Nucleophilic Fluorination and Radiofluorination via Aziridinium Intermediates: N-Substituent Influence, Unexpected Regioselectivity, and Differences between Fluorine-19 and Fluorine-18

Article abstract of DOI:10.1021/acs.joc.5b01714

The efficient dehydrofluorination and radiofluorination of N,N-disubstituted-β-aminoalcohols through an anchimeric-assisted mechanism was developed. An investigation into the influence of N-substituents on the ring opening of the aziridinium intermediate indicated differences in the isomeric ratio and the yields of fluorinated products obtained from N,N-disubstituted-phenylalaninol. This influence was substantial for ¹⁸F-radiofluorination, with yields varying from 0 to 71% at room temperature (RT). Although no significant effects were observed in the fluorine-19 chemistry when the reaction was heated to 90 °C, considerable changes appeared during radiofluorination. In the latter case, the radiochemical yields increased, and degradation of the 2-fluoro-propan-1-amine isomer (b) occurred, leading to a regiospecific reaction in the radiolabeling of [¹⁸F]-fluorodeprenyl. This method involving nucleophilic radiofluorination at RT was successfully applied to the radiolabeling of [¹⁸F]-2-fluoroethylamines in which the influence of the N-substituent was also observed.

Full text of DOI:10.1021/acs.joc.5b01714

Subscriber access provided by University of Sussex Library
Article
Nucleophilic Fluorination and Radiofluorination via Aziridinium
Intermediates: N-Substituent Influence, Unexpected Regioselectivity
and Differences Between Fluorine-19 and Fluorine-18
Marie Médoc, and Franck Sobrio
J. Org. Chem., Just Accepted Manuscript • Publication Date (Web): 25 Sep 2015
Downloaded from http://pubs.acs.org on September 27, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
The Journal of Organic Chemistry is published by the American Chemical Society.
1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 39
The Journal of Organic Chemistry
1
2
Nucleophilic Fluorination and Radiofluorination via Aziridinium Intermediates: N-
Substituent Influence, Unexpected Regioselectivity and Differences Between Fluorine-19
and Fluorine-18.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Marie Médoc, Franck Sobrio*
CEA, I2BM, LDMꢀTEP, UMR 6301 ISTCT / Centre National de la Recherche Scientifique,
UMR 6301 ISTCT, LDMꢀTEP / Université de Caen BasseꢀNormandie, UMR 6301 ISTCT,
LDMꢀTEP, GIP Cyceron, BP5229, Fꢀ14074 Caen, France. Normandie Univ, France
*Eꢀmail : franck.sobrio@cea.fr
ABSTRACT
The efficient dehydrofluorination and radiofluorination of N,Nꢀdisubstitutedꢀβꢀaminoalcohols
through an anchimericꢀassisted mechanism was developed. An investigation into the
influence of Nꢀsubstituents on the ring opening of the aziridinium intermediate indicated
differences in the isomeric ratio and the yields of fluorinated products obtained from N,Nꢀ
disubstitutedꢀphenylalaninol. This influence was substantial for 18Fꢀradiofluorination, with
yields varying from 0 to 71% at room temperature (RT). Although no significant effects were
observed in the fluorineꢀ19 chemistry when the reaction was heated to 90 °C, considerable
changes appeared during radiofluorination. In the latter case, the radiochemical yields
increased and degradation of the 2ꢀfluoroꢀpropanꢀ1ꢀamine isomer (b) occurred, leading to a
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 39
1
2
3
4
5
6
7
8
regiospecific reaction in the radiolabeling of [18F]ꢀfluorodeprenyl. This method involving
nucleophilic radiofluorination at RT was successfully applied to the radiolabeling of [18F]ꢀ2ꢀ
fluoroethylamines in which the influence of the Nꢀsubstituent was also observed.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
INTRODUCTION
Positron emission tomography (PET) is a powerful in vivo imaging technique used for
diagnostic and prognostic purposes, therapy monitoring and as a research tool in various
clinical fields.1ꢀ3 Fluorineꢀ18 remains the major positron emitting isotope for the development
of radiopharmaceuticals due to its physical properties. The low positron energy emission
associated with the decay profile (97% β+ emission) yields the most favorable properties for
PET imaging, which results in accurate absolute quantification and highꢀresolution images,
especially for preclinical dedicated cameras.4 Its radioactive decay of 109.7 min allows for the
distribution of radiopharmaceuticals in nuclear medicine departments, which has resulted in
the considerable development of this functional molecular imaging technique. The extensive
use of PET has generated a demand for new radiofluorinated tracers, resulting in new
challenges for 18Fꢀradiochemistry and leading to the development of new methods.5
Important criteria for the radiopharmaceutical production efficiency are a high radiochemical
yield (RCY), rapid reaction and processes, simplicity in the automation of the process, and the
highest possible specific activity.6 To satisfy these criteria, radiolabeling should typically
occur as late as possible during radiosynthesis to obtain the final radiolabeled compound
within a minimum number of steps. Moreover, radiofluorination at RT can be beneficial for
the development of radiopharmaceutical kits and the radiolabeling of fragile structures.
Radiofluorination typically occurs via nucleophilic substitution because fluorideꢀ18 can be
produced in larger quantities than [18F]F2, which is obtained with a low specific radioactivity.6
Except for enzymatic fluorination, only a few examples of nucleophilic radiofluorination at
RT have been reported.7ꢀ11 Therefore, aliphatic nucleophilic incorporation of fluorideꢀ18 is
ACS Paragon Plus Environment

Page 3 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
typically performed at elevated temperatures using sulfonates or halides as leaving groups.12
When an Oꢀ or Nꢀ2ꢀ[18F]ꢀfluoroethyl moiety could not be directly labeled from the
corresponding precursor, [18F]ꢀfluoroethylhalides or sulfonates have been extensively used,
resulting in multistep radiolabeling.13 This approach requires the preparation of a 18Fꢀ
fluoroethylating agent and involves its eventual purification followed by an alkylation
reaction. Only a few βꢀ[18F]ꢀfluoroamines have been labeled using aziridine ring opening but
in low radiochemical yields.14,15 More recent studies obtained superior yields using aziridine
activated by Nꢀbenzoyl, Nꢀbenzyloxycarbonyl16,17 or Nꢀphenylsulfonyl substituents.18,19 The
ring opening of aziridine leads to a secondary or primary amine and requires a high
temperature for radiofluorination to occur, except for the case of phenylsulfonamides, in
which a temperature of 50 °C is sufficient to achieve radiolabeling.19
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
SCHEME 1. Substitution reaction via an aziridinium intermediate.
Recently, we reported a new labeling method for βꢀ[18F]ꢀfluoroamines through the SN2 ring
opening of aziridinium by [18F]ꢀfluoride (Scheme 1) developed on a model compound (i.e.,
N,Nꢀdibenzylphenylalaninol (1a)).20 The anchimericꢀassisted mechanism involving
aziridinium ring formation yielded two fluoroamine isomers 2a and 2b (Scheme 2).21,22 The
reaction performed with isomeric alcohol precursor 1b afforded the same mixture of isomers
2 as that obtained starting from 1a with similar yields, reinforcing the presence of an
aziridinium intermediate. Radiolabeling at RT led to [18F]2a and [18F]2b in 56ꢀ58%
radiochemical yield (RCY) from [18F]ꢀfluoride incorporation as a K[18F]F/K2.2.2. complex or
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 4 of 39
1
2
3
4
5
6
7
8
as [18F]TBAF ([18F]ꢀtetrabutylammonium fluoride). Nucleophilic SN2 radiofluorination with
the formation of a carbonꢀfluorine bond at RT in one step can be considered significant
progress in the development of radiopharmaceuticals.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
SCHEME 2. Fluorination of 1a and 1b.
In this study, we report the examination of the influence of Nꢀsubstituents on the reactivity
and selectivity of the anchimericꢀassisted radiofluorination of phenylalaninols, the differences
observed between radioactive and nonꢀradioactive fluorination and the unexpected
regioselectivity observed during some radiofluorinations.
RESULTS AND DISCUSSION
To study and compare the influence of the amine substituents on the reactivity of aziridinium,
different analogues of N,Nꢀdibenzylphenylalaninol (1a) (Scheme 3) were prepared by Nꢀ
alkylation from phenylalaninol 3a. N,Nꢀdiallyl (4) and N,Nꢀdipropargylphenylalaninol (5a)
were prepared by reaction with allyl and propargyl bromide, respectively. An Eschweilerꢀ
Clarke reaction using the nonꢀcomplete dialkylation reaction intermediate (6a) afforded
precursor 7a20 for the radiolabeling of [18F]ꢀfluorodeprenyl (8a), which is
a
radiopharmaceutical used to image monoamine oxidase B (MAOꢀB).23 Starting from 3a or Nꢀ
monobenzylphenylalaninol 9a, the same methylating method yielded 10 and 11, respectively.
Piperidinyl derivative 12 was obtained by the reaction of 3a with diiodopentane under reflux.
ACS Paragon Plus Environment

