Investigation on the preparation of 2,2-difluoroethylamine by amination of 1-halo-2,2-difluoroethane

Article abstract of DOI:10.1016/j.jfluchem.2020.109690

The synthesis of 2,2-difluoroethylamine via the amination of 1-halo-2,2-difluoroethane (halogen = Cl or Br) without catalyst and solvent was studied by experiment and theoretical calculation. In order to compare the difficulty in the amination of 1-halo-2,2-difluoroethane with NH₃, experimental investigations were carried through changing reaction temperature, molar ratio and reaction time. And experimental evidence demonstrated that 1-chloro-2,2-difluoroethane was more prone to amination than 1-bromo-2,2-difluoroethane under the same conditions. In addition, the reaction mechanism was investigated, and the formation complexity evaluated. Calculation of the transition states and rate constants, and analysis of the frontier molecular orbitals, indicated that the reaction rate of the amination of 1-chloro-2,2-difluoroethane was higher than that of 1-bromo-2,2-difluoroethane, which was consistent with the results of the experiment.

Full text of DOI:10.1016/j.jfluchem.2020.109690

Journal Pre-proof
Investigation on the preparation of 2,2-difluoroethylamine by amination of
1-halo-2,2-difluoroethane
Xue Han, Feiyao Qing, Chengping Zhang, Xiaoqing Jia, Hengdao
Quan
PII:
S0022-1139(20)30388-2
DOI:
https://doi.org/10.1016/j.jfluchem.2020.109690
FLUOR 109690
Reference:
To appear in:
Journal of Fluorine Chemistry
Received Date:
Revised Date:
Accepted Date:
15 September 2020
17 October 2020
8 November 2020
Please cite this article as: Han X, Qing F, Zhang C, Jia X, Quan H, Investigation on the
preparation of 2,2-difluoroethylamine by amination of 1-halo-2,2-difluoroethane, Journal of
Fluorine Chemistry (2020), doi: https://doi.org/10.1016/j.jfluchem.2020.109690
This is a PDF file of an article that has undergone enhancements after acceptance, such as
the addition of a cover page and metadata, and formatting for readability, but it is not yet the
definitive version of record. This version will undergo additional copyediting, typesetting and
review before it is published in its final form, but we are providing this version to give early
visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal
pertain.
© 2020 Published by Elsevier.

Investigation on the preparation of 2,2-difluoroethylamine by amination
of 1-halo-2,2-difluoroethane
Xue Han a, Feiyao Qing a, Chengping Zhang a, Xiaoqing Jia a, Hengdao Quan b,*
a
Beijing Institute of Technology, 5 South Zhongguancun Street, Haidian District,
Beijing, 100081, China
b
National Institute of Advanced Industrial Science and Technology, 1-1-1Higashi,
Tsukuba, Ibaraki, 305-8565, Japan
*Corresponding
Author:
E-mail:
Hengdao-quan@aist.go.jp;
quanhengdao@hotmail.com
Graphical Abstract – Pictogram
F
F
NH4X
2 NH3
X
NH2
F
F
(X = Cl or Br)
Highlights

The amination of CHF2CH2X (X = Cl or Br) without catalyst and solvent was
reported.



The theoretical study on the amination of CHF2CH2X was performed.
The mechanism of the amination of CHF2CH2X with NH3 was proposed.
CHF2CH2Cl was easier to be aminated than CHF2CH2Br.

Abstract
The synthesis of 2,2-difluoroethylamine via the amination of 1-halo-2,2-
difluoroethane (halogen = Cl or Br) without catalyst and solvent was studied by experiment
and theoretical calculation. In order to compare the difficulty in the amination of 1-halo-
2,2-difluoroethane with NH3, experimental investigations were carried through changing
reaction temperature, molar ratio and reaction time. And experimental evidence
demonstrated that 1-chloro-2,2-difluoroethane was more prone to amination than 1-bromo-
2,2-difluoroethane under the same conditions. In addition, the reaction mechanism was
investigated, and the formation complexity evaluated. Calculation of the transition states
and rate constants, and analysis of the frontier molecular orbitals, indicated that the reaction
rate of the amination of 1-chloro-2,2-difluoroethane was higher than that of 1-bromo-2,2-
difluoroethane, which was consistent with the results of the experiment.
Keywords: 2,2-difluoroethylamine; 1-chloro-2,2-difluoroethane; Density functional
theory; 1-bromo-2,2-difluoroethane; Kinetic modeling.
1. Introduction
2,2-Difluoroethylamine is an important fluoroaliphatic compound of significant
industrial value, and is widely used as a synthetic raw material or intermediate in the
manufacture of numerous pharmaceuticals and pesticides, among other products [1]. A
difluoroethylamine group can replace the trifluoroethylamine group found in cathepsin K
inhibitors such as odanacatib [2]. Downstream products synthesized from 2,2-
difluoroethylamine, including amine amides sulfo acyl pyrrole derivatives,
aminobenzophenones, trifluoroethyl sulfide derivatives，etc, which are widely used, such

