19F NMR of linear N,N-difluoroaminoalkanes

Article abstract of DOI:10.1016/j.jfluchem.2012.02.013

Whereas most N,N-difluoroaminoalkanes exhibit a single ¹⁹F resonance at about +50 ppm, the tri-component mixtures of both N,N-difluoroaminopentanes and N,N-difluoroaminohexanes exhibited a more complex pattern. The individual 1-N,N-difluoroamino-, 2-N,N-difluoroamino- and 3-N,N-difluoroaminopentanes have been synthesized and their separated resonances are reported.

Full text of DOI:10.1016/j.jfluchem.2012.02.013

Journal of Fluorine Chemistry 137 (2012) 73–76
Contents lists available at SciVerse ScienceDirect
Journal of Fluorine Chemistry
journal homepage: www.elsevier.com/locate/fluor
19F NMR of linear N,N-difluoroaminoalkanes
*
Randolph K. Belter
Y-Not Chemical Consulting, Zachary, LA 70791, USA
A R T I C L E I N F O
A B S T R A C T
Whereas most N,N-difluoroaminoalkanes exhibit a single 19F resonance at about +50 ppm, the tri-
component mixtures of both N,N-difluoroaminopentanes and N,N-difluoroaminohexanes exhibited a
more complex pattern. The individual 1-N,N-difluoroamino-, 2-N,N-difluoroamino- and 3-N,N-
difluoroaminopentanes have been synthesized and their separated resonances are reported.
ß 2012 Elsevier B.V. All rights reserved.
Article history:
Received 24 February 2012
Accepted 27 February 2012
Available online 5 March 2012
Keywords:
N,N-difluoroaminoalkanes
Difluoroalkylamines
19F NMR
Nuclear magnetic resonance
1. Introduction
2. Results and discussion
1.1. 19F spectra of N,N-difluoroaminoalkanes
2.1. 19F spectra of N,N-difluoroaminopentane isomers
We have previously shown that Nitrogen Trifluoride (NF3) can
be reacted in a controlled manner with alkanes, cycloalkanes and
ethers in a 400 8C vapor phase reactor to generate N,N-difluor-
oaminoalkanes, -cycloalkanes and -ethers [1]. In most cases, a
singular resonance was observed in each 19F spectrum between
+25 and +55 ppm vs. CFCl3 (see Table 1). In the cases of n-
pentane + NF3 and n-hexane + NF3, three isomeric products were
generated. However, rather than three individual singlets, a
pattern of five equidistant peaks at about +39 ppm was observed
for both the N,N-difluoroaminopentanes mixture and the N,N-
difluoroaminohexanes mixture (see Fig. 1). In addition, a nearby
singlet at +56 ppm added more uncertainty to the assignment of
resonances to individual isomers.
To positively assign resonances and coupling values to each
individual isomer, and thus interpret the spectra of the mixed
isomers, we undertook to prepare each isomer of N,N-
difluoroaminopentane and N,N-difluoroaminohexane indepen-
dently and to observe the individual 19F spectra. The individual
N,N-difluoroaminopentanes and -hexanes were prepared by
reaction of the appropriate Grignard reagent with NF3 (see
Section 4).
2.1.1. 19F spectra of 1-N,N-difluoroaminopentane
1-N,N-difluoroaminopentane, 1, was prepared by the reaction
of NF3 with 1-bromomagnesiopentane. The 19F NMR spectrum of
1 shows that this compound is responsible for the single
resonance observed at +55.8 ppm in the mixed spectra. NMR
literature for 1-N,N-difluoroaminobutane [2] indicates a single
resonance at +54.6 ppm (sign inverted from
result.
F), supportive of our
2.1.2. 19F spectra of 2-N,N-difluoroaminopentane
2-N,N-difluoroaminopentane, 2, was prepared by the reaction
of NF3 with 2-bromomagnesiopentane. The 19F NMR spectrum of 2
is observed to be an AB pattern of two doublets, one for each
fluorine, at +43.1 ppm and +35.7 ppm. They are coupled
JF–F = 566 Hz.
Reference [3] reports that the spectrum of 2-N,N-difluoramino-
3-fluorobutane ‘‘had bands centered about +39.1’’ (again, sign
inverted from
result.
F and CF3CO2H standard), consistent with our
2.1.3. 19F spectra of 3-N,N-difluoroaminopentane
3-N,N-difluoroaminopentane, 3, was prepared by the reaction
of NF3 with 3-bromomagnesiopentane. The 19F NMR spectrum of 3
is observed to be a single resonance at +39.4 ppm. In light of the
widely separated doublets observed for 2-N,N-difluororamino-
pentane, we had expected a more complex pattern from 3,
superimposed upon the pattern of 2 in the mixed spectra. Such is
* Tel.: +1 225 658 8792.
