Stereodynamics of N-allyl-N-methyl-2-aminopropane. 1H and 13C{1H} DNMR studies. Molecular mechanics calculations

Article abstract of DOI:10.1021/jp970449n

N-Allyl-N-methyl-2-aminopropane (AMAP) is a relatively simple tertiary amine that has a chiral center at the pyramidal nitrogen. The ¹H and ¹³C{¹H} dynamic nuclear magnetic resonance (NMR) spectra of AMAP show a higher temperature decoalescence due to slowing inversion-rotation at nitrogen and a more complex decoalescence at lower temperatures due to slowing isolated rotation about carbon-nitrogen bonds. A well-defined ¹H NMR spectrum at 100 K is simulated accurately by invoking the presence of five equilibrium conformations including two conformers that interconvert rapidly at 100 K.

Full text of DOI:10.1021/jp970449n

5700
J. Phys. Chem. A 1997, 101, 5700-5706
1
13
Stereodynamics of N-Allyl-N-methyl-2-aminopropane. H and C{1H} DNMR Studies.
Molecular Mechanics Calculations
Jay H. Brown and C. Hackett Bushweller*
Department of Chemistry, UniVersity of Vermont, Burlington, Vermont 05405-0125
ReceiVed: February 5, 1997; In Final Form: May 16, 1997X
N-Allyl-N-methyl-2-aminopropane (AMAP) is a relatively simple tertiary amine that has a chiral center at
the pyramidal nitrogen. Racemization occurs by inversion-rotation at nitrogen. For each AMAP enantiomer,
conformational exchanges occur via isolated rotation about carbon-nitrogen and carbon-carbon bonds. The
1H and 13C{1H} dynamic NMR (DNMR) spectra of AMAP show a higher temperature decoalescence due to
slowing inversion-rotation at nitrogen (∆Gq ) 7.4 kcal/mol at 162 K) and a more complex decoalescence
at lower temperatures due to slowing isolated rotation about carbon-nitrogen bonds (∆Gq ) 5.1-5.5 kcal/
1
mol at 100-110 K). A well-defined H NMR spectrum at 100 K is simulated accurately by invoking the
presence of five equilibrium conformations including two conformers that interconvert rapidly at 100 K. A
two-letter designation is used to name the various conformations. The first letter defines the orientation of
the vinyl group with respect to the lone pair (G denotes gauche to the lone pair and to the N-methyl group;
G′ denotes gauche to the lone pair and the isopropyl group; A denotes anti to the lone pair). The second
letter defines the orientation of the isopropyl methine proton (G denotes gauche to the lone pair and to the
N-methyl group; G′ denotes gauche to the lone pair and to the allyl group; A denotes anti to the lone pair).
The major subspectrum at 100 K is assigned to a family of GG′ and G′G′ conformations (59 %) that interconvert
rapidly at 100 K. Other subspectra at 100 K are assigned to the GG (36 %), GA (≈3 %), and AA (≈2 %)
forms. The presence of dominant GG′, G′G′, and GG conformers is confirmed by the 13C{1H} NMR spectrum
at 100 K. Conformational energies and internal rotation barriers calculated by using Allinger’s MM2(87)
molecular mechanics computer program show good agreement with the NMR data. AMAP shows
stereodynamics very similar to N-ethyl-N-methyl-2-aminopropane.
Introduction
the tert-butyldialkylamines, the preferred route for exchange of
the tert-butyl methyl groups among molecular sites is via
concomitant tert-butyl rotation and nitrogen inversion.4
Steric crowding forces triisopropylamine into an equilibrium
conformation that has essentially C3h symmetry with an almost
trigonal planar nitrogen atom and all three methine C-H bonds
in the trigonal plane.22 In sterically crowded systems such as
(CH3CH2)2CHN(i-C3H7)223 and (t-C4H9)2CHN(CH3)2,24 there are
significant barriers to isolated rotation about carbon-nitrogen
bonds on an essentially trigonal planar nitrogen template. In
N-neopentyl-4-tert-butylpiperidine, the unusual but nevertheless
preferred conformation of the neopentyl group has the neopentyl
tert-butyl group eclipsing the nitrogen lone pair.25
Tertiary aliphatic amines institute an important class of
compounds. A complete understanding of the chemistry of such
amines requires insight into the molecular stereodynamics.
Interconversions among the equilibrium conformations of a
tertiary aliphatic amine occur via isolated rotation about single
bonds and via pyramidal inversion at nitrogen.1-3 The overall
inversion process is complex; it involves not only pyramidal
inversion at nitrogen but also concomitant or accompanying
rotation about carbon-nitrogen bonds.1-4 It is best character-
ized as an inVersion-rotation process. Isolated rotation barriers
(no inversion) about carbon-nitrogen bonds in simple aliphatic
amines are usually smaller than barriers to inversion-rotation
at nitrogen.1 The stereodynamics of many simple amines
including methylamine,5 dimethylamine,5 trimethylamine,5 ethyl-
amine,6 isopropylamine,7 ethylmethylamine,8 and dimethylethyl-
amine9 have been investigated by Raman, infrared, or micro-
wave spectroscopy and molecular orbital theory. Barriers to
pyramidal inversion have been measured in ammonia,10
methylamine,5a,11 dimethylamine,5c and trimethylamine.12
The combination of dynamic NMR (DNMR) spectroscopy13
and molecular mechanics calculations14 has been useful in
studying the stereodynamics of more sterically encumbered
tertiary aliphatic amines including diethylmethylamine,15 tri-
ethylamine,15,16 dibenzylmethylamine,1 tribenzylamine,17 iso-
propyldimethylamine,18 isopropylethylmethylamine,19 2-butyl-
ethylmethylamine,20 2-(diethylamino)propane21, 2-(dibenzyl-
amino)propane21, and a series of tert-butyldialkylamines.4 In
In some tertiary aliphatic amines that have relatively high
symmetry, inversion-rotation is “DNMR-invisible”.1 Chirality
induced by deuterium substitution in such amines has been used
successfully to observe inversion-rotation that is otherwise
DNMR-invisible.26
A number of factors will determine the relative energies of
the equilibrium conformations of a tertiary aliphatic amine
including the pyramidality at nitrogen and the steric and/or
electronic requirements of various substituents.1,2 In particular,
previous studies of tertiary aliphatic amines showed that the
methyl group of an N-ethyl substituent shows a strong, but not
exclusive, preference for the position gauche to the nitrogen
lone pair, as compared to the position anti to the lone
pair.15,16,19,21 The phenyl group on the N-benzyl substituent in
tertiary aliphatic amines shows a conformational preference
similar to methyl.17,21 While the stereodynamics of allylamine
have been studied by microwave spectroscopy,27 by gas phase
electron diffraction,28 and by ab initio SCF calculations,29 the
conformational preference of the N-allyl substituent in a tertiary
* Corresponding author. FAX: 802.656.8705. Internet: cbushwel@
zoo.uvm.edu
X Abstract published in AdVance ACS Abstracts, July 1, 1997.