as disease treatment and pest control. Amine amides sulfo acyl pyrrole derivatives have
excellent effect in the treatment of hepatitis B [3], while aminobenzophenones were applied
in the treatment of inflammatory ophthalmic diseases or cancer [4]. In addition,
trifluoroethyl sulfide derivatives played an important role in killing animal pests [5]. In
recent years, 2,2-difluoroethylamine were used as the raw material to produce
difluorodiazoethane (CF2HCHN2), a novel reagent applied in [3+2]-cycloadditions,
esterification and carbene transfer reaction[6-8]. The downstream products synthesized
from 2,2-difluoroethylamine were summarized in scheme 1.
Scheme 1 Downstream products synthesized from 2,2-difluoroethylamine.
Thus far, 2,2-difluoroethylamine was prepared from various starting materials via
different synthetic routes [9-18], which were summarized in scheme 2. The direct
fluorination of ethylamine with trifluoromethyl hypofluorite [9] was not suitable for
industrial production because of the high toxicity and pyrophoricity of diborane and
trifluoromethyl hypofluorite. In addition, reduction of 2,2-difluoroacetamide [10], 1,1-
difluoro-2-nitrosoethane [11], or 1,1-difluoro-2-nitroethane [12] were unattractive
synthetic routes because of the difficulty in obtaining the necessary raw materials. For the

same reason, synthetic routes involving hydrogenolysis of 2,2-difluoroethylbenzylamine
[12], or reaction of N-(2,2-difluoro-ethyl)prop-2-en-1-amine with 2-aminoethanol [13]
were disadvantageous. 2,2-difluoroethylamine can be also synthesized in two steps by
hydrogenation of 2,2-difluoroacetonitrile [14] followed by acidic hydrolysis of the
resulting N-(2,2-difluoroethyl)acetamide. However, this approach had the following
disadvantages: it was a longer route, and the end product was 2,2-difluoroethylamine
hydrochloride, such that 2,2-difluoroethylamine can only be obtained after neutralization
reaction of 2,2-difluoroethylamine hydrochloride with a base. Preparation of 2,2-
difluoroethylamine by addition of the appropriate amine to endo-cis-5-norbornene-2,3-
dicarboxylic anhydride [15] was tedious and not suitable for industrial application.
Furthermore, the amination of 1-bromo-2,2-difluoroethane with ammonium acetate [16]
was unattractive because of the deliquescent character of the latter, which affected reaction
efficiency, in addition to the long reaction time. Thus, the methods discussed above for the
synthesis of 2,2-difluoroethylamine were not suitable for industrial use.
In addition, 2,2-difluoroethylamine can be prepared by the amination of 1-chloro-2,2-
difluoroethane [17-18]. When the reaction was controlled at 140-145℃ for 9h over KBr
catalyst, the yield of 2,2-difluoroethylamine was 59% after extraction separation by N-
methlpyrroliding and water [14]. While the reaction was carried out at 143-145℃for 5.5h
over KI catalyst in N-methylpyrrolidine, the yield of 2,2-difluoroethylamine was 88% [18].
In the above amination method, solid catalysts and large amount of solvents were used,
which generated a lot of liquid waste and solid waste. Thus, this method was
environmentally unfriendly, and its reaction mechanism was not clear.

Scheme 2 Known routes for the preparation of 2,2-difluoroethylamine.
In this work, 2,2-difluoroethylamine was prepared by reacting 1-halo-2,2-
difluoroethane with excess NH3 in the absence of catalyst and solvent, and the reaction was
studied both experimentally and computationally. 1-halo-2,2-difluoroethane were reacted
with NH3 under the same reaction conditions. Compared with routes 9-12, the raw
materials used in this method were easily obtained and low toxicity. The route was short to
obtain 2,2-difluoroethylamine via the amination of 1-halo-2,2-difluoroethane, which was
better than routes 14-15. In addition, no catalysts and solvents were used in this method,
without waste generation, and was beneficial to the environment. The reaction conditions
were changed to obtain the conversion rate of reactants and the selectivity of products. A
comparison of the experimental results determined which 1-halo-2,2-difluoroethane was
more likely to react with NH3. Generally, various reaction pathways for the preparation of
2,2-difluoroethylamine were studied, and two transition states were predicted to be located
along the process. In addition, analysis of the frontier molecular orbitals, was performed in
order to compare the ease of the two reactions [19, 20]. The molecular electrostatic
potential (MESP) was used to visualize charged regions of molecules [21]. Furthermore,

thermal rate constants for temperatures between 298.15K and 2500K were calculated by
conventional TST [22, 23] and canonical variational TST (CVT) [24], considering the
Wigner and Eckart tunneling corrections.
2. Results and discussion
2.1 Synthesis of difluoroethylamine
In this experiment, 1-chloro-2,2-difluoroethane and 1-bromo-2,2-difluoroethane were
reacted with ammonia, with changing the material ratio, temperature conditions and
reaction time to prepare difluoroethylamine. The results were shown in Fig. 1, Fig. 2 and
Fig. 3, respectively.
Fig. 1 Conversion of 1-chloro-2,2-difluoroethane and 1-bromo-2,2-difluoroethane and selectivity
to 2,2-difluoroethylamine with changing molar ratio. (Reaction conditions: molar ratio
(NH3/CHF2CH2X) = 2/1-6/1, reaction temperature = 160℃, reaction time = 12h)

