The hydrogen bond directing effect in nitrile oxide cycloadditions to allylic substituted cyclopentenes

Article abstract of DOI:10.1016/j.tet.2017.03.042

A quantitative evaluation of the H-bond directing effect on the stereoselectivity in the cycloaddition of nitrile oxides to 2-cyclopenten-1-ol and allylic cyclopentenyl amides is reported. In apolar solvents the H-bond directing effect promotes a high syn stereoselectivity while H-bond acceptor solvents divert the reactions to the anti face of the dipolarophile. Taft's β parameter gives a good description of the solvent effect on the H-bond directing effect. The persistence of some syn stereoselectivity even in good H-bond acceptor solvents points out the existence of some residual hydrogen bond direction. The syn stereoselectivity in the presence of M(II) salts was also investigated and the results discussed in the light of the potential application of these scaffolds in nucleoside synthesis.

Full text of DOI:10.1016/j.tet.2017.03.042

Tetrahedron xxx (2017) 1e12
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
The hydrogen bond directing effect in nitrile oxide cycloadditions to
allylic substituted cyclopentenes
Anna Maria Cardarelli, Vera Fassardi, Misal Giuseppe Memeo, Paolo Quadrelli*
ꢀ
Universita degli Studi di Pavia, Dipartimento di Chimica, Viale Taramelli 12, 2700 Pavia, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
A quantitative evaluation of the H-bond directing effect on the stereoselectivity in the cycloaddition of
nitrile oxides to 2-cyclopenten-1-ol and allylic cyclopentenyl amides is reported. In apolar solvents the
H-bond directing effect promotes a high syn stereoselectivity while H-bond acceptor solvents divert the
Received 24 January 2017
Received in revised form
27 February 2017
Accepted 13 March 2017
Available online xxx
reactions to the anti face of the dipolarophile. Taft's
b parameter gives a good description of the solvent
effect on the H-bond directing effect. The persistence of some syn stereoselectivity even in good H-bond
acceptor solvents points out the existence of some residual hydrogen bond direction. The syn stereo-
selectivity in the presence of M(II) salts was also investigated and the results discussed in the light of the
potential application of these scaffolds in nucleoside synthesis.
Keywords:
Hydrogen bond
Nitrile oxide
© 2017 Elsevier Ltd. All rights reserved.
Cyclopentenols
Cyclopentene amide
Solvent effect
1. Introduction
oxidant which is less prone to HB effects9 and hexafluoroacetone
perhydrate (HFAH)10 which is a non-metal catalyst for higher syn-
Among the directing effects which assume relevance in the
control of stereoselectivity in many organic reactions, the inter-
molecular Hydrogen Bond (HB) between reactants is perhaps the
most known, powerful and extensively explored for synthetic
purposes.1 One of the earliest examples of HB directed reactions is
the Henbest peracid epoxidation2 of cyclic allylic alcohols (2-
cyclohexenol, 2-cyclopentenol) which causes a remarkably syn
stereoselection when the reaction is performed in apolar solvents,
while the substrates are stereorandomly oxidized when alcohols
are used as solvents. A quantitative assessment of the solvent effect
on the stereoselectivity of the peracid epoxidation is however
difficult since the solvent can interact either with the alcoholic
function or with the peracid hydroxyl group. In the latter case a rate
retardation is observed, as shown by kinetic measurements in the
epoxidation of cyclohexene.3 Other reagents have been developed
and of special interest are catalytic procedures which utilize
transition-metal catalysts, mainly TBHP/Ti(OiPr)4,4 VO(acac)2,5
methyl-trioxorhenium (MTO)6 and Mn(Salen)7 complexes,
achieving best results in term of stereoselection. Besides peracids,
dimethyl dioxirane (DMD)8 is another purely organic non-metal
selective oxidations via a hydrogen bonded transition structure
similar to that of a peracid with a catalytic oxygen transfer process.
HB directing effects have been observed in nitrile oxide cyclo-
additions and affect stereoselection remarkably.11e13 We have
exploited the HB directing effects for the stereoselective synthesis
of bicyclic isoxazolines, which serve as building blocks for the
synthesis of modified nucleosides.15 With the exception of one case
where an intramolecular HB prevents the establishment of the
intermolecular HB needed for the heteroatom directing effect,
stereoselectivity was satisfactorily tuned with the appropriate
solvents.
On pursuing our interests in cyclopentene-based nucleoside
analogues synthesis,16 we report here a quantitative evaluation of
the solvent effect on the HB directing effects in the cycloaddition of
nitrile oxide to 2-cyclopenten-1-ol and allylic N-cyclopentenyl
amides, which serve as reference data for more complex systems.
The persistence of some syn stereoselectivity even in good H-bond
acceptor solvents and comparisons with the stereoselectivities of
the O- and N-methyl derivatives point out the existence of some
residual hydrogen bond direction. The syn-stereoselectivity in the
presence of M(II) salts was also investigated and a brief discussion
is given, in light of potential applications in nucleoside analogue
synthesis using 2-cyclopenten-1-ol or allylic N-cyclopentenyl am-
ides as scaffolds for these targets.
* Corresponding author.
E-mail address: paolo.quadrelli@unipv.it (P. Quadrelli).
http://dx.doi.org/10.1016/j.tet.2017.03.042
0040-4020/© 2017 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Cardarelli AM, et al., The hydrogen bond directing effect in nitrile oxide cycloadditions to allylic substituted
cyclopentenes, Tetrahedron (2017), http://dx.doi.org/10.1016/j.tet.2017.03.042

2
A.M. Cardarelli et al. / Tetrahedron xxx (2017) 1e12
Table 1
2. Results
Reaction yields and regioisomeric ratios of cycloadducts 2a-d in the cycloadditions
to 2-cyclopenten-1-ol and 3-methoxy-cyclopentene.
2.1. Solvent effect in nitrile oxide cycloaddition reactions
a
Entry
Solvents (
b
)
Yield %
2a
2b
2c
2d
2d/2b
2.1.1. Cycloadditions to 2-cyclopenten-1-ol and 3-methoxy-
cyclopentene
2-cyclopenten-1-ol (1A)
1
Cy (0.00)
82
87
65
82
81
88
95
79
86
90
66
63
60
68
60
64
58
55
58
20
23
21
20
12
10
13
8
10
9
e
1
e
1
2
3
5
2
1
5
1
5
1
24
29
27
30
31
28
53
54
57
53
64
61
64
3.0
2.9
3.0
2.7
2.8
2.5
2.4
2.6
2.7
2.4
2.8
2.5
2.9
2
3
4
5
nHex (0.00)
CCl4 (0.00)
CH2Cl2 (0.00)
CHCl3 (0.00)
Bz (0.10)
Cycloadditions of benzonitrile oxide (BNO) to 2-cyclopenten-1-
ol (1A) and 3-methoxy-cyclopentene (1B) afforded the known four
regio and stereoisomeric cycloadducts 2Aa-d and 2Ba-d, respec-
tively (Scheme 1).12 The solvent effect was investigated by per-
forming the reactions in 13 solvents of different polarities and
Hydrogen Bond Acceptor (HBA) abilities and the product distribu-
tion was determined with quantitative HPLC analyses. Table 1 gives
the reaction yields and the regio and stereoisomeric ratios of 2a-
11
12
11
22
21
21
22
23
24
22
6
7
8
9
10
11
12
13
DIOX (0.37)
AcOEt (0.45)
Acetone (0.48)
THF (0.55)
MeOH (0.62)
DMF (0.69)
EtOH (0.77)
d along with the Taft's b
parameter,14 a descriptor of the HBA ability
of the solvents.
As a whole, cycloadditions to cyclopentenol give good yields
(60e95%) of the adducts, while the product distribution shows a
remarkable solvent dependence.
3-methoxy-cyclopentene (1B)
14
15
16
17
CH2Cl2 (0.00)
Bz (0.10)
AcOEt (0.45)
MeOH (0.62)
72
84
82
72
6
7
6
7
30
30
31
30
5
7
5
5
58
56
58
58
1.9
1.9
1.9
1.9
In non-HBA solvents (
b
¼ 0.00e0.10; entries 1e6), the 6syn
a
cycloadduct 2Aa, which derives from the HB directed cycloaddition
shown in 3, is the major regioisomer and amounts to 55e68% of the
cycloaddition products. In HBA acceptor solvents (entries 7e13) the
HB directing effect decreases because of the competing effect of the
solvent for the HB and the amount of cycloadduct 2Aa steadily
b
values, see ref. 14.
plotted against solvent
b
values. In non-HBA solvents (
b
equal or
proximal to 0.00), the HB directing effect is fairly active and con-
stant and all the points cluster together. HBA solvents cause the
drop of the points along a line with slope ꢀ1.02. The linear
drops with the increasing
b values of the solvent. Meanwhile, the
anti adducts 2Ad and 2Ab steadily increase maintaining between
them a rather constant ratio 2d/2b, ranging from 2.5 to 3.0. The
major anti cycloadduct is the 4anti stereoisomer 2Ad, in accordance
with Frontier Orbital (FO) and electrostatic expectations.12a The
4syn cycloadduct 2Ac is nearly negligible in all the cases owing to
the steric hindrance between the hydroxyl substituent and the
nitrile oxide phenyl substituent.11,12,17
regression is fair (r2 ¼ 0.95) and shows that the
b parameter
satisfactorily accounts for the solvent effect on the stereoselectivity.
Conversely, the related cycloaddition to 3-methoxy-cyclopentene
does not show any noticeable solvent dependence (Table 1, entries
14e17). At the bottom of Fig. 1 the red squares show the constance
of the 6syn adduct 2Ba in the mixtures. The major adducts are the
anti regioisomer 2Bd and 2Bb in a ratio ca. 2:1 similar to that re-
ported in the case of the alcohols.
A graphical illustration of the effect is given in Fig. 1, where the
log of the percentage of the adduct 6syn 2Aa in the mixture is
2.2. Cycloadditions to cyclopentenyl amides
We have investigated the solvent effect in the cycloadditions of
BNO to the N-cyclopentenyl amides 4A-D which afford the regioi-
someric adducts 5a-d (Scheme 2). Preparative cycloaddition of BNO
Scheme 1. BNO cycloaddition reaction to 2-cyclopenten-1-ol and 3-methoxy-
cyclopentene.
Fig. 1. Plot and linear regression of the log(2a%) vs solvents b values.
Please cite this article in press as: Cardarelli AM, et al., The hydrogen bond directing effect in nitrile oxide cycloadditions to allylic substituted
cyclopentenes, Tetrahedron (2017), http://dx.doi.org/10.1016/j.tet.2017.03.042

