The stereochemistry of isoquinoline Reissert compounds: A unique platform for observation of steric and electronic interactions

Article abstract of DOI:10.1016/j.tet.2012.06.008

Isoquinoline Reissert compounds (2-acyl-1,2-dihydroisoquinaldonitriles) with either 3-H (1) or 3-CH₃ (2) substituents and various N-acyl groups have been examined in detail by ¹H and ¹³C NMR spectroscopy and X-ray crystall

Full text of DOI:10.1016/j.tet.2012.06.008

Tetrahedron 68 (2012) 8052e8067
Contents lists available at SciVerse ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
The stereochemistry of isoquinoline Reissert compounds: a unique platform
for observation of steric and electronic interactions
y
*
Harry W. Gibson , Michael A.G. Berg, Terry L. Price, Jr., Zhenbin Niu, Minjae Lee ,
Mason A. Rouser, Jennifer Clifton Dickson z, Carla Slebodnick
Department of Chemistry, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061-0212, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Isoquinoline Reissert compounds (2-acyl-1,2-dihydroisoquinaldonitriles) with either 3-H (1) or 3-CH3 (2)
substituents and various N-acyl groups have been examined in detail by 1H and 13C NMR spectroscopy
and X-ray crystallography. In all cases the trans amide conformation, with reference to the carbonyl
oxygen and the 3-position of the isoquinoline ring, predominates in solution. In the solid state the nitrile
moieties are pseudo-axial and the amides exist almost exclusively in the trans form, except for the case of
2-isobutyryl-3-methyl-1,2-dihydroisoquinaldonitrile (2c), which exists exclusively as the cis amide form
in the solid state. In N-aroyl 3-CH3 compounds with two ortho-aroyl substituents both amide isomerism
and hindered aryl/carbonyl rotation are observed by 1H NMR spectroscopy. In other N-aroyl derivatives
only hindered aryl/carbonyl rotations are observed by NMR and in N-alkanoyl compounds amide
isomerism is observable only at very low temperatures. X-ray crystallography reveals the two rotamers in
the solid state in four cases of ortho-substituted benzoyl compounds; with one exception, the rotamer
with the larger ortho-aroyl substituent syn to the pseudo-axial cyano group is favored. Unusual solubility
and reactivity patterns observed with these compounds are rationalized in terms of the interplay be-
tween steric and electronic factors.
Received 19 April 2012
Received in revised form 2 June 2012
Accepted 4 June 2012
Available online 20 June 2012
This paper is dedicated to the memory of
Professor Frank D. Popp, who pioneered the
synthesis, study and use of Reissert com-
pounds, and to the memory of Professor
Ernest L. Eliel, an exemplary mentor in
stereochemistry
Keywords:
Isoquinolines
Reissert compounds
Stereochemistry
Amide isomerism
Atropisomerism
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
hydroxyalkyl)isoquinoline forms. The anions likewise react with
alkyl halides to produce derivatives such as 3 and 4. Via these
processes Reissert compounds have been employed in syntheses of
Reissert compounds are
a
-(acylamino)nitriles formed by the
novel monomers and polymers,7 and racemic isoquinoline alka-
loids.8e12 Their application in stereoselective syntheses, an ongoing
challenge,13 has now been reported,14e17 portending their in-
creasing importance in this field. In spite of their wide synthetic
application, the stereochemistry of the Reissert compounds
themselves has received only scant attention.17,18 Our interest in the
stereochemistry of 1-alkyl derivatives, e.g., 3 and 4,19 and the ne-
cessity of synthesizing a series of new Reissert compounds20,21 for
that effort led to this investigation of the stereochemistry of these
parent compounds.
formal addition of acyl nitriles to imine bonds via N-acylium ions.1
Isoquinoline Reissert compounds, 2-acyl-1-cyano-1,2-dihydroiso-
quinolines, 1 and 2, for example, are prepared by reaction of the
isoquinolines and acid chlorides in the presence of a cyanide ion
source.2e4 Aromatic and aliphatic acid chlorides and a variety of
heterocycles and acyclic imines may be employed to make Reissert
compounds.1,5 Proton H1 of 1 and 2 is acidic and may be abstracted
by a variety of bases and the resultant anions are excellent nucle-
ophiles that undergo a number of reactions with proven utility
in elaboration of nitrogen heterocycles.1,6 When an isoquinoline
Reissert anion is treated with an aldehyde, the ester of a 1-(a-
2. Results and discussion
It is known that the spatial requirement of the cyano group
is somewhat greater than that of hydrogen.22,23 In analogous
1,2-dihydronaphthalene systems, the conformationally larger
1-substituent, cyano, occupies the pseudo-axial position.24,25 Thus,
in Reissert compounds 1 and 2 the cyano moiety occupies the
* Corresponding author. Tel.: þ1 540 231 5902; e-mail address: hwgibson@vt.edu
(H.W. Gibson).
y
Current address: Samsung Total Petrochemicals, Co. Ltd, Seosan-si, Korea.
Current address: Wikoff Color Corp., 1886 Merritt Rd., Fort Mill, SC 29715-7707,
z
United States.
0040-4020/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tet.2012.06.008

H.W. Gibson et al. / Tetrahedron 68 (2012) 8052e8067
8053
H4
H4
H3
CH3
N COR
N COR
H1 CN
H1 CN
1
2
g.
h.
i.
R =o-C6H4CH3
a.
b.
c.
d.
e.
f.
R = CH3
m.
n.
o.
p.
q.
r.
R = -C10H7
R =m-C6H4CH3
R =p-C6H4CH3
R =o-C6H4Cl
R =p-C6H4Cl
R =o-C6H4F
R = C2H5
R = 2,6-C6H3(CH3)2
R = CH(CH3)2
R = (CH2)3Cl
R = (CH2)2Cl
R = 2,6-C6H3(OCH )
3
j.
R = 2,6-C6H3FCl
R = 2,6-C6H3ClCH3
R = OC2H5
k.
l.
R =C
6H5
s.
R = OCH2C6H5
Table 1
1H NMR spectra of 2-acyl-1,2-dihydroisoquinaldonitriles (1) in CDCl3 at 40 ꢁC (60 or
H4
80 MHz) unless otherwise noted
H3
N COR
CH CN
Cpd.
R
H1 (ppm)
H3 (ppm)
H4 (ppm)
J1e3 (Hz)
1a
CH3
C2H5
6.70
6.79
6.87
6.82
6.63
6.63g
6.37g
6.61
6.85
6.62
6.88
6.76
6.87
6.97
6.95
6.74
6.60
5.98
6.73
6.35
6.68
6.30
6.09
6.15
6.20
6.19
6.15
6.11
5.57
6.10
6.05
6.15
5.88
1
<1
0.6
1
0.8
1.0
1.0
d
1
1.2
d
d
d
0.8
1
1b
1ca
1d
CH(CH3)2
(CH2)3Cl
(CH2)2Cl
C6H5
o-C6H4CH3
p-C6H4CH3
o-C6H4Cl
p-C6H4Cl
CH3
CH3
1ef
1ff,b
1g
3
1i
1j
1k
g. R=o-C6H4CH3
1mh,c
1oi,d
1r
a-C10H7
H4
2,6-C6H3(OCH3)2 5.67, 6.84j 6.38,j 6.68 5.86,j 6.31
OC2H5
OCH2C6H5
d
d
d
6.43
6.43
6.61
6.27
d
7.03
7.03
6.61
6.80
7.37
6.07
6.08
6.08
6.16
6.27
CH3
1s
6f,k,l
7e
1
d
N COR
10: R¼H
CH CN
CH3
CH3
a
See Supplementary data Fig. S1.
See Supplementary data Fig. S2.
See Supplementary data Fig. S5.
See Supplementary data Fig. S3a.
See Supplementary data Fig. S17.
In CD3COOD.
b
c
d
e
f
4
f. R=C6H5
j. R=o-C6H4Cl
m. R=o-C10H7
g
h
i
Assignment verified by deuterium labeling, i.e., by taking the spectrum of analog 8.
400 MHz, 22 ꢁC.
500 MHz, 22 ꢁC; methoxy signals at 3.67 (major), 3.70 (minor), 3.86 (major), and
3.89 (minor) ppm.
j
Major signal.
25 ꢁC.
k
pseudo-axial position predominantly, if not exclusively, at room
temperature, consistent26 with the w1 Hz coupling constants be-
tween the 1- and 3-protons of compounds 1 (Table 1), in agreement
with previous work.18 Reissert compounds are tertiary amides and
as such display amide isomerism, defined in Scheme 1 as s-cis and
s-trans with respect to the R and H1 moieties.27 These isomers arise
from the partial double bond character of the NeC(O) bond of the
amide via resonance contribution from Nþ]CeOꢀ.28,29 Addition-
ally in compounds with unsymmetrically substituted N-aroyl
moieties, there is hindered rotation around the aryl/carbonyl bond
as shown in Scheme 1, yielding a total of four possible isomers, as
known with other aromatic amides.28e31 The interactions between
the 3-H or 3-CH3 substituent and the substituents on the nitrogen
atom may be considered in the sense of the classical allylic or A1e3
strain,32 and indeed these provide the steric driving forces that
seem to control the stereochemistry of these compounds to great
extent. These isomerization processes are elucidated in this work.
l
Reported in CDCl3: H1: 6.61 ppm, H3: 6.65 ppm, H4: 6.11 ppm, J1e3¼1 Hz (Knabe,
J.; Frie, A. Arch. Pharm. 1973, 306, 648e658).
2.1. Ambient temperature 1H NMR spectra
The 1H NMR spectra of the 2-acyl-1,2-dihydroisoquinaldoni-
triles (1 and 2) are summarized in Tables 1 and 2. The spectra of 68,9
and 733 (see Supplementary data Fig. S17 for the latter) are also
given in Table 1. The signals due to H1 were identified directly by
substitution of deuterium for hydrogen; treatment of compound 1
or 2 with sodium hydride in dimethylformamide in the presence of
carbon disulfide leads to formation of the orange dithiocarbamate
anion (Scheme 2), which is unstable in water and reverts to starting
1.34 By quenching the anion with D2O, 1-deutero-2-acyl-1,2-
dihydroisoquinaldonitriles (8 and 9) were prepared.35 For exam-
ple, from 1f, 8f was isolated in 92% yield. The signal present at
6.63 ppm (CDCl3) in the spectrum of 1f is absent in that of 8f, while