From Fig. 1 (corrected by gas chromatography), 160℃ was selected as the reaction
temperature, while the molar ratio (NH3/CHF2CH2X) increased from 2:1 to 6:1, the
conversion of 1-chloro-2,2-difluoroethane and 1-bromo-2,2-difluoroethane rose with the
increasing molar ratio. When the molar ratio reached 6:1, the upward trend tended to be
flat. The rate of formation of 2,2-difluoroethylamine in the amination of 1-chloro-2,2-
difluoroethane was always higher than that of 1-bromo-2,2-difluoroethane. Besides, the
selectivity of 2,2-difluoroethylamine always approached 100% in the two reactions.
Fig. 2 Conversion of 1-chloro-2,2-difluoroethane and 1-bromo-2,2-difluoroethane and selectivity
to 2,2-difluoroethylamine with changing temperature. (Reaction conditions: molar ratio
(NH3/CHF2CH2X) = 6/1, reaction temperature = 80-180℃, reaction time = 12h)
As shown in Fig. 2 (corrected by gas chromatography), the reaction temperature was
increased from 80℃to 180℃and the molar ratio was 6:1 (NH3/CHF2CH2X), the conversion
of 1-chloro-2,2-difluoroethane and 1-bromo-2,2-difluoroethane rose significantly as the
reaction proceeded in the two reactions. At the beginning, the progress of the each of the
two reactions was relatively low, as the conversion of the two raw materials was less than

10%. With the reaction progressing, the conversion increased significantly, and finally the
upward trend tended to be flat. The rate of formation of 2,2-difluoroethylamine in the
amination of 1-chloro-2,2-difluoroethane was higher than that of 1-bromo-2,2-
difluoroethane. Besides, the selectivity of 2,2-difluoroethylamine always approached 100%
in the two reactions, which indicated that the reaction produced no by-products.
Fig. 3 Conversion of 1-chloro-2,2-difluoroethane and 1-bromo-2,2-difluoroethane and selectivity
to 2,2-difluoroethylamine with changing time. (Reaction conditions: molar ratio (NH3/CHF2CH2X)
= 6/1, reaction temperature = 160℃, reaction time = 2-12h)
Choosing a representative temperature point of 160℃ and a molar ratio point of 6:1,
the reaction time was changed to observe the effect on the two groups of reactions. As
shown in Fig. 3 (corrected by gas chromatography), with the longer reaction time, the
conversion increased significantly, and finally the upward trend tended to be flat
respectively. And the conversion of reactant in the amination of 1-chloro-2,2-
difluoroethane was always higher than that of 1-bromo-2,2-difluoroethane, which
demonstrates that 1-chloro-2,2-difluoroethane was more prone to amination than 1-bromo-

2,2-difluoroethane. When the reaction time was 12h, the yield of 2,2-difluoroethylamine
was 59.5% in the amination of 1-chloro-2,2-difluoroethane without catalyst and solvents,
which was slightly higher than that in the presence of KBr catalyst (yield: 59%) [17]. Thus,
the present experiments provide another facile method to synthesize 2,2-
difluoroethylamine without catalyst and solvent.
In the above single factor experiments, it can be found that at the same reaction
condition, the conversion of 1-chloro-2,2-difluoroethane was always higher that of 1-
bromo-2,2-difluoroethane in the amination with NH3. Thus, 1-chloro-2,2-difluoroethane
was more easily aminated compared to 1-bromo-2,2-difluoroethane. In order to find the
reason for the above results, the chemical computations in the amination of 1-halo-2,2-
difluoroethane were then performed.
2.2 Computation
2.2.1 Geometries
The optimized structures of all located stationary points and selected geometrical
parameters at the B3LYP/6-311+G level were shown in Fig. 4 and Fig. 5. Attack on the
carbon atom of the –CH2X group by NH3 proceeded via the transition states leading to
+
+
intermediates CHF2CH2NH3 Cl- and CHF2CH2NH3 Br-, respectively. Subsequently,
another NH3 molecule reacted as a base to abstract the proton and deliver CHF2CH2NH2.

NH3
CHF2CH2Cl
+
TS1
CHF2CH2NH3 Cl-
NH3
CHF2CH2Br
+
TS2
CHF2CH2NH3 Br-
Fig. 4 Selected geometrical parameters of the optimized reactants, transition states and
products of the first step at the B3LYP/6-311+G level. Bond lengths are in Å, and angles
are in degrees.