A.M. Cardarelli et al. / Tetrahedron xxx (2017) 1e12
3
5Aa (6syn) is resolutely predominant (76e82%) while the isomer
4anti 5Ad stands around 18% along with small amounts of the other
isomers.
In moderate HBA solvents (entries 7e10) the HB directing effect
decreases and almost equimolecular amounts of 5Aa and 5Ad are
formed. In alcohols and DMF (entries 11e13) the 4anti isomer 5Ad
outweighs the 6syn isomer 5Aa (ratio 1.5:1 to 2.8:1) and small, but
noteworthy, amounts (~10%) of the other isomers are formed. The
regioisomeric ratio 5Ad/5Ab for the anti addition ranges around 6.
When plotting the log of percentage of the adduct 6syn 5Aa
against the solvent
b values (Fig. 2) we obtained a fair linear
regression (r2 ¼ 0.90) with a slope of ꢀ0.63 slightly smaller with
respect to the case of cyclopentenol. The related cycloaddition to
the N-methyl-N-cyclopentenyl-acetamide (4B) affords mainly the
anti adducts 5Bd and 5Bb in a ratio ranging from 2:1 to 2.6:1 and
does not show any remarkable solvent dependence. The red
squares in Fig. 2 show the level of the 6syn adduct 5Ba, with respect
to the line of the unsubstituted adduct 5Aa.
The N-benzoyl-amides 4C and 4D behave similarly. The cyclo-
addition of BNO to the benzoyl derivative 4C in methanol yielded a
mixture of the four cycloadducts 5Ca-d (85% overall yield) which
have been separated by column chromatography, characterized
spectroscopically and converted by methylation into the N-methyl
derivatives 5Da-d. The influence of the solvent effects on the cy-
cloadditions to 4C and 4B has been tested in a few cases (Table 2)
and is almost identical to the cases of the acetamide analogues.
Scheme 2. BNO cycloaddition reaction to cyclopentenyl amides.
to N-cyclopentenyl-acetamide (4A) in methanol gave the stereo-
isomeric cycloadducts 5Aa-d. Chromatographic separation of the
reaction mixture furnished the cycloadducts 5Aa (30%) and 5Ad
(41%), which are known compounds,13e and the unknown
regioisomers 5Ab (6%) and 5Ac (8%), whose structures were
assigned upon their analytical and spectroscopic data. Methylation
of the adducts with NaH/CH3I in THF afforded the N-methyl de-
rivatives 5Ba-d. The solvent effect on the cycloaddition to acet-
amide 4A was investigated in 13 solvents according to the
procedures previously reported. HPLC analyses of the reaction
mixtures gave the reaction yields and the isomeric ratios, which are
reported in Table 2.
3. Metal-mediated stereocontrol in cycloaddition reactions
In recent years some metal-mediated approaches have been
proposed for nitrile oxide cycloaddition reactions to acyclic allylic
alcohols. The use of magnesium allylic alkoxides as dipolarophiles
in nitrile oxide cycloadditions was proposed by Kanemasa.18 The
same approach was extended by Ryu to the cycloaddition of nitrile
All the reactions afford the cycloadducts in good yields
(73e96%). In apolar or non-HBA solvents (entries 1e6) cycloadduct
oxides to
a,b
-unsaturated carbonyl compounds.19 Other metal
alkoxides, such as lithium, zinc, nickel and aluminium, have been
tested but they were less effective.18 Nevertheless, magnesium salts
were successfully employed in ion-mediated 1,3-dipolar cycload-
ditions, showing moderate catalytic efficiency, ligand acceleration
effect and providing a promising access to a catalyzed version of the
1,3-dipolar cycloaddition reaction.20
Table 2
Reaction yields and isomeric ratios of cycloadducts 5a-d in the cycloaddition to
cyclopentenyl-amides.
a
Entry
Solvents (
b
)
Yield %
5a
5b
5c
5d
5d/5b
N-cyclopentenyl-acetamide (4A)
1
2
3
4
5
6
7
8
9
10
11
12
13
Cy (0.00)
91
94
94
96
95
95
90
94
96
88
94
73
91
82
79
81
76
77
78
47
43
43
40
32
20
33
e
e
e
e
e
e
5
7
6
8
9
3
3
3
4
4
3
7
5
7
6
10
10
9
15
18
16
20
19
19
41
45
44
46
49
57
49
nHex (0.00)
CCl4 (0.00)
CH2Cl2 (0.00)
CHCl3 (0.00)
Bz (0.10)
DIOX (0.37)
THF (0.55)
6.8
6.4
7.3
5.7
5.4
4.4
5.4
AcOEt (0.45)
Acetone (0.48)
MeOH (0.62)
DMF (0.69)
EtOH (0.77)
13
9
N-methyl-N-cyclopentenyl-acetamide (4B)
14
15
16
17
CH2Cl2 (0.00)
Bz (0.10)
AcOEt (0.45)
MeOH (0.62)
99
96
89
82
5
4
4
5
32
27
29
30
e
e
e
e
63
69
67
65
2.0
2.6
2.3
2.2
N-cyclopentenyl-benzamide (4C)
18
19
20
21
CH2Cl2 (0.00)
Bz (0.10)
AcOEt (0.45)
MeOH (0.62)
95
93
86
73
89
81
72
67
1
2
4
5
10
17
24
28
N-methyl-N-cyclopentenyl-benzamide (4D)
22
23
24
25
CH2Cl2 (0.00)
Bz (0.10)
AcOEt (0.45)
EtOH (0.77)
86
88
89
80
3
3
3
3
29
32
31
33
1
67
65
66
64
2.3
2.0
2.1
1.9
e
e
e
a
b
values, see ref. 14.
Fig. 2. Plot and linear regression of the log(5a%) vs solvents b values.
Please cite this article in press as: Cardarelli AM, et al., The hydrogen bond directing effect in nitrile oxide cycloadditions to allylic substituted
cyclopentenes, Tetrahedron (2017), http://dx.doi.org/10.1016/j.tet.2017.03.042

4
A.M. Cardarelli et al. / Tetrahedron xxx (2017) 1e12
In order to improve the regio- and stereocontrol of the 1,3-
of the metal ions or from a geometrical point of view, marginally
only influencing the reaction rate. The geometrical features are of
strong relevance for an efficient directing activity of the metal ions
and these are determined by the dimension of the metal centre and
the space occupied by the reagents that are the ligands of the metal.
When Mg(ClO4)2 was used (entry 6) the cycloaddition reaction
proceeds at 25 ꢁC quantitatively with high syn-selectivity affording
the adduct 2a nearly as single product (92%). Adducts 2b and 2d are
4% of the mixture while the adduct 2c was absent. In light of this
result, we explored the reaction temperature effect and the results
are collected in Table 3, entries 11e17. A series of experiments were
conducted at 0 ꢁC in DCM as solvent in the presence of Mg(II) and
Zn(II) (entries 11e14) without any significant change in regio- and
stereoselectivity. With Mg(ClO4)2 (entry 15) a slight improvement
was observed both in terms of reaction yield (97%) and selectivity:
adduct 2a was 96% of the mixture. Lower temperatures (ꢀ20 ꢁC
or ꢀ40 ꢁC) were also tested without significant changes with
respect to those obtained at 0 ꢁC.
It is worth noting that the M(II) salts were used in equimolar
amounts with respect to the reagents. For this reason, we per-
formed a couple of experiments reducing the amount of the metal
salts limiting the case to Mg(ClO4)2 (entries 9 and 10). While the
chemical yields remained good, a neet decrease of the regio- and
stereoselectivities was progressively observed upon the reduction
of the amount of salt from 0.5 to 0.2 equivs. The adduct 2a is nearly
halved in the reaction mixture and the 2d/2b ratio is nearly tripled
as a sign of the FO interactions prevailing over the metal directing
effect. It seems that the catalytic Mg(II), instead of directing the
cycloaddition, plays the opposite role, occupying the syn face and
reversing the selectivity outcome. These results indicate that Mg(II)
is the metal of choice for performing the reaction with a signifi-
cative increase of regio- and stereoselectivity when the metal is
used as a reagent and not as a catalyst. The counterion ClO4 has a
decisive role in the selectivity outcome; it is known to be a less-
coordinating anion leaving the naked metal cation able to coordi-
nate the cycloaddends for an efficient cycloaddition process. Other
perchlorates of Ni(II) and Zn(II) were tested both at 25 ꢁC (entries 7
and 8) and 0 ꢁC (entries 16 and 17); the regio- and stereo-
selectivities remained at the levels obtained with bromides or tri-
flates as a confirmation of the importance of the metal dimension
for the correct approach of the cycloaddends at reaction distance.
dipolar cycloaddition of nitrile oxides to 2-cyclopenten-1-ol and
allylic N-cyclopentenyl amides, we have also studied the opportu-
nity to perform the previously reported cycloadditions in the
presence of representative bivalent M(II) salts.
3.1. Cycloadditions of benzonitrile oxide to 2-cyclopenten-1-ol
We have investigated the effect of the presence of bivalent metal
M(II) salts on the cycloaddition outcome by performing the cyclo-
additions of BNO to 2-cyclopenten-1-ol (1A) in DCM solutions
(Scheme 1).12 The solvent choice is based on the necessary com-
parison with the results obtained in the absence of any metal ion.
DCM is a typical non-HBA (Taft's parameter
b
¼ 0.00) and the
directing effect of the dipolarophile (see 3 in Scheme 1) is the sole
effect at work. Moreover, DCM displays a good ability to dissolve
the inorganic salts under the experimental conditions applied.
Table 3 reports the reaction yields as well as regio- and stereo-
isomeric ratios of cycloadducts 2a-d when M(II) salts are present in
equimolar amounts. The values are the average results of at least
two independent experiments.
All the reactions gave nearly quantitative yields of the cyclo-
adducts after one day of reaction time instead of two, as a result of
increased reaction rate in accordance with literature observa-
tions.18e21 The regio- and stereoisomeric ratios of the cycloadducts
2a-d in the experiments of entries 2 and 3 performed with Mg(II)X2
salts (X ¼ Br, Tf) and those of entries 4, 5 and 8 conducted in the
presence of Zn(II)X2 salts (X ¼ Br, Tf, ClO4) strongly resemble the
result obtained in the absence of any metal catalyst as reported in
the entry 1. DCM being a typical non-HBA solvent, the syn selec-
tivity seems to be enforced by the HB directing effect of the OH
group of the dipolarophile 1 towards BNO, even in the presence of
salts. The anti isomers maintain a nearly constant ratios (2d/2b
ranges from 2.3 to 2.6) in fair accordance with the value of 2.7 in
entry 1. The anti isomer 2d remains the major one in keeping with
the FO and electrostatic expectations.12 The 4-syn cycloadduct 2c is
nearly negligible in all the cases because of the steric hindrance
between the substituents.11,12,17
These observations support the suggestion that only the HB
directing effect is at work in governing the selectivity or that the
coordination to the metal centre is imperfect from low availability
3.2. Cycloadditions of 4-chlorobenzonitrile oxide to 2-cyclopenten-
1-ol
Table 3
Reaction yields and regioisomeric ratios of cycloadducts 2a-d in the cycloadditions
to 2-cyclopenten-1-ol (1A).
We have investigated the cycloaddition of 4-chlorobenzonitrile
oxide (pClBNO) to 2-cyclopenten-1-ol (1A) in DCM solutions in the
presence of M(II) salts with the aim to have more insight into the
potential effect of Et3N and its hydrochloride salt when the base is
used to generate the nitrile oxide species as they can be potential
competitors in the HB directing effects or in the construction of the
metal complexes (Scheme 3). pClBNO can be prepared as a
moderately stable nitrile oxide upon base treatment of the corre-
Entry
M(II)X (1.0 equivs.)
Yield %
2a
2b
2c
2d
2d/2b
Reaction conducted at 25 ꢁC
1
2
3
4
5
6
7
/
82
98
99
98
84
99
98
98
93
89
58
51
52
56
59
92
57
60
65
51
11
14
13
12
12
4
12
12
11
13
1
2
2
2
2
e
2
1
1
2
30
33
33
30
27
4
29
27
23
34
2.7
2.4
2.5
2.5
2.3
1.0
2.4
2.3
2.1
2.6
MgBr2
MgTf2
ZnBr2
ZnTf2
Mg(ClO4)2
Ni(ClO4)2
Zn(ClO4)2
Mg(ClO4)2
Mg(ClO4)2
sponding a-chloroxime in diethyl ether as solvent. Evaporation of
the solvent affords the solid nitrile oxide, found to be stable for at
least one day without appreciable dimerization.
8
9a
10b
Reaction conducted at 0 ꢁ
C
The reaction afforded the four stereoisomeric cycloadducts 6a-
d in 80% overall yield that were separated through column chro-
matography and fully characterized via their analytical and spec-
troscopic data. Table 4 reports the relative amounts of the four
cycloadducts in the experiments performed; the data are the
average of at least two independent experiments.
All the reactions performed in the presence of M(II) salts gave
excellent yields, in many cases nearly quantitative after one day of
reaction. The analysis of these results reveals that the absence of
11
12
13
14
15
16
17
MgBr2
MgTf2
ZnBr2
ZnTf2
Mg(ClO4)2
Ni(ClO4)2
Zn(ClO4)2
98
99
99
91
97
92
98
51
51
54
59
96
58
61
14
14
13
12
2
2
2
1
1
e
2
e
33
33
32
28
2
2.4
2.4
2.5
2.3
1.0
2.6
2.3
11
12
29
27
a
With 0.5 equivs. of catalysts.
With 0.2 equivs. of catalysts.
b
Please cite this article in press as: Cardarelli AM, et al., The hydrogen bond directing effect in nitrile oxide cycloadditions to allylic substituted
cyclopentenes, Tetrahedron (2017), http://dx.doi.org/10.1016/j.tet.2017.03.042