8054
H.W. Gibson et al. / Tetrahedron 68 (2012) 8052e8067
Table 2
R3
Y
R3
X
1H NMR spectra of 2-acyl-3-methyl-1,2-dihydroisoquinaldonitriles (2) in CDCl3 at
40 ꢁC (60 or 80 MHz) unless otherwise noted
N
N
Cpd.
R
H1 (ppm)
3-CH3 (ppm) H4 (ppm)
X
H
CN
O
Y
H
CN
O
2a
CH3
6.75
2.31
6.35
5ta
2b
C2H5
6.82
2.36
2.40
2.32
1.81
1.72
1.80
1.85
1.82 (br)
1.85
6.44
6.47
6.47
6.47
6.20
6.15
6.33
6.30
2ca
2d
CH(CH3)2
(CH2)3Cl
C6H5
o-C6H4CH3
m-C6H4CH3
p-C6H4CH3
o-C6H4Cl
p-C6H4Cl
o-C6H4F
6.72j
5ts
X > Y
6.72
2f
6.53
6.65j
R3
2g
R3
2hk
2i
6.43
6.60
6.60 (br)
6.60
6.31
O
O
N
Y
2jb
N
X
H
CN
2k
6.36
6.01
6.20
H
CN
Y
2lk,c
2ml
2nk,d
2ok,e
2pk,n,f
1.80
1.58
X
a
-C10H7
6.75 (br)
2,6-C6H3(CH3)2
5.40,j 7.05j,m
1.52,m 2.52
1.68,m 2.55
1.72, 1.74,m
2.54, 2.56
1.58,m 1.69,
2.58, 2.60
2.34
6.15,m 6.37
6.17,m 6.33
6.27, 6.29,m
6.42, 6.45
6.25,m,o 6.41,
6.46
5cs
5ca
2,6-C6H3(OCH3)2 5.59,j 6.99j,m
2,6-C6H3FCl
5.48, 5.50,j
6.94, 6.96j,m
5.41, 5.52,j
7.09, 7.06j,m
6.43
Scheme 1. Isomers resulting from hindered rotation about the amide and carbonyl/
aryl bonds: t¼trans amide; c¼cis amide; s¼larger group X syn to CN; a¼larger group X
anti relative to the cyano moiety.
2qk,n,g
2,6-C6H3ClCH3
2rh
OC2H5
OCH2C6H5
d
6.17
6.17
6.28
all other signals are at the same positions (see Supplementary data
Fig. S2). Likewise 1g affords 100% of 8g, which lacks the 6.55 ppm
signal due to H1 of 1g. It is noteworthy that signals for H1 and 3-CH3
are broadened in the spectrum of 2j.
2si
6.44
d
2.31
2.36
10: R¼CH3
a
See Supplementary data Fig. S4.
b
c
For 600 MHz spectra at 25, ꢀ15, and ꢀ50 ꢁC, see Supplementary data Fig. S18.
For 600 MHz spectra at 25, ꢀ15, and ꢀ50 ꢁC, see Supplementary data Fig. S19.
See Supplementary data Fig. S9 for the 60 MHz spectrum and Fig. 2 for the
O
d
S
600 MHz spectrum.
e
N COC6H5
See Fig. 3.
See Fig. 4.
See Fig. 5.
N P(OCH3)2
f
O
g
H
CN
H
CN
h
500 MHz spectrum@22 ꢁC; H4 signal identified by COSY correlation with 3-CH3;
see Supplementary data Fig. S10.
6
7
i
500 MHz spectrum@22 ꢁC; H4 signal identified by COSY correlation with 3-CH3;
see Supplementary data Fig. S11.
j
Assignment verified by deuterium labeling, i.e., by taking the spectrum of the
analogous compound 9.
2.2. Low temperature spectra
k
At 25 ꢁC.
l
600 MHz at 20 ꢁC; see Supplementary data Figs. S6eS8.
Major signal.
500 MHz at 20 ꢁC; see Fig. 5.
The 1H NMR (80 MHz) spectra of the following compounds were
examined in CDCl3 solution down to ꢀ46 ꢁC and underwent no
significant changes: 1a, 1c, 1f, 1k. Likewise, compound 2f did not
exhibit any signal doubling, but significant peak broadening did
occur. Compound 1i showed no broadening or doubling of signals
down to ꢀ62 ꢁC in acetone-d6. Again the 3-methyl analog, 2i, while
not exhibiting signal doubling, did undergo significant broadening
of the 3-methyl signal at ꢀ62 ꢁC in acetone-d6 (full width at half
height 6.0 Hz). See Supplementary data Table S1. Not all com-
pounds were examined at low temperatures, only the ones in-
dicated here and in the following discussion.
However, the spectra of the following compounds revealed the
presence of more than one stereoisomeric form as evidenced by
doubling of NMR signals for certain protons: 2a, 2c, 1g, 2g, 2j, 2l,
2m, 2n, 2o, and 2q. The changes revealed in the spectra of 1g and 2g
as a function of temperature (Fig. 1) are typical of some of the
changes observed (see Supplementary data Table S1).
m
n
o
Two coincident signals.
undergo a small downfield (ꢀ) shift. Due to limited solubility of
many of these compounds in CCl4, CDCl3 was used; it was dem-
onstrated that the resultant differences are small and un-
ambiguous, the chemical shifts in CDCl3 being slightly upfield of
those in CCl4 (see Supplementary data Table S2, compounds 1c, 2c).
ASIS were determined for several of the compounds at low tem-
perature using chlorobenzene (C6H5Cl), or toluene-d8 as the aro-
matic solvent so that solvent freezing was not a problem. The shifts
were then those associated with the changes from CDCl3 to C6H5Cl
or toluene-d8 (see Supplementary data Table S2). Included are re-
sults for compounds 10, which can only exist in the cis amide
configuration; as a result the 3-H or 3-CH3 (R) signal undergoes
the expected small shift, while the methylene protons
a to the
1-position and analogous to H1 of 1 and 2 undergo large upfield
shifts. In each instance the only or major signal for R3 of the Reissert
compounds 1 and 2 undergoes a large upfield shift; this means that
the trans amides dominate in each case. This was previously ob-
served with benzothiazole Reissert compounds.5d
2.3. Determination of amide configuration
A diagnostic test for amide configuration is provided by aro-
matic solvent induced shifts (ASIS).36 This technique involves
comparison of chemical shifts of the compound when the solvent is
changed from carbon tetrachloride to benzene. Groups s-trans to
the carbonyl oxygen undergo large upfield (þ) shifts; those s-cis
undergo small downfield (ꢀ) shifts. These differential shifts origi-
nate in the anisotropy resulting from complexation of benzene to
the carbonyl moiety.36 In 5c (Scheme 1) changing from carbon
tetrachloride to benzene should cause a large upfield (þ) shift in H1
and a small downfield (ꢀ) in R3 and H4. Conversely in 5t, R3 and H4
are predicted to shift upfield (þ) significantly and H1 is expected to
R
N
O
NC
10

H.W. Gibson et al. / Tetrahedron 68 (2012) 8052e8067
8055
R3
CH3
NaH/DMF
CS2
D2O
OR
N COR
S C CN
N COR
CN
1 or 2
N COR
CN
D
D
8
S
9
c. R = CH(CH3)2
g. R = o-C6H4CH3
m. R = α-C10H7
f. R = C6H5
g. R = o-C6H4CH3
n. R = 2,6-C6H3(CH3)2
o. R = 2,6-C6H3(OCH3)2
q. R = 2,6-C6H3ClCH3
Scheme 2. Selective deuteration of isoquinoline Reissert compounds via reactions of their anions with CS2 and then D2O.
As noted above a number of the N-aroyl compounds exhibit
both amide and atropisomers; these will now be discussed in detail.
The new 20,60-dimethylbenzoyl-3-methyl compound 2n affords
a good reference point for comparison of chemical shift data for the
four possible individual isomers in ortho-substituted N-aroyl
compounds. The presence of six methyl signals in the room tem-
perature NMR spectrum of 2n (Fig. 2, Table 2, Scheme 3) indicated
that there were two slow isomerization processes involved. The
presence of four signals (1.7e2.6 ppm) for the N-aroyl methyl
groups is due to restricted rotation about the aryl-carbonyl bond
coupled with amide isomerism. The two signals for the 3-methyl
group (identified by the 1 Hz coupling to H4) and two signals for
H4 are attributed to slow amide isomerization. Accordingly, H1 also
displays two signals identified by examination of the deutero an-
alog 9n (Fig. 2 inset), the major one at 7.08 ppm, the minor at
5.46 ppm. From integration of the 3-methyl and ortho-methyl sig-
nals in the 1H NMR spectrum the trans/cis isomer ratio was found
to be 71:29(ꢂ1). The 13C NMR spectrum of 2n (see Supplementary
data Fig. S12) corroborates these results. Two signals at 44.58 and
49.62 ppm in intensity ratio 72:28, respectively, are assigned to the
C1 carbon of the isoquinoline ring. Matching these are two signals
at 116.97 and 117.64 ppm in intensity ratio 72:28, respectively,
assigned to the cyano carbons.
Clearly from the data for 2n, amide isomerism leads to large
chemical shift differences for the 3-substituent (R3) (1 ppm) and H1
(w1.6 ppm). However, only relatively small differences, w0.1 ppm,
are observed between syn methyl signals in the cis and trans amide
isomers, and likewise for the anti-methyl groups (Scheme 3).
However, the syn and anti positions lie in greatly different envi-
ronments; for example, the chemical shifts of the two ortho-methyl
groups of 2n in the major trans amide forms vary by w0.6 ppm
(Scheme 3).
Fig. 1. Methyl region of the 1H NMR spectra (60 MHz) of (a) (left) 2-o-toluyl-1,2-
dihydroisoquinaldonitrile (1g) and (b) (right) 2-o-toluyl-3-methyl-1,2-
dihydroisoquinaldonitrile (2g) in CDCl3 as a function of temperature.
2.4. Identification of isomerization processes with selected
compounds
Molecular modeling indicates that the aryl ring of this and
similar Reissert compounds cannot be coplanar with the carbonyl
group; instead, it is rotated out of plane in such a way that the ortho
substituents are projected above and below the plane of the iso-
quinoline nucleus as indicated in structures 5. Because of the tilt of
the aroyl moiety, the syn-substituent is tilted away from the cyano
group and is less hindered than the anti position, which lies under
the C3 and C4 positions of the isoquinolyl moiety. Thus, the syn-
substituent experiences a deshielding effect from the nearby car-
bonyl group, while the anti-substituent is shielded by the iso-
quinolyl nucleus. The solid state structure derived from X-ray
crystallography, discussed below, is consistent with this analysis.
Turning now to the new 20,60-dimethoxybenzoyl-3-methyl
compound 2o, in CDCl3 two signals are observed for the 3-methyl
group (Fig. 3, Table 2, Scheme 4), along with four methoxy signals
and two signals for H1, identified by preparation of 9o. These results
are again attributed to the combination of hindered aryl/carbonyl
rotation and cis/trans amide isomerism, 5t being the major (55%)
Compounds 1a, 1c, 1f, 1i, and 1k at low temperature show no
significant broadening or doubling of signals, indicating that for all
five compounds the rate of interconversion of cis and trans amide
isomers is rapid and in 1f, 1i, and 1k rotation about the aryl/car-
bonyl bond is also rapid. In the 3-methyl series the simple benzoyl
compound 2f and the para-toluyl analog 2i did not exhibit signal
doubling, only peak broadening; in these compounds the barriers
to rotation about the aryl/carbonyl and amide bonds are such as to
be not observable at the lowest temperature accessed. This is
expected, since the N-aroyl groups in these compounds are not
ortho-substituted.
In considering the N-alkanoyl compounds the only possible
observable dynamic process is amide isomerism. In 2a and 2c signal
doubling is observed for 3-CH3, H1, H4, and 2-alkanoyl groups at
low temperatures and the trans amide isomers predominate (see
Supplementary data Table S1).

8056
H.W. Gibson et al. / Tetrahedron 68 (2012) 8052e8067
1
Fig. 2. H NMR spectrum of 2-(20,60-dimethylbenzoyl)-3-methyl-1,2-dihydroisoquinaldonitrile (2n) in CDCl3 at 22 ꢁC; the signals marked with asterisks (*) are due to ethyl acetate.
Inset: partial spectrum of the 1-deuterio derivative 9n; the arrows indicate the chemical shifts of the two H1 signals in 2n.
2.38 ppm
Supplementary data Fig. S3a); amide isomers with hindered
arylecarbonyl rotation were present in an 86:14 ratio in this case as
proven by four signals for the methoxy protons, and two signals
each for H1, H3, and H4. The 13C NMR spectrum (see Supplementary
data Fig. S3b) confirmed the presence of two carbonyl peaks (164.7
and 165.8 ppm) with a height ratio of 1:9 and likewise four aro-
matic C(O) signals (156e158 ppm) in 1:1:9:9 peak height ratios, as
well as other signals with the same relative intensities. Assign-
ments were verified by 1H COSY (correlation spectroscopy, see
Supplementary data Fig. S3a) and HMBC (heteronuclear multiple
bond correlation, see Supplementary data Fig. S3c) experiments.
Another 20,60-disubstituted aroyl compound, the new 20-chloro-
60-fluorobenzoyl-3-methyl derivative 2p displays four 3-methyl
signals in its 500 MHz 1H NMR spectrum (Fig. 4, Table 2,
Scheme 5), indicating the presence of all four isomers. The chemical
shifts of the diagnostic 3-methyl group fall in line with the results for
2n; in the cis isomer it appears at w2.5 ppm and in the trans at
w1.7 ppm. The trans isomer predominates in a 79:21 ratio. From the
four signals each for the 3-methyl protons, H1 and H4 (Table 2), the
average ratio of the major to minor rotamers in the trans isomer was
56:44, as compared to 55:45 in the cis isomer (values good to ꢂ2%).
In the 13C NMR spectrum of 2p (see Supplementary data Fig. S14)
there are two peaks for the 3-methyl group (20.76 and 21.04 ppm),
two for C1 (45.15 and 49.69 ppm) and at least six signals in the range
155e162 ppm; the latter should include two carbonyl peaks and
2.56 ppm 2.48 ppm
1.50 ppm
Slow
@
CH3
CH3
CH3
O
N
N
CH3
+22 oC
CH3
1.73 ppm
H
CN
H
CN
O
7.08
CH
5.46
2n: 29% (cis)
31.93 ppm
2n: 71% (trans)
Scheme 3. 1H NMR results (400 MHz, CDCl3) for 2n.
amide isomer. The 13C NMR spectrum of 2o (see Supplementary
data Fig. S13) corroborates the 1H NMR results. Signals at 45.07
and 49.56 ppm in intensity ratio 55:45, respectively, are assigned to
the C1 carbon of the isoquinoline ring. Matching these are two
signals at 116.12 and 116.37 ppm in intensity ratio 53:47, re-
spectively, assigned to the cyano carbons. Likewise the two car-
bonyl signals at 164.97 and 165.46 have a similar intensity ratio.
There are two signals for the 3-methyl carbons at 20.49 and
20.59 ppm (intensity ratio 45:55, respectively) and four signals for
the methoxy carbons at 55.37, 55.82, 55.96, and 55.97 (intensity
ratios 25:20:27:28, respectively).
The only 3-H compound that displayed isomerism at room
temperature was the new dimethoxy analog 1o (Table 1; also see