Page 5 of 39
The Journal of Organic Chemistry
1
2
SCHEME 3. Synthesis of N,N-disubstituted phenylalaninol substrates.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The conditions previously described for the fluorination of 1a in 64% yield were directly
applied to the substrates without any optimization (Table 1). Briefly, the hydroxyl substrate
was reacted at RT for 1 h with triflic anhydride (1.1 eq.), and then, diisopropylethylamine
(DIPEA, 1.2 eq.) was added followed by the addition of 2 eq. of TBAF in solution in THF for
2 h. At RT, fluorination of 5a (Table 1, entry 7) was as efficient as that observed for 1a, and
the reaction yield was higher (77%) with diallylꢀsubstituted compound 4 (Table 1, entry 3).
The piperidinyl derivative afforded a slightly lower yield (Table 1, entry 5) compared with 1a.
The presence of one methyl substituent rather than a second benzyl or propargyl group
afforded higher fluorination yields (Table 1, entries 9 and 11), especially for compound 11
compared with 1a (Table 1, entries 1). Different ratios were also observed in favor of isomer
a from 34:66 to 48:52 for 5a and 7a, respectively, compared with 39:61 for 1a (Table 1,
entries 7, 9 and 1). This result may be due to the stronger electrodonating effect of the methyl
substituent, which is reflected by the increased pKb of 11 (7.9 instead 6.9 for 1a, calculated
using ACD/Labs Software V11.02). By heating the reaction to 90 °C, the fluorination yield
increased (Table 1, entries 2, 4 and 8) to 88% for 13. However, this yield decreased slightly
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 39
1
2
3
4
5
6
7
8
9
for 14 and 8 without affecting the isomer ratio. Except for the piperidine compound, no effect
of the temperature on the isomer ratio was observed, demonstrating the kinetic fluorination of
aziridinium. For piperidine 15, the yield remained stable, and the ratio progressed in favor of
thermodynamic isomer b (53 to 69%), which may be due to the rigidity of the piperidine ring.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
TABLE 1. Fluorination of phenylalaninols. a
Reaction
Isolated
Entry Substrate
R1
R2
Product
Ratio a:b
temperature
yield (%)
1
2
Bn
Bn
Bn
Bn
RT
90 °C
RT
64
79
77
88
45
45
61
72
69
58
80
74
39:61
37:63
40:60
38:62
47:53
31:69
34:66
34:66
48:52
48:52
47:53
46:54
1a
1a
4
2
2
3
Allyl
Allyl
Allyl
Allyl
13
13
15
15
16
16
8
4
90 °C
RT
4
5
Piperidinyl
12
12
5a
5a
7a
7a
11
11
6
Piperidinyl
90 °C
RT
7
Propargyl Propargyl
Propargyl Propargyl
8
90 °C
RT
9
Propargyl
Propargyl
Bn
Me
Me
Me
Me
10
11
12
90 °C
RT
8
14
14
Bn
90 °C
a Substrate (0.8 mmol), Tf2O (1.1 eq.), DIPEA (1.2 eq.), 1 h at RT then TBAF (2 eq.), 2 h.
The influence of the Nꢀsubstituents on aziridinium ring opening by nucleophiles has rarely
been studied. Cossy observed a slight steric effect from benzylic Nꢀsubstituents (i.e., benzyl,
trityl, pꢀmethoxybenzyl and benzhydryl) for the ring extensions of pyrrolidines to
piperidines.24,25 A reactivity comparison between Nꢀbenzyl and Nꢀallyl substitutedꢀ
ACS Paragon Plus Environment

Page 7 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
phenylalaninols26 was studied in the presence of different nucleophiles, and better yields were
observed for allylic substituents compared with benzylic substituents on the amine, with a
slight effect on the reaction selectivity, which is consistent with our observations using
fluorine (Table 1, entries 1 and 3). In contrast, Sharpless and collaborators did not observe
any effects from the Nꢀsubstituents (i.e., morpholine, 1ꢀphenylpiperazine, diallyl and dibenzyl
amines) of 2ꢀchlorophenylalanine esters during the nucleophilic opening of the aziridinium
ring.27 Thermodynamic effects could not be retained in the regioselectivity with fluorine
because fluorine is not a good leaving group.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Because the 16a/16b and 8a/8b mixtures were not separable by silica gel chromatography, we
prepared pure and isolated products using a different approach to obtain standard compounds
for establishing the identity of the radiolabeled products by coelution using HPLC. After
removal of the Nꢀallyl substituents of 13a and 13b, amines 17a and 17b were isolated and
subjected to alkylations (Scheme 4).
The preparation of the aziridinium intermediate and radiofluorination were performed
according to the conditions previously described.20 Briefly, the hydroxyl precursor was stirred
in the presence of triflic anhydride in dichloromethane (DCM) for 1 h to yield the aziridinium
intermediate. Then, this aziridiniumꢀcontaining solution was added to dry [18F]ꢀfluoride which
was cooled to RT after azeotropic evaporation with acetonitrile at 90 °C. To investigate the
influence of the [18F]ꢀfluoride source, we performed the radiofluorination of 2 using isomers
1a and 1b with either the [18F]KF/K2.2.2 complex or [18F]TBAF at RT and 90 °C in acetonitrile
(Table 2).
Starting from each hydroxyl isomer (1a and 1b), we obtained an identical mixture of both
isomers ([18F]2a and [18F]2b, respectively). This result confirmed the exclusive anchimericꢀ
assisted mechanism leading to the formation of the identical aziridinium intermediate. If
another mechanism was involved (e.g., direct SN2 on the triflate leaving group or SN1), then
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 8 of 39
1
2
3
4
5
6
7
8
9
the corresponding isomer [18F]2a or [18F]2b starting from 1a or 1b, respectively, would be
favored. The radiofluorination of 1a and 1b resulted in comparable yields and ratios, despite
the [18F]ꢀfluoride source, reaction temperature or hydroxyl isomer precursor. Heating of the
reaction led to an increase in the radiofluorination yield by approximately 25% (Table 2,
entries 1 vs 2 and 7 vs 8) and reached 83% [18F]ꢀfluoride incorporation (Table 2, entry 2).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
SCHEME 4: Synthesis of isolated fluorinated standards.
Substituent effects on radiofluorination were investigated (Table 3). In contrast to 19Fꢀ
fluorination at RT (Table 1), radiolabeling starting from N,Nꢀdipropargylamine 5a afforded a
higher yield (Table 3, entry 11) than N,Nꢀdibenzylamine 1a (Table 3, entry 1), whereas N,Nꢀ
diallylamine yielded only 17% radiofluorination yield (Table 3, entry 3). Piperidine 12 did not
react at RT (Table 3, entry 26), whereas the nonꢀradioactive fluorination of 12 yielded 45%
product (Table 1, entry 5). When one methyl substituent was present on the amine, the RCY
decreased compared with that of the disubstituted corresponding analogue (Table 3, entries 1
ACS Paragon Plus Environment