S1089-5639(97)00449-0 CCC: $14.00 © 1997 American Chemical Society

Stereodynamics of N-Allyl-N-methyl-2-aminopropane
J. Phys. Chem. A, Vol. 101, No. 31, 1997 5701
Figure 1. Experimental 1H DNMR spectra (250 MHz) of the aliphatic
protons of N-allyl-N-methyl-2-aminopropane (AMAP; 3 % v/v in
CBrF3) in the left column and theoretical simulations in the right
column. Spectra are progressively offset to the left with increasing
temperature. The rate constants k1-k4 are defined in Scheme 1; ki is
the rate constant for inversion-rotation at nitrogen.
1
Figure 2. Subspectra invoked to simulate the H NMR spectrum of
the aliphatic protons of N-allyl-N-methyl-2-aminopropane (AMAP; 3
% v/v in CBrF3) at 100 K. See Table 1 for a listing of chemical shifts.
1
amine has not been assessed. This paper reports H and 13C-
130 K, inversion-rotation is slow on the NMR chemical
exchange time scale, while all isolated rotations about carbon-
carbon and carbon-nitrogen bonds remain fast.
{1H} DNMR studies of N-allyl-N-methyl-2-aminopropane
(AMAP). Inversion-rotation interconverts AMAP between
enantiomers that respectively have R and S absolute configura-
tions at nitrogen. For each enantiomer, there is a family of
conformations that interconvert via isolated rotation about
carbon-nitrogen bonds. This paper reports evidence for a
strong preference for those conformations that have the isopropyl
methyl groups respectively anti and gauche to the nitrogen lone
pair with the vinyl group gauche to the lone pair. There is
additional evidence for two minor conformations that have both
isopropyl methyl groups gauche to the lone pair. Molecular
mechanics calculations using the Allinger-Profeta amine force-
field are in good agreement with experiment.30
Below 130 K, a second more complex decoalescence occurs
due to slowing isolated rotation about the N-CH and N-CH2
nitrogen-carbon bonds.1,19 A resharpened spectrum is observed
at 100 K (Figure 1). Accurate simulation of the 100 K spectrum
can be achieved by invoking the five subspectra illustrated in
Figure 2.31 Superposition of the properly weighted subspectra
produces the composite spectrum under conditions of no
chemical exchange (second spectrum from bottom in Figure 2).
The bottom composite spectrum in Figure 2 incorporates specific
rates of chemical exchange that are necessary for an accurate
simulation at 100 K.19,21,31 Each and every methyl group
resonance was simulated as three identical chemical shifts; that
is, all isolated methyl rotations are fast at 100 K.
DNMR Studies
1
The H NMR spectrum (250 MHz) of AMAP (3 % v/v in
The spectrum at 100 K is dominated by two well-resolved
subspectra; one accounts for 59 % of the total spectral area, the
other 36 %. The 36 % subspectrum shows an NCH3 singlet at
δO 2.18, an unresolved isopropyl methine proton septet at δF
3.00 [efficient transverse relaxation (T2*) and a small rate of
CBrF3) at 200 K shows a doublet at δ 1.00 (3JHCCH ) 6.1 Hz;
C(CH3)2), a singlet at δ 2.10 (NCH3), a septet at δ 2.84 (3JHCCH
) 6.1 Hz; NCH), a doublet of triplets at δ 2.98 (3JHCCH ) 6.7
4
4
Hz; JHCCHdCH ) JHCCdCH ) 1.2 Hz; NCH2), and multiplets
at δ 5.08 (dCH2) and at δ 5.89 (-CHd). Below 170 K,
decoalescence occurs. At 130 K, the signal due to the methylene
protons is decoalesced into two resonances; the lower frequency
(higher field) signal overlaps the methine proton resonance
(Figure 1). At 130 K, the isopropyl methyl signal is decoalesced
into two differentially broadened resonances (Figure 1). Using
a well-established rationale, this decoalescence is assigned to
slowing inversion-rotation at the pyramidal nitrogen.1,19 At
3
chemical exchange obscure the JHCCH splitting], isopropyl
methyl signals at δU 1.23 and δY 0.92, and methylene protons
3
signals at δE 3.40 (2JEL ) -12 Hz; JHCCH ≈ 4 Hz) and δL
2.37 (3JHCCH ) 12 Hz). Based on established chemical shift
trends, the isopropyl methyl resonance at δU 1.23 is assigned
to a methyl group that is gauche to the nitrogen lone pair.1,15,18,19,21
The isopropyl methyl signal at δY 0.92 is assigned to a methyl
group that is anti to the lone pair.1,15,18,19,21 In the conformation

5702 J. Phys. Chem. A, Vol. 101, No. 31, 1997
Brown and Bushweller
SCHEME 1: Stable Equilibrium Conformations of
N-Allyl-N-methyl-2-aminopropane (AMAP): The
Conformations Shown are Optimized Equilibrium Forms
Computed by Using the MM2(87) Molecular Mechanics
Force-Field30
preference of the vinyl group in the GG conformation. NMR
parameters and conformational assignments are listed in Table
1.