H.W. Gibson et al. / Tetrahedron 68 (2012) 8052e8067
8057
1
Fig. 3. H NMR spectrum (400 MHz) of 2-(20,60-dimethoxybenzoyl)-3-methyl-1,2-dihydroisoquinaldonitrile (2o) in CDCl3 at 22 ꢁC.
four doublets for CeF carbons with JCeF w250 Hz; however, only
two such doublets are clearly observed, corresponding to major
isomers 5ta and 5ts. In hopes of a clearer picture, we obtained the
19F NMR spectrum (see Supplementary data Fig. S15), which as
expected contains four signals (at ꢀ113.43, ꢀ112.18, ꢀ111.99,
and ꢀ111.52 ppm) in 37:10:11:42 proportion, in rough agreement
with the proton NMR spectrum (Fig. 4) whose peaks overlapped to
some extent, making precise determination of relative proportions
difficult. The most upfield (ꢀ113.43 ppm) 19F signal constituted 37%
and was assigned to F anti to the CN (5ts stereoisomer) based on the
chemical shifts of the methyl protons of 2g and 2n. In the more
Scheme 4. 1H NMR results (400 MHz, CDCl3) for 2o.
1
Fig. 4. H NMR spectrum (400 MHz) of 2-(20-chloro-60-fluorobenzoyl)-3-methyl-1,2-isoquinaldonitrile (2p) in CDCl3 at 22 ꢁC.

8058
H.W. Gibson et al. / Tetrahedron 68 (2012) 8052e8067
2.56 ppm
2.54 ppm
1.74, -113.43 ppm
1.72 ppm
CH3
CH3
CH3
Slow
Slow
CH3
F
Slow
O
Cl
O
N
N
F
Cl
N
CN
N
@
@
@
F
H
CN
Cl
H
CN
Cl
H
O
+22 oC
+22 oC
+22 oC
H
CN
O
F
5.48 -112.18 ppm
5.50 -111.99 ppm
6.94 ppm
2p: 37% (5ts)
6.96 -111.51 ppm
2p: 10% (5cs)
2p: 11% (5ca)
2p: 42% (5ta)
Scheme 5. 1H NMR (500 MHz, CDCl3) and 19F (376 MHz, CDCl3) results for 2p.
abundant trans amide isomer the signal was the most downfield
(ꢀ111.52 ppm) and, therefore, assigned to F syn to the cyano group,
i.e., isomer 5ta. In the cis amide the F signal of the major (11%)
rotational conformer was the downfield one (ꢀ111.99 ppm), corre-
sponding to F syn to CN (5ca). The minor (10%) cis amide isomer with
the F signal upfield (ꢀ112.18 ppm) thus corresponds to 5cs, having
the F anti to the CN. The slight preponderance of the 5ta isomer over
the 5ts isomer may be due to the higher dipole associated with the
CeCl versus CeF bond;37 having this dipole oppose the pseudo-axial
C1eCN dipole may lead to greater stability in the non-polar solvent.
It is noteworthy, though, that in the crystalline state the ts isomer
predominates greatly, as expected; see below.
These results prompted us to re-examine the ortho-halo com-
pounds 2j and 2l at low temperatures at 600 MHz. Although a de-
tailed study was not carried out in either case, at ꢀ50 ꢁC the
presence of both amide isomers and rotamers was detected. In line
with the results discussed above for 2p, at that temperature the
ortho-chloro compound 2j displayed two resolved signals at 1.57
and 1.65 (major, total 93%) and another at 2.56 ppm (minor, 7%) for
the 3-methyl signals of the trans and cis amides, respectively, along
with definite signals for individual rotamers in each of these; that
is, similar to the room temperature spectrum of 2p there are four
clear signals each for H1 and H4 (see Supplementary data Fig. S18),
indicative of all four isomers of 2j in a ratio of 58:35:5:2 for 5ts/5ta/
5cs/5ca (X¼Cl>Y¼H). Similarly at ꢀ50 ꢁC the ortho-fluoro com-
pound 2l displayed two broad signals at 1.57 and 2.49 ppm for the
3-methyl protons of the cis and trans amides, along with definite
signals for individual rotamers in that there are four clear signals
each for H1 and H4 (see Supplementary data Fig. S19), indicative of
all four isomers in a ratio of 49:41:9:1 for 5ts/5ta/5cs/5ca
(X¼H>Y¼F).
A fourth new 2,6-disubstituted aroyl compound, 20-chloro-60-
methylbenzoyl-3-methyl derivative 2q, displays four 3-methyl
signals and four signals for H1 and three for H4 (Fig. 5, Table 2,
Scheme 6). The H1 signal assignments were verified by deuteration,
producing 9q. The chemical shifts of the 3-methyl group are con-
sistent with the above results, the upfield signals at w1.6e1.7 ppm
denoting the trans amide and w2.6 ppm indicating the cis isomer.
Moreover, the chemical shifts of the methyl group on the benzoyl
ring are clearly indicative of its disposition relative to the cyano
group: w2.5 ppm if syn and 1.8e2.0 ppm if anti, in accord with the
results for the methyl groups in 2n. The trans isomer predominates
in a 71:29 (ꢂ1) ratio. Analysis of these signals indicates that the
ratio of the major to minor rotamers in both the cis and trans amide
isomers was 52:48 (values good to ꢂ1%). The larger chloro sub-
tituent38 preferentially occupies the syn position versus the methyl
substituent, in agreement with other results. The 13C NMR spec-
trum of 2q (see Supplementary data Fig. S16) corroborates the 1H
NMR results: four signals each for the 3-methyl group, C1 and the
benzoyl methyl group; analysis of their intensities in each case
affords the following isomer ratios: 14:15:34:37 in excellent
agreement with the 1H results. This conclusion is consistent with
the X-ray crystallographic results below.
The ortho-toluyl benzothiazole Reissert compound displays
similar chemical shifts for the tolyl methyl group in the four iso-
mers (1.87, 2.40, and 2.53, the latter overlapped), but the assign-
ments made earlier for this compound are wrong;5d in that case as
in the present isoquinoline analogs the larger ortho substituent
1
Fig. 5. H NMR spectrum (500 MHz) of 2-(20-chloro-60-methylbenzoyl)-3-methyl-1,2-isoquinaldonitrile (2q) in CDCl3 at 22 ꢁC. Inset: downfield region of deutero analog 9q.

H.W. Gibson et al. / Tetrahedron 68 (2012) 8052e8067
8059
Slow
@
Slow
@
2.60 ppm
2.58 ppm
1.69 ppm
1.58 ppm
Slow
@
CH3
CH3
CH3
CH3
Cl
CH3
O
N
O
N
CH3
Cl
+22 oC
+22 oC
N
N
+22 oC
H
CN
O
H
CN
Cl
H
CN
Cl
CH3
H
CN
O
CH3
5.41 2.02 ppm
5.52 2.53 ppm
7.06 1.80 ppm
2q: 37% (5ts)
7. 09 2.46 ppm
2q: 15% (5cs)
2q: 14% (5ca)
2q: 34% (5ta)
Scheme 6. 1H NMR results (500 MHz, CDCl3) for 2q.
(CH3) is syn to the cyano group in the major aryl/carbonyl isomers
of both amide isomers.
above discussion, indicate that 1g exists in the trans amide form ex-
clusively, but undergoes slow arylecarbonyl rotation in the low
temperature range examined. The same conclusion may be drawn for
2g: only the trans amide exists in solution; in fact the chemical shift
difference for the tolyl methyls (0.6 ppm, Fig. 1b, Scheme 7) is about
the same as for trans-2n (Scheme 3). These conclusions are supported
by the low temperature results, which show the same sign of ASIS for
each of the doubled signals for the H3 in 1g and the 3-methyl moiety
(R3) in 2g (see Supplementary data Table S2).
For naphthoyl analog 2m at room temperature only broad signals
are observed (See Supplementary data Fig. S6); however, at low
temperature four distinct isomers are detected, although the per-
centages of the cis amide isomers are quite low (Fig. 6, Scheme 8).
Again the chemical shifts of the 3-methyl and H1 moieties are di-
agnostic. The 3-methyl and H4 signals were examined via a 600 MHz
COSYexperiment at ꢀ34 ꢁC (see Supplementary data Fig. S7), taking
advantage of the w1 Hz allylic coupling between them to make the
assignments. The H1 signals were also verified at ꢀ34 ꢁC by com-
parison to the deutero analog 9m (see Supplementary data Fig. S8).
The predominance of 5ts isomer of 2m is consistent with its X-ray
structure (see below).
In contrast to the behavior of these compounds, at low tempera-
ture (ꢀ46 ꢁC) the doubled signals for H3 and H4 of 1g are only slightly
different (0.03e0.06 ppm) and H1 is only broadened (see
Supplementary data Table S1); moreover, the two tolyl methyl signals
vary by only 0.32 ppm (Fig.1a, Scheme 7). These results, in light of the
Slow
@
2.32 ppm
CH3
H
H
-32 oC
N
N
H
CN
O
CH3
H
CN
O
2.00 ppm
1g: 55% (5ts)
1.55 ppm
1g: 45% (5ta)
1.50 ppm
CH3
Slow
@
2.50 ppm
CH3
CH3
H
-37 oC
N
N
H
H
CN
Based on the chemical shifts of the 3-methyl groups in urethane
Reissert compounds 2r and 2s in comparison to those of com-
pounds 2m, 2n, 2o, 2p, 2q, and 10 it is apparent that these com-
pounds both exist predominantly in the form of the cis isomers
around the C(O)eN bond. Interestingly in 2r the methylene protons
O
CH3
H
CN
O
1.88 ppm
2g: 31% (5ta)
2g: 69% (5ts)
Scheme 7. 1H NMR results (80 MHz, CDCl3) for 1g and analog 2g.
1
Fig. 6. H NMR spectrum (600 MHz) of 2m at ꢀ34 ꢁC in CDCl3. H4 signals distinguished by COSY experiments to correlate with the 3-CH3 signals via the w1 Hz allylic coupling (see
Supplementary data Fig. S7). H1 signals verified with the deuterated analog 9m at ꢀ34 ꢁC in CDCl3 (see Supplementary data Fig. S8).