Page 9 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
vs 23 and 11 vs 17). For the N,Nꢀdipropargyl and N,Nꢀmethylpropargyl amines (5a and 7a,
respectively), the radiofluorination yield decreased from 67 to 19%. Dimethylamine 10 did
not react under any of these conditions. It is important to note the absence of the [18F]ꢀfluoride
source effect using [18F]TBAF or the [18F]KF/K2.2.2 complex.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
TABLE 2. Radiofluorination starting from isomers 1a and 1b.
Reaction
RCY [18F]2a+[18F]2b
Ratiod
Entry Substratea 18FꢀꢀSourceb
temperature
(%)c
[18F]2a:[18F]2b
1
2
3
4
5
6
7
8
[18F]KF
[18F]KF
RT
90 °C
RT
58 ± 3
83 ± 1
46 ± 9
73 ± 4
56 ± 6
66 ± 4
40 ± 1
72 ± 3
33:67
33:67
32:68
35:65
34:66
35:65
34:66
34:64
1a
1a
1b
1b
1a
1a
1b
1b
[18F]KF
[18F]KF
90 °C
RT
[18F]TBAF
[18F]TBAF
[18F]TBAF
[18F]TBAF
90 °C
RT
90 °C
a Substrate (33 µmol). b [18F]KF : [18F]ꢀfluoride / K2CO3 (15 µmol) / K222 (18 µmol); [18F]TBAF : [18F]ꢀfluoride /
TBAHCO3 (23 µmol).c 18FꢀRCY established by radioꢀTLC (n = 3). d Established by radioꢀHPLC.
The order of reactivity at RT was as follows: N,Nꢀdipropargylamine > N,Nꢀ
methylpropargylamine = N,Nꢀdibenzylamine >> N,Nꢀdiallylamine > N,Nꢀmethylbenzylamine
>> piperidine > N,Nꢀdimethylamine. This order is in good agreement with the electronic
effect of the substituents on the electronic density of the nitrogen, which could be represented
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 10 of 39
1
2
3
4
5
6
7
8
9
by the respective pKb values (calculated using ACD/Labs Software V11.02) of precursors 5a,
7a, 1a, 4, 11, 12 and 10, which are 4.29, 6.58, 6.98, 7.14, 7.93, 8.96 and 8.88, respectively. It
is important to note that these results are different from those obtained by fluorination with
nonꢀradioactive [19F]ꢀfluorine (Table 1) in which the methyl substituent had a positive effect.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
SCHEME 5: Synthesis of phenylpropanol precursors.
TABLE 3. Radiofluorination of phenylpropanol and phenylalaninol isomers.
18Fꢀ
18Fꢀꢀ
Reaction
RCY (%)c
Ratiod
Entry Substratea
R1
R2
Product Sourceb temperature
[18F]a+b
[18F]a:b
1
Bn
Bn
Bn
Bn
[18F]TBAF
[18F]TBAF
[18F]TBAF
[18F]TBAF
[18F]KF
RT
90 °C
RT
56 ± 6
66 ± 4
17 ± 6
36 ± 3
10 ± 2
36 ± 5
0
34:66
35:65
36:64
44:56
35:65
42:58
ꢀ
1a
2
2
1a
2
3
Allyl
Allyl
Allyl
Allyl
Me
Allyl
Allyl
Allyl
Allyl
Me
4
13
13
13
13
18
18
18
4
90 °C
RT
4
5
4
6
[18F]KF
90 °C
RT
4
7
[18F]TBAF
[18F]TBAF
[18F]KF
10
8
Me
Me
90 °C
RT
12 ± 1
0
100:0
ꢀ
10
9
10
Me
Me
ACS Paragon Plus Environment

Page 11 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
10
5a
5a
5a
5a
5b
5b
7a
7a
7a
7a
7b
7b
11
11
12
12
Me
Me
18
16
16
16
16
16
16
8
[18F]KF
[18F]TBAF
[18F]TBAF
[18F]KF
90 °C
RT
6 ± 3
67 ± 5
59 ± 5
60 ± 6
60 ± 4
71 ± 2
57 ± 7
31 ± 6
19 ± 2
41 ± 4
17 ± 5
40 ± 3
21 ± 3
10 ± 1
21 ± 2
1 ± 1
100:0
38:62
44:56
39:61
47:53
33:67
41:59
55:45
100:0
60:40
96:4
Propargyl Propargyl
Propargyl Propargyl
Propargyl Propargyl
Propargyl Propargyl
Propargyl Propargyl
Propargyl Propargyl
90 °C
RT
[18F]KF
90 °C
RT
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
[18F]TBAF
[18F]TBAF
[18F]TBAF
[18F]TBAF
[18F]KF
90 °C
RT
Propargyl
Propargyl
Propargyl
Propargyl
Propargyl
Propargyl
Bn
Me
Me
Me
Me
Me
Me
Me
Me
90 °C
RT
8
8
[18F]KF
90 °C
RT
8
[18F]TBAF
[18F]TBAF
[18F]TBAF
[18F]TBAF
[18F]TBAF
[18F]TBAF
55:45
95:5
8
90 °C
RT
8
48:52
79:21
37:63
49:51
14
14
15
15
Bn
90 °C
RT
Piperidinyl
Piperidinyl
90 °C
44 ± 5
Substrate (33 µmol). [18F]KF: [18F]ꢀfluoride / K2CO3 (15 µmol) / K222 (18 µmol); [18F]TBAF: [18F]ꢀfluoride /
a
b
TBAHCO3 (23 µmol). c RCY established by radioꢀTLC (n = 3). d Established by radioꢀHPLC.
To confirm the anchimericꢀassisted mechanism, the reaction was performed starting from
isomers of 5 and 7. Compounds 5b and 7b were prepared following Scheme 5.
Radiofluorinations starting from 5b and 7b led to identical results for 5a and 7a with
[18F]TBAF and the [18F]KF/K2.2.2 complex, with similar yields and isomer ratios (Table 3,
entries 3 vs 5, 7 vs 9, 11 vs 13 and 17 vs 19).
At RT, the a:b isomeric ratios were identical for [18F]2, [18F]13, [18F]16 and [18F]15 (i.e.,
approximately 35:65). For [18F]8 and [18F]15, the a:b ratios (55:45 and 37:63, respectively)
were slightly different from those observed in Table 1 (48:52 and 47:53, respectively). The
ratios obtained for [18F]2, [18F]13, [18F]16 and [18F]14 were the same as those measured for
nonꢀradioactive fluorinations (Table 1).
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 12 of 39
1
2
3
4
5
6
7
8
At 90 °C, an increase in the yield was observed for all compounds, except for amines
containing a propargyl group, [18F]16 and [18F]8. For piperidine [18F]15, the radiofluorination
yield substantially increased to 44%, and only traces were observed at RT (Table 3, entries 25
9
and 26). For
N,Nꢀdimethylamine 10, small quantities were obtained (up to 12%) at 90 °C,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
whereas no reaction was detected at RT. Starting from
N,Nꢀdipropargylamine 5a or isomer
5b, the RCY of [18F]ꢀfluoride remained stable or exhibited a slight decrease (Table 3, entries
11 to 16). For
N,Nꢀmethylpropargylamine 7a and isomer 7b (Table 3, entries 17 to 22),
radiofluorination decreased from 41% to 17% (Table 3, entries 19 and 20).
Except for [18F]2, the isomer ratio varied when the reaction was heated to 90 °C. When the
yield increased, the ratio changed in favor of isomer a from 35:65 to 43:57 for [18F]13 and
from 37:63 to 49:51 for [18F]15. For
N,N
ꢀmethylbenzylamine, the [18F]14a:[18F]14b ratio
increased to 79:21 compared with 48:52 at RT. When the RCYs decreased at 90 °C,
substantial changes in the isomer ratio were observed. For example, the [18F]8a isomer
represented 55% of the products at RT but more than 95% at 90 °C (Table 3, entries 17, 19,
21 vs 18, 20, 21). This unexpected phenomenon occurred independently of the [18F]ꢀfluoride
source. The same results were obtained with isomers 7a and 7b, and a direct SN2 mechanism
via a triflate intermediate corresponding to isomer a can be dismissed. The influence of the
temperature on the radiofluorination yield was studied using three compounds (i.e., [18F]2,
[18F]16 and [18F]8) (Figure 1). An increase in the yield was observed for [18F]2, even at 90 °C
for both isomers. For compound [18F]16, the yield reached a plateau at RT for isomer
[18F]16a, whereas the yield of isomer [18F]16b decreased with heating. The same
phenomenon was observed with [18F]8.
ACS Paragon Plus Environment