E-mail address: randolphbelter@gmail.com.
0022-1139/$ – see front matter ß 2012 Elsevier B.V. All rights reserved.
doi:10.1016/j.jfluchem.2012.02.013

74
R.K. Belter / Journal of Fluorine Chemistry 137 (2012) 73–76
Table 1
Representative 19F resonances.
2-N,N-difluoroamino-2-methylpropane
d
d
+28.3 (s)
a-N,N-difluoroaminotetrahydrofuran
+31.9 (d, J = 66.5 Hz)
(a relatively small splitting)
N,N-difluoroaminocyclohexane
N,N-difluoroaminocyclopentane
N,N-difluoroaminobenzene
d
d
d
+42.3 (s)
+52.9 (s)
ꢀ63.2 (s)
not the case. However, the resonance of 3 is precisely positioned at
the center point of the 2-N,N-difluororaminopentane pattern! The
simplicity of the resonance for 3 can be attributed to the symmetry
of the molecule, where each
environments.
F is in equivalent magnetic
2.2. 19F spectra of N,N-difluoroaminohexane isomers
Fig. 1. 19F NMR of difluoroaminopentanes.
2.2.1. Analogous interpretation of the spectral pattern of mixed N,N-
difluoroaminohexane isomers
The 19F spectrum of the mixed N,N-difluoroaminohexane
isomers is identical to that of the pentane isomers (completely
superimposable!). By direct analogy, we can interpret the singlet
observed at +55.9 ppm to be the resonance of 1-N,N-difluoroami-
nohexane, 4. The spectrum of 2-N,N-difluoroaminohexane, 5, can
be extracted as two coupled AB doublets at +46.7 and +35.7 ppm,
JF–F = 566 Hz. The presumed spectrum of 3-N,N-difluoroaminohex-
ane, 6, gives us pause. As this molecule is NOT symmetrical like 3,
we would predict two resonances, one for each fluorine, as in 2 and
5. Since the mixed spectra has a central peak, we could explain this
by predicting the spectra of 6 to be two overlapping coupled
doublets at +40.6 and +38.1 ppm, JF–F = 566 Hz, but integration of
various fractions of the mixed isomers did not support this idea. To
be confident, we had no choice but independently prepare 5 and 6
and observe their spectra.
2.2.2. 19F spectra of 2-N,N-difluoroaminohexane
2-N,N-difluoroaminohexane, 5, was prepared by the reaction of
NF3 with 2-chloromagnesiohexane. As predicted, the 19F NMR
spectrum of 5 is observed to be two doublets, one for each fluorine,
at +43.1 ppm and +35.7 ppm. They are coupled JF–F = 566 Hz.
2.2.3. 19F spectra of 3-N,N-difluoroaminohexane
3-N,N-difluoroaminohexane, 6, was prepared by the reaction of
NF3 with 2-chloromagnesiohexane. The 19F NMR spectrum of 6 is
observed to be a single resonance at +39.3 ppm. Despite a formal
lack of symmetry in the molecule, the two fluorines of 6 are
apparently in identical environments. How the molecule is
attaining a symmetric configuration, at this point, we do not
know. None-the-less, the resonance of 6 is precisely positioned at
the center point of the 2-N,N-difluororaminopentane, 5, pattern
just as the resonance of 3 is centered on that of 2!
2.3. Analysis of the 2-NF2 vs. 3-NF2 spectral difference
In seeking an explanation for the stark difference in the 19F
resonances of 2-difluoroamino- and 3-difluoroaminoalkanes
Fig. 2. Calculated F–H distances of difluoroaminopentanes and difluoroaminohexanes.

R.K. Belter / Journal of Fluorine Chemistry 137 (2012) 73–76
75
Table 2
(particularly the lack of an AB pattern in 6), we evaluated each
F–C coupling constants from 13C spectra (
a carbon in bold).
molecule for differentiated interaction between the fluorine atoms
and nearby hydrogen atoms. This was first attempted theoretically
by availing ourselves of a configurational model for each isomer as
determined by Gaussian ab initio calculations [4]. Fig. 2 shows
Newman projections of the four molecules in question based on
the Gaussian results. It can be seen that for both 2-substituted
compounds, 2 and 5, each F atom is nearly equidistant and of equal
dihedrals to their respective vicinal protons. One would not expect
differentiation in the 19F signals from the slight differences.