8060
H.W. Gibson et al. / Tetrahedron 68 (2012) 8052e8067
1.50 ppm
1.23 ppm
2.59 ppm
2.71 ppm
Slow
@
Slow
@
Slow
@
CH3
CH3
CH3
CH3
O
O
N
N
N
N
-34 oC
-34 oC
-34 oC
H
CN
H
CN
H
CN
O
H
CN
O
5.65
2m: 5% (5cs)
ppm
5.41 ppm
6.95 ppm
2m: 75% (5ts)
7.13 ppm
2m: 17% (5ta)
2m: 2% (5ca)
Scheme 8. 1H NMR results (600 MHz, CDCl3) for 2m.
are diastereotopic as revealed by their classic ABX3 signature (see
Supplementary data Fig. S10); they differ in chemical shift by
0.07 ppm. Likewise the diastereotopic benzylic methylene protons
of 2s display a classic AB quartet and differ by 0.07 ppm (see
Supplementary data Fig. S11).
The free energy differences (
the various Reissert compounds were determined from the equi-
librium constants evaluated by NMR. G values range from near
DG) between the stereoisomers of
D
zero to 1.2 kcal/mol (see Supplementary data Table S1).
2.5. Single crystal X-ray structures
Single crystals of some of the Reissert compounds were examined
by X-ray crystallography. Geometric factors derived from these
analyses are gathered in Table 3. In general these results confirm that
the trans amide isomers are preferentially formed in the solid state,
that the rotamers with the larger ortho substituent syn to the 1-cyano
group are favored and that the cyano moiety occupies a pseudo-axial
position (dihedral angles with respect to the benzo plane: 82e96ꢁ).
The ortho-toluyl Reissert compound 1g exists in the solid state
entirely as the trans amide, as also determined from the solution
NMR studies; the amide linkage is very nearly planar, only out by
0.4ꢁ. There are, however, two rotamers present in a ratio of 78:22
(Fig. 7). However, unlike the structure determined in solution (55%
ts/45% ta), in the crystalline phase the rotamer with the ortho-
methyl group anti to the cyano moiety, i.e., 5ta, is favored. The other
noteworthy aspect of the solid state structure of the major rotamer
is that the aryl/carbonyl dihedral angle (48ꢁ) is the lowest of any of
the substituted N-aroyl Reissert compounds examined in the
present work (Table 3). It is known that in some cases packing ef-
fects can cause solid state structures to differ from solution struc-
tures. This compound seems to fall into this category.
Fig. 7. Two views of one enantiomer from the X-ray crystal structure of 2-(o-toluyl)-
1,2-dihydroisoquinaldonitrile (1g) showing the pseudo-axial location of the cyano
moiety, the trans orientation of the amide linkage and the twisted disposition of the
tolyl group. In the major [78.2 (ꢂ0.4)%] rotamer (top views) the ortho-methyl group is
anti to the cyano group (5ta), while in the minor [21.8 (ꢂ0.4)%] isomer (bottom views)
ꢀ
it is syn to the cyano group (5ts). The 3-proton is 3.43 A away from the centroid of the
ꢀ
tolyl ring in the major rotamer and 3.15 A away in the minor rotamer.
This result can be compared to the structure of the analogous
3-methyl compound 2g, which exists solely in the trans amide
form, but again with two aryl/carbonyl rotamers (Fig. 8). However,
in this case, as in all of the other systems, the larger ortho sub-
stituent is preferentially syn to the cyano moiety, i.e., as in 5ts and
as determined in the solution phase to the extent of 69%. The amide
linkage is out of planarity by 14ꢁ in this case, the largest variation in
all of the N-aroyl compounds examined, and the pyridyl ring is the
most distorted of the aroyl compounds in terms of N being out of
the benzo plane (Table 3).
Table 3
Structures of Reissert compounds from X-ray diffraction
Cpd.
Conf of 5
maj/min
Ar/CO
dihedral (ꢁ)
C1eNeCeO amide
dihedral (ꢁ)
Ar/benzo plane
angle (ꢁ)
Benzo/C1eC(N)
angle (ꢁ)
N out of benzo
plane (A)
C1 out of benzo
plane (A)
C3 out of benzo
plane (A)
3-Me HeAr
centroid (A)
ꢀ
ꢀ
ꢀ
ꢀ
1a
1f46
1g
1g
1m
2c
2g
2g
2j
trans
trans
ts (minor)
ta (major)
ts only
d
d
0.2
1.7
0.45
0.39
3.6
162.2
13.7
13.7
5.6
d
82.27
72.50
83.28
83.28
86.44
95.81
90.32
90.32
89.11
87.62
85.44
90.28
87.67
89.66
0.402
0.373
0.517
0.517
0.642
0.762
0.698
0.698
0.619
0.684
0.602
0.680
0.589
0.636
0.129
0.009
0.060
0.060
0.017
0.048
0.040
0.040
0.074
0.009
0.050
0.058
0.100
0.092
0.263
0.125
0.291
0.292
0.372
0.423
0.368
0.368
0.356
0.244
0.232
0.392
0.353
0.411
d
45.0
56.2
48.2
49.7
d
41.24
50.9
52.4
47.5
d
d
3.15a
3.43b
d
d
ta (minor)
ts (major)
ts/ta
ts/ta
ts only
trans
53.6
56.0
69.5
60.4
57.4
67.5
71.3
69.6
52.3
55.6
58.9
51.9
48.8
54.7
59.0
56.0
2.99
2.99c
2.78
2.75
2.87
2.79
2.79
2.82
2l
8.6
2m
2n
2p
2q
10.1
12.6
9.5
ts/ta
ts/ta
12.5
a
ꢀ
3-H to Ar centroid distance; 3-H to CH3 distance 3.65 A.
b
c
ꢀ
3-H to Ar centroid distance; 3-H to CH3 distance 2.72 A.
ꢀ
CH3 to CH3 distance 2.60 A.

H.W. Gibson et al. / Tetrahedron 68 (2012) 8052e8067
8061
ꢀ
ꢀ
ring at 2.87 A, a distance (2.7e2.9 A) typically considered to indicate
an ArCHep
interaction.45
The X-ray structure for ortho-chlorobenzoyl-3-methyl de-
rivative 2j (Fig. 11), however, reveals the presence of two aryl/car-
bonyl rotamers in the form of disorder of the location of the
chlorine atom in the trans amide isomer, which exists to the ex-
clusion of the cis. In the most populated (92%) rotamer, the chlorine
atom is syn to the cyano group, corresponding to isomer 5ts, con-
sistent with the general rule deduced from NMR results and in good
agreement with the 93% ts observed in solution. It is noteworthy
ꢀ
that one of the 3-methyl protons is 2.78 A from the centroid of the
chlorobenzoyl ring, once more indicative of a possible CHe
p
interaction.45
Fig. 8. Two views of one enantiomer from the X-ray crystal structure of 2-(o-toluyl)-3-
methyl-1,2-dihydroisoquinaldonitrile (2g) showing the pseudo-axial location of the
cyano moiety, the trans orientation of the amide linkage and the twisted disposition of
the tolyl group. In the major [78.9 (ꢂ0.5)%] rotamer (top views) the ortho-methyl group
is syn to the cyano group (5ts), while in the minor [21.1 (ꢂ0.5)%] isomer (bottom views)
ꢀ
it is anti to the cyano group (5ta). In 5ts a 3-Me proton is 2.99 A away from the centroid
ꢀ
of the tolyl ring but only 2.60 A away from the closest ortho-methyl proton.
In the solid state 10-naphthoyl Reissert compound 1m (Fig. 9)
exists totally in the trans amide form and there is a 50ꢁ rotation
about the aryl/carbonyl bond, placing the benzo ring in the con-
formation described by 5ts. However, the amide unit is very nearly
coplanar, deviating only by 4ꢁ. The protons on the 3- and 20-posi-
tions are in very close proximity, as deduced from the solution
phase NMR results. Addition of the 3-methyl substituent in 2m
causes the amide bond to deviate more (10ꢁ) from planarity
(Fig. 10), but the other aspects of the structure are very similar to
those in 1m. There is no evidence of disorder in either structure,
indicating that only one aryl/carbonyl rotamer exists in the solid
state; in both cases the benzo ring of the naphtho unit is syn to the
cyano moiety. This establishes the distance between one of the
3-methyl protons of 2m and the centroid of the closest naphtho
Fig. 11. Twoviews of one enantiomer of each of the two C]O/arylrotational isomers from
the X-raycrystal structure of 2-(o-chlorobenzoyl)-3-methyl-1,2-dihydroisoquinaldonitrile
(2j). In the major [92.3(ꢂ0.2)%] rotamer (top) thechlorine atom(purple) is syn tothecyano
group (5ts) and in the minor [7.7(ꢂ0.2)%] isomer (bottom) it is anti (5ta), but otherwise the
structure is the same. The cyano moiety is pseudo-axial, the amide linkage is trans and the
ꢀ
aroyl group is twisted. The closest 3-methyl proton is 2.78 A away from the centroid of the
benzoyl moiety.
The X-ray crystal structure of the ortho-fluorobenzoyl compound
2l (Fig. 12) also reveals two rotameric forms of the trans amide. The
60-proton is larger than the 20-fluorine39a and the major (69%)
rotamer thus again corresponds to structure 5ts, but the proportions
Fig. 9. Two views of one enantiomer from the X-ray crystal structure of 2-(10-naph-
thoyl)-1,2-dihydroisoquinaldonitrile (1m) showing the pseudo-axial location of the
cyano moiety, the trans orientation of the amide linkage and the twisted disposition of
the naphthoyl group and the location of the benzo ring syn to the cyano group (5ts). H3
0
ꢀ
and the naphthyl 2 -proton are only 2.78 A.
Fig. 12. Twoviewsofone enantiomerofeach of the twoC]O/aryl rotational isomersfrom
theX-ray crystal structure of 2-(o-fluorobenzoyl)-3-methyl-1,2-dihydroisoquinaldonitrile
(2l). In the major [69.2(ꢂ0.2)%] rotamer (top) the fluorine atom (green) is anti to the cyano
group (5ts) and in the minor [30.8(ꢂ0.2)%] isomer (bottom) it is syn (5td), but otherwise
the structure is the same. The cyano moiety is pseudo-axial, the amide linkage is trans and
Fig. 10. Two views of one enantiomer from the X-ray crystal structure of 2-(10-
naphthoyl)-3-methyl-1,2-dihydroisoquinaldonitrile (2m) showing the pseudo-axial
location of the cyano moiety, the trans orientation of the amide linkage, the twisted
disposition of the naphthoyl group and the location of the benzo ring syn to the cyano
ꢀ
ꢀ
group (5ts). The closest 3-methyl proton is 2.87 A away from the centroid of the A-ring
the aroyl group is twisted. The closest 3-methyl proton is 2.75 A away from the centroid of
of the naphthoyl moiety.
the benzoyl moiety.