Page 13 of 39
The Journal of Organic Chemistry
1
2
3
4
5
FIGURE 1. Effects of temperature on the radiofluorination with [18F]TBAF of [18F]2,
[18F]16 and [18F]8.
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 14 of 39
1
2
3
4
5
6
7
8
9
To confirm the hypothesis that the b isomer was degraded, we performed radiofluorination of
[18F]8 at RT and measured the yield and isomer ratio at different times. After 10 min at RT,
the [18F]8a:[18F]8b isomer ratio was 43:57, which remained stable after 30 min at RT.
However, the radioactive yield continued to increase (Figure 2, red curves). When the
reaction mixture was heated to 90 °C after 10 min at RT (Figure 2, blue curves), a rapid
degradation of isomer [18F]8b (Figure 2, dashed blue curve) was observed until it completely
disappeared. Simultaneously, the [18F]8a isomer continued to form more rapidly than at RT
(Figure 2, dotted blue curve). The degradation was not due to the heat itself. When the
reaction mixture was quenched with methanol after 10 min at RT, heating at 90 °C for an
additional 20 min did not result in degradation of the [18F]8b isomer. Therefore, the
degradation of [18F]8b was due to a secondary reaction in the reaction mixture during heating.
The difference in the stability of the fluorinated products from fluorineꢀ19 chemistry and
radiochemistry should be due to the difference in the concentration. The radioactive [18F]ꢀ
fluoride and [18F]-product are present in the reaction mixture in trace amounts (less than 0.1
µmol), but the precursor (33 µmol) and reactants (TBAHCO3, 22 µmol) are present in large
excess. These differences in the reaction conditions may explain the instability observed in
the radiochemistry. Due to these highly diluted conditions, some phenomena can be observed
in the radiochemistry but not in the fluorineꢀ19 chemistry. Therefore, the stable or decreased
RCYs observed at 90 °C were not due to a difference in reactivity but to the rapid degradation
of the b isomer in the reaction mixture, modifying the measured isomer ratio. If the isomer
ratio changes were due to degradation at 90 °C, then the order of the effect of the amine
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
substituent on the stability of the product would be as follows:
methylpropargylamine > ꢀmethylbenzylamine >> piperidine =
ꢀdiallylamine >> ꢀdibenzylamine.
N,
N
ꢀdimethylamine =
N,Nꢀ
N,
N
N,Nꢀdipropargylamine =
N
,N
N,N
ACS Paragon Plus Environment

Page 15 of 39
The Journal of Organic Chemistry
1
2
3
4
5
FIGURE 2. Kinetics and effects of heating during radiofluorination with [18F]TBAF
starting from 7a.
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
N,NꢀMethylpropargylphenylalaninol (7a) (Scheme 4) corresponds to the precursor of [18F]ꢀ
fluorodeprenyl ([18F]8a), which is a radiopharmaceutical used to image monoamine oxidase B
(MAOꢀB).23,28 The [18F]ꢀfluoride RCY was 23% for the labeling of this radiotracer, and the
isomer ratio was inverted in favor of 1ꢀ[18F]ꢀfluoropropyl isomer ([18F]8a), which resulted in
the selective production of [18F]ꢀfluorodeprenyl. [18F]ꢀFluorodeprenyl was previously
prepared by heating a mixture of two chloride isomers in the presence of [18F]ꢀfluoride,
resulting in 50% RCY.25 However, the 18Fꢀisomer ratio was not specified. In this case, no
evidence of aziridinium ring opening by [18F]ꢀfluoride was observed because the competitive
SN2 mechanism would occur at the temperature used in the reaction (120 °C).
SCHEME 6. Radiolabeling of [18F]-fluoroethylamines.
R
1) Tf2O
2) TEA
R
Bn
Bn
Bn
NH
N
18/19F-
N
Bn
R
RBr, K2CO3
CH3CN
N
OH
18/19F
OH
25 R=Bn; 26 R=propargyl
27 R=allyl; 28 R=Me
21 R=Bn; 22 R=propargyl
23 R=allyl; 24 R=Me
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 16 of 39
1
2
TABLE 4. Radiofluorination of ethylamines.
3
4
5
6
7
8
Reaction
Entry Substratea
R
18FꢀProduct
18Fꢀꢀsourceb
18FꢀRCY (%)c
temperature
9
1
Benzyl
Benzyl
Benzyl
Benzyl
Propargyl
Propargyl
Propargyl
Propargyl
Allyl
[18F]KF
[18F]KF
RT
90 °C
RT
23 ± 2
35 ± 4
26 ± 2
48 ± 2
22 ± 6
29 ± 4
26 ± 4
24 ± 2
14 ± 3
16 ± 3
14 ± 3
9 ± 2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
21
25
25
25
25
26
26
26
26
27
27
27
27
28
28
30
30
2
21
3
[18F]TBAF
[18F]TBAF
[18F]KF
21
4
90 °C
RT
21
5
22
6
[18F]KF
90 °C
RT
22
7
[18F]TBAF
[18F]TBAF
[18F]KF
22
8
90 °C
RT
22
9
23
10
11
12
13
14
15
16
23
23
23
24
24
29
29
Allyl
[18F]KF
90 °C
RT
Allyl
[18F]TBAF
[18F]TBAF
[18F]TBAF
[18F]TBAF
[18F]TBAF
[18F]TBAF
Allyl
90 °C
RT
Methyl
Methyl
0
90 °C
RT
0
12 ± 1d
21 ± 1d
90 °C
a Substrate (33 µmol).20 b [18F]KF: [18F]ꢀfluoride / K2CO3 (15 µmol) / K222 (18 µmol); [18F]TBAF: [18F]ꢀfluoride /
TBAHCO3 (23 µmol). c RCY established by radioꢀTLC (n = 3). d n = 2.
To apply our method to a larger number of radiopharmaceuticals, the radiofluorination of 2ꢀ
[18F]ꢀfluoroethylamines was investigated (Scheme 6). Because the aziridinium ring is
symmetrical in these cases, the ring opening yielded only one radiofluorinated product. The
reaction conditions developed for the radiolabeling of 1a were applied to the radiofluorination
of fluoroethylamines 25-30 (Table 4). At RT, the radiofluorination yields of [18F]25 and
[18F]28 were lower than those for the corresponding phenylalaninol analogues (23ꢀ26% for
[18F]25 compared with 58 and 46% for [18F]2, no reaction was observed for [18F]28 and 10%
ACS Paragon Plus Environment