A.M. Cardarelli et al. / Tetrahedron xxx (2017) 1e12
5
Scheme 4. BNO cycloaddition reaction to cyclopentenyl amide (4A).
Scheme 3. pClBNO cycloaddition reaction to 2-cyclopenten-1-ol (1A).
in DCM as solvent. The reactions afforded the regioisomeric adducts
5Aa-d (Scheme 4) and the data are reported in Table 5.
With respect to the reaction performed in the absence of any
metal ion (Table 5, entry 1), the chemical yields are slightly better
with the remarkable exception of the cases of MgBr2 (entries 2 and
11). The effect of the metal coordination on the regio- and stereo-
selectivity is modest. The major adduct 5Aa never surpasses 86% at
25 ꢁC and 87% at 0 ꢁC with Mg(ClO4)2 (entries 6 and 15). Adduct 5Ab
is totally absent in all the reaction mixtures, 5Ac ranges around
1e4% while 5AdB is always present in 14e27% in the reaction
mixtures. Temperature does not give substantial improvements in
terms of selectivities.
Cyclopentenyl amides seem to be less apt to obey the strict rules
for an efficient direction in the nitrile oxide cycloaddition reactions
conducted in the presence of metals. Reasonably, the amide group
strongly competes as a ligand in the coordination of metal ions,
diverting the metal centre far from the orientation of the cyclo-
addends. Here the coordination to the metal can be defined as
imperfect.
the couple Et3N/Et3N$HCl smoothly increases the syn selectivity.
Stereoisomer 6a, the major component, increases its percentage
within the reaction mixture with good performances even with
Mg(II) bromides or triflates (Table 4, entries 2e5) as well as with
the same Zn(II) salts.
The use of Mg(ClO4)2 (entry 6) afforded the best result for the
reaction conducted at 25 ꢁC: 99% yield, 96% of isomer 6a. The other
perchlorates (entries 7 and 8) did not give the same performances
and a reduction of the amount of salts (entries 9 and 10) is detri-
mental for the syn-selectivity, as previously reported.
Further improvements were obtained at 0 ꢁC for the reaction
conducted in the presence of Mg(ClO4)2 (entry 15) where the ste-
reoisomer 6a was obtained nearly as single component (99%) with a
nearly quantitative yield. For the other cases the temperature did
not influence the stereochemical outcome very much (entries
11e14 and 16e17).
3.3. Cycloadditions to cyclopentenyl amide 4A
To complete the picture, we have investigated the effect of M(II)
salts on the cycloadditions of BNO to the N-cyclopentenyl amide 4A
4. Discussion
The fair correlations between the percentage of the directed
cycloadducts 2Aa and 5Aa in the cycloaddition mixtures and the
Table 4
Reaction yields and regioisomeric ratios of cycloadducts 6a-d in the cycloadditions
to 2-cyclopenten-1-ol (1A).
Taft's
b parameter support the idea of a HB direction in these
Entry
M(II)X (1.0 equivs.)
Yield %
6a
6b
6c
6d
6d/6b
Table 5
Reaction conducted at 25 ꢁC
Reaction yields and regioisomeric ratios of cycloadducts 5a-d in the cycloadditions
to N-cyclopentenyl acetamide (4A).
1
e
80
98
99
97
97
99
92
98
97
89
55
67
52
66
74
96
61
59
77
59
12
10
13
10
8
3
1
2
2
2
e
2
3
1
3
30
22
33
22
16
2
27
27
15
27
2.5
2.2
2.5
2.2
2.0
1.0
2.7
2.5
2.1
2.5
2
MgBr2
Entry
M(II)X (1.0 equivs.)
Yield %
5a
5b
5c
5d
5d/5b
3
MgTf2
4
ZnBr2
Reaction conducted at 25 ꢁC
5
ZnTf2
1
2
3
4
5
6
7
8
e
96
67
99
99
97
95
98
98
76
75
79
81
85
86
79
69
e
e
e
e
e
e
e
e
4
3
3
2
1
1
2
4
20
22
19
17
14
13
19
27
6
7
8
Mg(ClO4)2
Ni(ClO4)2
Zn(ClO4)2
Mg(ClO4)2
Mg(ClO4)2
2
MgBr2
MgTf2
ZnBr2
e
e
e
e
e
e
e
10
11
7
9a
10b
ZnTf2
11
Mg(ClO4)2
Ni(ClO4)2
Zn(ClO4)2
Reaction conducted at 0 ꢁ
C
11
12
13
14
15
16
17
MgBr2
MgTf2
ZnBr2
ZnTf2
Mg(ClO4)2
Ni(ClO4)2
Zn(ClO4)2
99
99
99
99
99
95
98
66
62
66
76
99
61
60
10
10
9
8
e
1
2
1
2
e
2
3
23
26
24
14
1
2.3
2.6
2.7
1.8
e
Reaction conducted at 0 ꢁ
C
11
12
13
14
15
16
17
MgBr2
MgTf2
ZnBr2
ZnTf2
Mg(ClO4)2
Ni(ClO4)2
Zn(ClO4)2
68
98
99
96
95
97
98
75
79
76
84
87
78
69
e
e
e
e
e
e
e
3
2
3
1
1
2
4
23
19
21
15
12
19
27
e
e
e
e
e
e
e
10
10
27
27
2.7
2.3
a
With 0.5 equivs. of catalysts.
With 0.2 equivs. of catalysts.
b
Please cite this article in press as: Cardarelli AM, et al., The hydrogen bond directing effect in nitrile oxide cycloadditions to allylic substituted
cyclopentenes, Tetrahedron (2017), http://dx.doi.org/10.1016/j.tet.2017.03.042

6
A.M. Cardarelli et al. / Tetrahedron xxx (2017) 1e12
reactions and point out that
practical tuning of stereoselectivity.
b
is the descriptor of choice for the
preference has been ascribed to the lack of the repulsion between
the oxygen lone pairs and the
bond.26 Solvent effects do not
p
The directing effect of the acetamido group is slightly larger
than that of the OH, well in keeping with the slightly higher HB
acidity of the amide group, as determined by thermochemical
studies22 and theoretical calculations.23 Somewhat surprisingly,
however, the sensitivity to the solvent effect is slightly larger for the
alcohol than in the case of amides, as shown by the different slopes
in Figs. 1 and 2.
sizeably alter the conformational equilibrium of 1A according to the
COSMO calculations in cyclohexane, methanol and water.27
In the case of the amide 4A steric effects determine the
conformational preferences. The acetamide substituent is more
rigid and the equatorial and axial conformers with the NH bond
directed outside the ring are about 3 kcal/mol higher in energy, the
bulky N-acetyl moiety staggering the CeC bonds of the ring.
The Transition State 3 (TS) shown in Scheme 1 nicely accounts
for the HB directing effect for both the dipolarophiles. This
arrangement requires for the cyclopentenol 1A in conformer to be
populated, a relatively easy process taking into account the small
energy differences involved. Nevertheless, the same small energetic
differences among the various conformers are partly the reason for
There are some other intriguing aspects in the results discussed
above, which deserve attention. While synthetically useful, the size
of the solvent effect is somewhat smaller than what we had ex-
pected. The strength of a moderate HB, e.g. between an ether or an
amide with n-butanol, is usually estimated at 3e4 kcal/mol (-D
Hf)24
while the change in stereoselectivity (DDGs) on going from the
apolar solvents to DMF is only 1.04 kcal/mol in the case cyclo-
pentenol and even less in the case of amides (0.76 kcal/mol).
Moreover Figs. 1 and 2 seem to indicate a persistence of the HB
directivity even in good HBA solvents, since the points relative to
alcoholic solvents and DMF lie well above the reference points of
the methyl derivatives. Although entropic effects on the stereo-
selectivity (DDGs) may be involved and the steric bulk of the O-
and N-methylated substituents may overcome that of the solvated
directing groups, the most immediate impression is that only a part
of the HB directing effect is lost in the solvation of the directing
groups. We have already noted a similar case of persistence of syn-
direction in HBA solvents in the cycloaddition of BNO to 4-
benzoylamino-cyclopentenone15 and we attributed it to the Cie-
plak effect by analogy to the explanation given for the moderate
stereoselection of the Diels-Alder reaction of cyclopentadiene with
4-substituted cyclopentenones.25
the imperfect HB directivity in non-HBA solvents (
b
¼ 0.00e0.10;
Table 1, entries 1e6) where the amounts of the anti isomers are not
negligible. This situation is repeated in the case of the amide 4A.
The lower sensitivity to the solvent effect is due to the conforma-
tional preference that leaves the acetyl group outside the ring
pointing the NeH bond in the correct direction to orient the
incoming nitrile oxide. Moreover, polar and protic solvents can be
diverted from interacting with the NH group by the carbonyl group
other than the lone pair on the nitrogen atom, determining a
consistent level of direction in the cycloaddition reaction.
The persistence of the HB even in good HBA solvents can then be
attributed to the residual interaction between the XH and the
nitrile oxide oxygen. The directing groups give rise to a sort of
bifurcated hydrogen bond with two interactions of different
magnitude and presumably influence to each other. Bifurcated
hydrogen bonds have been frequently noted in theoretical calcu-
lations28 and the main evidence of their role comes from structural
studies, which show that they are very common (Fig. 3).24 Spec-
troscopic evidence for their importance has been recently
provided.29
The conformational preferences of the dipolarophiles are sum-
marized in Scheme 5. In the case of cyclopentenol (1A) two axial
and two equatorial outside staggered conformers with the OH bond
near or far with respect to the double bond can be drawn and lie ca.
1 kcal/mol above the inside conformation. This conformational
The results obtained in the studies of the metal-mediated
cycloaddition reaction between nitrile oxides and 2-cyclopenten-
1-ol (1A) and cyclopentenyl amide 4A as dipolarophiles showed a
general increase of the reaction rates due to the simple coordina-
tion of dipole/dipolarophile. This coordination does not always
influence the reaction selectivity since it remains at the same level
reached by simple H-bonding effect, implying an incorrect geom-
etry of the complexes.
As far as the Mg-complexed cycloaddends are concerned as
drawn in Fig. 4, the metal ion must have the correct size to be co-
ordinated by the cyclopentenol (1A) that is prone to receive the
nitrile oxide on the syn face as shown in TS_A. Mg(II) ion has an
ionic radius of 0.65 Å while Ni(II) and Zn(II) possess larger values
(0.72 Å and 0.74 Å, respectively).
This implies a distortion of the resulting complex or an
Scheme 5. Conformers of 2-cyclopenten-1-ol (1A) and cyclopentenyl amide 4A.
Fig. 3. Bifurcated hydrogen bonding.
Please cite this article in press as: Cardarelli AM, et al., The hydrogen bond directing effect in nitrile oxide cycloadditions to allylic substituted
cyclopentenes, Tetrahedron (2017), http://dx.doi.org/10.1016/j.tet.2017.03.042