The subspectrum that accounts for 59 % of the spectral area
shows an NCH3 singlet at δ 2.09, an isopropyl methine proton
resonance at δ 2.92, and isopropyl methyl signals at δ 1.16 and
δ 0.81. The isopropyl methyl resonance at δ 1.16 is assigned
to a methyl group that is gauche to the nitrogen lone pair and
that at δ 0.81 to a methyl group that is anti to the lone
pair.1,15,18,19,21 The isopropyl methyl chemical shifts reveal
unequivocally that, in the conformation or conformations that
give this subspectrum, one isopropyl methyl group is gauche
and the other anti to the lone pair. Having assigned the 36 %
subspectrum to the GG conformer, the 59 % subspectrum must
be assigned to the GG′ or G′G′ conformer, or to a family of
rapidly exchanging GG′ and G′G′ forms (Scheme 1).1,15,19
The methylene protons resonances in the 59 % subspectrum
occur in a narrow range between 2.9 and 3.1 ppm. This is
inconsistent with the presence of a single conformation (GG′
or G′G′) in which the methylene protons oriented anti and
gauche to the lone pair whould show a large chemical shift
difference.1,15,19,21,32 The observed small chemical shift differ-
ence is consistent with the presence of a family of GG′ and
G′G′ conformations that are exchanging rapidly at 100 K.1,15,19
Exchange occurs by isolated rotation about the N-CH2 bond
with vinyl passing the lone pair. Molecular mechanics calcula-
tions (Vide infra) predict that the isolated rotation barrier for
interconversion between essentially isoenergetic GG′ and G′G′
conformations (vinyl passes lone pair) is a minuscule 2.8 kcal/
mol. Consistent with this interpretation, an accurate theoretical
simulation of the methylene protons signals at 100 K can be
achieved by invoking exchange of magnetization between
equally populated AJ (δA 3.82, δJ 2.68, 2JHCH ≈ -10 Hz, 3JHCCH
(J to vinyl methine proton) ≈ 4 Hz) and NB (δN 2.25, δB 3.51,
3
2JHCH ≈ -10 Hz, JHCCH (B to vinyl methine proton) ≈ 4 Hz)
spin systems with a rate constant k1 of 104 s-1 at 100 K (Figure
1). The methylene protons spectrum at 100 K reveals the onset
of slowing exchange between GG′ and G′G′ conformations. The
approximate free energy of activation is 3.8 kcal/mol at 100 K.
The signals for the diastereotopic methyl groups fall, respec-
tively, in narrow chemical shift ranges1,15,18,19,21 and are not
perceptibly affected at 100 K by the onset of slowing exchange
between the GG′ and G′G′ conformations. Thus, the 59 %
subspectrum is rationalized in terms of two subspectra that are
undergoing rapid magnetization exchange at 100 K. The two
subspectra employed in the composite theoretical simulations
of the aliphatic region of the spectrum over a wide temperature
range (Figure 1) are illustrated at the top of Figure 2. The
various NMR parameters are compiled in Table 1.
that gives this subspectrum, one isopropyl methyl group is
gauche and the other anti to the lone pair. The chemical shift
difference between the methylene protons (∆δ 1.03) is char-
acteristic of NCH2 protons that are “locked” in positions gauche
and anti to the lone pair.1,15,19,21,32 The resonance at δE 3.40 is
assigned to a methylene proton that is gauche to the lone pair;
the signal at δL 2.37 is due to a methylene proton that is anti to
the lone pair. Based on the Karplus relationship between 3JHCCH
3
and the HCCH torsion angle, the large JHCCH value (12 Hz)
reveals that the methylene proton that is anti to the lone pair is
also oriented anti to the vinyl methine proton.
The 36 % subspectrum is assigned to the GG conformer
(Scheme 1).33 In the GG form, the isopropyl methine proton
and the vinyl group are both gauche to the lone pair and to the
N-methyl group (G orientation).33 The vinyl moiety is “locked”
in the G position by the isopropyl methyl group that is gauche
to the lone pair. This is reflected by the large chemical shift
difference between the methylene protons (∆δ 1.03).1,15,19,21,32
While isolated rotation about the N-CH2 bond that interconverts
GG and G′G forms (vinyl group passes lone pair) is expected
to have a barrier lower than 4.5 kcal/mol (exchange is fast at
100 K),15,16,19 any significant concentration of the G′G conformer
is precluded by highly destabilizing syn-1,5-repulsions between
isopropyl methyl and vinyl groups. Any exchange between the
GG conformer and a minuscule concentration of the G′G
conformer at 100 K will not result in any perceptible time-
averaging of the methylene protons signals.15,19 The large
coupling of the vinyl methine proton to the methylene proton
that is anti to the lone pair shows that these two protons prefer
to be anti to each other, thus establishing the conformational
1
After an exhaustive attempt to do so, the experimental H
DNMR spectra could not be simulated accurately over a wide
temperature range by invoking only the three subspectra
discussed above. A line-on-line fit could not be achieved at
100 K, and serious discrepancies could not be corrected at higher
temperatures and higher rates of exchange. In a manner strictly
analogous to that reported for N-ethyl-N-methyl-2-aminopro-
pane,19 accurate fits were achieved by invoking two additional
minor subspectra. In one of the minor subspectra (≈2 % at
100 K) internally consistent fits over a temperature range
incorporated both methylene protons chemical shifts near δ 3.42
(both gauche to lone pair; the Vinyl group is anti to lone pair)
and both isopropyl methyl chemical shifts at δ 1.00. The
presence of the methylene protons signals is manifested by a
“distortion” (increased intensity) of the higher frequency
component of the gauche methylene proton doublet at δE 3.40
and the GG conformer (Figure 1; 100 K). An N-methyl

Stereodynamics of N-Allyl-N-methyl-2-aminopropane
J. Phys. Chem. A, Vol. 101, No. 31, 1997 5703
1
TABLE 1: Conformational Assignments and Aliphatic Proton NMR Parameters Employed To Simulate the H DNMR
Spectra of AMAPa
conformationsc
proton(s)b
GG (36 %)
G′G′ or GG′ (29.5%)
GG′ or G′G′ (29.5%)
GA (3%)
AA (2%)
d
d
CH(CH3*)2
δU 1.23e
δV 1.16e
δ
δ
V′ 1.16e
Z′ 0.81f
δG 2.94
δ
δ
W 1.00e
δX 1.00e
δY 0.92f
δZ 0.81f
W′ 1.00e
δ
X′ 1.00e
CH*(CH3)2
NCH3
NCH2
δF 3.00
δH 2.90
δS 2.00
δT 2.00
δO 2.18
δQ 2.09
δR 2.09
δP 2.16
δK 2.40
δE 3.40e
δA 3.82e
2JAJ ≈ -10
3JHCCH ≈ 4
δJ 2.68f
δB 3.51e
δI 2.86e
δC 3.42e
2JEL ) -12g
3JHCCH ≈ 4
δL 2.37f
2JBN ≈ -10
3JHCCH ≈ 4
δN 2.25f
2JIM ≈ -12
2JCD ≈ -10
3JHCCH ≈ 4
δD 3.42e
3JHCCH ≈ 4
δ
M 2.37f
3JHCCH ) 12
3JHCCH ≈ 4
3JHCCH ≈ 4
3JHCCH ≈ 11
3JHCCH ≈ 4
a 100 K in CBrF3; TMS at 0.0 ppm. b Indicated by an asterisk. c See Scheme 1; see Figure 2 for a spectral decomposition at 100 K.