Page 17 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
was obtained for [18F]14). The effect of the benzylic substituent on the aziridinium ring
appears to be important for the reactivity of aziridinium. At 90 °C, the radiofluorination yield
increased for [18F]25 (Table 4, entries 2, 4 and 1, 3), whereas no [18F]28 was obtained from
precursor 24. For analogues [18F]27 and [18F]22, heating did not affect the incorporation of
[18F]ꢀfluoride, regardless of the [18F]ꢀfluoride source. These radiochemical yields can be
compared with the yields from the classical twoꢀstep radiosynthesis, which involves the
preparation of 2ꢀ[18F]ꢀfluoroethylsulfonate or a halide followed by Nꢀalkylation, leading to the
Nꢀ[18F]ꢀfluoroethylamine product with poor radiochemical yields.12 The method was applied
to the radiolabeling of a radiopharmaceutical used to image dopamine transporters in the brain
(i.e., [18F]ꢀFECNT).29,30 The current radiosyntheses involve the alkylation of a nortropane
precursor by different [18F]ꢀfluoroethylsulfonates or bromide in overall radiochemical yields
of 16.5% to 40% (decay corrected at the end of bombardment).31ꢀ33 Recently, radiosyntheses
in one step from a mesylate precursor were reported with similar radiochemical yields of
approximately 45%.34,35 However, the precursor developed for this synthesis is unstable and
requires specific conditions for storage. Here, we applied our method starting from stable
alcohol 29, which afforded 12% and 21% of the desired product ([18F]ꢀFECNT (30)) at RT
and 90 °C, respectively, without any optimization of the reaction conditions (Scheme 7)
(Table 4, entries 15 and 16).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
SCHEME 7. Radiolabeling of [18F]-FECNT.
Although the radiochemical yield remains low compared with the other method, this method
offers the advantages of a stable precursor and a oneꢀstep reaction, which facilitate
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 18 of 39
1
2
3
4
5
6
7
8
9
radiosynthesis automation. In addition, the instability of the previously reported mesylate
precursor may be related to the fact that this anchimericꢀassisted reaction occurs over time.
The radiosyntheses with high levels of radioactivity were performed on a commercially
available GE TRACERLab® FX module to prepare more than 1 GBq of a mixture of [18F]2a
and [18F]2b at RT with a 70:30 ratio within 70 min. The specific radioactivity of [18F]2b and
[18F]2a was similar to typical fluorineꢀ18 specific radioactivities obtained in our lab (i.e., 130ꢀ
320 GBq/µmol at the end of bombardment).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
CONCLUSION
We developed a nucleophilic radiofluorination method using [18F]ꢀfluoride via an anchimericꢀ
assisted mechanism for the radiolabeling of βꢀ[18F]ꢀfluoroamines. Aziridinium intermediates
were obtained from stable phenylalaninol precursors. Oneꢀpot 19Fꢀfluorination using TBAF
yielded two isomers resulting from aziridinium ring opening. The Nꢀsubstituents resulted in a
variation of the isomer ratios (a:b) between 31:69 and 48:52, with fluorination yields ranging
from 45% to 88%. The reaction temperature did not significantly affect the efficiency or
regioselectivity of the fluorination. Radioactive βꢀ[18F]ꢀfluoroamines were obtained in one
step by nucleophilic substitution starting from [18F]ꢀfluoride. The [18F]ꢀfluoride forms (i.e.,
[18F]TBAF or [18F]KF/K2.2.2 complex) did not affect the radiochemical yield. We
demonstrated the neighboringꢀgroup mechanism by performing the radiofluorination starting
from both of the phenylalaninol precursors (a:b isomers of alcohols 1, 5 and 7), which led to
similar [18F]ꢀfluorine RCYs and an identical a:b ratio. The N,Nꢀsubstituents exerted a
substantial influence on the radiofluorination. [18F]ꢀFluoride incorporations varied from no
reaction (for N,Nꢀdimethylamine or piperidine) to 71% yield (for N,Nꢀdipropargylamine) at
RT. The isomer ratio was mostly comparable to the ratios obtained using 19Fꢀfluorine.
Heating during radiofluorination substantially improved the radiochemical yield in some
cases (from 1% to 44% for piperidine). However, when the radiochemical yield increased
ACS Paragon Plus Environment

Page 19 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
slightly or even decreased when heated at 90 °C, we observed a change in the a:b ratio in
favor of 1ꢀfluoroꢀpropanꢀ2ꢀamine isomer a. The ratio changes were due to the instability of
the [18F]b isomer in the reaction mixture. The Nꢀmethyl substituent resulted in a very unstable
b isomer, leading to selective radiofluorination of the [18F]a isomer at 90 °C. This
radiolabeling method was extended to βꢀ[18F]ꢀfluoroethylamines, which afforded a unique
radioactive product in moderate to low yields without any optimization of the reaction
conditions. The method was successfully applied to the radiolabeling of two
radiopharmaceuticals, [18F]ꢀfluorodeprenyl and [18F]ꢀFECNT. For [18F]ꢀfluorodeprenyl,
radiofluorination only yielded the a isomer at 90 °C. This new radiofluorination method may
be useful for the routine preparation of 18Fꢀradiopharmaceuticals starting from stable
precursors in a fewer number of steps. The possibility to perform the nucleophilic
radiofluorination at RT also affords new opportunities in the field of PET chemistry.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
EXPERIMENTAL SECTION
Materials and methods
Reagents and solvents used, unless stated otherwise, were of commercially available reagents
grade quality and were used without further purification. Thin layer chromatographies (TLC)
were run on preꢀcoated aluminum plates of silica gel 60F254 and retention factor (Rf) were
established using a UVꢀlamp at 254 nm or visualization with ninhydrin solution. Silica gel
flashꢀchromatographies were performed on prepacked columns (20–40 µm). Optical rotations
were determined on a polarimeter and are given in cm3 gꢀ1 dmꢀ1 while the concentrations are
given in g cmꢀ3. Melting points were determined on a digital melting point apparatus and are
1
uncorrected. H, 13C and 19F NMR spectra were recorded at 400 MHz (1H), 100.6 MHz (13C)
and 376.5 MHz (19F). Fluorineꢀ19 NMR spectra are coupled to proton. Chemical shifts were
reported as parts per million (δ in ppm) using tetramethylsilane (TMS) as internal standard or
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 20 of 39
1
2
3
4
5
6
7
8
9
by reference to proton resonances resulting from incomplete deuteration of the NMR solvent.
Coupling patterns are abbreviated as: s (singlet), bs (broad singlet), d (doublet), t (triplet), m
(multiplet), dd (doublet of doublet), dt (doublet of triplet). IR spectra were recorded on a FTꢀ
IR spectrometer and are given in cmꢀ1. High resolution mass spectra (HRMS) were obtained
on a QꢀTOF spectrometer by electrospray ionization (ESIꢀTOF). Analytical HPLC was
realized with a photodiode arrays detector (198ꢀ380 nm) coupled with a NaI probe radioactive
detector. Purity was determined by HPLC on an analytical column (MachereyꢀNagel,
Nucleodur C18 Gravity, 250 × 4.6mm, 5µm; flow rate: 1 mL/min; UVꢀdetection λ = 210 nm).
The purity of compounds was found to be more than 98%. Radioactivity measurements were
carried out with an ionization chamber.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(S)
-2-(N-Benzylamino)-3-phenylpropan-1-ol (9) and
(S)-2-(N,N-dibenzylamino)-3-
phenylpropan-1-ol (1a)
To a solution of 3a (2.02 g, 13.2 mmol) in DCM (5 mL) at RT was added Na2CO3 (0.95 g, 8.9
mmol) followed by benzylbromide (1.06 mL, 8.9 mmol). The mixture was heated to 110 °C
and stirred overnight. After cooled to RT, H2O (20 mL) was added to the mixture and aqueous
layer was extracted with DCM (3 × 20 mL). The combined organic layers were dried over
MgSO4, filtered and concentrated under vacuum. Purification by chromatography on silica gel
(heptane/AcOEt, 90/10 to 40/60) gave compound 9 as a white solid (1.52 g, 48%) and 1a (306
20
mg, 7%) as yellow oil. 9: m.p.: 67ꢀ68 °C (litt. 54ꢀ56 °C);36 [α]D ꢀ18.3 (c 1.00, CHCl3).
20
Characterization data are in accordance to the published data.36 1a: [α]D = +40 (c 1.90,
CHCl3); HPLC purity: eluent MeOH/H2O 85/15, tr = 10.6 min. Characterization data are in
accordance to the published data.37
(S)-2-(N,N-Diallylamino)-3-phenylpropan-1-ol (4)
ACS Paragon Plus Environment