8062
H.W. Gibson et al. / Tetrahedron 68 (2012) 8052e8067
differ somewhat from those observed in solution (54% tu, 46% ta).
Other features are similar to those of 2j, including a 2.75 A distance
conformers about the arylecarbonyl bond are present, in w2:1
ratio. The more stable conformer is the one that places the chloro
substituent syn to the cyano group and the smaller39a fluoro moiety
trans to it, a repetition of the pattern observed by NMR, but with
different proportions (1.2:1) in solution.
ꢀ
from the centroid of the fluorobenzoyl ring to one of the 3-methyl
protons, again consistent with a CHep
interaction.45
The X-ray results for the 20,60-dimethylbenzoyl-3-methyl Reis-
sert compound 2n (Fig. 13) support the conclusion that the trans
amide isomer is more stable than the cis isomer, since the latter is
not observed in the solid state, even though it comprises 29% in
solution; however, the amide unit is non-planar by 13ꢁ. Moreover,
the influence of the two ortho-methyl substituents is clearly seen in
the large twist about the aryl/carbonyl bond, amounting to 68ꢁ.
Even with the twist, two of the ortho- and 3-methyl protons are still
In 2-(20-chloro-60-methylbenzoyl)-3-methyl-1,2-dihydroiso-
quinaldonitrile (2q) (Fig. 15) again only the trans amide form is
observed, but out of planarity by 12.5ꢁ and accompanied by a 56ꢁ
aryl/carbonyl rotation. The chloro moiety is sterically larger than
the methyl group,38 so the dominant (56%) structure corresponds
to isomer 5ts, consistent with the general rule that 5ts pre-
dominates and remarkably similar to the solution result (52% ts
vs 48% ta). Closer inspection reveals that in this case one of the
ortho-methyl protons in 5ta is quite close to a 3-methyl proton,
ꢀ
quite close to each other: 2.49 A. And from the centroid of the
dimethylbenzoyl ring to one of the 3-methyl protons the distance is
ꢀ
ꢀ
only 2.79 A, again suggesting CHe
p
interaction as one of the con-
2.44 A. In the predominant 5ts rotamer the chlorine atom is
ꢀ
tributing forces in the solid state.
3.18 A away from the same 3-methyl proton. As noted with the
other compounds, a CHe
proximity (2.82 A) of one of the 3-methyl protons to the centroid
of the chloromethylbenzoyl ring.
p
interaction45 is indicated by the
ꢀ
Fig. 13. Two views of one enantiomer from the X-ray crystal structure of 2-(20,60-di-
methylbenzoyl)-3-methyl-1,2-dihydroisoquinaldonitrile (2n) showing the pseudo-
axial location of the cyano moiety, the trans orientation of the amide linkage (5t)
and the highly twisted disposition of the aroyl group. The 3-methyl carbon and the
ꢀ
ꢀ
upper ortho-methyl carbon are 3.84 A apart, putting the hydrogens within 2.49 A of
ꢀ
each other. The closest 3-methyl proton is 2.79 A away from the centroid of the aroyl
moiety.
The X-ray structure of the 2-(20-chloro-60-fluorobenzoyl)-3-
methyl Reissert compound 2p (Fig. 14) confirms the trans amide
solely as opposed to 21% of the cis amide in solution. The amide unit
in this case is non-coplanar by 71ꢁ, the largest such deviation ob-
ꢀ
served. One of the 3-methyl protons is 2.79 A from the centroid of
the chlorofluorobenzoyl ring (CHe
p interaction). And again two
Fig. 15. Two views of one enantiomer of each of the two C]O/aryl rotational isomers
from the X-ray crystal structure of 2-(20-chloro-60-methylbenzoyl)-3-methyl-1,2-
dihydroisoquinaldonitrile (2q). The cyano moiety is pseudo-axial, the amide is trans
and the aroyl group is twisted. In the major [56.0(ꢂ0.2)%] rotamer (top) the chlorine
atom (green) is syn to the cyano group and the methyl group is anti to it (5ts). In the
minor [44.0(ꢂ0.2)%] isomer (bottom) the positions of the chlorine atom and the
methyl group are reversed (5ta), but otherwise the structure is the same. The closest 3-
ꢀ
methyl proton is 2.82 A away from the centroid of the benzoyl moiety. Two Hs on the
ꢀ
ꢀ
CH3 groups are 2.44 A apart in the minor isomer and >3 A apart in the major isomer.
The aryl/carbonyl twist angles of 1m, 2m, 2n, 2p, and 2q, i.e.,
50ꢁ, 57ꢁ, 68ꢁ, 71ꢁ, and 70ꢁ, respectively (Table 3), can be compared
to that of the analogous N-benzoyl compound, 1f, in which the aryl
and amide planes are situated at a 45ꢁ angle to each other,46
reflecting the smaller steric interaction of the phenyl group with
the 3-proton. In all of the Reissert compounds with un-
symmetrically substituted benzoyl moieties, except for 1g, in the
solid state the larger ortho-substituent lies syn to the cyano group,
as deduced from the NMR solution spectra.
Turning to the N-alkanoyl Reissert compounds, the X-ray crystal
structure of the acetyl Reissert compound 1a (Fig. 16) reveals the
expected trans amide conformation and nearly planar isoquinoline
ring system and amide linkages. On the other hand, the X-ray
crystal structure of the isobutyryl Reissert compound 2c (Fig. 17)
demonstrates the result of the steric interactions brought about by
the 3-methyl group and the bulky N-isobutyryl unit: (1) the amide
linkage is now cis, (2) the dihydropyridyl ring is more puckered, and
(3) the amide linkage is non-planar by 18ꢁ, the largest such
Fig. 14. Two views of one enantiomer of each of the two C]O/aryl rotational isomers
from the X-ray crystal structure of 2-(20-chloro-60-fluorobenzoyl)-3-methyl-1,2-
dihydroisoquinaldonitrile (2p). The cyano moiety is pseudo-axial, the amide is trans
and the aroyl group is twisted. In the major [67.6(ꢂ0.5)%] rotamer (top) the fluorine
atom (green) is anti to the cyano group and the chlorine atom (purple) is syn to it (5ts).
In the minor [32.4(ꢂ0.5)%] isomer (bottom) the positions of the two halogens are
reversed (5ta), but otherwise the structure is the same. The closest 3-methyl proton is
ꢀ
2.79 A away from the centroid of the benzoyl moiety.

H.W. Gibson et al. / Tetrahedron 68 (2012) 8052e8067
8063
2g, however, there is no close contact of the oxygen atom with H8;
ꢁ
ꢀ
instead there is an H5eN(^C) interaction (2.73 A/148.13 ), but the
ꢀ
H1eO(]C) interaction remains (2.60 A at a CeHeN angle of
147.81ꢁ). Indeed, while short H1eO(]C) distances appear in many
of these compounds, such an interaction is not universal; com-
pounds 2j, 2n, 2p, and 2q with large aroyl substituents are excep-
tions (see Supplementary data Table S3). Other close contacts
involve protons H3, H4, H5, H6, and H7 of the isoquinolyl unit with
the nitrogen atom of the nitrile, and the carbonyl oxygen with H8.
Fig. 16. Two views of one enantiomer from the X-ray crystal structure of 2-acetyl-1,2-
dihydroisoquinaldonitrile (1a) showing the pseudo-axial location of the cyano moiety
and the trans orientation of the amide linkage. H3 and the closest two acetyl protons
ꢀ
are both 2.37 A apart.
2.6. Rationalization of structural effects
Increased steric hindrance due to ortho N-aroyl and 3-methyl
substituents inhibits resonance of type 11a, which requires co-
planarity of the COAr group, tending to raise the amide isomerism
barrier by increasing the contribution of resonance form 11b.
However, as the hindrance becomes more severe the coplanarity of
the NCO group required for 11b is also diminished and it becomes
less important and the non-polar, less conjugated resonance
structure 11c becomes more important; moreover, the energy dif-
ference between the cis and trans isomers decreases as steric
crowding increases.
While such arguments are known to be somewhat tenuous,
melting point observations are consistent with the above amide
resonance considerations. Four of the five highest melting Reissert
compounds [2p (212 ꢁC)>2q (208 ꢁC)>2n (202 ꢁC)>2m (199 ꢁC)>
2o (184 ꢁC)] possess two bulky ortho N-aroyl substituents; the
naphthoyl derivative 2m is the exception. The high melting points
[relative, e.g., to simple N-benzoyl analog 2f (140 ꢁC)], we believe,
are indicative of increased solid state bonding due to the presence
of stronger molecular dipoles arising from increased contributions
of 11b and 11c at the expense of 11a as the substituent size(s) in-
crease(s). On the same basis the ortho-, meta-, and para-toluyl de-
rivatives 2g, 2h, and 2i are exemplary in terms of their lower
trending melting points (146, 134, 130 ꢁC, respectively).
Fig. 17. Two views of one enantiomer from the X-ray crystal structure of 2-isobutyryl-3-
methyl-1,2-dihydroisoquinaldonitrile (2c) showing the pseudo-axial location of the
cyano moiety and the cis orientation of the amide linkage. One of the 3-methyl protons
ꢀ
and the carbonyl oxygen atom are 2.39 A apart. H1 is very close to the isopropyl methine
ꢀ
proton,1.94 A. The origin of the chemical shift difference of the isopropyl methyl signals
(0.17 ppm in CDCl3, previously unpublished result) is obvious in this conformation.
deviation observed in this study. Note in contrast, however, that the
cis amide structure is the minor isomer in solution, apparently
a result of different intermolecular interactions in the solution
versus solid states.
Considering the packing in these crystalline compounds, some
display interesting close contacts. For example, in 1g there are close
in-plane contacts between the carbonyl oxygens and H1 and H8 of
neighboring molecules (Fig. 18) in both the major and minor amide
isomers. Moreover, the nitrogen atom of the cyano group is close to
X
X
X (+)
R3
R3
R3
N
N
N
(+)
O
Y
ꢁ
H
ꢀ
O
Y
H
CN
H4 of a neighboring molecule, 2.54 A at a CeHeN angle of 159.03 .
The reported46 packing of parent Reissert compound 1f is very
similar (see Supplementary data Table S3). In the 3-methyl analog
O
Y
CN
H
CN
11a
11b
11c
Moreover, this view is consistent with the observed solubilities
of these compounds. The 3-H/3-Me pairs of ortho- and para-chloro
compounds 1j/2j and 1k/2k, respectively, reveal the decreased
solubility behavior of the ortho-substituted N-aroyl derivatives: the
ortho-chloro compound 1j and its 3-methyl analog 2j are soluble in
chloroform to less than 1% by weight/volume as opposed to the
para-chloro compounds 1k and 2k, which are soluble to the extent
of w20%. Again this is attributed to the increased contribution of
polar resonance form 11b for the ortho-substituted compounds 1j
and 2j relative to the para-substituted compounds 1k and 2k.
2.7. Reactivity patterns
A number of rather puzzling reactivity patterns of Reissert
compounds are now explicable in terms of their stereochemistry.
For example, the deuterated 3-H N-benzoyl Reissert compound 8f
does not exchange its deuterium atom when recrystallized from
boiling ethanol, whereas the deuterated 3-CH3 N-(o-toluyl) Reissert
compound 9g undergoes nearly 100% exchange under such con-
ditions. The latter compound would be expected to be less acidic on
purely electronic grounds because of the electron donating
Fig. 18. Close contacts in the solid state of 1g. Distance/CeHeX angle (a):
2.52 A/143.71 ; (b): 2.53.
ꢁ
ꢀ

8064
H.W. Gibson et al. / Tetrahedron 68 (2012) 8052e8067
character of the methyl groups. Therefore, the unexpected differ-
ence in exchange must be due to some other factor.
aroyl compounds 1g, 2g, 2j), which exchange most rapidly do not
rearrange as rapidly, while those which exchange most slowly
(unsubstituted or para-substituted aroyl compounds 1f, 2f, 2k) rear-
range rapidly. The rearrangement results are understandable on the
same grounds as the exchange results, namely the variable contri-
butions of resonance forms 11a, 11b, and 11c. The rearrangement of
Reissert compounds to the 1-acylisoquinolines 12 is an intra-
molecular reaction47 involving attack of the C1 carbanion on the
amide carbonyl group. This attack must occur from the equatorial
disposition of the carbanion as shown in Scheme 9. A prerequisite for
the reaction is rotation about the NeCO bond to put the carbonyl
carbonintoa positionfavorableforattackina cisamideconformation,
leading to the anti-periplanar configuration required39b for elimina-
tion of cyanide ion. This rotation is easier if the contribution from
structure 11b is minimized; rotation in 11a and 11c will be relatively
more facile. As noted above, ortho N-aroyl substituents reduce con-
tributions from 11a and 11b, thus facilitating the tendency to rear-
range. Indeed the cis amide isomers of 1g and 2g were not observed.
The tendency for 1g, 2g, and 2j to rearrange slowly relative to alkyl-
ation is attributable to a higher barrier for formation of the cis amide
because of their ortho-substituents, relative to non-ortho-substituted
1f, 2f, and 2k, which possess much lower barriers for formation of the
cis amide geometry.
The acidity of H1 or D1 is enhanced by greater contribution of
resonance form 11b, in which the positive charge resides upon the
nearby N atom. Conversely, increased contribution of resonance
structure 11a causes diminished acidity due to the greater electron
density on the N. 11a requires coplanarity of the NCOAr moiety; 11b
does not. As shown above, the H1 analog (1f) of 8f does not exhibit
hindered arylecarbonyl rotation, but in the analog (2g) of 9g
arylecarbonyl rotation is slow. Both compounds exist pre-
dominantly in the trans amide form. Therefore, the increased
acidity of 2g and deutero analog 9g relative to 1f and 8f is attributed
to increased contribution of non-coplanar (aryl) resonance form
11b due to steric hindrance in 2g; the positively charged nitrogen in
11b enhances the acidity of H1. This is clear in the solid state as the
amide linkage is nearly planar in 1f (dihedral 1.7ꢁ),46 indicating an
enhanced contribution from 11b, while in 2g the dihedral angle is
14ꢁ (Table 3), signaling a decrease in the contribution of resonance
structure 11b.
Yet another anomalous reactivity pattern is observed in the
rearrangement reactions1,20 of 1f and 1g in the presence of sodium
hydride. Compound 1f readily rearranges to 12a, while 1g does not
rearrange, even though the red color due to its anion is observed.
Similarly, in the corresponding 3-methyl analogs 2f and 2g a striking
difference was observed. N-Benzoyl-3-methyl compound 2f rear-
ranges to ketone 12b in 89% yield even in the presence of 2 equiv of
The alkylation products 3 and 4 from alkanoyl Reissert com-
pounds such as 1ae1c and 2ae2c upon treatment with base readily
undergo cyclization to 14 by means of intramolecular attack of the
N
N
O
N
R
N
NaH
R3
- CN-
R
H
O
N
R
N
R
N
N
N
R3
R
R3
R3
O
R3
O
O
Scheme 9. The proposedmechanism for base-promotedconversionofisoquinolineReissert compoundsto 1-acylisoquinolines, involvingisomerization of the trans amide tothecis isomer.
isopropyl iodide; in the presence of 8 equiv only 15% of the alkylated
product 4f was isolated.19 In contrast, even though it too tends to
rearrange to ketone 12c, ortho-toluyl-3-methyl compound 2g in the
presence of 8 equiv of isopropyl iodide afforded 70% of the isopropyl
compound 3g.20 Thus, the apparent ratio of the rates of rearrange-
ment to alkylation is 6 for 2f and 0.5 for 2g, a factor of 10 difference
caused by the ortho-CH3 moiety of 2g. Inpart the lower propensity of
2g for rearrangement might be attributed to the destabilization of
the intermediate anion by the electron releasing tolyl methyl group.
However, the results for the o-Cl- and p-Cl-benzoyl-3-methyl de-
rivatives 2j and 2k also bear on this point. It was not possible to
isolate the alkylation product of the para-chloro 2k; even in the
presence of 8 equiv of isopropyl iodide, the ketone 12d was isolated
in 93% yield.20 ortho-Chloro2j, on the otherhand, afforded47% of the
alkylated product 4j under these conditions.20
enolate anion 13 on the equatorial cyano group (Scheme 10).19b,48
This reaction demands that the amide group exist in the cis form
as shown in 13. This is consistent with the observation of inter-
converting cis and trans amide isomers in 2a and 2c in solution and
the cis structure of 2c observed exclusively in the solid state.
R3
R3
base
O
N
3 or 4
R = CH2R’
N
O
HC R'
HN
R'
N
14
13
Scheme 10. The proposed mechanism for base-promoted cyclization of 1-alkyl-2-
alkanoyl-1,2-isoquinaldonitriles.
R3
A similar situation exists in the conversion of 1d and 2d to 10.21
The anion derived from cis amide isomer of 1d and 2d is required
for effective ring closure, presumably. The high yields of this
transformation are consistent with a mobile cis%trans equilibrium
in these N-alkanoyl Reissert compounds as noted in the present
work.
N
12
O
R
a. R3 = H, R = C6H5
b. R3 = CH3, R = C6H5
c. R3 = CH3, R = o-C6H5CH3
d. R3 = CH3, R = p-C6H5Cl
3. Experimental
3.1. General
The rearrangement versus alkylation results parallel the deute-
rium exchange rate results; the compounds (mono-ortho-substituted
Melting points were taken in capillaries on a Thomas-Hoover
apparatus and are corrected. Elemental analyses were carried out