6 Hz. e Gauche to nitrogen lone pair. f Anti to nitrogen lone pair. g Hz.
J
)
d 3
HCCH
1
TABLE 2: Relative Free Energies of the Stable
TABLE 3: DNMR-Visible H Magnetization Transfers
1
Equilibrium Conformations of AMAP Determined by H
NMR Spectroscopy
Associated with Internal Rotation in AMAP
rate
relative free energy (kcal/mol)a
magnetization constant conformational
∆Gq
conformation
substituent
transfera
(s-1
)
exchangeb
GG to GA
(kcal/mol)
GG′/G′G′ family
0.0
0.1
0.6
0.7
CH(CH3)2 FUY to SWW′
(or SW′W)
k2
5.1 ( 0.3c
GG
GA
AA
HVZ to SWW′
(or SW′W)
GV′Z′ to SWW′
(or SW′W)
k3
k3
k4
G′G′/GG′
to GA
G′G′/GG′
to GA
5.5 ( 0.3d
5.5 ( 0.3d
5.5 ( 0.5d
a At 100 K.
resonance at δK 2.40 (Table 1) was also invoked to achieve
accurate fits of the spectral region near δL 2.37, i.e., near the
GG anti methylene proton triplet (Figure 1). This subspectrum
must be assigned to the AA conformation (Scheme 1). In the
remaining subspectrum (≈3 % at 100 K), the chemical shifts
invoked for the isopropyl methyl groups are identical at δW 1.00
and δW′ 1.00. Methylene protons resonances were invoked at
δI 2.86 (gauche to lone pair) and δM 2.37 (anti to lone pair)
with a geminal coupling constant of about -12 Hz and a vicinal
coupling of the anti proton to the vinyl methine proton of about
11 Hz. Invoking the methylene proton signal at δL 2.37 and
the N-methyl resonance at δK 2.40 assigned to the AA
conformer (Vide supra) was required for accurate fits of the
spectral region near δL 2.37, i.e., near the GG methylene proton
triplet (Figure 1). The chemical shift difference between the
methylene protons signals suggests methylene protons that are
oriented anti and gauche to the lone pair. This subspectrum is
logically assigned to the GA conformation (Scheme 1). In fact,
the GA conformer is the only remaining stable conformation
of AMAP that does not have destabilizing 1,5-repulsions
between methyl or vinyl groups. It can be assigned by default.
The estimated errors in measurement of all conformational
populations is (1 % at 100 K. Because the error in temperature
measurement at these low temperatures is (3 K and because it
is not possible to measure conformer populations over a
significant temperature range under conditions of slow exchange,
no attempt was made to determine thermodynamic parameters
(∆H° and ∆S°) for the various conformational equilibria. The
relative free energies of the stable, NMR-detectable conforma-
tions at 100 K are compiled in Table 2.
SWW′ to TXX′
(or TX′X)
O to P
Q to P
GA to AA
NCH3
NCH2
k2
k3
GG to GA
G′G′/GG′
to GA
G′G′/GG′
to GA
GA to AA
GG to GA
G′G′/GG′
to GA
G′G′/GG′
to GA
5.1 ( 0.3c
5.5 ( 0.3d
R to P
k3
5.5 ( 0.3d
P to K
EL to IM
BN to IM
k4
k2
k3
5.5 ( 0.5d
5.1 ( 0.3c
5.5 ( 0.3d
AJ to MI
k3
k4
5.5 ( 0.3d
5.5 ( 0.5d
IM to CD
(or DC)
GA to AA
a See Table 1 and Figures 1 and 2. b See Scheme 1. c At 100 K. d At
110 K.
the DNMR spectra was to start with the 100 K spectrum and,
as the sample temperature increased, to turn on or increase only
those rates that are necessary and sufficient to achieve accurate
line shape simulations (Figure 1). The magnetization transfers
used to simulate the DNMR spectra up to about 130 K, assigned
conformational interconversions, and free energies of activation
are listed in Table 3. Because the error in temperature
measurement at these low temperatures is (3 K and because
the decoalescence phenomena occur over a relatively short
temperature range, we hesitate to report ∆Hq and ∆Sq values
for the rate processes detected. However, using the rate
constants derived from the line shape simulations, ∆Sq values
for all processes are equal to 0 ( 3 cal/(mol K).
The primary internal motion associated with all of these
DNMR-visible processes involves isolated rotation about
carbon-nitrogen bonds (Scheme 1). As temperature increased,
it was necessary to progressively increase the populations of
the minor species. For example, the populations of the GA
conformer are 4 % and 5 % at 115 and 130 K, respectively.
The populations of the AA form are 3 % and 4 % at 115 and
130 K, respectively. Attempts to simulate the exchange-
broadened DNMR spectra above 100 K without invoking the
minor subspectra failed, providing strong circumstantial evi-
dence for the presence of the minor conformations. In simulat-
ing the DNMR spectra, it was not necessary to invoke the k5,
k6, k7, and corresponding reverse processes (Scheme 1) to
All of the stable, equilibrium conformations of one invertomer
of AMAP (S configuration at nitrogen) are illustrated in Scheme
1. The conformations in Scheme 1 interconvert by isolated
rotation about carbon-nitrogen bonds with accompanying
geometrical optimization of the vinyl group.