Fig. 4 showed that the C–Cl bond in 1-chloro-2,2-difluoroethane must elongate by
0.410 Å to reach TS(1), while from CHF2CH2Br to TS(2) the corresponding value for the
C–Br bond was 0.482 Å. In contrast, the lengths of the forming C–N bonds in the TS1 and
TS2 geometries (2.216 Å and 2.145 Å respectively) were both longer than those in the
corresponding products (1.496 Å and 1.497 Å respectively). The fact that the C–N bond
shortened from transition state to product was possibly attributed to an interaction between
the unpaired electron of the N atom of ammonia and the C atom, leading to an electron
transfer. During the reaction, the C–C–N bond angle increased from 93.33° in TS1 to 111.4°
+
in CHF2CH2NH3 Cl-, forming a relatively stable structure, while it increased from 94.56°
+
in TS2 to 111.5° in CHF2CH2NH3 Br-.
As displayed in Fig. 5, the C–N and the C–C bond lengths in the transition state were
1.486 Å and 1.505 Å respectively, and the C–C–N bond angle were 112.5°. While the C–
N bond shortened to 1.459 Å in CHF2CH2NH2, the length of the C–C bond changed only
slightly to 1.506 Å, and the C–C–N bond angle increased to 113.6°, leading to a relatively
stable structure.

+
CHF2CH2NH3
NH3
CHF2CH2NH2
TS
Fig. 5 Selected geometrical parameters of the optimized reactants, transition states and
products of the second step at the B3LYP/6-311+G level. Bond lengths are in Å, and angles
are in degrees.
2.2.2 Molecule Electrostatic Potential Analysis
Molecule electrostatic potential was associated with the electronic density, and was a
well-suited descriptor for identification of sites for nucleophilic and electrophilic reactions,
in addition to hydrogen bonding interactions [25]. Mapped electrostatic potential (ESP)
surfaces of the optimized structures of the three reactants and the product at DFT-B3LYP
level were shown in Fig. 6.

(A)
(C)
(B)
(D)
Fig. 6 B3LYP/6-311+G optimized structures and Multiwfn calculated mapped electrostatic
potential (MESP) surfaces of CHF2CH2Cl (A), CHF2CH2Br (B), NH3 (C), and
CHF2CH2NH2 (D). Red, white and blue represent positive, zero and negative electrostatic
potential values (kcal mol-1) respectively.
For 1-chloro-2,2-difluoroethane (A), the highest electron density was localized on the
carbon atom which is attached to the chlorine atom (62.00kcal/mol), and was higher than
that of the corresponding carbon atom in 1-bromo-2,2-difluoroethane (B) (57.69kcal/mol).
The higher the ESP value, the more likely a nucleophilic reactionwas to occur, otherwise,
the electrophilic reaction would contribute to the reaction. Thus, the carbon atom of 1-
chloro-2,2-difluoroethane was more prone to reaction with an electron donating group [26].
Among all calculated values, the nitrogen atom of NH3 (C) had the lowest electron density
(-48.70kcal/mol), which indicated its ability to donate electrons. In contrast, the carbon
atoms which were bound to the halogen atoms in 1-halo-2,2-difluoroethane (halogen = Cl
or Br) had a relatively strong ability to attract electrons, which was reflected in their high

electrostatic potential.
2.2.3 Energetics
TS2
TS1
87.44
59.33
0.00
0.00
-82.46
-
reaction A
reaction B
Fig. 7 Potential energy profile for the reaction of CHF2CH2Cl+NH3 (reaction A) and
CHF2CH2Br+NH3 (reaction B) at the B3LYP/6-311+G level of theory.
In order to understand the reaction mechanism, DFT calculations were performed to
elucidate possible reaction paths. Based on B3LYP/6-311+G, the ZPVEs, energies,
enthalpies and Gibbs free energies of various stationary points were calculated. The
energies of the reaction pathways for the two substrates were shown Fig. 7. The reactions
were predicted to occur in two steps, and the energy values of the reactants of each step
+
(CHF2CH2Cl + NH3/CHF2CH2Br + NH3 and CHF2CH2NH3 + NH3) were set as zero for
reference.
As shown in Fig. 7, the first step of the reaction for either substrate proceeded through
transition states TS1 and TS2, forming intermediates IM1 and IM2, respectively. From
Fig.7, the energy of TS1 was higher than that of the reactants CHF2CH2Cl + NH3 by

59.33kJ/mol, while the difference in energy between TS2 and CHF2CH2Br + NH3 was
87.44kJ/mol. As such, the activation energy barrier of the reaction 2 was higher than that
of the reaction 1 by 28.11kJ/mol. In the second step, shown in Fig. 8, either reaction
proceeds via the same transition state, which was higher in energy than the reactants
+
CHF2CH2NH3 + NH3 by 2.01kJ/mol.
2.01
0.00
-14.52
+
Fig. 8 Potential energy profile for the reaction of CHF2CH2NH3 +NH3 at the B3LYP/6-
311+G level of theory.
According to the data and discussion provided above, it was concluded that the
reaction proceeded preferentially by attack of the NH3 molecule to the carbon atom of the
+
–CH2X group, forming intermediate CHF2CH2NH3 X-, with subsequent reaction of a
second NH3 molecule as a base to deprotonate the intermediate and deliver CHF2CH2NH2.
The discussion of the reaction energetics was consistent with structural implications and
the Hammond’s postulate, confirming the earlier character of TS1 when compared to TS2.
2.2.4 HOMO-LUMO analysis