H.W. Gibson et al. / Tetrahedron 68 (2012) 8052e8067
8065
by Spang Microanalytical Laboratories, Ann Arbor, MI and Atlantic
Microlabs, Norcross, GA. Some NMR spectra were recorded on
Varian A-60-4, Jeolco C-60H, and Bruker WP-80 instruments,
equipped with variable temperature probes, and the temperature
was calibrated using methanol. Chemical shifts are relative to tet-
ramethylsilane internal standard. Recent spectra were recorded on
Varian Unity or Inova 400/100 MHz, JEOL Eclipse 500/125 MHz or
Bruker Avance III 600/150 MHz instruments. The electrospray
ionization mass spectrometry (ESI MS) was carried out on an Agi-
lent 6220 Accurate Mass TOF LC/MS Spectrometer. X-ray crystal-
lography was done using an Oxford Diffraction SuperNova X-ray
diffractometer. Molecular modeling was carried out over the years
with several different commercial molecular mechanics software
packages, including Serena Software PC Model (MMX), Cam-
bridgeSoft Chem 3D (MM2), CambridgeSoft ChemBioDraw 3D
(MM3) and Wave Function Mac Spartan (MM2).
possible). Anal. Calcd: C, 79.44; H, 6.00; N, 9.27. Found: C, 79.42; H,
6.04; N, 9.26.
3.2.5. 2-(20,60-Dimethoxybenzoyl)-3-methyl-1,2-dihydroisoquinaldo-
nitrile (2o). Yield 44%, mp 182.5e183.5 ꢁC (ethanol); 1H NMR
(400 MHz, CDCl3, 23 ꢁC):
d 1.67 (s, 1.65H), 2.55 (s, 1.35H), 3.28 (s,
1.65H), 3.64 (s, 1.35H), 3.89 (s, 1.65H), 3.93 (s, 1.35H), 5.59 (s, 0.46H),
6.16 (br s, 0.54H), 6.33 (br s, 0.46H), 6.41 (d, J¼8, 0.54H), 6.57 (d,
J¼8, 0.54H), 6.60 (d, J¼8, 0.54H), 6.67 (d, J¼8, 0.46H), 6.99 (s,
0.54H), 6.99 (d, J¼8, 0.46H), 7.10e7.45 (m, 4.5H), including 7.40 (t,
13
J¼8, 0.54H) ppm; C NMR (100.6 MHz, CDCl3, 23 ꢁC):
d 20.59,
20.69, 45.1, 49.6, 55.4, 55.8, 56.0 (two peaks: 55.96. 55.97), 103.6,
104.0 (two peaks: 103.96, 104.03), 116.1, 116.4, 116.7, 117.3, 124.6,
124.8, 125.1, 125.9, 126.3, 126.7, 127.2, 127.6, 129.5, 129.8, 131.5,
132.0, 134.7, 137.1, 157.0, 157.3, 157.4, 157.8, 165.0, 165.5 ppm (35
peaks; for two isomers total of 40 possible). Anal. Calcd: C, 71.84; H,
5.43; N, 8.38. Found: C, 71.83; H, 5.39; N, 8.39.
3.2. Reissert compounds (1 and 2)
3.2.6. 2-(20-Chloro-60-fluorobenzoyl)-3-methyl-1,2-dihydroisoquinal-
donitrile (2p). Yield 83%, mp 211.5e212.0 ꢁC (ethanol); 1H NMR
Most of the Reissert compounds used in this work have been
previously reported.1,3,20 However, several new compounds were
prepared by the methylene chloride procedure,2 as summarized
below.
(500 MHz, CDCl3, 23 ꢁC):
d
1.72 (br s,1.4H), 1.74 (d, J¼1, 1.0H), 2.54 (d,
J¼1, 0.37H), 2.56 (br s, 0.26H), 5.48 (br s, 0.09H), 5.50 (br s, 0.12H),
6.27 (br s, 0.34H), 6.29 (br s, 0.45H), 6.43 (br s, 0.12H), 6.45 (br s,
0.09H), 6.94 (s, 0.33H), 6.96 (s, 0.46H), 7.0e7.6 (m, 7H) ppm; 13C
3.2.1. 2-(20,60-Dimethoxybenzoyl)-1,2-dihydroisoquinaldonitrile
NMR (125 MHz, CDCl3, 23 ꢁC):
d 20.1, 20.3, 20.8, 21.0, 45.2, 49.7,114.4,
(1o). Yield 88%; mp 160.5e161.5 ꢁC (ethanol); 1H NMR (500 MHz,
114.6, 114.7, 114.9, 115.2, 115.4, 116.6 (two peaks: 116.59, 116.61),
116.6, 117.9, 118.0, 118.1, 118.2, 124.6, 124.7, 124.8, 124.9, 125.0, 125.3,
125.4, 125.5 (three peaks: 125.48, 125.51, 125.53), 125.6, 125.7, 126.0
(three peaks: 125.95, 126.00, 126.03), 126.2 (two peaks: 126.22,
125.25), 126.4, 128.0 (two peaks: 127.97. 128.01), 128.3, 130.1, 130.4,
130.5, 130.8, 130.9, 131.0, 132.1 (two peaks: 132.10, 132.14), 132.2,
132.3 (two peaks: 132.28, 132.31), 132.4, 132.5, 132.6, 133.6 (two
peaks: 133.56, 133.60), 157.9, 158.9, 160.0, 160.9, 161.7, 161.8 (CeF
doublets at 157.9/160.0 and 158.9/160.9) ppm (62 peaks; for four
isomers total of 76 possible including CeF doublets); 19F NMR
CDCl3, 25 ꢁC):
d 3.67 (s, 2.6H), 3.70 (s, 0.4H), 3.86 (s, 2.6H), 3.89 (s,
0.4H), 5.67 (br s, 0.14H), 5.86 (d, J¼7, 0.86H), 6.31 (d, J¼7, 0.14H),
6.38 (d, J¼7, 0.86H), 6.56 (d, J¼8, 0.86H), 6.61 (d, J¼8, 0.86H), 6.69
(d, J¼7, 0.14H), 6.84 (br s, 0.86H), 7.06 (d, J¼8, 0.14H), 7.13 (d, J¼8,
13
0.86H), 7.2e7.5 (m, 4.3H) ppm; C NMR (125 MHz, CDCl3, 22 ꢁC):
d
43.8, 48.4, 55.8, 55.9, 56.0, 56.2, 104.0 (two peaks: 103.98, 104.01),
109.2, 111.2, 112.2, 116.4 (two peaks: 116.39, 116.45), 125.1, 125.6,
126.8, 128.1, 129.9, 130.4, 132.0, 156.9, 157.2, 157.3, 157.7, 164.7, 165.8
(27 peaks; expected for two amide isomers: total of 38). Anal.
Calcd: C, 71.24; H, 5.03; N, 8.74. Found: C, 71.43; H, 5.00; N, 8.66.
(376 MHz, CDCl3, 23 ꢁC):
d
ꢀ113.43 (t, J¼7, 0.37F), ꢀ112.18 (t, J¼7,
0.10F), ꢀ111.99 (t, J¼7, 0.11F), ꢀ111.51 (t, J¼7, 0.42F). Anal. Calcd: C,
66.16; H, 3.70; N, 8.57. Found: C, 65.88; H, 3.84; N, 8.71. ESI MS:
m/z 300.0605, 100% [MꢀCN]þ; 327.0686 [MþH]þ; calcd for
M¼C18H12ClFN2O: m/z [MꢀCN]þ, 300.0586, error 6.3 ppm; m/z
[MþH]þ, 327.0700, error 4.2 ppm.
3.2.2. 2-(m-Toluyl)-3-methyl-1,2-dihydroisoquinaldonitrile (2h). Yield
69%, mp 132.5e134.5 ꢁC (ethanol); 1H NMR (500 MHz, CDCl3,
23 ꢁC):
d 1.82 (s, 3H), 2.38 (s, 3H), 6.26 (s, 1H), 6.51 (s, 1H),
13
7.2e7.5 (m, 8H) ppm; C NMR (125 MHz, CDCl3, 22 ꢁC):
d 21.3,
22.0, 115.5, 116.9, 125.2, 125.5, 125.8,126.2, 127.9, 128.5, 129.0,
130.0, 131.3, 132.8, 135.0, 135.5, 138.8, 169.0 (19 peaks, as
required). Anal. Calcd: C, 79.14; H, 5.59; N, 9.72. Found: C, 79.14;
H, 5.57; N, 9.71.
3.2.7. 2-(20-Chloro-60-methylbenzoyl)-3-methyl-1,2-dihydroisoquinal-
donitrile (2q). Yield 88%; mp 207.4e208.1 ꢁC (ethyl acetate); 1H NMR
(500 MHz, CDCl3, 23 ꢁC):
d
1.58 (d, J¼1, 1.11H), 1.69 (d, J¼1, 1.01H),
1.80 (s, 1.01H), 2.02 (s, 0.42H), 2.46 (s, 1.11H), 2.53 (s, 0.46H), 2.58 (d,
J¼1, 0.46H), 2.60 (d, J¼1, 0.42H), 5.41 (br s, 0.14H), 5.52 (br s, 0.15H),
6.25 (br s, 0.72H), 6.41 (br s, 0.15H), 6.46 (br s, 0.13H), 6.98 (br s,
0.14H), 6.99 (s, 0.15H), 7.02 (br m, 0.34H), 7.04 (br m, 0.37H), 7.06 (b,
0.37H), 7.08 (br m, 0.46H), 7.09 (b, 0.34H), 7.1e7.5 (m, 7H) ppm; 13C
3.2.3. 2-(ortho-Fluorobenzoyl)-3-methyl-1,2-dihydroisoquinaldonitrile
(2l). Yield 65%, mp 166.5e168.5 ꢁC (ethanol); 1H NMR (500 MHz,
CDCl3, 22 ꢁC):
d 1.85 (br s, 3H), 6.28 (s, 1H), 6.57 (br s, 1H), 7.0e7.6
13
(m, 8H) ppm; C NMR (125 MHz, CDCl3, 22 ꢁC):
d 21.4, 46.6, 116.4,
116.6 (two peaks: 116.59, 116.64), 123.7, 124.7, 125.4, 125.7, 126.2,
128.0, 129.9, 130.1, 131.1, 133.4, 134.3, 158.3, 160.4, 164.5 (19 peaks, as
required, including the doublet peaks at 158.3 and 160.4 for CeF).
Anal. Calcd: C, 73.96; H, 4.48; N, 9.59. Found: C, 74.19; H, 4.58;
N, 9.68.
NMR (125 MHz, CDCl3, 23 ꢁC):
d 18.4, 19.2, 19.3, 19.9, 20.2, 20.6 (two
peaks: 20.59, 20.62), 21.0, 44.6, 44.9, 49.6 (two peaks: 49.56, 49.63),
115.8, 116.2, 116.6, 117.0, 117.5, 117.8 (three peaks: 117.79, 117.82,
117.83), 124.8, 125.2, 125.3, 125.5 (two peaks: 125.46, 125.47), 125.7,
125.9, 126.0, 126.1, 126.2, 127.2, 127.5, 127.8, 127.9 (two peaks: 127.87,
127.90), 128.2, 128.4, 128.7, 129.0, 129.1, 129.6, 130.0, 130.2, 130.3,
130.4, 130.8, 130.9 (two peaks 130.89, 130.90), 131.0 (two peaks:
130.97, 130.99), 131.3, 132.0, 133.1, 133.9, 135.3, 136.7, 138.4, 166.2,
166.3 ppm (59 peaks; for four isomers total of 76 possible). Anal.
Calcd for C19H15ClN2O: C, 70.70; H, 4.68; N, 8.68. Found: C, 70.67; H,
4.65; N, 8.81.
3.2.4. 2-(20,60-Dimethylbenzoyl)-3-methyl-1,2-dihydroisoquinaldoni-
trile (2n). Yield 47%, mp 201.0e202.5 ꢁC (ethanol); 1H NMR
(400 MHz, CDCl3, 22 ꢁC): d 1.50 (d, J¼1, 2.12H), 1.73 (s, 2.12H), 1.93
(s, 0.88H), 2.38 (s, 2.12H), 2.48 (s, 0.88H), 2.56 (d, J¼1, 0.88H), 5.46
(s, 0.31H), 6.17 (q, J¼1, 0.70H), 6.40 (q, J¼1, 0.30H), 6.89 (d, J¼4,
0.88H), 6.94 (d, J¼7, 0.30H), 7.05e7.40 (m, 6.7H, including 7.08,
13
0.69H) ppm; C NMR (100.6 MHz, CDCl3, 23 ꢁC):
d
18.2, 18.9, 19.6,
3.2.8. 2-Ethoxycarbonyl-3-methyl-1,2-dihydroisoquinaldonitrile
20.6, 20.9, 116.2, 116.9, 116.9, 117.5, 124.8, 125.1, 125.5, 125.9, 127.6,
128.00, 128.1, 128.5, 129.8, 130.0, 130.3, 131.1, 134.2, 135.7, 135.8,
136.6, 169.1, 169.5 ppm (27 peaks; for two isomers total of 40
(2r). Yield 36%, mp 46.5e49.0 ꢁC (hexane); 1H NMR (500 MHz,
CDCl3, 22 ꢁC):
d
1.33 (t, J¼7, 3H), 2.34 (d, J¼1, 3H), 4.27 (16-line AB of
ABX3, JAB¼7.5, JAx¼JBx¼10, nA¼4.24, nB¼4.31, 2H), 6.17 (br s, 1H),