A.M. Cardarelli et al. / Tetrahedron xxx (2017) 1e12
7
The use of bivalent metal salts was also investigated as a tool for
increasing the directing effect. Magnesium perchlorate was found
to be the salt of choice for efficient syn selectivity and the reaction
conditions can be properly tuned up by decreasing the reaction
temperature and removing hydrogen-bond donors or acceptors,
such as the bases and their hydrochlorides used to generate the 1,3-
dipoles. These results open the way to the introduction of chiral
ligands on Mg(II) for an enantioselective route to cycloadducts of
the type 2Aa.
Cycloadduct 2Aa is the ideal candidate for our synthetic stra-
tegies towards nucleoside analogues. Scheme 6 shows the oppor-
tunities that previous investigations open in this field. Cycloadducts
of type 2Aa, bearing a simple aromatic group or polyaromatic
substituents with fluorescent properties, can be easily converted
into derivatives of type 7 containing a good leaving group (LG) for a
stereoselective introduction of nucleophilic heterobases to give the
bicyclic nucleoside analogues 8, with an anti-relationship between
the heterobases and the isoxazoline ring, these latter conformers
can be eventually elaborated into the simple carbocyclic derivatives
9 by cleavage of the NeO bond of the isoxazoline ring.16,32
On the other hand, cycloadducts 2Aa obtained from bromoni-
trile oxide undergo easy substitution of the bromine atom with
nucleophilic heterobases to give the syn-substituted nucleoside
analogues of type 10.33
Fig. 4. TSs of Mg-complexed cycloaddends.
impossibility to build a complex and as a consequence the impos-
sibility for the nitrile oxide moiety to add according to the direction
imposed by the interacting FOs.30 Hence, the directing effects are
just determined by the H-bonding producing the same selectivities
as observed in the absence of any metal ion. This type of preference
of metal ions is often found in catalyzed reactions and mainly in
1,3-dipolar cycloadditions.31 In parallel, perchlorates are often
preferred as counterions because of their low coordinating ability,
leaving the metal free to act as an aggregation centre for the
reacting ligands.31 The need for a stable and correctly built complex
is part of the reason why the required amount of magnesium salt is
equimolar with the reagents. Mg(II) is not a catalyst and the con-
struction of the complex, the consequent cycloaddition and release
of the metal ion are time consuming with respect to the simple
cycloaddition governed by H-bonding effects that still are
competitive with the coordinative process.
Finally, cycloadducts 2Aa obtained from the carbetoxynitrile
oxide bearing a good LG, as in 11, can be submitted to heterobases
substitution reaction to afford 12 that can undergo two different
pathways: a basic hydrolysis cleaves the isoxaoline ring to give the
b-hydroxy nitrile derivatives 13 while the reduction of the
The case of the cyclopentenyl amide 4A is similar although with
some potential complications: TS_B in Fig. 4 shows a single type of
coordination that excludes the carbonyl oxygen that can have a role
in diverting the metal ion from the possibility to orientate the
addition of the nitrile oxide. This cannot be excluded or avoided and
for this reason the results were less impressive than the cases of
cyclopentenol (1A).
If it is easy to depict the temperature effect in stabilizing the
complexes without slowing down the reaction rates (0 ꢁC seems a
good compromise), it is more complex to define the effect of the
couple Et3N/Et3N$HCl on the selectivity. When nitrile oxides are
generated in situ from the corresponding
a-chloroxime in the
presence of a base and its conjugated acid cannot be avoided and
this aspect has never been thoroughly investigated with the
exception of a single control experiment.32 This adds complexity to
the system under investigation and Et3Nþ-H is surely able to H-
bond to both the cycloaddends, competing with the factor playing
in favour of a syn selectivity in the cycloaddition. The results ob-
tained with the isolable pClBNO demonstrated that the absence of
Et3N and Et3N$HCl simplifies the system and slightly ameliorated
both chemical yields and selectivities in the reaction with the
cyclopentenol (1A).
5. Conclusions
The solvent effect on the HB directing effect of allylic hydroxy
and acylamino substituents in the cycloaddition of nitrile oxide to
cyclopentenol and N-cyclopentenyl amide has been studied. The
syn stereoselectivity decreases with the increasing HBA ability of
solvents and Taft's
b parameter is a satisfactory descriptor. The
directing effect decreases but does not vanish completely even in
good HBA solvents and the residual reactivity is likely due to the
instalment of a bifurcated HB between the HB proton with the HBA
solvent and the nitrile oxide oxygen. Bifurcated hydrogen bonds are
well known in structural chemistry and we suggest their involve-
ment in a case of reactivity.
Scheme 6. Synthetic strategies from cycloadduct 2Aa towards nucleoside syntheses.
Please cite this article in press as: Cardarelli AM, et al., The hydrogen bond directing effect in nitrile oxide cycloadditions to allylic substituted
cyclopentenes, Tetrahedron (2017), http://dx.doi.org/10.1016/j.tet.2017.03.042

8
A.M. Cardarelli et al. / Tetrahedron xxx (2017) 1e12
carboethoxy functionality allows for the introduction of a polar/
protic hydroxymethylene group suitable for increasing solubility of
these compounds.34
Due to the low level of selectivity obtained with the cyclo-
pentenyl amides, the corresponding cycloadducts will not be used
to access other nucleoside analogues.
(m, 2H, arom.); 13C NMR (
d, CDCl3): 28.5, 31.3 (CH2), 50.9 (CH), 77.6
(CHeOH), 91.9 (CHeO), 126.8, 128.6, 129.7, 129.8 (arom.), 158.7 (C]
N).
2Ac. 0.20 g (2%). 1H NMR (
d, CDCl3): 1.45 (s, 1H, OH), 1.80e1.84
(m, 1H, CH2), 2.20e2.24 (m, 1H, CH2), 2.01e2.04 (m, 2H, CH2),
4.10e4.12 (m, 1H, H4-isox.), 4.60 (dd, 1H, J ¼ 9.4, 7.2 Hz, CHeOH),
5.22e5.26 (m, 1H, H5-isox.), 7.40e7.44 (m, 3H, arom.), 7.76e7.80
6. Experimental section
(m, 2H, arom.); 13C NMR (
d, CDCl3): 30.6, 33.8 (CH2), 56.7 (CH), 75.4
(CHeOH), 88.0 (CHeO), 126.8, 128.6, 129.7, 130.0 (arom.), 156.0 (C]
All melting points are uncorrected. Elemental analyses were
done on a C. Erba 1106 elemental analyzer. 1H and 13C NMR spectra
were recorded on a Bruker AVANCE 300 and Bruker AC 200 spec-
trometers in CDCl3 solutions unless otherwise stated. Chemical
N).
2Ad. 2.34 g (23%). 1H NMR (
d, CDCl3): 1.56 (s, 1H, OH), 1.79e1.82
(m, 2H, CH2), 2.26e2.30 (m, 2H, CH2), 4.00 (dd, 1H, J ¼ 8.8, 1.0 Hz,
H4-isox.), 4.44e4.48 (m, 1H, CHeOH), 5.34 (ddd, 1H, J ¼ 8.8, 5.0,
1.0 Hz, H5-isox.), 7.41e7.45 (m, 3H, arom.), 7.76e7.80 (m, 2H,
shifts are expressed in ppm from internal tetramethylsilane (d). IR
spectra (nujol mulls) were recorded on an FT-IR Perkin-Elmer
Paragon 1000. HPLC quantitative analyses have been performed
with a Waters 510 HPLC apparatus equipped with an UV 490 E
arom.); 13C NMR (
d, CDCl3): 32.4, 33.0 (CH2), 61.6 (CH), 76.5
(CHeOH), 87.2 (CHeO), 126.7, 128.7, 129.0, 129.8 (arom.), 155.8 (C]
N).
detector: column RP C-18 Intersil ODS-2 (2.5
m
m, id ¼ 4.6 mm,
250 mm length); eluant water/acetonitrile. Column chromatog-
raphy and TLC: silica gel H60 and GF254 (Merck), respectively,
eluant cyclohexane/ethyl acetate 9:1 to 5:5. The identification of
samples from different experiments was secured by mixed mps and
superimposable IR spectra.
6.3. Cycloaddition of benzonitrile oxide to 3-methoxy-cyclopentene
(1B)12
To a stirred solution of 3-methoxy-cyclopentene (1B) (5.0 g,
51 mmol) and benzhydroximoyl chloride (6.7 g, 43 mmol) in ben-
zene (120 mL), 7 mL of triethylamine (51 mmol) in the same solvent
(20 mL) were added over a 0.5 h period. After stirring the reaction 2
days at r.t., the mixture was evaporated under reduced pressure
leaving a residue which was separated by column chromatography
affording the stereoisomeric cycloadducts 2Ba-d.12 These cyclo-
adducts served as references for the analyses conducted for the
investigation of the solvent effect. Hereby, we report the chemical
yields and NMR data for reference.
6.1. Materials
Benzhydroximoyl chloride, the precursor of BNO,35 was ob-
tained by treatment of benzaldoxime with sodium hypochlorite.36
4-Chlorobenzhydroxymoyl chloride, the precursor of pClBNO,35
was obtained by treatment of the corresponding 4-
chlorobenzaldoxime with sodium hypochlorite.36
2-Cyclopenten-1-ol (1A) and 3-methoxy-cyclopentene (1B)
were prepared from freshly distilled cyclopentadiene by addition of
gaseous HCl and basic hydrolysis of the obtained 3-chloro-cyclo-
pentene with NaHCO3/H2O for 1A and NaHCO3/MeOH for 1B,
respectively.37
2Ba. 0.56 g (6%) 1H NMR (
d, CDCl3): 1.59e1.62 (m, 1H, CH2),
1.92e1.94 (m, 3H, CH2), 3.53 (s, 3H, OCH3), 3.90 (qi, 1H, J ¼ 5.2 Hz,
CHeOH), 4.09 (dt, 1H, J ¼ 8.5, 2.6 Hz, H4-isox.), 5.10 (dd, 1H, J ¼ 8.5,
4.6 Hz, H5-isox.), 7.38.7.42 (m, 3H, arom.), 7.68e7.72 (m, 2H, arom.);
13C NMR (
d, CDCl3): 26.1, 27.2 (CH2), 50.4 (CH), 57.9 (OCH3), 83.9
N-Cyclopentenyl-amides 4A-D were prepared according to the
(CHeOH), 85.2 (CHeO), 126.9, 128.4, 128.7, 129.8 (arom.), 158.5 (C]
procedures reported by Curran.13e
N).
Mg(II), Zn(II) and Ni(II) salts were purchased from Sigma-
Aldrich; MgBr2 and ZnBr2 were dried at 140 ꢁC for one day under
vacuum.
2Bb. 2.62 g (28%). 1H NMR (
d, CDCl3): 1.60e1.63 (m, 1H, CH2),
1.91e1.94 (m, 2H, CH2), 2.18e2.22 (m, 1H, CH2), 3.40 (s, 3H, OCH3),
3.94 (d, 1H, J ¼ 3.8 Hz, CHeOH), 4.17 (dt, 1H, J ¼ 9.0, 2.0 Hz, H4-
isox.), 5.05 (d, 1H, J ¼ 9.0 Hz, H5-isox.), 7.42e7.45 (m, 3H, arom.),
6.2. Cycloaddition of benzonitrile oxide to 2-cyclopenten-1-ol
7.70e7.73 (m, 2H, arom.); 13C NMR (
d, CDCl3): 28.4, 28.8 (CH2), 51.1
(1A)12
(CH), 56.7 (OCH3), 86.8 (CHeOH), 89.5 (CHeO), 126.9, 128.7, 128.8,
129.8 (arom.), 158.8 (C]N).
To a stirred solution of 2-cyclopenten-1-ol (1A) (5.0 g, 59 mmol)
and benzhydroximoyl chloride (7.7 g, 50 mmol) in benzene
(120 mL), 8 mL of triethylamine (57 mmol) in the same solvent
(20 mL) were added over a 0.5 h period. After stirring the reaction 2
days at r.t., the mixture was evaporated under reduced pressure
leaving a residue which was separated by column chromatography
affording the stereoisomeric cycloadducts 2Aa-d.12 These cyclo-
adducts served as references for the analyses conducted for the
investigation of the solvent effect. Hereby, we report the chemical
yields and NMR data for reference.
2Bc. 0.47 g (5%). 1H NMR (
d, CDCl3): 1.89e1.93 (m, 3H, CH2),
2.15e2.18 (m, 1H, CH2), 3.17 (s, 3H, OCH3), 4.08e4.12 (m, 2H,
CHeOH and H4-isox.), 5.18e5.22 (m, 1H, H5-isox.), 7.36e7.40 (m,
3H, arom.), 7.70e7.74 (m, 2H, arom.); 13C NMR (
d, CDCl3): 29.6, 30.6
(CH2), 55.9 (CH), 57.5 (OCH3), 84.1 (CHeOH), 87.4 (CHeO), 127.0,
128.1, 129.1, 130.6 (arom.), 156.7 (C]N).
2Bd. 4.67 g (50%). 1H NMR (
d, CDCl3): 1.57e1.60 (m, 1H, CH2),
1.95e1.99 (m, 1H, CH2), 2.12e2.16 (m, 2H, CH2), 3.37 (s, 3H, OCH3),
3.89 (dt, 1H, J ¼ 4.4, 1.0 Hz, CHeOH), 4.04 (dt, 1H, J ¼ 8.8, 1.0 Hz, H4-
isox.), 5.26e5.30 (m, 1H, J ¼ 8.8, 3.0, 1.2 Hz, H5-isox.), 7.42e7.44 (m,
2Aa. 5.28 g (52%). 1H NMR (
d, CDCl3): 1.51e154 (m, 2H, CH2),
3H, arom.), 7.74e7.78 (m, 2H, arom.); 13C NMR (
d, CDCl3): 27.9, 32.5
1.92e1.95 (m, 2H, CH2), 2.40 (bs, 1H, OH), 4.10e4.15 (m, 1H,
CHeOH), 4.22e4.25 (m, 1H, H4-isox.), 5.20 (dd, 1H, J ¼ 9.3, 5.0 Hz,
H5-isox.), 7.42e7.44 (m, 3H, arom.), 7.69e7.71 (m, 2H, arom.); 13C
(CH2), 56.4 (CH), 59.3 (OCH3), 85.3 (CHeOH), 87.1 (CHeO), 126.6,
128.7, 129.2, 129.8 (arom.), 155.7 (C]N).
NMR (
(CHeO), 126.9, 128.0, 128.7, 130.0 (arom.), 159.4 (C]N).
2Ab. 1.02 g (10%). 1H NMR (
, CDCl3): 1.68 (s, 1H, OH), 1.77e1.81
d
, CDCl3): 26.6, 30.9 (CH2), 50.8 (CH), 76.0 (CHeOH), 85.3
6.4. Cycloaddition of benzonitrile oxide to N-cylopentenyl-
acetamide (4A)
d
(m, 2H, CH2), 1.94e1.96 (m, 1H, CH2), 2.34e2.38 (m, 1H, CH2),
4.20e4.24 (m, 1H, H4-isox.), 4.48 (dt, 1H, J ¼ 9.0, 2.0 Hz, CHeOH),
4.99 (d,1H, J ¼ 9.0 Hz, H5-isox.), 7.41e7.45 (m, 3H, arom.), 7.70e7.74
To a stirred solution of N-cyclopentenyl-acetamide (4A) (6.0 g,
48 mmol) and benzhydroximoyl chloride (6.2 g, 40 mmol) in
methanol (120 mL), 7 mL of triethylamine (45 mmol) in the same
Please cite this article in press as: Cardarelli AM, et al., The hydrogen bond directing effect in nitrile oxide cycloadditions to allylic substituted
cyclopentenes, Tetrahedron (2017), http://dx.doi.org/10.1016/j.tet.2017.03.042