Page 21 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Compound 4 was prepared following the procedure described for the synthesis of 1b using
(S)ꢀ2ꢀaminoꢀ3ꢀphenylꢀpropanꢀ1ꢀol (3a)37 (752 mg, 4.98 mmol), K2CO3 (1.51 g, 10.93 mmol)
and allylbromide (0.88 mL, 10.17 mmol). Product 4 was obtained after purification by
chromatography on silica gel (heptane/AcOEt, 90/10) as an yellow oil (790 mg, 69%). [α]D20 ꢀ
35.4 (c 1.00, CHCl3); HPLC purity: eluent MeOH/H2O 85/15, tr = 6.8 min. Characterization
data are in accordance to the published data.38
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(S)-2-(N-Propargylamino)-3-phenylpropan-1-ol (6a) and (S)-2-(N,N-dipropargylamino)-
3-phenylpropan-1-ol (5a)
Compound 6a was prepared following the procedure described for the synthesis of 1b,
starting from (S)ꢀ2ꢀaminoꢀ3ꢀphenylꢀpropanꢀ1ꢀol (3a)37 (1.93 g, 12.8 mmol), K2CO3 (3.88 g,
10.9 mmol) and a solution of propargylbromide in toluene (80 wt.%, 1.4 mL, 13.0 mmol).
After purification by chromatography on silica gel (heptane/AcOEt, 80/20 to 60/40), product
6a was obtained as a white solid (1.32 g, 55%) and compound 5a (640 mg, 22%) as an yellow
20
1
oil. 6a: m.p. 71ꢀ72 °C; [α]D +9.4 (c 1.00, CHCl3); H NMR (400 MHz, CDCl3, TMS) δ
7.32ꢀ7.20 (m, 5H), 3.66 (dd, 2JHH = 11.2 Hz, 3JHH = 3.6 Hz, 1H), 3.44 (d, 4JHH = 2.4 Hz, 2H),
2
3
3.38 (dd, JHH = 11.2 Hz, JHH = 4.8 Hz, 1H), 3.16ꢀ3.10 (m, 1H), 2.83ꢀ2.73 (m, 2H), 2.20 (t,
4JHH = 2.4 Hz, 1H), 1.97 (bs, 2H); 13C NMR (100.6 MHz, CDCl3, TMS) δ 138.5, 129.6,
128.9, 126.8, 82.2, 71.9, 62.5, 58.9, 38.2, 36.1; IR (ATR) νmax: 3294, 2929, 1494, 1451, 1338,
1053, 743, 699; HRMS (ESIꢀTOF) m/z: [M+H]+ Calcd. for C12H16NO 190.1232; Found
190.1223. 5a [α]D20 ꢀ8.2 (c 0.72, CHCl3); 1H NMR (400 MHz, CDCl3, TMS) δ 7.30ꢀ7.16 (m,
2
4
5H), 3.65 and 3.60 (ABX system, JAB = 17.2 Hz, 4JAX = JBX = 2.4 Hz, 4H), 3.47ꢀ3.37 (m,
4
2H), 3.26ꢀ3.18 (m, 2H), 2.72 (bs, 1H), 2.48ꢀ2.42 (m, 1H), 2.29 (t, JHH = 2.4 Hz, 2H); 13C
NMR (100.6 MHz, CDCl3, TMS) δ 139.1, 129.3, 128.9, 126.7, 80.2, 73.5, 65.0, 60.7, 39.3,
33.5; IR (ATR) νmax: 3288, 2928, 1030, 742, 632; HRMS (ESIꢀTOF) m/z: [M+H]+ Calcd. for
C15H18NO 228.1388; Found 228.1377; HPLC purity: eluent MeOH/H2O 70/30, tr = 7.2 min.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 22 of 39
1
2
(S)-2-(N
,
N
-Benzylmethyl-amino)-3-phenylpropan-1-ol (11)
3
4
5
6
7
8
Compound 11 was prepared following the procedure described for the synthesis of 10,
starting from 9 (720 mg, 3.0 mmol), formic acid (0.57 mL, 15.1 mmol) and formaldehyde
(37% in H2O) (0.67 mL, 9.4 mmol). Compound 11 was obtained after purification by
chromatography on silica gel (AcOEt/Hept, 85/15) as colorless oil (659 mg, 87%). [α]D20 ꢀ6.0
(c 1.00, CHCl3). Characterization data are in accordance to the published data.39
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(S)-3-Phenyl-2-piperidin-1-yl-propan-1-ol (12)
To a solution of 3a (500 mg, 3.3 mmol) in CH3CN (10 mL) at 0 °C was added K2CO3 (1 g,
7.3 mmol) followed by diiodopentane (0.52 mL, 3.5 mmol). The mixture was heated to reflux
and stirred overnight. The solvent was evaporated before CH2Cl2 (20 mL) and water (20 mL)
were added to the residue. The layers were separated and the aqueous layer was extracted
further with CH2Cl2 (2 × 20 mL). The combined organic layers were dried over MgSO4,
filtered and concentrated under vacuum. Purification by chromatography on silica gel
(heptane/AcOEt, 65/25) gave compound 12 as a white solid (455 mg, 63%). m.p.: 50ꢀ51 °C;
[α]D20 ꢀ3.3 (c 1.0, CHCl3). Characterization data are in accordance to the published data.40
(S)-2-(N,N-Methylpropargylamino)-3-phenylpropan-1-ol (7a)
A mixture of 6a (200 mg, 1.06 mmol), formic acid (0.2 mL, 5.30 mmol) and formaldehyde
(37% in water) (0.12 mL, 1.6 mmol) was stirred at 105 °C overnight. The mixture was cooled
to RT and acidified to pH 1 with HCl (2 N). The solvent was evaporated and CH2Cl2 (20 mL)
and water (20 mL) were added to the residue. The layers were separated and the aqueous layer
was washed further with water (2 × 20 mL). The combined aqueous layers were brought to
pH 8ꢀ9 by the addition of aqueous ammonia (28%) and extracted with CH2Cl2 (3 × 30 mL).
The combined organic layers were dried over MgSO4, filtered and concentrated under
vacuum. Compound 7a was obtained after purification by chromatography on silica gel
ACS Paragon Plus Environment