Theoretical simulations of the 1H DNMR spectra over a wide
temperature range allowed identification of the preferred
pathways for conformational exchange in AMAP. The spectrum
at 100 K is simulated accurately by invoking the five subspectra
illustrated in Figure 2 with rapid exchange of magnetization at
100 K between the subspectra due to the G′G′ and GG′
conformations (k1 ≈ 104 s-1). The approach used in simulating

5704 J. Phys. Chem. A, Vol. 101, No. 31, 1997
Brown and Bushweller
achieve accurate fits of the spectra. It is apparent that the rates
of these exchanges are sufficiently slower than the DNMR-
detected processes (Table 3) that they do not affect the DNMR
line shape.
Above 130 K, it became necessary to turn on random
exchange among all N-methylene protons signals and among
all isopropyl methyl protons resonances for all conformations.
This reflects the onset of inversion-rotation and is associated
with rate constant ki in Figure 1 (∆Gq ) 7.4 kcal/mol at 162
K).1
The preferred pathways for conformational exchange (via
isolated rotation) involve the k1-k4 and reverse rate processes
in Scheme 1. As described above, the barrier for the k1 process
is below the lower limit for accurate DNMR measurement in
our laboratory (4.5 kcal/mol). The k2-k4 processes all involve
one approximately 120° rotation about a carbon-nitrogen bond.
During the k4 conversion, vinyl must pass N-methyl and C-H
bonds respectively pass the one pair and isopropyl group. The
k2 process involves C-methyl passing N-methylene, C-H
passing N-methyl, and C-methyl passing the lone pair. The k3
conversion (GG′ to GA) involves C-methyl passing N-methyl,
C-H passing N-methylene, and C-methyl passing the loan pair.
During all of these processes, at least one C-C bond must pass
an N-C bond, leading to rotation barriers high enough to be
DNMR-visible (5.1-5.5 kcal/mol).15,16,18-21 The direct k5
conversion (i.e., GG to GG′) must occur via a pathway that
involves essentially two simultaneous C-methyl/N-methyl and
C-methyl/N-methylene eclipsings. The GG to GG′ conversion
can also occur via the GA form as an unstable intermediate;
this requires two sequential 120° rotations, each of which
involves just one C-C/N-C passing. Molecular mechanics
calculations (Vide infra) for AMAP predict a significantly higher
barrier (8.4 kcal/mol) for the former process (k5) than each of
the latter sequential conversions (5.5 kcal/mol). The direct
conversion is predicted to occur at a relative rate too slow to
affect the DNMR line shape, as observed. Similar behavior is
observed for N,N-dimethyl-2-aminopropane.18 The k6 and k7
processes require two sequential 120° rotations and, for reasons
of probability and possibly the intervention of highly unstable
equilibrium conformations, occur at rates slower than the k1-
k4 conversions. They do not perceptibly affect the DNMR line
shape. Thus, the preferred exchange manifold for AMAP
involves primarily a series of single 120° rotations about C-N
bonds. For example, starting with the G′G′ form (Scheme 1),
a preferred interconversion itinerary involves G′G′ going to GG′
to GA, which can proceed to either AA or GG. During all these
processes, the vinyl group rotates to adopt an optimum equi-
librium geometry. The family of GG′ and G′G′ conformations
interconverts with the GG form via the less stable GA conformer
as an intermediate.
Figure 3. 13C{1H} DNMR spectra (62.898 MHz) of the aliphatic
carbons of N-allyl-N-methyl-2-aminopropane (AMAP; 3 % v/v in
CBrF3). Spectra are progressively offset to the left with increasing
temperature. The resonances at 100 K labeled with open circles are
assigned to the GG conformation. The resonances labeled with solid
circles are assigned to the family of rapidly interconverting GG′ and
G′G′ conformations.
populations of approximately 65 % and 35 %. The 35 %
subspectrum shows isopropyl methyl signals at δ 12.00 and
21.68 (methyl groups anti and gauche to lone pair1,18,19,21) and
other resonances at δ 41.65, 51.89, 55.44, 119.05, and 139.91.
These aliphatic carbon signals are labeled with open circles on
the 100 K spectrum in Figure 3. Consistent with the analysis
In contrast to the isolated rotation barriers, the inversion-
rotation barrier in AMAP is significantly higher at 7.4 kcal/
mol.
The 13C{1H} NMR spectrum (62.9 MHz) of AMAP (3 %
v/v in CBrF3) at 210 K shows six singlets at δ 17.81 (isopropyl
methyl carbons), δ 37.04 (N-methyl carbon), δ 53.58 (N-
methylene carbon), δ 58.51 (isopropyl methine carbon), δ
117.97 (vinyl methylene carbon), and δ 138.32 (vinyl methine
carbon). At 162 K, the resonance due to the prochiral isopropyl
methyl groups is differentially broadened and almost decoa-
lesced, while all other signals remain sharp (Figure 3). This is
consistent with the onset of slowing nitrogen inversion while
all isolated rotation processes remain fast.1 Below 162 K,
another decoalescence occurs and a sharpened spectrum is
observed at 100 K (Figure 3). Simulation of the 100 K spectrum
confirmed the presence of two dominant subspectra with
1
of the H NMR spectrum at 100 K (Vide supra), these signals
are assigned to the GG conformation. The resonances due to
the 65% subspectrum are labeled with solid circles on the 100
K spectrum in Figure 3. This subspectrum shows two isopropyl
methyl signals at δ 10.77 and 21.68, revealing methyl groups
that are respectively anti and gauche to the lone pair1,18,19,21 and
other resonances at δ 34.50, 49.42, 59.78, 118.76, and 138.95.