8066
H.W. Gibson et al. / Tetrahedron 68 (2012) 8052e8067
1e25; Popp, F. D. Adv. Heterocycl. Chem. 1979, 24, 187e214; (d) Popp, F. D. In
Quinolines, Part II; Jones, G., Ed.; J. Wiley and Sons: New York, NY, 1982;
pp 353e375;
(e) Cooney, J. V. J. Heterocycl. Chem. 1983, 20, 823e837; (f) Kielland, N.; Lavilla, R.
Top. Heterocycl. Chem. 2010, 25, 127e168.
2. (a) Popp, F. D.; Blount, W. Chem. Ind. 1961, 550; (b) Popp, F. D.; Blount, W.;
Melvin, P. J. Org. Chem. 1961, 26, 4930e4932; (c) Popp, F. D.; Blount, W. J. Org.
Chem. 1962, 27, 297e298; (d) Popp, F. D.; Soto, A. J. Chem. Soc. 1963, 1760e1763.
3. (a) Ruchirawat, S.; Phadungkul, N.; Chuankamnerdkarn, M.; Thebtaranonth, C.
Heterocycles 1977, 6, 43e46; (b) Bhattacharjee, D.; Popp, F. D. J. Heterocycl.
Chem. 1980, 17, 1211e1212.
4. (a) Gibson, H. W.; Pandya, A.; Guilani, B.; Rasco, M. L.; Jois, Y. H. R. Polym.
Commun. 1991, 32, 13e16; (b) Gibson, H. W.; Guilani, B.; Rasco, M. L. Macro-
molecules 1991, 24, 3700e3703.
5. (a) Jois, Y. H. R.; Berg, M. A. G.; Merola, J. S.; Gibson, H. W. Tetrahedron Lett. 1991,
32, 2997e3000; (b) Jois, Y. H. R.; Gibson, H. W. J. Org. Chem. 1991, 56, 865e867;
(c) Gibson, H. W.; Pandya, A.; Rasco, M. L.; Guilani, B.; Hermann, C. K. F.; Leb-
lanc, J.-P.; Jois, Y. H. R. Makromol. Chem., Macromol. Symp. 1992, 54/55, 413e421;
(d) Pandya, A.; Gibson, H. W. J. Org. Chem. 1993, 58, 2851e2855; (e) Leblanc, J.
P.; Rorrer, K.; Gibson, H. W. J. Org. Chem. 1994, 59, 674e675.
6.43 (s, 1H), 7.14 (d, J¼8, 1H), 7.22e7.27 (m, 2H), 7.31e7.37 (m, 1H)
13
ppm; C NMR (125 MHz, CDCl3, 23 ꢁC):
d 14.4, 21.4, 47.5, 63.3,
114.6, 117.0, 125.0, 125.5 (two peaks: 125.48, 125.54), 127.5, 129.9,
131.3, 135.2, 152.7 (14 peaks, as required). Anal. Calcd: C, 69.40; H,
5.82; N, 11.57. Found: C, 69.48; H, 5.79; N, 11.63.
3.2.9. 2-Benzyloxycarbonyl-3-methyl-1,2-dihydroisoquinaldonitrile
(2s). Yield 41%, mp 72.5e74.5 ꢁC (ethyl acetate-hexane); 1H NMR
(500 MHz, CDCl3, 22 ꢁC):
d
2.31 (br s, 3H), 5.22 (AB q, JAB¼10,
nA¼5.19, nB¼5.26, 2H), 6.17 (br s, 1H), 6.44 (s, 1H), 7.14 (d, J¼8, 1H),
7.20e7.26 (m, 2H), 7.31e7.43 (m, 6H) ppm; 13C NMR (125 MHz,
CDCl3, 23 ꢁC):
d 21.4, 47.6, 69.0, 114.9,116.9, 125.1,125.5, 127.6, 128.5,
128.7 (two peaks: 128.71,128.74),130.0, 131.2, 135.0,135.1,152.6 (16
peaks, 17 required). Anal. Calcd: C, 74.98; H, 5.30; N, 9.21. Found: C,
74.86; H, 5.26; N, 9.12.
6. (a) Graulich, A.; Dilly, S.; Farce, A.; Scuvee-Moreau, J.; Waroux, O.; Lamy, C.;
Chavatte, P.; Seutin, V.; Liegeois, J.-F. J. Med. Chem. 2007, 50, 5070e5075;
(b) Fuchs, C.; Bender, C.; Ziemer, B.; Liebscher, J. J. Heterocycl. Chem. 2008, 45,
1651e1658; (c) Bittermann, A.; Baskakov, D.; Herrmann, W. A. Organometallics
2009, 28, 5107e5111; (d) Kim, S. H.; Kim, Y. M.; Park, B. R.; Kim, J. N. Bull. Korean
Chem. Soc. 2010, 31, 3031e3034; (e) Ahamed, M.; Todd, M. H. Eur. J. Org. Chem.
2010, 31, 5935e5942; (f) Kim, S. H.; Lee, S.; Kim, Y. M.; Kim, J. N. Bull. Korean
Chem. Soc. 2010, 31, 3822e3825.
7. (a) Gibson, H. W. Macromolecules 1974, 7, 711e712; (b) Gibson, H. W. Macro-
molecules 1975, 8, 89e90; (c) Gibson, H. W.; Bailey, F. C. Macromolecules 1976, 9,
10e15; (d) Gibson, H. W.; Bailey, F. C. J. Polym. Sci., Polym. Chem. Ed. 1976, 14,
1661e1669; (e) Gibson, H. W.; Bailey, F. C. Macromolecules 1976, 9, 221e226;
(f) Gibson, H. W.; Guilani, B. Macromolecules 1990, 23, 4339e4340; (g) Pandya,
A.; Gibson, H. W. Polym. Commun. 1991, 32, 134e136; (h) Jois, Y. H. R.; Gibson, H.
W. Polym. Commun. 1991, 32, 168e170; (i) Gibson, H. W.; Guilani, B. Polym.
Commun. 1991, 32, 324e328; (j) Pandya, A.; Gibson, H. W. Polym. Bull. 1991, 25,
17e24; (k) Leblanc, J. P.; Gibson, H. W. Macromolecules 1993, 26, 4953e4955;
(l) Jois, Y. H. R.; Gibson, H. W. Macromolecules 1993, 26, 6151e6154; (m) Pandya, A.;
Yang, J.;Gibson, H. W. Macromolecules 1994, 27,1367e1375; (n)Jois, Y.; Gibson, H. W.
Macromolecules 1994, 27, 2912e2916; (o) Gibson, H. W.; Dotson, D. L. Polymer 1998,
39, 6483e6487; (p) Yang, J.; Tyberg, C. S.; Gibson, H. W. Macromolecules 1999, 32,
8259e8268; (q) Gibson, H. W.; Brumfield, K. K.; Grisle, R. A.; Hermann, C. K. F. J.
Polym. Sci. Part: A, Polym. Chem. 2010, 48, 3856e3867; (r) Gibson, H. W.; Rasco, M. L.;
Niu, Z. J. Polym. Sci., Polym. Chem. 2011, 49, 3842e3851.
3.3. Deuteration of Reissert compounds (9)
The following is a representative procedure. A solution of 1.30 g
(5.00 mmol) of 1f and 0.35 mL (5.8 mmol) of CS2 in 10 mL of DMF
was treated with 0.42 g (5.0 mmol) of 30% NaH in mineral oil. The
solution turned red-orange. After it had been stirred for 45 min, the
solution was poured into 18 mL (0.90 mol) of D2O with stirring. The
solid 8f was filtered and dried (1.2 g, 92%) and recrystallized from
ethanol, mp 126e128 ꢁC (reported 124e125 ꢁC,2d 125e126 ꢁC2c for
the H-analog 1f). Note that some deuterated Reissert compounds
revert to the protonated form upon recrystallization from ethanol;
these compounds were recrystallized from hexane/ethyl acetate.
3.4. Growth of single crystals
Single crystals of the Reissert compounds suitable for diffrac-
tometry were grown by slowly cooling hot solutions in hexane/
ethyl acetate.
8. Popp, F. D.; McEwen, W. E. J. Polym. Sci., Part A, Polym. Chem. 1957, 79,
3773e3777.
9. (a) Gibson, H. W.; Popp, F. D.; Catala, A. C. J. Heterocycl. Chem. 1964, 1, 251e252;
(b) Gibson, H. W.; Popp, F. D.; Noble, A. C. J. Heterocycl. Chem. 1966, 3, 99e100;
(c) Popp, F. D.; Gibson, H. W.; Noble, A. C. J. Org. Chem. 1966, 31, 2296e2299.
10. Neumeyer, J. L.; Oh, K. H.; Weinhardt, K. K.; Neustadt, B. R. J. Org. Chem. 1969, 34,
3786e3788.
11. Jackson, A. H.; Stewart, G. W. J. Chem. Soc., Chem. Commun. 1971, 149e150.
12. (a) Popp, F. D. Heterocycles 1973, 1, 165e180; (b) Popp, F. D. Heterocycles 1980,
14, 1033e1043; (c) Blasko, G.; Kerkes, P.; Makleit, S. In The Alkaloids: Chemistry
and Pharmacology; Brossi, A., Ed.; Academic: NY, 1987; Vol. 21, pp 1e28; (d) Ni,
F.; Shang, H.; Yan, M.; Shi, X.; Yang, K. Zhongguo Yiyao Gongye Zazhi 2006, 37,
505e507.
13. Chrzanowska, M.; Rozwadowska, M. D. Chem. Rev. 2004, 104, 3341e3370.
14. (a) Takamura, M.; Funabashi, K.; Kanai, M.; Shibasaki, M. J. Am. Chem. Soc. 2000,
122, 6327e6328; (b) Takamura, M.; Funabashi, K.; Kanai, M.; Shibasaki, M. J.
Am. Chem. Soc. 2001, 123, 6801e6808; (c) Funabashi, K.; Ratni, H.; Kanai, M.;
Shibasaki, M. J. Am. Chem. Soc. 2001, 123, 10784e10785; (d) Ichikawa, E.; Suzuki, M.;
Yabu, K.; Albert, M.; Kanai, M.; Shibasaki, M. J. Am. Chem. Soc. 2004, 126,
11808e11809.
Acknowledgements
The authors are grateful for the technical assistance of former
colleagues Dr. John H. Fager (Union Carbide Corp.), Mr. F.C. Bailey
(Xerox Corp., deceased), Mr. Ronald J. Weagley (Xerox Corp.)
and Dr. Samuel Kaplan (Xerox Corp.) in obtention of some of the
NMR spectra. We acknowledge the National Science Foundation for
funds to purchase the Varian Unity 400 MHz, Inova 400 MHz and
Bruker 600 MHz NMR spectrometers (DMR-8809714, CHE-0131124
and CHE-0619382, respectively), the Agilent 6220 Accurate Mass
TOF LC/MS Spectrometer (CHE-0722638) and the Oxford Diffrac-
tion SuperNova X-ray diffractometer (CHE-0131128) at Virginia
Tech. We are also very appreciative of the constructive comments
and criticisms of the reviewers, who brought about significant
improvement in the precision and clarity of this paper.
15. Frisch, K.; Jorgensen, K. A. Org. Biomol. Chem. 2007, 58, 2966e2974.
16. (a) Reimann, E.; Erdle, W.; Weigl, C.; Polborn, K. Monatsh. Chem. 1999, 130,
313e326; (b) Reimann, E.; Grasberger, F.; Polborn, K. Monatsh. Chem. 2000, 131,
73e84; (c) Reimann, E.; Grasberger, F.; Polborn, K. Monatsh. Chem. 2000, 134,
991e1014; (d) Evain, M.; Pauvert, M.; Collet, S.; Guingant, A. Acta Crystallogr.,
Sect. E 2002, E58, o1121eo1122; (e) Sieck, O.; Schaller, S.; Grimme, S.; Liebscher,
J. Synlett 2003, 337e340; (f) Pauvert, M.; Collet, S. C.; Bertrand, M.-J.; Guingant,
Supplementary data
1H NMR spectra of 1c, 1f, 1o, 2c, 1m, 2m, 2n, 2r, 2s, 7, 8f, 9m, and
9n; low temperature 1H NMR spectra of 2j, 2l, and 2m; 13C NMR
spectra of 1o, 2n, 2o, 2p, and 2q; 19F NMR spectrum of 2p; COSY
results for 1o, 2m, and 2r; HMBC results for 1o; tables of ASIS re-
sults, low temperature 1H NMR results and solid state packing
parameters. Supplementary data associated with this article can be
found in the online version, at http://dx.doi.org/10.1016/
j.tet.2012.06.008.
ꢁ
A. Y.; Evain, M. Tetrahedron Lett. 2005, 46, 2983e2987; (g) Brozda, D.;
Hoffman, K.; Rozwadowska, M. D. Heterocycles 2006, 67, 119e122;
(h) Kanemitsu, T.; Yamashita, Y.; Nagata, K.; Itoh, T. Synlett 2006, 1595e1597;
(i) Reimann, E.; Renz, M. Monatsh. Chem. 2007, 138, 211e218; (j) Reimann, E.;
Hertel, R.; Krauss, J. Monatsh. Chem. 2008, 139, 673e684; (k) Wouters, J.;
Elasaad, K.; Norberg, B.; Graulich, A.; Liegois, J.-F. Eur. J. Med. Chem. 2010, 45,
3240e3244.
17. Gibson, H. W.; Berg, M. A. G.; Dickson, J. C.; Lecavalier, P. R.; Wang, H.; Merola, J.
S. J. Org. Chem. 2007, 72, 5759e5770.
18. (a) Chhabra, S. R.; Kershaw, J. R.; Uff, B. C. Tetrahedron Lett. 1967, 8, 3199e3200;
(b) Uff, B. C.; Kershaw, J. R.; Chhabra, S. R. J. Chem. Soc., Perkin Trans. 1 1974,
1146e1153.
19. (a) Gibson, H. W. Tetrahedron Lett. 1968, 5549e5551; (b) Gibson, H. W. J. Org.
Chem. 1973, 38, 2851e2857.
20. Gibson, H. W. J. Heterocycl. Chem. 1970, 7, 1169e1172.
References and notes
1. (a) Reissert, A. Chem. Ber. 1905, 38, 1603e1614; (b) McEwen, W. E.; Cobb, R. L.
Chem. Rev. 1955, 55, 511e549; (c) Popp, F. D. Adv. Heterocycl. Chem. 1968, 9,