A.M. Cardarelli et al. / Tetrahedron xxx (2017) 1e12
9
solvent (20 mL) were added over a 0.5 h period. After stirring the
reaction 2 days at r.t., the mixture was evaporated under reduced
pressure leaving a residue which was separated by column chro-
matography affording the stereoisomeric cycloadducts 5Aa-d.
5Aa. 2.93 g (30%). Colorless solid, mp 163-4 ꢁC from i-Pr2O. Rf
5Bb. 0.17 g (65%). Colorless solid, mp 116e118 ꢁC from ethanol. Rf
(1:1 cyclohexane/AcOEt) 0.30. IR (cmꢀ1): nC 1643. 1H NMR (
d,
]
O
CDCl3): 1.89e1.92 (m, 2H, CH2), 2.38e2.42 (m, 2H, CH2), 2.20 (s, 3H,
COCH3), 3.00 (s, 3H, NeCH3), 4.23e4.27 (m, 1H, H4-isox.),
5.00e5.03 (m, 1H, CHeNH), 5.38e5.42 (m, 1H, H5-isox.), 7.40e7.44
(1:1 cyclohexane/AcOEt) 0.67. IR (cmꢀ1): nNH 3246, nC
NMR (
]
O
1637. 1H
(m, 3H, arom.), 7.66e7.70 (m, 2H, arom.); 13C NMR (
d, CDCl3): 22.8,
d
, CDCl3): 2.04 (s, 3H, COCH3), 1.0e1.06 (m, 2H, CH2),
28.9 (CH2), 29.6 (NeCH3), 36.0 (COCH3) 51.9 (CH), 67.3 (CHeN), 89.4
(CHeO), 126.9, 128.3, 129.8, 130.1 (arom.), 159.3 (C]N), 170.9 (C]
O). Elemental analysis: C 69.69%, H 6.98%, N 10.82%; calcd. for
1.97e2.01 (m, 2H, CH2), 4.13e4.17 (m, 1H, H4-isox.), 4.44e4.46 (m,
1H, CHeNH), 5.03 (dd, 1H, J ¼ 8.4, 5.1 Hz, H5-isox.), 6.11 (d, 1H,
J ¼ 8.4 Hz, NH), 7.41e7.43 (m, 3H, arom.), 7.68e7.71 (m, 2H, arom.).
C15H18N2O2 (258.3) C 69.74%, H 7.02%, N 10.85%.
13C NMR (
d, CDCl3): 23.2 (COCH3), 27.7, 28.3 (CH2), 50.8 (CH), 55.3
5Bc. 0.13 g (52%); Straw colored oil. Rf (1:1 cyclohexane/AcOEt)
(CHNH), 84.2 (CHeO), 126.9, 128.0, 128.7, 130.1 (arom.), 159.2 (C]
N), 169.9 (C]O). Elemental analysis: C 68.81%, H 6.61%, N 11.50%;
calcd. for C14H16N2O2 (244.3) C 68.83%, H 6.60%, N 11.47%.
5Ab. 0.59 g (6%). Colorless solid, mp 150-1 ꢁC from benzene. Rf
0.55. IR (cmꢀ1): nC 1634. 1H NMR (
] d, CDCl3): 1.67e1.72 (m, 2H,
O
CH2), 1.89e1.92 (m, 2H, CH2), 1.64 (s, 3H, COCH3), 2.42 (s, 3H,
NeCH3), 4.36 (t, 1H, J ¼ 1.3 Hz, H4-isox.), 5.02e5.06 (m, 1H,
CHeNH), 5.18e5.22 (m, 1H, H5-isox.), 7.36e7.40 (m, 3H, arom.),
(1:1 cyclohexane/AcOEt) 0.33. IR (cmꢀ1): nNH 3285, nC
]
O 1645. 1H
7.64e7.68 (m, 2H, arom.). 13C NMR (
d, CDCl3): 22.6, 25.1 (CH2), 29.3
NMR (
d
, CDCl3): 2.04 (s, 3H, COCH3), 1.78e1.82 (m, 2H, CH2),
(NeCH3), 30.8 (COCH3) 51.0 (CH), 58.1 (CHeN), 87.2 (CHeO), 126.7,
128.4, 129.5, 129.7 (arom.), 157.3 (C]N), 171.8 (C]O). Elemental
analysis: C 69.75%, H 6.98%, N 10.86%; calcd. for C15H18N2O2 (258.3)
C 69.74%, H 7.02%, N 10.85%.
2.16e2.20 (m, 2H, CH2), 4.18 (dt, 1H, J ¼ 9.1, 2.6 Hz, H4-isox.),
4.39e4.42 (m, 1H, CHeNH), 5.10 (dd, 1H, J ¼ 9.1, 1.8 Hz, H5-isox.),
5.66 (d, 1H, J ¼ 6.0 Hz, NH), 7.40e7.44 (m, 3H, arom.), 7.68e7.72 (m,
2H, arom.). 13C NMR (
d, CDCl3): 23.3 (COCH3), 29.3, 29.4 (CH2), 51.0
5Bd. 0.12 g (48%); Straw colored oil. Rf (1:1 cyclohexane/AcOEt)
(CH), 58.2 (CHNH), 90.6 (CHeO), 126.9, 128.6, 128.7, 129.9 (arom.),
158.4 (C]N), 169.8 (C]O). Elemental analysis: C 68.79%, H 6.58%, N
11.46%; calcd. for C14H16N2O2 (244.3) C 68.83%, H 6.60%, N 11.47%.
5Ac. 0.78 g (8%). Colorless solid, mp 155-6 ꢁC from i-Pr2O. Rf (1:1
0.33. IR (cmꢀ1): nC 1643. 1H NMR (
] d, CDCl3): 1.98e2.02 (m, 2H,
O
CH2), 2.32e2.37 (m, 2H, CH2), 2.11 (s, 3H, COCH3), 2.95 (s, 3H,
NeCH3), 4.12e4.16 (m, 1H, H4-isox.), 4.60e4.64 (m, 1H, CHeNH),
5.33e5.37 (m, 1H, H5-isox.), 7.36e7.40 (m, 3H, arom.), 7.58e7.62
cyclohexane/AcOEt) 0.45. IR (cmꢀ1): nNH 3257, nC
]
O 1643. 1H NMR
(m, 2H, arom.). 13C NMR (
d, CDCl3): 22.7, 29.2 (CH2), 32.8 (NeCH3),
(
d
, CDCl3): 1.41 (s, 3H, COCH3), 1.50e2.30 (m, 4H, CH2), 4.26 (t, 1H,
34.3 (COCH3) 56.4 (CH), 63.7 (CHeN), 87.8 (CHeO), 126.7, 128.1,
128.9, 130.2 (arom.), 157.8 (C]N), 170.3 (C]O). Elemental analysis:
C 69.74%, H 6.98%, N 10.87%; calcd. for C15H18N2O2 (258.3) C 69.74%,
H 7.02%, N 10.85%.
J ¼ 8.8 Hz, H4-isox.), 4.70e4.72 (m, 1H, CHeNH), 5.10 (d, 1H,
J ¼ 9.0 Hz, NH), 5.25 (dd, 1H, J ¼ 8.8, 4.0 Hz, H5-isox.), 7.38e7.42 (m,
3H, arom.), 7.64e7.66 (m, 2H, arom.). 13C NMR (
d, CDCl3): 22.5
(COCH3), 29.5, 31.0 (CH2), 52.7 (CH), 53.5 (CHNH), 88.0 (CHeO),
126.7, 128.5, 129.6, 129.7 (arom.), 156.9 (C]N), 169.8 (C]O).
Elemental analysis: C 68.82%, H 6.59%, N 11.45%; calcd. for
6.6. Cycloaddition of benzonitrile oxide to N-cyclopentenyl-
benzamide (4C)
C
14H16N2O2 (244.3) C 68.83%, H 6.60%, N 11.47%.
5Ad. 4.01 g (41%). Colorless solid, mp 199e200 ꢁC from benzene.
To a stirred solution of N-cyclopentenyl-benzamide (4C) (4.0 g,
21 mmol) and benzhydroximoyl chloride (5.0 g, 32 mmol) in
methanol (120 mL), 5 mL of triethylamine (36 mmol) in the same
solvent (20 mL) were added over a 0.5 h period. After stirring the
reaction 2 days at r.t., the mixture was evaporated under reduced
pressure leaving a residue which was separated by column chro-
matography affording the stereoisomeric cycloadducts 5Ca-d.
5Ca. 2.45 g (25%). Colorless solid, mp 195e196 ꢁC from ethyl
Rf (1:1 cyclohexane/AcOEt) 0.42. IR (cmꢀ1): nNH 3289, nC
] 1637.
O
1H NMR (
d, CDCl3): 2.03 (s, 3H, COCH3), 1.78e1.82 (m, 2H, CH2),
2.17e2.21 (m, 2H, CH2), 4.23 (d, 1H, J ¼ 9.0 Hz, H4-isox.), 4.39 (t, 1H,
J ¼ 5.0, CHeNH), 5.26 (dd, 1H, J ¼ 9.0, 5.0 Hz, H5-isox.), 5.82 (d, 1H,
J ¼ 5.0 Hz, NH), 7.40e7.44 (m, 3H, arom.), 7.80e7.82 (m, 2H, arom.).
13C NMR (
d, CDCl3): 23.2 (COCH3), 29.2, 33.7 (CH2), 55.6 (CH), 59.5
(CHNH), 85.9 (CHeO), 127.2, 128.5, 128.6, 129.8 (arom.), 156.2 (C]
N), 170.2 (C]O). Elemental analysis: C 68.79%, H 6.61%, N 11.49%;
calcd. for C14H16N2O2 (244.3) C 68.83%, H 6.60%, N 11.47%.
acetate. Rf (1:1 cyclohexane/AcOEt) 0.40. IR (cmꢀ1): nNH 3349, nC
1638. 1H NMR (
, DMSO): 1.55e2.35 (m, 4H, CH2), 4.33 (t, 1H,
]
O
d
J ¼ 8.5 Hz, H4-isox.), 4.40e4.44 (m, 1H, CHeNH), 5.09 (dd, 1H,
6.5. Methylation of cycloadducts 5Aa-d
J ¼ 8.5, 5.0 Hz, H5-isox.), 7.48e7.52 (m, 6H, arom.), 7.66e7.71 (m,
2H, arom.), 7.91e7.94 (m, 2H, arom.), 8.47 (d, 1H, J ¼ 7.0 Hz, NH). 13
C
To a stirred solution of the cycloadducts 5Aa-d (0.24 g, 1 mmol)
in anhydrous THF (10 mL), NaH (0.029 g, 1.2 mmol) was added
portionwise in the presence of a slight excess of CH3I (1 mL,
2 mmol). After stirring the reactions 5 h at r.t., methanol was added
first, followed by water and the reaction mixtures extracted with
chloroform. The dry organic phases were evaporated and affording
in quantitative yields the crude methylated compounds 5Ba-d,
which were recrystallized from suitable solvents and characterized.
5Ba. 0.19 g (75%). Colorless solid, mp 104e105 ꢁC from ligroin. Rf
NMR (d, DMSO): 26.8, 27.2 (CH2), 50.3 (CH), 56.3 (CHNH), 84.4
(CHeO), 126.7, 127.5, 128.1, 128.4, 128.9, 129.8, 131.1, 134.2 (arom.),
158.7 (C]N), 166.2 (C]O). Elemental analysis: C 74.50%, H 5.91%, N
9.19%; calcd. for C19H18N2O2 (306.4) C 74.49%, H 5.92%, N 9.15%.
5Cb. 0.29 g (3%). Colorless solid, mp 188e189 ꢁC from ethyl ac-
etate. Rf (1:1 cyclohexane/AcOEt) 0.51. IR (cmꢀ1): nNH 3295, nC
]
O
1675. 1H NMR (
d, DMSO): 1.85e2.35 (m, 4H, CH2), 4.30e4.34 (m,
2H, H4-isox. and CHeNH), 5.09 (dd, 1H, J ¼ 9.2, 1.0 Hz, H5-isox.),
7.46e7.50 (m, 6H, arom.), 7.70e7.72 (m, 2H, arom.), 7.85e7.88 (m,
(1:1 cyclohexane/AcOEt) 0.50. IR (cmꢀ1): nC
]
O
1651. 1H NMR (
d
,
2H, arom.), 8.49 (d, 1H, J ¼ 6.5 Hz, NH). 13C NMR (
d, DMSO): 29.7,
CDCl3): 1.70e2.00 (m, 4H, CH2), 2.17 (s, 3H, COCH3), 3.09 (s, 3H,
NeCH3), 4.07e4.10 (m, 1H, H4-isox.), 4.82e4.86 (m, 1H, CHeNH),
5.20 (dd, 1H, J ¼ 8.5, 4.4 Hz, H5-isox.), 7.38e7.42 (m, 3H, arom.),
32.9 (CH2), 55.4 (CH), 58.3 (CHNH), 86.7 (CHeO), 126.8, 127.5, 128.1,
128.8, 128.9, 129.8, 131.2, 134.3 (arom.), 156.2 (C]N), 166.5 (C]O).
Elemental analysis:
19H18N2O2 (306.4) C 74.49%, H 5.92%, N 9.15%.
5Cc. 1.18 g (12%). Colorless solid, mp 204e205 ꢁC from benzene.
Rf (1:1 cyclohexane/AcOEt) 0.70. IR (cmꢀ1): nNH 3325, nC
1660.
1H NMR (
, DMSO): 1.77e2.09 (m, 4H, CH2), 4.38 (t, 1H, J ¼ 7.6 Hz,
H4-isox.), 4.67e4.70 (m, 1H, CHeNH), 5.18 (dd, 1H, J ¼ 8.5, 2.0 Hz,
C 74.48%, H 5.89%, N 9.12%; calcd. for
7.65e7.69 (m, 2H, arom.). 13C NMR (
d
, CDCl3): 22.3, 22.7 (CH2), 27.2
C
(NeCH3), 33.4 (COCH3) 50.3 (CH), 58.9 (CHeN), 85.8 (CHeO), 126.7,
128.2, 128.6, 129.8 (arom.), 158.7 (C]N), 171.5 (C]O). Elemental
analysis: C 69.73%, H 7.01%, N 10.82%; calcd. for C15H18N2O2 (258.3)
C 69.74%, H 7.02%, N 10.85%.
]
O
d
Please cite this article in press as: Cardarelli AM, et al., The hydrogen bond directing effect in nitrile oxide cycloadditions to allylic substituted
cyclopentenes, Tetrahedron (2017), http://dx.doi.org/10.1016/j.tet.2017.03.042