Page 23 of 39
The Journal of Organic Chemistry
1
2
3
(CH2Cl2/MeOH, 99/1) as colorless oil (154 mg, 72%). [α]D20 ꢀ20.1 (c 1.00, CHCl3); 1H NMR
4
5
6
(400 MHz, CDCl3, TMS) δ 7.30ꢀ7.16 (m, 5H), 3.50ꢀ3.32 (m, 4H), 3.11ꢀ3.06 (m, 2H), 2.43 (s,
4
3H), 2.42ꢀ2.35 (m, 1H), 2.29 (t, JHH = 2.4 Hz, 1H); 13C NMR (100.6 MHz, CDCl3, TMS) δ
7
8
9
139.3, 129.2, 128.7, 126.4, 80.5, 73.2, 65.5, 60.4, 43.8, 36.1, 32.4; IR (ATR) νmax: 3289,
2935, 2799, 1453, 1030,740, 699; HRMS (ESIꢀTOF) m/z: [M+H]+ Calcd. for C13H18NO
204.1388; Found 204.1387; HPLC purity: eluent MeOH/H2O 65/35, tr = 8.9 min.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(S)-N,N-Dimethyl-2-amino-3-phenylpropan-1-ol (10)
A mixture of 3a (503 mg, 3.3 mmol), formic acid (0.63 mL, 17 mmol) and formaldehyde
(37% in water) (0.76 mL, 10 mmol) was stirred at 105 °C overnight. The mixture was cooled
to RT and acidified to pH 1 with HCl (2N). The solvent was evaporated and CH2Cl2 (20 mL)
and water (20 mL) were added to the residue. The layers were separated and the aqueous layer
was washed further with water (2 × 20 mL). The combined aqueous layers were brought to
pH 8ꢀ9 by the addition of aqueous ammonia (28%) and extracted with CH2Cl2 (3 × 30 mL).
The combined organic layers were dried over MgSO4, filtered and concentrated under
vacuum. Compound 10 was obtained after purification by chromatography on silica gel
(CH2Cl2/MeOH, 98/2) as white solid (372 mg, 62%). m.p.: 48ꢀ50 °C (lit. 49ꢀ51 °C);41 [α]D20 ꢀ
1
1.2 (c 1.0, CHCl3); H NMR (400 MHz, CDCl3, TMS) δ 7.28ꢀ7.11 (m, 5H), 3.40ꢀ3.29 (m,
2H), 3.16 (bs, 1H), 2.92ꢀ2.81 (m, 2H), 2.33 (s, 6H), 2.31ꢀ2.25 (m, 1H); 13C NMR (100.6
MHz, CDCl3, TMS) δ 139.6, 129.1, 128.6, 126.3, 67.0, 60.6, 40.31, 40.29, 30.7; IR (ATR)
νmax: 3334, 2929, 2859, 2813, 1602, 1103, 1026, 732, 698; HRMS (ESIꢀTOF) m/z: [M+H]+
Calcd. for C11H18NO, 180.1388, found 180.1386.
Typical procedure for fluorination
To a solution of alcohol (0.6 mmol) in dry CH2Cl2 (4.5 mL) in a conic vial under nitrogen
atmosphere was added a solution of trifluoromethanesulfonic anhydride (1 M in CH2Cl2, 0.66
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 24 of 39
1
2
3
4
5
6
7
8
9
mL). The solution was stirred at RT for 1 h. N,NꢀDiisopropylethylamine (125 µL, 0.72 mmol)
was then added and after 1 minute stirring, a solution of TBAF (1 M in THF, 1.2 mL) was
added. The reaction mixture was stirred further for 2 h at RT and quenched with aqueous
NaOH (15%, 5 mL). The aqueous layer was extracted with CH2Cl2 (2 × 10 mL), the
combined organic layers were concentrated under reduced pressure and the residue was
purified by chromatography on silica gel.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(
R N,N-Dibenzyl-2-fluoro-3-phenylpropan-1-amine (2b) and (S)-N,N-dibenzyl-1-fluoro-
)-
3-phenylpropan-2-amine (2a)
Following the typical procedure for fluorination, starting from (S)ꢀ2ꢀ(N,Nꢀdibenzylamino)ꢀ3ꢀ
phenylpropanꢀ1ꢀol 1a,42 isomers 2b (50 mg, 25%) and 2a (78 mg, 39%) were obtained after
purification by chromatography on silica gel (heptane/AcOEt, 100/1) as colorless oils.
20
2b: [α]D ꢀ1.9 (c 2.48, CHCl3); HPLC purity: eluent MeOH/H2O 85/15, tr = 25.5 min.
Characterization data are in accordance to the published data.22
20
2a: [α]D ꢀ19.9 (c 1.36, CHCl3); HPLC purity: eluent MeOH/H2O 85/15, tr = 28.7 min.
Characterization data are in accordance to the published data.22
(
R N,N-Diallyl-2-fluoro-3-phenylpropan-1-amine (13b) and (S)-N,N-diallyl-1-fluoro-3-
)-
phenylpropan-2-amine (13a)
Following the typical procedure for fluorination, starting from 4 (184 mg, 0.8 mmol), isomers
13b (58 mg, 31%) and 13a (86 mg, 46%) were obtained after purification by chromatography
20
1
on silica gel (heptane/AcOEt, 98/2) as colorless oils. 13b: [α]D +0.3 (c 2.2, CHCl3); H
NMR (400 MHz, CDCl3, TMS) δ 7.32ꢀ7.21 (m, 5H), 5.88ꢀ5.78 (m, 2H), 5.16ꢀ5.10 (m, 4H),
3
4.92ꢀ4.74 (m, 1H), 3.15 (d, JHH = 6.0 Hz, 4H), 2.96ꢀ2.88 (m, 2H), 2.73ꢀ2.59 (m, 2H); 13C
3
NMR (100.6 MHz, CDCl3, TMS) δ 137.7 (d, JCF = 4.2 Hz), 135.8, 129.7, 128.7, 126.9,
ACS Paragon Plus Environment

Page 25 of 39
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
1
2
2
118.0, 93.7 (d, JCF = 173.1 Hz), 58.0, 56.7 (d, JCF = 21.5 Hz), 40.1 (d, JCF = 21.2 Hz); 19F
NMR (376.5 MHz, CDCl3) δ ꢀ180.2; IR (ATR) νmax: 3077, 3029, 3026, 1643, 1454, 1030,
995, 918, 744, 699; HRMS (ESIꢀTOF) m/z: [M+H]+ Calcd. for C15H21NF 234.1658; Found
9
20
234.1653; HPLC purity: eluent MeOH/H2O 80/20, tr = 7.7 min. 13a: [α]D ꢀ22.6 (c 1.0,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
CHCl3); H NMR (400 MHz, CDCl3, TMS) δ 7.30ꢀ7.17 (m, 5H), 5.82ꢀ5.72 (m, 2H), 5.20ꢀ
5.07 (m, 4H), 4.53ꢀ4.32 (m, 2H), 3.30ꢀ3.13 (m, 5H), 2.85 (dd, 2JHH = 13.2 Hz, 3JHH = 5.2 Hz,
1H), 2.72 (dd, JHH = 13.2 Hz, JHH = 9.2 Hz, 1H); 13C NMR (100.6 MHz, CDCl3, TMS) δ
2
3
1
2
140.0, 137.5, 129.6, 128.7, 126.4, 117.0, 83.9 (d, JCF = 172.0 Hz), 61.0 (d, JCF = 17.5 Hz),
4
19
53.9 (d, JCF = 1.8 Hz), 33.2; F NMR (376.5 MHz, CDCl3) δ ꢀ226.2; IR (ATR) νmax: 3078,
3027, 3026, 1641, 1455, 991, 918, 742, 699; HRMS (ESIꢀTOF) m/z: [M+H]+ Calcd. for
C15H21NF 234.1658; Found 234.1660; HPLC purity: eluent MeOH/H2O 80/20, tr = 10.2 min.
N
-Benzyl-
N-methyl-(1-benzyl-2-fluoroethyl)amine (14a) and N-benzyl-N-methyl-(2-
fluoro-3-phenylpropyl)amine (14b)
Following the typical procedure for fluorination, starting from 11 (153 mg, 0.6 mmol),
isomers 14a (58 mg, 38%) and 14b (65 mg, 42%) were obtained after purification by
20
chromatography on silica gel (heptane/AcOEt, 95/5) as colorless oils. 14a: [α]D ꢀ38.6 (c
1
1.00, CHCl3); H NMR (400 MHz, CDCl3, TMS) δ 7.31ꢀ7.18 (m, 10H), 4.62ꢀ4.40 (m, 2H),
2
3
3.74 (s, 2H), 3.14ꢀ3.04 (m, 1H), 2.93 (dd, 1H, JHH = 13.6 Hz, JHH = 6.0 Hz), 2.78 (dd, 1H,
2JHH = 13.6 Hz, JHH = 8.8 Hz), 2.37 ( s, 3H); C NMR (100.6 MHz, CDCl3, TMS) δ 140.1,
3
13
1
2
139.8, 129.6, 128.9, 128.7, 127.2, 126.5, 83.4 (d, JCF = 171.1 Hz), 64.3 (d, JCF = 17.3 Hz),
59.3 (d, 4JCF = 1.8 Hz), 38.0 (d, 4JCF = 1.6 Hz), 33.1 (d, 3JCF = 6.6 Hz); 19F NMR (376.5 MHz,
CDCl3) δ ꢀ225.4; IR (ATR) νmax: 2951, 2792, 1453, 1010, 732, 697; HRMS (ESIꢀTOF) m/z:
[M+H]+ Calcd. for C17H21NF 258.1658; Found 258.1662; HPLC purity: eluent MeOH/H2O
20
1
85/15, tr = 9.5 min. 14b: [α]D +1.5 (c 1.00, CHCl3); H NMR (400 MHz, CDCl3, TMS) δ
7.20ꢀ7.16 (m, 10H), 4.91ꢀ4.73 (m, 1H), 3.52 (s, 2H), 2.91 (dd, 2H, 2JHH = 23.2 Hz, 3JHH = 6.4
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 26 of 39
1
2
3
Hz), 2.67ꢀ2.50 (m, 2H), 2.24 (s, 3H); 13C NMR (100.6 MHz, CDCl3, TMS) δ 139.1, 137.5 (d,
4
5
6
1
4JCF = 4.3 Hz), 129.7, 129.3, 128.7, 128.6, 127.4, 126.9, 93.5 (d, JCF = 171.7 Hz), 63.0 (d,
2
4
2
4JCF = 1.2 Hz), 60.4 (d, JCF = 21.5 Hz), 43.4 (d, JCF = 1.5 Hz), 40.0 (d, JCF = 21.1 Hz); 19F
NMR (376.5 MHz, CDCl3) δ ꢀ179.9; IR (ATR) νmax: 2945, 2787, 1453, 1020, 738, 697;
HRMS (ESIꢀTOF) m/z: [M+H]+ Calcd. for C17H21NF 258.1658; Found 258.1659; HPLC
purity: eluent MeOH/H2O 85/15, tr = 7.7 min.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(S
)-1-(1-Benzyl-2-fluoro-ethyl)-piperidine (15a) and (
R)-1-(2-fluoro-3-phenyl-propyl)-
piperidine (15b)
Following the typical procedure for fluorination, starting from 12 (131mg, 0.6 mmol), isomers
15a (28 mg, 21%) and 15b (32 mg, 24%) were obtained after purification by chromatography
20
1
on silica gel (heptane/AcOEt, 85/5) as colorless oils. 15a: [α]D ꢀ12.5 (c 1.00, MeOH); H
NMR (400 MHz, CD3OD) δ 7.34ꢀ7.21 (m, 5H), 4.61ꢀ4.34 (m, 2H), 2.96ꢀ2.65 (m, 7H), 1.65ꢀ
1.57 (m, 4H), 1.52ꢀ1.46 (m, 2H); 13C NMR (100.6 MHz, CD3OD, TMS) δ 140.1, 129.6,
128.7, 126.4, 82.7 (d, 1JCF = 171.0 Hz), 67.1 (d, 2JCF = 17.3 Hz), 51.1 (d, 4JCF = 1.5 Hz), 32.9
(d, 3JCF = 6.7 Hz), 26.9, 25.0; 19F NMR (376.5 MHz, CD3OD) δ ꢀ224.6; IR (ATR) νmax: 2931,
1467, 1000, 1029, 734, 698; HRMS (ESIꢀTOF) m/z: [M+H]+ Calcd. for C14H21NF 222.1658;
20
Found 222.1664; HPLC Purity: eluent : MeOH/H2O/TFA 15/85/0.1, tr = 22.8 min. 15b: [α]D
1
+0.6 (c 1.00, MeOH); H NMR (400 MHz, CD3OD) δ 7.34ꢀ7.20 (m, 5H), 4.99ꢀ4.81 (m, 1H),
3.00ꢀ2.93 (m, 2H), 2.64ꢀ2.51 (m, 2H), 2.48ꢀ2.44 (m, 4H), 1.63ꢀ 1.55 (m, 4H), 1.46ꢀ1.42 (m,
2H); 13C NMR (100.6 MHz, CD3OD, TMS) δ 137.4 (d, JCF = 4.5 Hz), 129.7, 128.7, 126.8,
4
93.0 (d, 1JCF = 171.6 Hz), 62.6 (d, 2JCF = 21.0 Hz), 55.4 (d, 4JCF = 1.2 Hz), 40.3 (d, 2JCF = 21.3
Hz), 26.3, 24.5; 19F NMR (376.5 MHz, CD3OD) δ ꢀ178.2; IR (ATR) νmax: 2931, 1468, 1120,
1029, 777, 698; HRMS HRMS (ESIꢀTOF) m/z: [M+H]+ Calcd. for C14H21NF 222.1658;
Found 222.1665; HPLC Purity: Column: MachereyꢀNagel, Nucleodur C18 Polartech, 250
×4.6 mm, 5µm; flow rate: 1 mL/min; eluent : MeOH/H2O/TFA 15/85/0.1, tr = 20.1 min.
ACS Paragon Plus Environment