These signals are logically assigned to the family of rapidly
exchanging GG′ and G′G′ conformations. It is noteworthy that
the signals in this subspectrum are broader than those in the 35
% subspectrum, consistent again with the onset of slowing

Stereodynamics of N-Allyl-N-methyl-2-aminopropane
J. Phys. Chem. A, Vol. 101, No. 31, 1997 5705
TABLE 4: Relative Energies of the Stable Equilibrium
Conformations of AMAP Calculated by the MM2(87)
Molecular Mechanics Force-Field
TABLE 5: Selected Torsion Angles, Bond Angles, and
Bond Lengths for the Stable, Equilibrium Conformations of
AMAP Calculated by the MM2(87) Molecular Mechanics
Force-Field
conformation
relative energy (kcal/mol)
torsion angle (deg)
bond angle (deg)
GG′ Conformera
bond length (Å)
GG′
G′G′
GG
GA
AA
0.00a
0.07
0.13
0.21
0.26
3-2-1-8
4-2-1-8
2-1-5-6
1-5-6-7
61.9
-64.3
160.7
121.0
2-1-5
2-1-8
5-1-8
112.3
113.7
110.2
1-2
1-5
1-8
1.466
1.461
1.456
a ∆Hf ) 0.30 kcal/mol.
G′G′ Conformera
3-2-1-8
4-2-1-8
2-1-5-6
1-5-6-7
64.0
-62.0
60.7
2-1-5
2-1-8
5-1-8
113.0
113.5
109.7
1-2
1-5
1-8
1.467
1.461
1.457
interconversion between the GG′ and G′G′ conformations (Vide
supra). There are no low-intensity signals observed that could
be assigned to the GA or AA forms. Any such signals that are
below 3 % the intensity of the major resonances could easily
be lost in the noise on the 100 K spectrum. Improving the
signal-to-noise on the 100 K spectrum was precluded by the
danger to the magnet system of remaining at such a very low
temperature for the required spectrum acquisition time. The
13C NMR spectrum of AMAP does confirm the presence of
dominant GG, GG′, and G′G′ conformations.
-121.1
GG Conformera
2-1-5
2-1-8
5-1-8
3-2-1-8
4-2-1-8
2-1-5-6
1-5-6-7
-66.6
167.2
165.1
120.4
114.0
112.0
110.2
1-2
1-5
1-8
1.466
1.460
1.458
AA Conformera
2-1-5
2-1-8
5-1-8
3-2-1-8
4-2-1-8
2-1-5-6
1-5-6-7
175.1
53.9
-59.7
117.0
113.4
112.5
110.6
1-2
1-5
1-8
1.469
1.464
1.458
Molecular Mechanics Calculations
GA Conformera
Molecular mechanics calculations based on the Allinger-
Profeta amine force-field in the MM2(87) computer program
agree with the stereodynamics deduced from the DNMR
studies.30 Five stable, equilibrium conformations were identified
(Scheme 1). All five are predicted to be present at concentra-
tions high enough to be NMR-detectable. For each stable form,
the geometry was optimized by using the standard Newton-
Raphson energy minimization scheme. MM2(87) predicts the
GG′ conformation (∆Hf ) 0.30 kcal/mol) to be most stable,
with the G′G′ form only 0.07 kcal/mol less stable than GG′.
The GG′ and G′G′ conformations are essentially isoenergetic.
The GG, GA, and AA conformers are calculated to be 0.13,
0.21, and 0.26 kcal/mol, respectively, less stable than the GG′
form. The relative conformational energies are listed in Table
4. The GG′, G′G′, and GG conformations are predicted to be
dominant, consistent with the 1H DNMR data. Selected torsion
angles, bond angles, and bond lengths are compiled in Table 5.
In the GG′, G′G′, and GG conformers, MM2(87) predicts the
torsion angle between the methylene proton that is anti to the
lone pair and the vinyl methine proton to be 177°, 176°, and
177°, respectively. Based on the Karplus relationship for
3-2-1-8
4-2-1-8
2-1-5-6
1-5-6-7
179.4
58.1
-60.8
-107.2
2-1-5
2-1-8
5-1-8
114.0
112.3
109.8
1-2
1-5
1-8
1.469
1.463
1.459
a The atomic numbering scheme is given in Scheme 1.
TABLE 6: Selected Barriers to Isolated Rotation in AMAP
Calculated by Using the Dihedral Driver Option in the
MM2(87) Molecular Mechanics Force-Field
processa
barrier (kcal/mol)
G′G′ to GG′
GG′ to GA
GG to GA
GA to AA
GG to GG′
2.79
5.56
5.49
4.55
8.42
a The various conformations are illustrated in Scheme 1.
By using the general dihedral angle driver option in MM2-
(87), isolated rotation barriers about nitrogen-carbon bonds in
AMAP were calculated (Table 6). Consistent with the time-
averaged character of the major 1H NMR subspectrum of AMAP
(Figure 2) and the presumption of rapid exchange between
essentially isoenergetic GG′ and G′G′ conformations (vide
supra), the MM2(87) barrier for direct interconversion of the
GG′ and G′G′ conformations (vinyl passes lone pair) is a
DNMR-invisible 2.79 kcal/mol. Those isolated rotations that
involve methyl passing methyl (GG′ to GA) and methyl passing
allyl (GG to GA) have MM2(87) barriers that are high enough
to be DNMR-visible (Table 6). The GA to AA conversion
(vinyl passes methyl) has a barrier that is calculated to be just
above the lower limit for DNMR detection.18 Consistent with
the DNMR data, isolated rotations that involve CCH3/NCH3,
CCH3/NCH2 or CCHdCH2/NCH3 eclipsings are predicted to
be high enough to be DNMR-visible.
The stereodynamics of AMAP and N-ethyl-N-methyl-2-
aminopropane (EMAP) and very similar.19 The populations of
dominant GG, GG′, and G′G′ conformations are comparable.
The minor GA and AA forms are also present at comparable
concentrations. Methyl and vinyl groups prefer to locate in the
sterically less crowded position gauche to the lone pair and not
in the sterically more crowded position anti to the lone pair. If
one uses the A-value as a measure of the steric size of a
substituent,34 the similarity in stereodynamics for EMAP and
AMAP is not surprising in light of the comparable A-values
3
3JHCCH, the large, directly observed JHCCH coupling (12 Hz)
between the N-methylene proton that is anti to the lone pair
and the vinyl methine proton for the GG form is consistent with
the MM2(87) calculation. Due to the rapid exchange between
3
GG′ and G′G′ conformations, such JHCCH values cannot be
measured directly for these species. In the GA form, the torsion
angle between the methylene proton that is anti to the lone pair
and the vinyl methine proton is calculated to be 178°. In the
AA conformation, the torsion angle between the N-methylene
proton that is proximate to the methyl group and the vinyl
methine proton is calculated to be 172°.