Compound
JF–C1
JF–C2
JF–C3
JF–C4
JF–C5
JF–C6
2
3
5
6
9.8
10.0
9.8
0
6.1
8.2
6.2
8.6
6.5
5.3
6.7
5.5
0
0
10.0
0
–
–
0
0
8.2
0
7.8
0
Gaussian. In each case, all C–C bond lengths are 1.53–1.54 A. No
differentiation in signals could be expected from such identical
distances. Evaluation of the dihedral angle of the C3–C4 bonds of 2
and of 4 show each to be ꢁ1768 and nearly optimal for orbital
interaction with the N–F bond. The C1 side obviously has no such
possibility for interaction. Again, however, we are disappointed
when we evaluate the interactions of 3 and 6. We see almost no
potential for interaction with the C4–C5 bond as the dihedral
angles with the nearest F are close to the minimally interactive 908
(898 and 928). However, this time, there is the strong potential for
interaction with the C1–C2 bond with the nearest F with dihedral
angles of ꢁ1668 for both 3 and 6. Of course, the C1–C2 and C4–C5
interactions are inversed for the more distant F’s, but the
expectation of interaction is the same. From the calculated
Conversely, both 3-substituted compounds,
3 and 6, have
distinctly different distances between the F atoms and their
vicinal neighbors (2.3A vs. 2.8A). However, a proton on the 5
position of each is nearly equidistant with the proton at the 2
position, thus possibly affording a magnetically symmetrical
environment to each F atom and equalizing their signals. As it
is, then, theoretical analysis of the vicinal F–H interactions does not
explain the observed resonance patterns.
We sought to circumvent theoretical analysis by repeating the
19F NMR spectra of compounds 2, 3, 5 and 6 in non-decoupled
mode (H–F coupled) and observing for tell-tale F–H couplings. As it
3
were, the only couplings observed at 235 MHz were the JF–H
coupling to the respective geminal protons. The 19F NMR signals of
2 exhibited a doublet 3JF–H = 18.0 for the upfield F and a doublet 3JF–
H = 27.7 for the downfield F and 5 similarly exhibited doublets 3JF–
dihedral angles, compounds 3 and 6 should have differentiated 19
F
signals as well as compounds 2 and 3. Again, this is not the case
and our modeling is not helpful.
3
H = 19.8 and JF–H = 27.6. This suggests a variation from the
perfectly staggered configuration calculated and shown in Fig. 2
and is in excess of that which could be attributed to the electronic
effect of the extended sidechain [5]. This vicinal coupling
information is otherwise uninformative. The 19F signal for 3
exhibited a doublet JF–H = 24.6 and for 6 exhibited a doublet JF–
H = 21.6, with no other couplings observable. This (along with the
original fact that there is only one F resonance), indicates that both
F’s in both molecules are symmetrically disposed in the staggered
conformation and that distances to neighboring protons are more
or less equivalent or are too distant to couple.
A follow-up to this theoretical analysis of the F-sidechain
interactions is an evaluation of the observed F–C couplings for each
molecule. Table 2 displays those coupling constants as measured
from the 13C spectra. One observes, as expected, that the clearly
unsymmetrical compounds 2 and 5 show distinctly different
3
3
coupling constants to the
evidence of unsymmetrical interactions with the
Likewise, completely symmetrical compound 3 shows equal F
interactions with the
and b0 carbons, as expected. However, the
anomalous compound 6, with its single F resonance, exhibits much
more similar couplings ( J = 0.8 Hz) than its five carbon analog 5.
b
and b0 carbons (
D
J = 3.2 Hz) and thus
F
atoms.
b
One final and slightly more fruitful look into the possible
causes for the lack of differentiation of the F signal in 6 was an
evaluation of the F interactions with neighboring C atoms. Fig. 3
D
In this small difference from one side of the molecule to the other,
we finally find supporting evidence for a magnetically symmetric
environment around the NF2 group of an otherwise unsymmetric
molecule.
0
shows C –C and C –C bond distances as well as the dihedral
a
b
a
b
angles to the rest of the carbon chains, again as calculated by
Fig. 3. C–C bond lengths and dihedral angles for 2- and 3-difluoroaminoalkanes.

76
R.K. Belter / Journal of Fluorine Chemistry 137 (2012) 73–76
3. Conclusion
added dropwise. The reaction was stirred a further 1–2 h. Stirring
was ceased and the solution allowed to settle. The clear solution
was transferred via non-metallic cannula to a 250 ml pressure-
rated glass round bottomed flask and held at 0 8C.
