr Maurizio Passaponti and Mrs Brunella Innocenti are
acknowledged for their technical support; Dr Cristina Faggi is
acknowledged for X-ray structure determination. The authors wish
to thank Dr S. Zhao for advice.
DCM–MeOH + 0.5% Et3N; Rf = 0.31. Colourless oil. ½a D26
¼ ꢁ73:8 (c
ꢄ
0.48, CHCl3). 1H NMR d (ppm): 6.30–6.20 (br s, 1H), 5.50 (s, 1H),
4.66 (t, J 4.8 Hz, 1H), 4.42 (t, J 4.8 Hz, 1H), 4.35–4.30 (m, 1H),
4.19–4.08 (m, 2H), 3.93–3.84 (m, 1H), 3.10–3.00 (m, 2H), 2.82–
2.64 (m, 3H), 2.48 (d, J 11.0 Hz, 1H). 13C NMR d (ppm): 98.8 (d),
80.9 (dt, JCF 168.5 Hz), 78.4 (d), 74.3 (d), 59.2 (t), 56.6 (dt, JCF
19.8 Hz), 56.4 (t), 53.1 (t). MS m/z (%): 191 (M+, 6). Anal. Calcd
for C8H14FNO3: C, 50.25; H, 7.38; N, 7.33. Found: C, 49.95; H,
7.46; N, 7.14.
References
1. Some recent examples are: (a) Parrott, R. W., II; Hitchcock, S. R. Tetrahedron:
Asymmetry 2008, 19, 19–26; (b) Harada, A.; Fujiwara, Y.; Katagiri, T.
Tetrahedron: Asymmetry 2008, 19, 1210–1214; (c) Roudeau, R.; Pardo, D. G.;
Cossy, J. Tetrahedron 2006, 62, 2388–2394;; (d) Szakonyi, Z.; Balázs, Á.;
Martinek, T. A.; Fülöp, F. Tetrahedron: Asymmetry 2006, 17, 199–204; (e)
Huang, Z.; Lai, H.; Qin, Y. J. Org. Chem. 2007, 72, 1373–1378; (f) Tanyeli, C.;
Sünbül, M. Tetrahedron: Asymmetry 2005, 16, 2039–2043; (g) Da, C. S.; Ni, M.;
Han, Z. J.; Yang, F.; Wang, R. J. Mol. Catal. A 2006, 245, 1–7; (h) Hari, Y.;
Aoyama, T. Synthesis 2005, 583–587; (i) Hajji, C.; Testa, M. L.; Zaballos-García,
E.; Sepúlveda-Arques, J. J. Chem. Res. 2005, 420–421.
4.3.8. (1S,5S,7R)-(3-(2,2,2-Trifluoroethyl)-6,8-dioxa-3-azabi-
cyclo[3.2.1]oct-7-yl)methanol 18h
Prepared according to Method B, starting from 17h (100 mg,
0.32 mmol) and obtaining 18h (69 mg, 95%). Compound 18h:
DCM–MeOH, 30:1 + 0.5% Et3N; Rf = 0.22. Pale yellow oil.
2. Oguni, N.; Omi, T. Tetrahedron Lett. 1984, 25, 2823–2824.
3. (a) Noyori, R.; Kitamura, M. Angew. Chem., Int. Ed. Engl. 1991, 30, 34–48; (b)
Yamakawa, M.; Noyori, R. J. Am. Chem. Soc. 1995, 117, 6327–6335.

350
D. Scarpi et al. / Tetrahedron: Asymmetry 20 (2009) 340–350
4. (a) de las Casas Engel, T.; Maroto, B. L.; García Martínez, A.; de la Moya Cerero,
13. Quach, T. D.; Batey, R. A. Org. Lett. 2003, 5, 4397–4400.
14. Parenty, A. D. C.; Smith, L. V.; Pickering, A. L.; Long, D.-L.; Cronin, L. J. Org. Chem.
2004, 69, 5934–5946.
15. Burdon, J.; McLaughlin, V. C. R. Tetrahedron 1965, 21, 1–4.
16. Reifenrath, W. G.; Roche, E. B.; Al-Turk, W. A. J. Med. Chem. 1980, 23, 985–990.
17. Hay, M. P.; Lee, H. H.; Wilson, W. R.; Roberts, P. B.; Denny, W. A. J. Med. Chem.
1995, 38, 1928–1941.