It is reasonable to assume that the entropy difference between
any two AMAP conformations will be small. At temperatures
as low as 100 K, entropy contributions (T∆S°) to the free energy
difference will be small. Therefore, the free energy differences
compiled in Table 2 are good estimates of the enthalpy
differences between conformers. While the NMR data do not
allow an accurate measure of the relative free energies of the
GG′ and G′G′ forms, there is no question that they are
comparably populated. A perusal of Tables 2 and 4 shows good
agreement between the conformational energies derived from
the NMR data and those calculated by the MM2(87) force field.

5706 J. Phys. Chem. A, Vol. 101, No. 31, 1997
Brown and Bushweller
for methyl (1.7 kcal/mol)35 and vinyl (1.5-1.7 kcal/mol).36
N-Ethyl and N-allyl substituents show very similar conforma-
tional preferences in analogous tertiary amines.
(3) Orville-Thomas, W. J., Ed. Internal Rotation in Molecules; J. Wiley
and Sons: New York, 1974.
(4) Bushweller, C. H.; Anderson, W. G.; Stevenson, P. E.; Burkey, D.
L.; O’Neil, J. W. J. Am. Chem. Soc. 1974, 96, 3892. Also see: Bushweller,
C. H.; Anderson, W. G.; Stevenson, P. E.; O’Neil, J. W. J. Am. Chem. Soc.
1975, 97, 4338.
(5) (a) Tsuboi, M.; Hirakawa, A. Y.; Tamagake, K. J. J. Mol. Spectrosc.
1967, 22, 272. (b) Nishikawa, T.; Itoh, T.; Shimoda, K. J. Am. Chem. Soc.
1977, 99, 5570. (c) Wollrab, J. W.; Laurie, V. W. J. Chem. Phys. 1968,
48, 5058. (d) Erlandson, G.; Gurdy, W. Phys. ReV. 1957, 10, 513. (e)
Lide, D. R.; Mann, D. E. J. Chem. Phys. 1958, 28, 572.
(6) Durig, J. R.; Li, Y. S. J. Chem. Phys. 1975, 63, 4110. Tsuboi, M.;
Tamagake, K. J.; Hirakawa, A. Y.; Yamaguchi, J.; Nakagava, H.; Manocha,
A. S.; Tuazon, E. C.; Fateley, W. G. J. Chem. Phys. 1975, 63, 5177.
(7) Krueger, P. J.; Jan, J. Can. J. Chem. 1970, 48, 3229.
(8) Durig, J. R.; Compton, D. A. C. J. Phys. Chem. 1979, 83, 2873.
(9) Durig, J. R.; Cox, F. O. J. Mol. Struct. 1982, 95, 85.
(10) Swalen, J. D.; Ibers, J. A. Chem. Phys. 1962, 36, 1914.
Experimental Section
NMR Spectra. The NMR spectra were recorded by using a
Bruker WM-250 NMR system with a pole gap modified to allow
safe operation (no magnet O-ring freezing) down to 93 K. NMR
sample temperature was varied by using a custom-built cold
nitrogen gas delivery system used in conjunction with the Bruker
BVT-1000 temperature control unit. Temperature measurement
is accurate to (3 K. NMR samples were prepared on a vacuum
line in precision 5 or 10 mm tubes and sealed after four freeze-
pump-thaw cycles. All spectra are referenced to tetramethyl-
silane at 0 ppm.
(11) Tsuboi, M.; Hirakawa, A. Y.; Takamitsu, I.; Sasaki, T.; Tamagake,
K. J. J. Chem. Phys. 1964, 41, 2721.
(12) Weston, R. E., Jr. J. Am. Chem. Soc. 1954, 76, 2645.
N-Allyl-N-methyl-2-aminopropane (AMAP). Allylamine
(4.0 g, 0.07 mol) was added to a 500 mL three-neck round
bottom flask equipped with a drying tube. Methanol (250 mL)
was added to the flask. With cooling and stirring, acetone (8.2
g, 0.14 mol) and sodium cyanoborohydride (9.0 g, 0.14 mol)
were added to the flask. The mixture was stirred at room
temperature for 3 days. With cooling and stirring, the mixture
was acidified to pH < 3 by using concentrated HCl (20 mL).
The liquid was removed under vacuum. The resulting yellow
residue was dissolved in 50 mL of water, and the mixture was
washed with five 50 mL portions of diethyl ether. The aqueous
layer was separated and added to a three-neck round bottom
flask fitted with an efficient condenser. Diethyl ether (50 mL)
was added to the flask. With cooling and stirring, solid KOH
(15 g) was added to pH > 10. NaCl (5 g) was then added to
the mixture. The mixture was stirred at room temperature for
24 h and filtered. The ether layer was separated, dried over
Na2SO4 for 2 h, and filtered. The presence of N-allyl-2-
(13) Oki, M. In Applications of Dynamic NMR Spectroscopy to Organic
Chemistry; VCH Publishers: New York, 1985. Sandstrom, J. Dynamic
NMR Spectroscopy; Academic Press: New York, 1982. Cotton, F. A.
Dynamic Nuclear Magnetic Resonance Spectroscopy; Academic Press: New
York, 1975.
(14) Burkert, U.; Allinger, N. L. Molecular Mechanics; American
Chemical Society: Washington, DC, 1984. Allinger, N. L. AdV. Phys. Org.
Chem. 1976, 13, 1.
(15) Bushweller, C. H.; Fleischman, S. H.; Grady, G. L.; McGoff, P.;
Rithner, C. D.; Whalon, M. R.; Brennan, J. G.; Marcantonio, R. P.;
Domingue, R. P. J. Am. Chem. Soc. 1982, 104, 6224.