At 400 8C, NF3 reacts in the vapor phase with n-pentane and n-
hexane to generate a mixture of N,N-difluoroaminopentanes and
N,N-difluoroaminohexanes, respectively. By generating each indi-
vidual isomer of N,N-difluoroaminopentane and -hexane by
solution chemistry, we have recorded the individual 19F spectra
of 1-N,N-difluoroaminopentane, 2-N,N-difluoroaminopentane, 3-
N,N-difluoroaminopentane, 2-N,N-difluoroaminohexane and 3-
N,N-difluoroaminohexane. The separate 19F spectra clarify the
interpretation of the spectral pattern observed for the mixed N,N-
difluoroaminopentanes and mixed N,N-difluoroaminohexanes
isomers generated by the high temperature vapor phase reaction.
The 1-substituted difluoroaminoalkanes have an associated
resonance frequency of ꢁ + 56 ppm which clearly distinguishes
this substitution from the others.
A 500 ml pressure-rated glass round bottomed flask, with
Teflon coated magnetic stir bar was charged with 200 ml diethyl
ether. This solution was cooled to 0 8C and sparged with NF3. The
solution was cooled to ꢀ25 8C and pressurized with NF3 to 140 psi.
The Grignard solution in the 250 ml flask was slowly
transferred under pressure of N2 to the NF3 solution in the
500 ml flask. The transfer cannula was non-metallic and positioned
with the outlet end below the surface of the NF3 solution. The
solution was stirred 1 hour, then vented and quenched with sat. aq.
NH4Cl solution. After warming to room temperature, the phases
were separated. The ether layer was dried with MgSO4 and filtered.
Excess ether was removed by fractional distillation through a glass
ring packed column at ꢁ18 8C under slight vacuum (400 mmHg).
The difluoroaminoalkane product was isolated in various fractions
at about 20 8C (ꢁ1 mmHg) without concern for yield. The purest
fraction as per GC was analyzed by NMR.
The 2-substituted and 3-substitute difluoroaminoalkanes share
identical resonance ‘‘centers of gravity’’ at +39.4 ppm. However, 2-
substituted compounds exhibit an AB pattern of doublets whereas
the 3-substituted compounds exhibit a single resonance. This
makes differentiation of the isomers by NMR easy.
It is logical that the unsymmetrical 2-substituted difluoroami-
noalkanes should exhibit individual resonances for each of the two
fluorine atoms. It is equally logical that the symmetric 3-
substituted difluoroaminopentane should exhibit a single reso-
nance for the two equivalent fluorine atoms. However, the
observation of a single resonance for the unsymmetrical 2-
difluoroaminohexane is not immediately logical and insinuates
4.3. Spectral details
1-N,N-difluoroaminopentane, 1 13C NMR (62 MHz CDCl3):
13.7 (s, C-5), 22.5 (s, C-4), 23.8 (t, J = 7.8 Hz, C-2), 29.1 (s, C-3), 66.2
(t, J = 6.3 Hz, C-1); 19F NMR (235 MHz CDCl3):
d
d +55.8 (s).
2-N,N-difluoroaminopentane, 2 1H NMR (250 MHz CDCl3/TMS):
0.95 (t, 3H, J = 7.2), 1.26 (d, 3H, J = 6.2 Hz), 1.43 (m, 2H), 1.73 (mm,
d
a
somehow magnetically symmetrical conformation of the
2H), 3.48 (tp, 1H, J = 22.4, 6.5 Hz); 13C NMR (62 MHz CDCl3):
d 13.1
molecule around the NF2-group. Gaussian conformational analysis
resulted in the modeling of such a conformation, but it also
resulted in similar models for the other asymmetric isomers.
Evaluation of the actual F–C coupling constants gives supporting
evidence of a magnetic symmetry within 3-difluoroaminohexane.
It is apparent that to the extent that it is observable by 19F NMR, an
excess of at least two carbon atoms to one side of the – NF2
attachment point is necessary to put the – NF2 group in an
unsymmetrical environment.
(t, J = 9.8 =Hz, C-1), 14.0 (s, C-5), 18.8 (s, C-4), 33.0 (t, J = 6.5 Hz, C-
3), 70.0 (t, J = 6.1 Hz, C-2); 19F NMR (235 MHz CDCl3):
J = 565.7 Hz), +43.1 (d, J = 565.7 Hz).
d +35.7 (d,
3-N,N-difluoroaminopentane, 3 1H NMR (250 MHz CDCl3/TMS):
0.96 (sext, 6H, J = 7.5 Hz), 1.64 (sept, 2H, J = 7.2 Hz), 1.72 (m, 2H),
d
3.17 (tp, 1, J = 26.3, 6.0 Hz); 13C NMR (62 MHz CDCl3):
d
10.0 (s, C-
1), 20.9 (t, J = 8.2 Hz, C-2), 76.7 (t, J = 5.3 Hz, C-3); 19F NMR
(235 MHz CDCl3): +55.9
+39.4 (s); 19F NMR (235 MHz CDCl3):
(s), +39.2 (p, J = 579.3 Hz); IR 2958, 2875, 1462, 1370, 953, 860,
844, 810; GC/MS 70 eV, m/z (rel. int.): 71(100), 55(40).