Frontier molecular orbitals, namely highest occupied molecular orbital (HOMO) and
lowest unoccupied molecular orbital (LUMO), were the most important orbitals in a
molecule. These orbitals determine how a molecule interacted with other species [27], and
the frontier molecular energy gap aids the characterization of the chemical reactivity [28].
The higher the HOMO energy, the more facile it was for it to donate electrons, whereas
LUMO was more prone to accept electrons when its energy was low [29]. In order to
simulate the energetic behavior of the title reaction, HOMO-LUMO energy calculations of
transition states were carried out by DFT theory level calculations using the B3LYP method
with 6-311+G basis set.
Band gap(Eb )  ELUMO  EHOMO
(1)

LUMO
LUMO
6.8130 eV
5.8623 eV
HOMO
IM2
HOMO
IM1
Fig. 9 Calculated molecular orbital energy levels of transition states and intermediates with
molecular orbital maps of LUMO and HOMO. Red and blue correspond to different phases
of the molecular wave functions for the frontier molecular orbitals.
1-Halo-2,2-difluoroethanes were electrophilic in nature. The HOMO of NH3 reacted
with the LUMO of the 1-halo-2,2-difluoroethane molecules, a reaction in which the
nitrogen atom participates as the electron donor. As shown in Fig. 9, the HOMO-LUMO
band gap of TS1 was wider than that of TS2, and the difference in the band gaps was
0.7370eV. This contributed to the higher stability and indicateed that a lower energy barrier
had to be overcome in order for the reaction to occur. In addition, the band gap of IM1 was
larger than that of IM2 suggesting that IM1 possessed a lower energy, which was consistent
with the energy calculations. According to I. Fleming, the HOMO represented the
molecule’s ability to donate electrons, while the LUMO represented the molecule’s ability
to accept electrons [30]. Consequently, by comparing the HOMO-LUMO band gap and
stability of two transition states, it was concluded that preparation of 2,2-
difluoroethylamine from amination of 1-chloro-2,2-difluoroethane was more facile than

from that of 1-bromo-2,2-difluoroethane.
2.2.5 Rate Constant
Wigner and Eckart tunneling corrections were plotted in Fig. 10. Analysis of the plot
reveals that Wigner and Eckart tunneling contributions become more significant as the
temperature decreases. While all values were in the range of 1.00–1.25, Eckart corrections
were consistently larger than the corresponding Wigner values. Within the temperature
range of 1000–2500K, both values were nearly the same for a given temperature, and vary
only slightly within this range. In contrast, tunneling values increased sharply from
approximately 1.02 to 1.24 as the temperature decreased from 1000K to 298K.
Fig. 10 Plots of tunneling coefficients for CHF2CH2Cl+NH3 (A) and CHF2CH2Br + NH3
(B) evaluated by different theoretical methods. W and E referred to Wigner and Eckart
corrections, respectively.
Rate constants were calculated by TST (conventional transition state theory) and CVT
(canonical variational transition state theory) at the B3LYP/6-311+G level for a wide
temperature range of 298.15K–2500K. Fig. 11 showed the profiles of the logarithm of TST,

CVT, TST/W, CVT/W, TST/E, and CVT/E rate constants versus the reciprocal (1000/T) of
reaction temperatures, taking into account the Wigner and Eckart tunneling corrections.
Fig. 11 Plots of rate constant expressions for the reactions CHF2CH2Cl+NH3(A) and
CHF2CH2Br + NH3(B) evaluated by different theoretical methods, highlighting the sections
in which the temperature ranges from 298.15K to 500K and from 1250K to 2500K in the
bottom and right amplifications, respectively.
As displayed in Fig. 11, the data obtained by either calculation method for both
reactions was similar. For the reaction CHF2CH2Cl+NH3 (A), the distinction between the
TST and the CVT rate constants was more pronounced than for reaction B in the wide
temperature range. It was noticeable that the evaluated kTST values are larger than kCVT
in both cases. In contrast, the amplified figure at the bottom (temperature range: 298.15 K–