10
A.M. Cardarelli et al. / Tetrahedron xxx (2017) 1e12
H5-isox.), 7.16e7.20 (m, 7H, arom.), 7.32e7.36 (m, 1H, arom.),
6.8. Cycloaddition of 4-chlorobenzonitrile oxide to 2-cyclopenten-
1-ol (1A)
7.55e7.58 (m, 2H, arom.), 8.11 (d, 1H, J ¼ 8.0 Hz, NH). 13C NMR (
d
,
DMSO): 27.9, 30.6 (CH2), 49.4 (CH), 52.2 (CHNH), 87.7 (CHeO),
126.5, 126.9, 127.4, 128.1, 128.9, 129.8, 130.5, 134.2 (arom.), 158.7
(C]N),166.6 (C]O). Elemental analysis: C 74.51%, H 5.91%, N 9.11%;
calcd. for C19H18N2O2 (306.4) C 74.49%, H 5.92%, N 9.15%.
To a stirred solution of 2-cyclopenten-1-ol (1A) (5.0 g, 59 mmol)
and 4-chlorobenzhydroxymoyl chloride (9.5 g, 50 mmol) in ben-
zene as solvent (120 mL), 8 mL of triethylamine (57 mmol) dis-
solved in the same solvent (20 mL) were added over a 0.5 h period.
After stirring the reaction 2 days at room temperature, the mixture
was evaporated under reduced pressure leaving a residue that was
submitted to chromatographic separation to isolate the cyclo-
adducts 6a-d, which were fully characterized.
5Cd. 5.00 g (51%). Colorless solid, mp 193e195 ꢁC from ethanol.
Rf (1:1 cyclohexane/AcOEt) 0.60. IR (cmꢀ1): nNH 3401, nC
] 1663.
O
1H NMR (
d
, DMSO): 1.57e2.39 (m, 4H, CH2), 4.28 (d, 1H, J ¼ 9.0 Hz,
H4-isox.), 4.34 (t, 1H, J ¼ 5.8 Hz, CHeNH), 5.27 (dd, 1H, J ¼ 9.0,
5.0 Hz, H5-isox.), 7.48e7.52 (m, 6H, arom.), 7.90e7.93 (m, 4H,
arom.), 8.57 (d, 1H, J ¼ 5.8 Hz, NH). 13C NMR (
d
, DMSO): 29.7, 32.9
6a. 5.23 g (44%); Colorless solid, mp 199 ꢁC (dec.) from ethanol.
(CH2), 55.4 (CH), 58.3 (CHNH), 86.7 (CHeO), 126.8, 127.5, 128.1,
128.8, 128.9, 129.8, 131.2, 134.3 (arom.), 156.2 (C]N), 166.5 (C]O).
Rf (1:1 cyclohexane/AcOEt) 0.36. IR (cmꢀ1): nOH 3380, nC
] 1589.
N
1H NMR (
d, CDCl3): 1.49e1.52 (m, 1H, CH2), 1.90e1.93 (m, 3H, CH2),
Elemental analysis:
C
74.47%,
H
5.90%,
N
9.20%; calcd. for
4.06e4.09 (m, 1H, H4-isox.), 4.19e4.22 (m, 1H, CHeOH), 5.03 (dd,
C
19H18N2O2 (306.4) C 74.49%, H 5.92%, N 9.15%.
1H, J ¼ 9, 5 Hz, H5-isox.), 7.31e7.70 (AA’BB0 syst., 4H, arom.). 13C
NMR (d, CDCl3): 26.2, 27.5 (CH2), 50.0 (CH), 81.9 (CHeOH), 85.3
6.7. Methylation of cycloadducts 5Ca-d
(CHeO), 126.9, 127.0, 128.7, 135.7 (arom.), 157.6 (C]N). Elemental
analysis: C 60.62%, H 5.04%, N 5.90%; calcd. for C12H12NO2Cl (237.7)
C 60.64%, H 5.09%, N 5.89%.
To a stirred solution of the cycloadducts 5Ca-d (0.31 g, 1 mmol)
in anhydrous THF (10 mL), NaH (0.029 g, 1.2 mmol) was added
portionwise in the presence of a slight excess of CH3I (1 mL,
2 mmol). After stirring the reactions 5 h at r.t., methanol was added
first, followed by water and the reaction mixtures extracted with
chloroform. The dry organic phases were evaporated and affording
in quantitative yields the crude methylated compounds 5Da-d,
which were recrystallized from suitable solvents and characterized.
5Da. 0.25 g (66%). Colorless solid, mp 137e138 ꢁC from ethanol.
6b. 1.19 g (10%); Colorless solid, mp 143e144 ꢁC from ethanol. Rf
(1:1 cyclohexane/AcOEt) 0.52. IR (cmꢀ1): nOH 3366, nC N 1589. 1H
]
NMR (d, CDCl3): 1.77e1.81 (m, 3H, CH2), 2.21e2.24 (m, 1H, CH2),
4.16 (dt,1H, J ¼ 5,1 Hz, H4-isox.), 4.43 (d,1H, J ¼ 2 Hz, CHeOH), 4.99
(d, 1H, J ¼ 5 Hz, H5-isox.), 7.31e7.70 (AA’BB0 syst., 4H, arom.). 13C
NMR (d, CDCl3): 28.6, 31.5 (CH2), 50.8 (CH), 77.1 (CHeOH), 92.3
(CHeO), 128.1, 128.6, 129.0, 135.8 (arom.), 157.9 (C]N). Elemental
analysis: C 60.60%, H 5.06%, N 5.86%; calcd. for C12H12NO2Cl (237.7)
C 60.64%, H 5.09%, N 5.89%.
Rf (1:1 cyclohexane/AcOEt) 0.36. IR (cmꢀ1): nC O 1633. 1H NMR (
] d,
CDCl3): 1.99 (m, 4H, CH2), 3.12 (s, 3H, NeCH3), 4.17 (bs, 1H, H4-
isox.), 4.92 (bs, 1H, CHeNH), 5.44 (bs, 1H, H5-isox.), 7.42.7.46 (m,
6c. 0.24 g (2%); Colorless solid, mp 169e170 ꢁC from ethanol. Rf
(1:1 cyclohexane/AcOEt) 0.30. IR (cmꢀ1): nOH 3395, nC N 1590. 1H
]
6H, arom.), 7.70e7.73 (m, 4H, arom.); 13C NMR (
d
, CDCl3): 24.4, 27.3
NMR (
H4-isox.), 4.55e4.58 (m, 1H, CHeOH), 5.20e5.23 (m, 1H, H5-isox.),
7.31e7.70 (AA'BB0 syst., 4H, arom.). 13C NMR (
, CDCl3): 30.6, 33.8
d
, CDCl3): 1.70e2.30 (m, 4H, CH2), 4.04 (dd, 1H, J ¼ 9, 7 Hz,
(CH2), 29.6 (NeCH3), 50.5 (CH), 60.0 (CHeN), 85.9 (CHeO), 126.8,
128.3, 128.7, 129.3, 130.0, 136.7 (arom.), 158.7 (C]N), 172.2 (C]O).
d
Elemental analysis:
20H20N2O2 (320.4) C 74.97%, H 6.29%, N 8.74%.
5Db. 0.22 g (57%). Colorless solid, mp 141e144 ꢁC from ethanol.
Rf (1:1 cyclohexane/AcOEt) 0.18. IR (cmꢀ1): nC O 1626. 1H NMR (
C
74.95%,
H
6.30%,
N
8.72%; calcd. for
(CH2), 56.3 (CH), 75.5 (CHeOH), 88.3 (CHeO), 128.2, 128.6, 128.8,
135.6 (arom.), 155.3 (C]N). Elemental analysis: C 60.66%, H 5.07%,
N 5.86%; calcd. for C12H12NO2Cl (237.7) C 60.64%, H 5.09%, N 5.89%.
C
ꢁ
]
d
,
6d. 2.85 g (24%); Colorless solid, mp > 200 C (dec.) from
CDCl3): 1.80e2.50 (m, 4H, CH2), 3.08 (s, 3H, NeCH3), 4.22 (bs, 1H,
H4-isox.), 4.36.4.40 (m, 1H, CHeNH), 5.40 (bs, 1H, H5-isox.),
7.40e7.44 (m, 6H, arom.), 7.47e7.50 (m, 2H, arom.), 7.86e7.89 (m,
ethanol. Rf (1:1 cyclohexane/AcOEt) 0.44. IR (cmꢀ1): nOH 3386,
nC
] d, CDCl3): 1.73e1.76 (m, 2H, CH2), 2.21e2.24
1599. 1H NMR (
N
(m, 2H, CH2), 3.93 (d, 1H, J ¼ 5 Hz, H4-isox.), 4.36 (d, 1H, J ¼ 2 Hz,
2H, arom.). 13C NMR (
d, CDCl3): 28.9, 29.6 (CH2), 31.8 (NeCH3), 51.5
CHeOH), 5.28e5.31 (m, 1H, H5-isox.), 7.31e7.70 (AA'BB0 syst., 4H,
(CH), 67.4 (CHeN), 84.9 (CHeO), 126.9, 127.1, 127.2, 128.4, 128.5,
128.7, 129.6, 129.9137.9 (arom.), 156.9 (C]N), 169.5 (C]O).
arom.). 13C NMR (
d, CDCl3): 32.4, 32.9 (CH2), 61.4 (CH), 76.4
(CHeOH), 87.6 (CHeO), 128.0, 128.2, 129.0, 135.8 (arom.), 155.1 (C]
Elemental analysis:
20H20N2O2 (320.4) C 74.97%, H 6.29%, N 8.74%.
5Dc. 0.23 g (58%). Colorless solid, mp 107e108 ꢁC from ethanol.
Rf (1:1 cyclohexane/AcOEt) 0.33. IR (cmꢀ1): nC O 1634. 1H NMR (
C
74.92%,
H
6.28%,
N
8.73%; calcd. for
N). Elemental analysis: C 60.65%, H 5.10%, N 5.88%; calcd. for
C12H12NO2Cl (237.7) C 60.64%, H 5.09%, N 5.89%.
C
]
d
,
6.9. Solvent effect and HPLC determinations
CDCl3): 1.80e2.40 (m, 4H, CH2), 2.54 (s, 3H, NeCH3), 4.51 (t, 1H,
J ¼ 8.5 Hz, H4-isox.), 5.10e5.14 (m, 1H, CHeNH), 5.31 (dd, 1H,
J ¼ 8.5, 3.5 Hz, H5-isox.), 7.27e7.31 (m, 6H, arom.), 7.40e7.43 (m,
The dipolarophiles 1A,B and 4A-D (1.2 equivs.) were dissolved in
the desired solvent (25 mL) and 1 equiv. of benzhydroximoyl
chloride (62.4 mg, 0.4 mmol), the precursor of BNO, was added to
the solution, followed by 1.2 equivs. of triethylamine and stirred for
2 days until complete consumption of BNO. The reaction mixtures
were evaporated to dryness, taken up with acetonitrile and sub-
mitted to HPLC analyses for quantitative determinations. Yields and
regioisomeric ratios have been determined upon external standard
method or internal normalization.
2H, arom.), 7.76e7.78 (m, 2H, arom.); 13C NMR (
d, CDCl3): 24.4, 27.0
(CH2), 29.3 (NeCH3), 51.3 (CH), 59.0 (CHeN), 87.8 (CHeO), 126.7,
126.8, 128.0, 128.7, 129.4, 129.8 (arom.), 157.6 (C]N), 172.2 (C]O).
Elemental analysis:
20H20N2O2 (320.4) C 74.97%, H 6.29%, N 8.74%.
5Dd. 0.24 g (61%). Colorless solid, mp 148e150 ꢁC from ethanol.
Rf (1:1 cyclohexane/AcOEt) 0.24. IR (cmꢀ1): nC O 1607. 1H NMR (
C 75.00%, H 6.29%, N 8.72%; calcd. for
C
]
d,
CDCl3): 1.99e2.50 (m, 4H, CH2), 3.06 (s, 3H, NeCH3), 4.21 (bs, 1H,
H4-isox.), 4.43 (bs, 1H, CHeNH), 5.38 (bs, 1H, H5-isox.), 7.20e7.70
Cycloaddition of BNO to 1A: isocratic elution H2O/CH3CN 70/30
(1.2 mL/min); UV detection at
Cycloaddition of BNO to 1B: isocratic elution H2O/CH3CN 60/40
(1.3 mL/min); UV detection at
¼ 264 nm.
Cycloaddition of BNO to 4A: elution gradient from H2O/CH3CN
80/20 to 60/40 (1.3 mL/min); UV detection at
¼ 264 nm.
Cycloaddition of BNO to 4B: isocratic elution H2O/CH3CN 70/30
l
¼ 262 nm.
(m, 10H, arom.); 13C NMR (
d, CDCl3): 29.6, 29.7 (CH2), 33.0 (NeCH3),
55.3 (CH), 55.4 (CHeN), 87.5 (CHeO), 126.4, 126.9, 128.3, 128.8,
129.5, 130.1 (arom.), 157.9 (C]N), 171.7 (C]O). Elemental analysis:
C 74.96%, H 6.25%, N 8.69%; calcd. for C20H20N2O2 (320.4) C 74.97%,
H 6.29%, N 8.74%.
l
l
Please cite this article in press as: Cardarelli AM, et al., The hydrogen bond directing effect in nitrile oxide cycloadditions to allylic substituted
cyclopentenes, Tetrahedron (2017), http://dx.doi.org/10.1016/j.tet.2017.03.042