H.W. Gibson et al. / Tetrahedron 68 (2012) 8052e8067
8067
21. Gibson, H. W.; Chesney, D. K.; Popp, F. D. J. Heterocycl. Chem. 1972, 9, 541e544.
22. (a) Allinger, N. L.; Szkrybalo, W. J. Org. Chem. 1962, 27, 4601e4603; (b) Rickborn,
B.; Jensen, F. R. J. Org. Chem. 1962, 21, 4606e4608.
23. Eliel, E. L.; Allinger, N. L.; Angyal, S. J.; Morrison, G. A. Conformational Analysis;
Interscience: New York, NY, 1965; (a) pp 436e442; (b) p 291.
37. Oh, J. J.; Park, I.; Wilson, R. J.; Peebles, S. A.; Kuczkowski, R. L.; Kraka, E.; Cremer,
D. J. Chem. Phys. 2000, 113, 9051e9059.
38. Although in substituted cyclohexanes, the methyl group is ‘larger’,39a because
the longer CeCl bond places the Cl atom further from neighboring pseudo-axial
protons relative to the shorter CeC bond in the methyl analog, comparison of
the rotational barriers of CeC bonds reveals that the chlorine atom exerts
a larger impediment than the methyl group, e.g., n-propane (3.4 kcal/mol)39b
versus chloroethane (3.56 kcal/mol)40 and 1-chloro-3,3-dimethylbutane
versus 2,3,3-trimethylbutane (0.90 kcal/mol difference).41 Likewise, in terms
of atropisomerism due to aryl-N rotation in N-arylsuccinimides42 and 3-
arylhydantoins the aryl-N rotational activation energy is higher (by 0.7e1.
24. Cooks, M. J.; Katritsky, A. R.; Pennington, F. C.; Semple, B. M. J. Chem. Soc. B 1969,
523e526.
25. Ayres, D. C.; Harris, J. A. J. Chem. Soc., Chem. Commun. 1969, 1135e1136.
26. In the case of coplanar protons in the w-relationship J values of w1 Hz are
expected. A pseudo-axial H1 would place it nearly orthogonal to H3 and most
likely result in a larger absolute value of the coupling constant with H3. These
differences arise because of different coupling mechanisms in the two cases. For
discussion see: Barfield, M.; Dean, A. M.; Fallick, C. J.; Spear, R. J.; Sternhell, S.;
Westerman, P. W. J. Am. Chem. Soc. 1975, 97, 1482e1492; Barfield, M.; Spear, R. J.;
Sternhell, S. Chem. Rev. 1976, 76, 593e624; Contreras, R. H.; Peralta, J. E. Prog.
Nucl. Magn. Reson. Spectrosc. 2000, 37, 321e425.
27. We have chosen to use the cis/trans nomenclature for clarity. Use of the E/Z
system is confusing because the CahneIngoldePrelog priority rules lead to
opposite designations of Z and E for the trans structures of the 3-H Reissert
compounds 1 and the 3-CH3 analogs 2.
5 kcal/mol43
Moreover, in diastereomeric (5S)-methyl-3-aryl-2,4-oxazlinediones the ortho-
chloro compound produces 58.8%
G¼0.210 kcal/mol) of the major di-
) for the ortho-Cl compound than the ortho-methyl analog.
(D
astereomer, while the ortho-methyl analog yields 57.0% (
D
G¼0.167 kcal/mol) of
the analogous diastereomer.44 On this basis Cl is considered to be ‘larger’ than
methyl.
39. Eliel, E. L.; Wilen, S. H. Stereochemistry of Organic Compounds; John Wiley and
Sons: New York, NY, 1994; (a) pp 696e697; (b) p 599.
40. Lide, D. R. J. Chem. Phys. 1959, 30, 37e39.
28. Stewart, W. E.; Siddall, T. H., III. Chem. Rev. 1970, 70, 517e551.
29. Siddall, T. H., III; Stewart, W. E.; Marston, A. L. J. Phys. Chem. 1968, 72,
2135e2141.
41. Anderson, J. E.; Pearson, H. J. Chem. Soc., Chem. Commun. 1972, 908e909.
42. Srivastav, K. K.; Verma, A. K.; Verma, S. M. Indian J. Chem., Sect. B: Org. Med.
Chem. 1993, 32, 1143e1148.
30. (a) Siddall, T. H., III; Garner, R. H. Tetrahedron Lett. 1966, 7, 3513e3517;
(b) Siddall, T. H., III; Garner, R. H. Can. J. Chem. 1966, 44, 2387e2394.
31. Siddall, T. H., III; Stewart, W. E. J. Chem. Soc., Chem. Commun. 1967, 393e394.
32. (a) Johnson, F. Chem. Rev. 1968, 68, 375e413; (b) Yokoyama, Y. Chem.dEur. J.
43. Colebrook, L. D.; Giles, H. G.; Granata, A.; Icli, S.; Fehlner, J. R. Can. J. Chem. 1973,
51, 3635e3639.
44. Demir, O; Dogan, I. Chirality 2003, 15, 242e250.
45. (a) Takahashi, H.; Tsuboyama, S.; Umezawa, Y.; Honda, K.; Nishio, M. Tetrahe-
dron 2000, 56, 6185e6191; (b) Nishio, M. CrystEngComm 2004, 6, 130e158;
(c) Nishio, M. Tetrahedron 2005, 61, 6923e6950; (d) Takahashi, O.; Kohno, Y.;
Nishio, M. Chem. Rev. 2010, 110, 6049e6076; (e) Nishio, M. Phys. Chem. Chem.
Phys. 2011, 13, 13873e13900.
€
ꢁ
2004, 10, 4388e4394; (c) Coombs, T. C.; Lushington, G. H.; Douglas, J.; Aube, J.
Angew. Chem., Int. Ed. 2011, 50, 2734e2737; (d) Volchkov, I.; Park, S.; Lee, D. Org.
Lett. 2011, 13, 3530e3533.
33. Spatz, D. M.; Popp, F. D. J. Heterocycl. Chem. 1968, 5, 497e499.
34. Popp, F. D.; Wefer, J. M. J. Heterocycl. Chem. 1967, 4, 183e187.
35. The authors wish to thank Dr. John M. Wefer for suggesting this procedure.
36. (a) Shvo, Y.; Taylor, E. C.; Mislow, K.; Raban, M. J. Am. Chem. Soc. 1967, 89,
4910e4917; (b) Kalyanam, N. J. Chem. Educ. 1983, 60, 635e636.
46. Tykarska, E.; Jaskolski, M.; Kosturkiewicz, Z. Acta Crystallogr., Sect. C: Cryst.
Struct. Commun. 1985, C41, 1792e1794.
47. Boekelheide, V.; Weinstock, J. J. Am. Chem. Soc. 1952, 74, 660e663.
48. Gibson, H. W. J. Heterocycl. Chem. 1989, 26, 361e364.