S. Tetrahedron: Asymmetry 2008, 19, 2003–2006; (b) de las Casas Engel, T.;
Maroto, B. L.; García Martínez, A.; de la Moya Cerero, S. Tetrahedron: Asymmetry
2008, 19, 269–272; (c) Olsson, C.; Hegelsson, S.; Frejd, T. Tetrahedron:
Asymmetry 2008, 19, 1484–1493; (d) García Martínez, A.; Vilar, E. T.; García
Fraile, A.; de la Moya Cerero, S.; Martínez Ruiz, P.; Díaz Morillo, C. Tetrahedron:
Asymmetry 2007, 18, 742–749; (f) Paolucci, C.; Rosini, G. Tetrahedron:
Asymmetry 2007, 18, 2923–2946; (g) Jimeno, C.; Vidal-Ferran, A.; Pericás, M.
A. Org. Lett. 2006, 8, 3895–3898; (h) García Martínez, A.; Vilar, E. T.; García
Fraile, A.; de la Moya Cerero, S.; Maroto, B. L. Tetrahedron 2005, 61, 3055–3064.
5. Pu, L.; Yu, H.-B. Chem. Rev. 2001, 101, 757–824.
18. Milosevic, P.; Hecht, S. Org. Lett. 2005, 7, 5023–5026.
19. Boucheron, C.; Desvergnes, V.; Compain, P.; Martin, O. R.; Lavi, A.; Mackeen, M.;
Wormald, M.; Dwek, R.; Butters, T. D. Tetrahedron: Asymmetry 2005, 16, 1747–
1756.
6. (a) Scarpi, D.; Lo Galbo, F.; Occhiato, E. G.; Guarna, A. Tetrahedron: Asymmetry
2004, 15, 1319–1324; (b) Scarpi, D.; Lo Galbo, F.; Guarna, A. Tetrahedron:
Asymmetry 2006, 17, 1409–1414.
7. Bessodes, M.; Shamsazar, J.; Antonakis, K. Synthesis 1988, 560–562.
8. Trabocchi, A.; Menchi, G.; Rolla, M.; Machetti, F.; Bucelli, I.; Guarna, A.
Tetrahedron 2003, 59, 5251–5258.
9. Zhao, S.; Petrus, L.; Serianni, A. S. Org. Lett. 2001, 3, 3819–3822.
10. Cini, N.; Danieli, E.; Menchi, G.; Trabocchi, A.; Bottoncetti, A.; Raspanti, S.; Pupi,
A.; Guarna, A. Bioorg. Med. Chem. 2006, 14, 5110–5120.
11. Meyers, A. I.; Tomioka, K.; Roland, D. M.; Comins, D. Tetrahedron Lett. 1978, 16,
1375–1378.
20. Duréault, A.; Portal, M.; Carreaux, F.; Depezay, J. C. Tetrahedron 1993, 49, 4201–
4210.
21. Traces of metal halides, acting as Lewis acids themselves or in situ generating
ethylzinc halides, could be responsible for the non-enantioselective addition.
22. Crystallographic data (excluding structure factors) for structure 1ꢀHCl have
been deposited with the Cambridge Crystallographic Data Centre as
supplementary publication numbers CCDC711875. Copies of the data can be
obtained, free of charge, on application to CCDC, 12 Union Road, Cambridge
CB2 lEZ, UK (fax: +44(0)-1223-336033 or e-mail:deposit@ccdc.cam.ac.uk).
23. It should be mentioned that in the case of the CF3 group, the steric effect ES
makes it more comparable to a sec-butyl rather than to a CH3. However, the ES
should be less pronounced in the case of a single H–F substitution. See Ref. 1b
and also: Taft, R. W. In Steric effects in Organic Chemistry; Newman, M. S., Ed.;
Wiley: New York, NY, 1956; pp 556–675.
12. (a) Chan, D. M. T.; Monaco, K. L.; Wang, R.-P.; Winters, M. P. Tetrahedron Lett.
1998, 39, 2933–2936; (b) Lam, P. Y. S.; Clark, C. G.; Saubern, S.; Adams, J.;
Winters, M. P.; Chan, D. M. T.; Combs, A. Tetrahedron Lett. 1998, 39, 2941–2944.