500 K) shows that nearly the same values were obtained by the two computing methods
for reaction B, while a small difference was observed for reaction A. In addition, the order
kTST ≧ kCVT for the reactions described herein conforms with the theoretical definition
by Fernández-Ramos et al. [31].
From Figure 11, the rate constants of the two reactions obtained by the TST method
and the CVT method can be obtained. Among them, the rate constant of CHF2CH2Cl+NH3
(A) was much larger than that of CHF2CH2Br+NH3 (B), between 298.15-500K. And as the
temperature increased, the difference gradually decreased (1250-2500K), but the rate
constant of CHF2CH2Br+NH3 (B) is always smaller than CHF2CH2Cl+NH3 (A). Therefore,
the calculation result of the rate constant was consistent with the above calculation and
experimental results, indicating that 1-chloro-2,2-difluoroethane was easily aminated than
1-bromo-2,2-difluoroethane.
3. Conclusions
The experiments on 1-chloro-2,2-difluoroethane and 1-bromo-2,2-difluoroethane
reacting with ammonia were carried out. The reaction temperature, molar ratio, and
reaction time were changed to obtain the conversion rate and selectivity of the two reactions,
respectively. By comparing the experimental results, it was found that the 1-chloro-2,2-
difluoroethane was easier to be aminated with NH3 than 1-bromo-2,2-difluoroethane. In
order to understand the detailed mechanism and the kinetics of the formation of 2,2-
difluoroethylamine by amination of 1-halo-2,2-difluoroethane with ammonia, the two
reactions were calculated and simulated by various analysis theories, including energy
calculation, geometry analysis, and band gap comparison. Results obtained by
thermodynamic calculations indicated that the amination of 1-chloro-2,2-difluoroethane
with ammonia was more facile than that of 1-bromo-2,2-difluoroethane. The results

reported herein can facilitate the screening for synthetic routes toward 2,2-
difluoroethylamine.
4. Experimental
4.1 Chemicals
1-Chloro-2,2-difluoroethane and 1-bromo-2,2-difluoroethane (purity > 98%) were
purchased from Baokai Chemical Co., Ltd. (Jiangsu, China). Ammonia was purchased
from Xinghua Group Co., Ltd. (Shaanxi, China).
4.2. Instruments
The synthetic reaction system consisted of a reaction kettle, a thermometer, and a
programmable temperature control device.
Gas chromatography (GC) was performed on a GC9560 thermal conductivity detector
(TCD) instrument equipped with a GS-GASPRO column of 30 m length and 0.320 mm
inner diameter. The operating conditions for the gas chromatography were as follows: 40°C
for 3min; 15°C min-1 until 230°C; hold for 40min. Both the injection port and the thermal
conductively were maintained at 200°C, and the nitrogen (N2) carrier gas was introduced
at a rate of 0.5mL min-1. The content of the product was analyzed on-line.
To calculate the conversion rate of CHF2CH2NH2 and the selectivity of CHF2CH2Cl
and CHF2CH2Br, their GC relative response factors needed to be measured. Placed each
10mmol sample into a 5ml glass reagent bottle and mixed well. The above-mentioned
mixture liquid was carried to the GC by a carrier gas under warm atmosphere and their
relative GC areas were measured. Finally, the following ratio was:

CHF2CH2Br/CHF2CH2NH2/CHF2CH2Cl = 0.92/1/1.34, and obtain the GC relative
response factor by relative calculation. The data were used to calculate the conversion of
CHF2CH2NH2 and the selectivity of CHF2CH2Cl and CHF2CH2Br in amination
fluorination reaction.
4.3 Synthesis of 2,2-difluoroethylamine
Experiments were carried out in a reactor which was subjected to a leak test prior to
introduction of the reactants. Once the absence of a leak was ensured, the container was
evacuated. 1-Halo-2,2-difluoroethane and liquid ammonia were introduced sequentially.
The reaction was carried out under stirring and certain temperature conditions and reaction
conditions were: (NH3/CHF2CH2X) = 2/1-6/1, reaction temperature = 80-180℃, reaction
time = 2-12h. After the reaction was finished, the aliquot was collected and analyzed online
by gas chromatography and the results are shown in Figures 1-3.
4.4 Computation methods
4.4.1 Theoretical calculations
In order to elucidate the reaction mechanism, theoretical calculations were performed
with Gaussian 09 software packages [32]. Geometries of the reactants, products and
transition states of the examined reactions were fully optimized with the density functional
theory (DFT) with Becke three-parameter Lee-Yang-Parr correlation B3LYP [33] at the
B3LYP/6-311+G level. Furthermore, single point energy including the zero-point
vibrational energy (ZPE) corrections was performed on the optimized structures. By
analyzing vibrational frequencies, the located stationary points could be confirmed at the
same level. The intrinsic reaction coordinates (IRC) calculations [34, 35] with a step size
of 0.100 Bohr confirmed the reaction from reactants to products. In addition, the HOMO-

LUMO energy calculations of transition states were carried out by DFT theory level B3LYP
method with 6-311+G basis set to simulate the energetic behavior of the title reaction.
Finally, electrostatic potential (ESP) surfaces were calculated with the Multiwfn [36]
package as well as the molecular surface quantificationally [37] analysis, which was a
valuable technique for analyzing non-covalent interactions, studying molecular recognition
modes, and predicting reactive sites and reactivity.
4.4.2 Rate constant calculations
Thermal rate constants in the temperature range of 298K–2500K were calculated
using conventional transition-state theory (TST) and canonical variational TST(CVT).
In the conventional transition-state theory (TST), the equation of thermodynamic
equivalent was employed:
G0
kbT