A.M. Cardarelli et al. / Tetrahedron xxx (2017) 1e12
11
(1.3 mL/min); UV detection at
Cycloaddition of BNO to 4C: isocratic elution H2O/CH3CN 55/45
(1.0 mL/min); UV detection at
¼ 264 nm.
Cycloaddition of BNO to 4D: isocratic elution H2O/CH3CN 50/50
(1.3 mL/min); UV detection at
l
¼ 264 nm.
CUP: F11J12000210001) is gratefully acknowledged. Many tranks
are due to CINMPIS for a research grant to AMC.
l
Appendix A. Supplementary data
l
¼ 264 nm.
Cycloadditions to 2-cyclopenten-1-ol (1A) in the presence of
M(II) salts and hplc determinations: general methods.
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.tet.2017.03.042.
6.9.1. Method A
References
In a 5 mL vial the dipolarophile 1A (1.2 equivs.) and 1.2 equivs. of
the M(II) salts were dissolved in anhydrous DCM (1 mL) at the
desired temperature. Benzhydroxymoyl chloride (64 mg, 1 equiv.),
dissolved in 1 mL of DCM, was added to the solution, followed by
1.2 equivs. of triethylamine and left under stirring for 1 day. The
reaction mixtures were then diluted with DCM, washed carefully
with water and the organic phase dried over anhydrous Na2SO4.
Evaporation of DCM leaves a residue that was taken up with
acetonitrile and the resulting solution filtered on cationic and
anionic exchange resins (SUPELCO LC-SAX SPE 504815 and 504920)
to ensure complete elimination of inorganic salts. The solution
volume was then adjusted to 25 mL and analyzed through HPLC for
quantitative determinations. Yields and stereoisomeric ratios were
determined upon internal normalization.
1. Hoveyda AH, Evans DA, Fu GC. Chem Rev. 1993;93:1307e1370.
2. Henbest HB, Wilson RAL. J Chem Soc. 1959:4136e4138.
3. (a) Curci R, Di Prete RA, Edwards JO, Modena G. J Org Chem. 1970;35:740e745;
(b) Kavcic R, Plesnicar B. J Org Chem. 1970;35:2033e2035.
4. Adam W, Corma A, Reddy TI, Renz M. J Org Chem. 1997;62:3631e3637.
5. (a) Sharpless KB, Verhoeven TR. Aldrichim. Acta. 1979;12:63e74;
(b) Rossiter BE, Verhoeven TR, Sharpless KB. Tetrahedron Lett. 1979;19:
4733e4737.
6. Adam W, Mitchell C. Angew Chem Int Ed Engl. 1996;35:533e535.
7. (a) Johnson RA, Sharpless KB. In: Ojima I, ed. Catalytic Asymmetric Synthesis.
New York: VCH; 1993. Chp 4.1;
(b) Jacobsen EN. In: Ojima I, ed. Catalytic Asymmetric Synthesis. New York: VCH;
1993. Chp 4.2;
(c) Katsuki T. Coord Chem Rev. 1995;140:189e214.
8. Adam W, Smerz AK. J Org Chem. 1996;61:3506e3510.
ꢁ
9. Freccero M, Gandolfi R, Sarzi-Amade M, Rastelli A. J Org Chem. 1999;64:
3853e3860.
€
10. Adam W, Degen HG, Saha-Moller CR. J Org Chem. 1999;64:1274e1277.
HPLC conditions: Isocratic elution H2O/MeCN 70/30 (1.2 mL/
11. Caramella P, Grünanger P. In: Padwa A, ed. 1,3-Dipolar Cycloaddition Chemistry.
New York: John Wiley and Sons Inc; 1984:291; vol. 1.
12. (a) Caramella P, Cellerino G. Allylic OH directing effects. Tetrahedron Lett. 1974:
229e232;
min); UV detection at
l
¼ 262 nm.
6.9.2. Method B
(b) De Micheli C, Gamba A, Gandolfi R, Scevola L. J Chem Soc Chem Comm. 1976:
246e247;
In a 5 mL vial the dipolarophile 1A (1.2 equivs.) and 1.2 equivs. of
the M(II) salts were dissolved in anhydrous DCM (1 mL) at the
desired temperature. 4-Chlorobenzhydroxymoyl chloride (63 mg,
1 equiv.), dissolved in 1 mL of DCM, was added to the solution,
followed by 1.2 equivs. of triethylamine and left under stirring for 1
day. The reaction mixtures were then diluted with DCM, washed
carefully with water and the organic phase dried over anhydrous
Na2SO4. Evaporation of DCM leaves a residue that was taken up
with acetonitrile and the resulting solution filtered on cationic and
anionic exchange resins (SUPELCO LC-SAX SPE 504815 and 504920)
to ensure complete elimination of inorganic salts. The solution
volume was then adjusted to 25 mL and analyzed through HPLC for
quantitative determinations. Yields and stereoisomeric ratios were
determined upon internal normalization.
(c) Caramella P, Marinone Albini F, Vitali D, et al. Tetrahedron Lett. 1984:
1875e1879;
ꢀ
(d) Corsaro A, Buemi G, Chiacchio U, Perrini G, Pistara V, Romeo R. Tetrahedron.
1996;52:7885e7892;
(e) Corsaro A, Chiacchio U, Pistara V, Rescifina A, Buemi G, Romeo G. J Chem Soc
ꢀ
Perkin Trans. 2000;2:1761e1766.
13. (a) Marinone Albini F, Vitali D, Oberti R, Caramella P. Allylic NHCOR directing
effects. J Chem Res (S). 1980:348;
(b) Corsaro A, Chiacchio U, Caramella P, Purrello G. J Heterocycl Chem. 1984;21:
949e952;
(c) Corsaro A, Chiacchio U, Caramella P, Purrello G. J Heterocycl Chem. 1985;22:
797e799;
(d) Curran DP, Choi SM, Gothe SA, Lin F. J Org Chem. 1990;55:3710e3712;
(e) Curran DP, Gothe SA, Choi SM. Heterocycles. 1993;35:1371e1395.
14. (a) Reichardt C. In: Solvents and Solvent Effects in Organic Chemistry. Weinheim:
VCH; 1988:392;
(b) Reichardt C. Chem Rev. 1994;94:2319e2358 (and references therein);
(c) Kamlet MJ, Taft RW. J Am Chem Soc. 1976;98:377e383;
(d) Kamlet MJ, Taft RW. J Am Chem Soc. 1976;98:2886e2894;
(e) Kamlet MJ, Abbaud JLM, Abraham MH, Taft RW. J Org Chem. 1983;48:
2877e2887.
HPLC conditions: Isocratic elution H2O/MeCN 70/30 (1.3 mL/
min); UV detection at
l
¼ 268 nm.
6.9.3. Method C
15. Quadrelli P, Fassardi V, Cardarelli A, Caramella P. Eur J Org Chem. 2002:
2058e2065.
16. Mantione D, Olaizola Aizpuru O, Memeo MG, Bovio B, Quadrelli P. Eur J Org
Chem. 2016:983e991.
17. (a) Caramella P, Reami D, Falzoni M, Quadrelli P. Tetrahedron. 1999;55:
7027e7044;
(b) Toma L, Quadrelli P, Perrini G, et al. Tetrahedron. 2000;56:4299e4309.
18. (a) Kanemasa S, Nishiuchi M, Kamimura A, Hori K. J Am Chem Soc. 1994;116:
2324e2339;
(b) Kamimura A, Kaneko Y, Ohta A, Kakehi A, Matsuda H, Kanemasa S. Tetra-
hedron Lett. 1999;40:4349e4352.
19. Kim HR, Song JH, Rhie SY, Ryu EK. Synth Commun. 1995;25:1801e1807.
20. (a) Kanemasa S, Okuda K, Yamamoto H, Kaga S. Tetrahedron Lett. 1997;38:
4095e4098;
(b) Yamamoto H, Watanabe S, Kadotami K, Hasegama M, Noguchi M,
Kanemasa S. Tetrahedron Lett. 2000;41:3131e3136.
21. (a) Kanemasa S, Oderaotoshi Y, Tanaka J, Wada E. J Am Chem Soc. 1998;120:
12355e12356;
(b) Shimizu M, Ukaji Y, Inomata K. Chem Lett. 1996:455e456;
(c) Ding X, Taniguchi K, Ukaji Y, Inomata K. Chem Lett. 2001:468e469.
22. Arnett EM, Joris L, Mitchell E, Murty TSSR, Gorrie TM, Schleyer PVR. J Am Chem
Soc. 1970;92:2365e2377.
In a 5 mL vial the dipolarophile 4A (1.2 equivs.) and 1.2 equivs. of
the M(II) salts were dissolved in anhydrous DCM (1 mL) at the
desired temperature. Benzhydroxymoyl chloride (64 mg, 1 equiv.),
dissolved in 1 mL of DCM, was added to the solution, followed by
1.2 equivs. of triethylamine and left under stirring for 1 day. The
reaction mixtures were then diluted with DCM, washed carefully
with water and the organic phase dried over anhydrous Na2SO4.
Evaporation of DCM leaves a residue that was taken up with
acetonitrile and the resulting solution filtered on cationic and
anionic exchange resins (SUPELCO LC-SAX SPE 504815 and 504920)
to ensure complete elimination of inorganic salts. The solution
volume was then adjusted to 25 mL and analyzed through HPLC for
quantitative determinations. Yields and stereoisomeric ratios were
determined upon internal normalization.
HPLC conditions: Elution gradient from H2O/CH3CN 80/20 to 60/
40 (1.3 mL/min); UV detection at
l
¼ 264 nm.
23. Rablen PR, Lockman JW, Jorgensen WL. J Phys Chem A. 1998;102:3782e3797.
24. Jeffrey GA. In: An Introduction to Hydrogen Bonding. New York: Oxford Press;
1997.
Acknowledgments
25. Dols P, Klunder A, Zwanenburg B. Tetrahedron. 1950;28:8515e8538.
26. (a) Abraham RJ, Bakke JM. Acta Chem Scand. 1983;B37:865e869;
Financial support of University of Pavia and MIUR (PRIN 2011,
Please cite this article in press as: Cardarelli AM, et al., The hydrogen bond directing effect in nitrile oxide cycloadditions to allylic substituted
cyclopentenes, Tetrahedron (2017), http://dx.doi.org/10.1016/j.tet.2017.03.042