1-N,N-difluoroaminohexane,
by subtraction 13C NMR
(62 MHz CDCl3): 14.1 (s, C-6), 22.6 (s, C-5), 24.1 (t, J = 7.7 Hz,
C-2), 26.6 (s, C-4), 31.6 (s, C-3), 66.2 (t, J = 6.3 Hz, C-1); 19F NMR
d
d
d
4. Experimental
4
d
4.1. General
(235 MHz CDCl3):
d +55.9 (s).
The generation of the individual isomers of N,N-difluoroami-
nopentane and N,N-difluoroaminohexane also allowed us to
generate individual 1H and 13C NMR spectra. These were in
agreement with the partial assignments given in reference [1] and
2-N,N-difluoroaminohexane, 5 1H NMR (250 MHz CDCl3/TMS):
0.92 (brs, 3H), 1.26 (d, 3H, J = 5.6 Hz), 1.36 (brm, 4H), 1.43 (m,
d
1H), 1.73 (mm, 1H), 3.41 (tp, 1H, J = 21.6, 6.5 Hz); 13C NMR
(62 MHz CDCl3):
d 12.9 (t, J = 9.8 Hz, C-1), 13.5 (s, C-6), 22.3 (s, C-5),
are repeated here complete. Products were identified by 1H and 13
and 19F NMR performed on a Bruker DPX-250. Older literature 19
values reported in
values relative to external CF3CO2H were converted to relative to
internal CFCl3 by addition of 78 ppm [6].
C
F
27.5 (s, C-4), 30.4 (t, J = 6.7 Hz, C-3), 70.0 (t, J = 6.2 Hz, C-2); 19F
NMR (235 MHz CDCl3): d +35.7 (d, J = 565.8 Hz), +43.1 (d,
F
have had their signs inverted (ꢀ for +) and
J = 565.8 Hz).
3-N,N-difluoroaminohexane, 6 1H NMR (250 MHz CDCl3/TMS):
0.95 (t, 3H, J = 7.1 Hz), 1.00 (t, 3H, J = 7.1 Hz), 1.43 (sext, 2H,
d
The difluoroaminopentanes were prepared from the reaction of
theappropriatepentanomagnesiumbromideswithNF3 asdescribed
below. TheprerequisitebromopentaneswereacquiredfromAldrich.
The difluoroaminohexanes were prepared from the reaction of the
appropriate hexanomagnesium chlorides with NF3 as described
below. The prerequisite chloropentanes were prepared from the
reaction of SOCl2 with the appropriate hexanol which was acquired
from Aldrich. Magnesium turnings were Aldrich 98%. Diethyl ether
was distilled from benzophenone/sodium ketyl.
J = 7.4 Hz), 1.67 (mm, 4H), 3.28 (tp, 1, J = 26.4, 5.8 Hz); 13C NMR
(62 MHz CDCl3):
d 10.1 (s, C-1), 14.0 (s, C-6), 19.1 (s, C-5), 21.5 (t,
J = 8.6 Hz, C-2), 30.0 (t, J = 7.8 Hz, C-4), 75.4 (t, J = 5.5 Hz, C-3); 19
F
NMR (235 MHz CDCl3):
d +39.3 (s).
References
[1] R.K. Belter, J. Fluorine Chem. 132 (2011) 961.
[2] R.C. Petry, J. Am. Chem. Soc. 89 (1967) 4600.
[3] C.L. Bumgardner, Tetrahedron Lett. 48 (1964) 3683.
[4] McFerrin, C.A., Dept. of Chemistry, Louisiana State University, Baton Rouge, LA
70803.
4.2. General procedure for the synthesis of N,N-difluoroanimoalkanes
[5] W.R. Dolbier, Guide to Fluorine NMR for Organic Chemists, Wiley, Hoboken, NJ,
2009, p. 18.
[6] R.E. Banks, Fluorocarbons and their Derivatives, 2nd ed., MacDonald Technical and
Scientific, London, 1970, p. 232.
5 g of magnesium turnings were suspended with stirring in
200 ml diethyl ether at 0 8C. The requisite haloalkane (0.2 mol) was