(4)
kbT RT
P0
KTST (T)  
(
)n e
h
This was the reaction path degeneracy, in which kb (J/K) was the Boltzmann constant,
T (K) was the temperature, h (J/s) was the Planck constant, NA was the Avogadro number,
and G≠0 represents the standard Gibbs free energy of activation for the considered reaction
(RT/P0 has a unit of the inverse of a concentration).
The canonical variational TST (CVT) was summarized in equation (5) and involved
a generalized transition state [38] rate constant, which was obtained by performing an IRC
calculation [39] using Gaussian 09:

v（s）
kbT QTS (T, S)
h NAQR (T)

kbT
KGT 
e
(5)

In the above equation, s was the distance along the MEP in isoinertial coordinates. In
this theory (CVT), the rate constant kCVT(T) was obtained by minimizing kGT with respect
to s along the reaction path:
KCVT  mins KGT (T,s)
(6)
Taking quantum mechanical effects into consideration in the theoretical kinetic
models, the TST and CVT rate constants were improved by a temperature-dependent
tunneling coefficient, κ (T), which was approximated using the Eckart [40] and Wigner [41]
correction methods. The corrected rate constants were computed by the following
expressions:
KTST /T  (T) KTST (T)
(7)
KCVT /T  (T) KCVT (T)
(8)
Here, TST/T and CVT/T denote TST and CVT including the tunneling corrections,
respectively.
References
[1] N. Lui, J.D. Heinrich, C. Funke, G. Schlegel, T.N. Müller, Process for the preparation
of 2,2-difluoroethylamine starting from a benzylamine compound, US Patent 8450528B2
(2013).
[2] E. Isabel, C. Mellon, M.J. Boyd, N. Chauret, D. Deschênes, S. Desmarais, et al.,
Difluoroethylamines as an amide isostere in inhibitors of cathepsin K, Bioorg. Med. Chem.
Lett. 21 (2011) 920–923.

[3] W.Q. Yao, D.M. Burns, J.C. Zhuo, Azetidinyl phenyl, pyridyl or pyrazinyl carboxamide
derivatives as jak inhibitors, US Patent 9023840B2 (2015).
[4] E.R. Autson, A.M. Honnemann, X.F. Liang, S.C. Shu, S.E. Awei, T.P. Sabler, Novel
aminobenzophenone compounds, Chinese Patent 1826310B (2010).
[5] A. Kohler, B. Ellie, A. Becker, A. Worcester, U. Johans, et al, Trifluoroethyl sulfide
derivatives as the substitution of the N-aryl amidines of acaricide and insecticide, Chinese
Patent 108276358A (2018).
[6] J. Li, X.L. Yu, J. Cossy, et al, Synthesis of CF2H-Substituted Pyrazolines by [3+2]
Cycloaddition between CF2HCHN2 and Electron-Deficient Alkenes, Eur. J. Org. Chem.
2(2017) 266–270.
[7] P.K. Mykhailiuk, In Situ Generation of Difluoromethyl Diazomethane for [3+2]
Cycloadditions with Alkynes, Angew. Chem. Int. Ed. 54(2016), 6558–6561
[8] P.K. Mykhailiuk, R.M. Koenigs, Difluorodiazoethane (CF2HCHN2):a new reagent for
the introduction of the difluoromethyl group, Chem. Eur. J. 25(2019), 6053-6063
[9] Swart, Ueber einige fluorhaltige alkylamine, Chem. Zentralblatt. 75 (1904) 944–945.
[10] J. Kollonitsch, Substitutive fluorination of organic compounds, US Patent 4030994A
(1977).
[11] A. Donetti, E. Cereda, A. Ezhaya, R. Micheletti, N-(Fluoroethyl) (imidazolylphenyl)
formamidines. The issue of the active species of mifentidine, J. Med. Chem. 32 (1989)
957–961.
[12] N. Lui, S. Antons, W.A. Moradi, Process for preparing 2,2-difluoroethylamine and
salts thereof proceeding form difluoroacetonitrile, US Patent 20110166388A1 (2012).
[13] S. Antons, N. Lui, A. Gerlach, Process for preparing 2,2-difluoroethylamine by
hydrogenating 1,1-difluoro-2-nitroethane, US Patent 20110082318A1 (2011).
[14] N. Lui, C. Funke, J.D. Heinrich, T.N. Muller, Process for the preparation of 2,2-
difluoroethylamine starting from prop-2-en-1-amine, US Patent 20120142971A1 (2012).