(16) Fleischman, S. H.; Weltin, E. E.; Bushweller, C. H. J. Comput.
Chem. 1985, 6, 249.
(17) Fleischman, S. H.; Whalon, M. R.; Rithner, C. D.; Grady, G. L.;
Bushweller, C. H. Tetrahedron Lett. 1982, 4233.
(18) Brown, J. H.; Bushweller, C. H. J. Am. Chem. Soc. 1992, 114,
8153.
(19) Brown, J. H.; Bushweller, C. H. J. Phys. Chem. 1994, 98, 11411.
(20) Danehey, C. T., Jr.; Grady, G. L.; Bonneau, P. R.; Bushweller, C.
H. J. Am. Chem. Soc. 1988, 110, 7269.
(21) Brown, J. H.; Bushweller, C. H. J. Am. Chem. Soc. 1995, 117,
1
aminopropane was confirmed by H NMR. The solution of
12567.
N-allyl-2-aminopropane in diethyl ether was added to a three-
neck round bottom flask fitted with an efficient condenser.
Formaldehyde (37 %; 11.5 g, 0.14 mol) was then added to the
diethyl ether solution. With cooling and stirring, formic acid
(97 %; 6.6 g, 0.14 mol) was added dropwise. The mixture was
refluxed for 24 h. With cooling and stirring, concentrated HCl
(10 g) was added. The bulk of the liquid was removed under
vacuum leaving a wet orange solid amine hydrochloride. The
amine hydrochloride was placed in a three-neck round bottom
flask equipped with an efficient condenser. With cooling and
stirring, the amine hydrochloride was neutralized by the slow
addition of aqueous 40 % NaOH to pH > 10. The mixture
was filtered. The top layer was separated and dried over Na2-
SO4 for 2 h. N-Allyl-N-methyl-2-aminopropane (AMAP) was
purified on a 25 % SF-96/5 % XE-60 on Chromosorb W GLC
column (20 ft × 3/8 in.) at 423 K. The structure of AMAP
was confirmed by 1H and 13C NMR (see text) and mass
spectrometry m/e (M+): 113.
(22) Bock, H.; Coebel, I.; Havlas, Z.; Liedle, S.; Oberhammer, H. Angew.
Chem., Int. Ed. Engl. 1991, 30, 187.
(23) Lunazzi, L.; Macciantelli, D.; Grossi, L. Tetrahedron 1983, 39,
305.
(24) Berger, P. A.; Hobbs, C. F. Tetrahedron Lett. 1978, 1905.
(25) Forsyth, D. A.; Johnson, S. M. J. Am. Chem. Soc. 1993, 115, 3364.
(26) Forsyth, D. A.; Johnson, S. M. J. Am. Chem. Soc. 1994, 116, 11481.
(27) Roussy, G.; Demaison, J. J. Mol. Spectrosc. 1971, 38, 535. Botskor,
I.; Rudolph, H. D. J. Mol. Spectrosc. 1974, 53, 15. Botskor, I.; Rudolph,
H. D.; Roussy, G. J. Mol. Spectrosc. 1974, 52, 457. Botskor, I. J. Mol.
Spectrosc. 1978, 71, 430.
(28) Hamada, Y.; Tsuboi, M.; Yamanouchi, K.; Matsuzawa, T.; Ku-
chitsu, K. J. Mol. Struct. 1990, 224, 345.
(29) Kao, J.; Seeman, J. I. J. Comput. Chem. 1984, 5, 200.
(30) Allinger, N. L. QCPE 1987, Program No. MM2(87). Profeta, S.,
Jr.; Allinger, N. L. J. Am. Chem. Soc. 1985, 107, 1907.
(31) Brown, J. H.; Bushweller, C. H. QCPE 1993, Program No. 633.
For a PC-based program to plot the DNMR spectra, see: Brown, J. H.
QCPE 1993, Program No. QCMP 123.
(32) Hamlow, H. P.; Okuda, S. Tetrahedron Lett. 1964, 37, 2553.
Lambert, J. B.; Keske, R. G.; Carhart, R. E.; Jovanovich, A. P. J. Am. Chem.
Soc. 1967, 89, 3761.
Acknowledgment. C.H.B. is grateful to the National Science
Foundation (Grant CHE80-24931) and to the University of
Vermont Committee on Research and Scholarship for partial
support of this research.
(33) A two-letter designation is used to name the various conformations.
The first letter defines the orientation of the vinyl group with respect to the
lone pair (G denotes gauche to the lone pair and to the N-methyl group; G′
denotes gauche to the lone pair and to the isopropyl group; A denotes anti
to the lone pair). The second letter defines the orientation of the isopropyl
methine proton (G denotes gauche to the lone pair and to the N-methyl
group; G′ denotes gauche to the lone pair and to the allyl group; A denotes
anti to the lone pair).
References and Notes
(34) Bushweller, C. H. In Conformational BehaVior of Six-Membered
Rings. Analysis, Dynamics, and Stereoelectronic Effects; Juaristi, E., Ed.;
VCH Publishers: New York, 1995.
(1) For a recent review, see: Bushweller, C. H. In Acyclic Organo-
nitrogen Stereodynamics; Lambert, J. B., Takeuchi, Y., Eds.; VCH
Publishers: New York, 1992.
(35) Anet, F. A. L.; Bradley, C. H.; Buchanan, G. W. J. Am. Chem.
Soc. 1971, 93, 258. Booth, H.; Everett, J. R. J. Chem. Soc., Perkin Trans.
2 1980, 255.
(36) Eliel, E. L.; Manoharan, M. J. Org. Chem. 1981, 46, 1959.
Buchanan, G. W. Can. J. Chem. 1982, 60, 2908.
(2) For previous reviews, see: Lambert, J. B. Top Stereochem. 1971,
6, 19. Lehn, J. M. Fortschr. Chem. Forsch. 1970, 15, 311. Payne, P. W.;
Allen, L. C. In Applications of Electronic Structure Theory; Schaefer, H.
F., Ed.; Plenum Press: New York, 1977; Vol. 4. Rauk, A.; Allen, L. C.;
Mislow, K. Angew. Chem., Int. Ed. Engl. 1970, 9, 400.