12
A.M. Cardarelli et al. / Tetrahedron xxx (2017) 1e12
8509e8524.
(b) Bakke JM. Acta Chem Scand. 1986;B40:703e710.
27. COSMO calculations at the B3LYP/6-31g* level show only a moderate (0.2 kcal/
mol) stabilization of the outside conformer.
28. Newton MD, Jeffrey GA, Takagi S. J Am Chem Soc. 1979;101:1997e2001.
29. Goutev N, Matsuura H. J Phys Chem A. 2001;105:4741e4748.
30. Caramella P, Grünanger P. In: Padwa A, ed. Nitrile Oxides and Imines. New York:
Wiley, J and Sons Eds.; 1984.
32. Curran DP, Ghote SA, Choi SM. Heterocycles. 1993;35:1371e1395.
33. Memeo MG, Lapolla F, Bovio B, Quadrelli P. Molecules. 2014;19:8661e8678.
34. (a) Quadrelli P, Bovio B, Piccinini A, Caramella P, De Sarlo F, Machetti F. Tet-
rahedron. 2009;65:10679e10684;
(b) Scagnelli L, Memeo MG, Carosso S, Bovio B, Quadrelli P. Eur J Org Chem.
2013:3835e3846.
31. (a) Kanemasa S, Oderaotoshi Y, Sakaguchi S, et al. J Am Chem Soc. 1998;120:
3074e3088;
35. (a) Grundmann C, Grünanger P. In: The Nitrile Oxides. Heidelberg: Springer-
Verlag; 1971;
(b) Desimoni G, Faita G, Mortoni A. Righetti P P Tetrahedron Lett. 1999;40:
2001e2004;
(c) Crosignani S, Desimoni G, Faita G, et al. Tetrahedron Lett. 1999;40:
7007e7010;
(b) Grünanger P, Vita Finzi P. In: . Isoxazoles, the Chemistry of Heterocyclic
Compounds, Part 1 and 2. vol. 49. New York: Wiley; 1991 and 1999.
36. Corsico Coda A, Tacconi G. Gazz Chim Ital. 1984;114:131e132.
37. (a) Org Synth Coll, vol. IV, 239; (b) Alder K, Flok FH. Chem Ber. 1993;89:
1732e1737.
(d) Desimoni G, Faita G, Mella M, Righetti PP, Zema M. Tetrahedron. 1999;55:
Please cite this article in press as: Cardarelli AM, et al., The hydrogen bond directing effect in nitrile oxide cycloadditions to allylic substituted
cyclopentenes, Tetrahedron (2017), http://dx.doi.org/10.1016/j.tet.2017.03.042