[15] R. Kluger, J. Chin, Carboxylic acid participation in amide hydrolysis. Evidence that
separation of a nonbonded complex can be rate determining, J. Am. Chem. Soc. 104 (1982)
2891–2897.
[16] R. Kluger, M. Brandl, Structural effects on reactivity in intramolecular catalysis. The
hydrolysis of N-alkylated monoamides derived from norbornene-2,3-dicarboxylic acid,
Can. J. Chem. 62 (1984) 2515–2520.
[17] N. Lui, R. Warsitz, C. Funke, C. SEVERINS, Process for preparing 2,2-
difluoroethylamine from 2,2-difluoro-1-chloroethane and ammonia, US Patent
20120184777A1 (2012).
[18] N. Lui, S. Pazenok, Y.G. Shermolovich, Process for preparing 2,2-difluoroethylamine,
US Patent 8,188,319B2 (2011).
[19] P. Muniappan, R. Meenakshi, G. Rajavel, M. Arivazhagan, Vibrational spectra and
theoretical calculations (dimerization, UV-Vis, multinuclear NMR and pes analyses) of
3,4-dimethylbenzamide and 3,4,5-trihydroxybenzamide, S. A. P: M&B. S. 117 (2014)
739–753.
[20] Ö. Tamer, N. Dege, G. Demirtaş, D. Avcı, Y. Atalay, M. Macit, A.A. Ağar, An
experimental and theoretical study on the novel (Z)-1-((naphthalen-2-ylaminomethylene)
-naphthalen-2(1H)-one crystal, S. A. P: M&B. S. 117 (2014) 13–23.
[21] B.P.S. Gautam, M. Srivastava, R. Prasad, R. Yadav, Synthesis, characterization and
quantum chemical investigation of molecular structure and vibrational spectra of 2,5-
dichloro-3,6-bis-(methylamino)1,4-benzoquinone, S. A. P: M&B. S. 129 (2014) 241–254.
[22] H. Eyring, The Activated Complex in Chemical Reactions, J. Phys. Chem. 3 (1935)
107–115.
[23] K.J. Laidler, M.C. King, Development of transition-state theory, J. Phys. Chem. 87
(1983) 2657–2664.

[24] B.C. Garrett, D.G. Truhlar, Generalized transition state theory. Bond energy-bond
order method for canonical variational calculations with application to hydrogen atom
transfer reactions, J. Am. Chem. Soc. 101 (1979) 4534–4548.
[25] E. Scrocco, J. Toması, in: P. Lowdin (Ed.), Advances Quantum Chemistry, Academic
Press, New York. 1978.
[26] Y. Luo, Z. Chang, B.J. Blamo, et al. Efficiency of fluorinated alcohol for extraction of
organic acid from its dilute aqueous solution: Amolecular optimization study of extractant,
J. Fluorine Chem. 180 (2015) 234–239.
[27] V. Balachandran, V. Karpagam, G. Santhi, B. Revathi, G. Ilango, M. Kavimani,
Conformational stability, vibrational (FT-IR and FT-Raman) spectra and computational
analysis of m-trifluoromethyl benzoic acid, S. A. P: M&B. S. 137 (2015) 165–175.
[28] K. Fukui, Role of frontier orbitals in chemical reactions Science 218 (1982) 747–754.
[29] B.J. Powell, T. Baruah, N. Bernstein, K. Brake, R.H. McKenzie, P. Meredith, M.R.
Pederson, A first-principles density-functional calculation of the electronic and vibrational
structure of the key melanin monomers, J. Chem. Phys. 120 (2004) 8608–8615.
[30] I. Fleming, Frontier Orbitals and Organic Chemical Reactions, John Wiley & Sons,
New York 5 (1976).
[31] A. Fernández-Ramos, J.A. Miller, S.J. Klippenstein, D.G. Truhlar, Modeling the
Kinetics of Bimolecular Reactions, Chem. Rev. 106 (2006) 4518–4584.
[32] M.J. Frisch, G.W. Trucks, H.B. Schlegel, et al., Gaussian 09, Revision C.01, Gaussian,
Inc., Wallingford CT, 2013.
[33] C. Lee, W. Yang, R.G. Parr, Density-functional exchange-energy approximation with
correct asymptotic behaviour, Phys. Rev. B. 37 (1988) 785–789.
[34] H.P. Hratchian, H.B. Schlegel, Accurate reaction paths using a Hessian based
predictor–corrector integrator, J. Chem. Phys. 120 (2004) 9918–9924.

[35] H.P. Hratchian, H.B. Schlegel, Using Hessian Updating To Increase the Efficiency of
a Hessian Based Predictor-Corrector Reaction Path Following Method, J. Chem. Theory
Comput. 1 (2005) 61–69.
[36] T. Lu, F.W. Chen, Multiwfn: A Multifunctional Wavefunction Analyzer, J. Comput.
Chem. 33 (2012) 580–592.
[37] T. Lu, F. Chen, Quantitative analysis of molecular surface based on improved
Marching Tetrahedra algorithm, J. Mol. Graph. Model. 38 (2012) 314–323.
[38] B.C. Garrett, D.G. Truhlar, Generalized transition state theory. Quantum effects for
collinear reactions of hydrogen molecules and isotopically substituted hydrogen molecules,
J. Phys. Chem. 83 (1979) 1079–1112.
[39] K. Fukui, The path of chemical reactions - the IRC approach, Acc. Chem. Res. 14
(1981) 363–368.
[40] C. Eckart, The Penetration of a Potential Barrier by Electrons, Phys. Rev. 35 (1930)
1303–1309.
[41] S. Canneaux, F. Bohr, E. Henon, KiSThelP: A program to predict thermodynamic
properties and rate constants from quantum chemistry results, J. Comput. Chem. 35 (2013)
82–93.