Page 27 of 39
The Journal of Organic Chemistry
1
2
(S)-1-Fluoro-3-phenylpropan-2-ammonium chloride (17a)
3
4
5
6
A
solution of 1,4ꢀbis(diphenylphosphino)butane (110 mg, 0.26 mmol) and
7
8
9
tris(dibenzylideneacetone)dipalladium (196 mg, 0.34 mmol) in dry THF (9 mL) was stirred
under argon for 15 minutes at RT. The orange mixture was added to a solution of 13a (200
mg, 0.86 mmol) and thiosalicylic acid (572 mg, 3.71 mmol) in dry THF (15 mL). The mixture
was then stirred under argon at 60 °C for 5 h. After cooling to RT, the reaction was quenched
with HCl (1 N, 10 mL) and AcOEt (10 mL) was added to the mixture. The two layers were
separated and the organic layer was extracted with HCl (1 N, 2 × 5 mL). The combined
aqueous layers were alkalized with a solution of NaOH (1 N) and extracted with CH2Cl2 (3 ×
20 mL). The combined organic layers were dried over MgSO4, filtered and an excess of HCl
(1 N in diethylether) was added (2 mL, 2 mmol). The solvent was evaporated under reduced
pressure and the residue was washed with Et2O to give 17a (120 mg, 74%) as white solid.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
20
1
m.p. 136ꢀ138 °C (litt. 143ꢀ144 °C from CH3CN)43; [α]D ꢀ10.2 (c 0.76, MeOH); H NMR
(400 MHz, CD3OD) δ 7.43ꢀ7.32 (m, 5H), 4.89 (bs, 3H), 4.72ꢀ4.42 (m, 2H), 3.83ꢀ3.71 (m,
3
1H), 3.04 (d, JHH = 8.0 Hz, 2H); 13C NMR (100.6 MHz, CD3OD, TMS) δ 137.1, 131.2,
1
2
3
19
131.0, 129.5, 83.5 (d, JCF = 171.4 Hz), 54.8 (d, JCF = 18.4 Hz), 36.3 (d, JCF = 5.2 Hz); F
NMR (376.5 MHz, CD3OD) δ ꢀ234.9; IR (ATR) νmax: 2868, 1492, 1619, 1023, 741, 701;
HRMS (ESIꢀTOF) m/z: [M+H]+ Calcd. for C9H13NF 154.1032; Found 154.1028
(S)-N,N-Dimethyl-1-fluoro-3-phenylpropan-2-amine (18a)
Compound 18a was prepared following the procedure described for the synthesis of 10,
starting from 17a (20 mg, 0.1 mmol), formic acid (0.071 mL, 9.5 mmol) and formaldehyde
(37% in H2O) (0.06 mL, 9.0mmol). 18a was obtained without further purification as a
1
colorless oil (12 mg, 63%). H NMR (400 MHz, CD3OD) δ 7.36ꢀ7.24 (m, 5H), 4.59ꢀ4.34 (m,
2H), 2.99ꢀ2.88 (m, 2H), 2.75ꢀ2.68 (m, 1H), 2.48 (s, 6H); 13C NMR (100.6 MHz, CD3OD,
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 28 of 39
1
2
3
1
2
TMS) δ 139.7, 129.5, 128.9, 126.6, 82.4 (d, JCF = 171.0 Hz), 66.2 (d, JCF = 17.2 Hz), 42.0
4
5
6
7
8
9
4
3
(d, JCF = 1.5 Hz), 32.2 (d, JCF = 6.8 Hz); 19F NMR (376.5 MHz, CD3OD) δ ꢀ226.0; IR
(ATR) νmax: 2935, 2781, 1603, 1454, 1008, 737, 698. HRMS (ESIꢀTOF) m/z: [M+H]+ Calcd.
for C11H17NF 182.1345; Found 182.1351.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(
S N,N-Dipropargyl-1-fluoro-3-phenylpropan-2-amine (16a) and (S)-N,N-propargyl-1-
)-
fluoro-3-phenylpropan-2-amine (19a)
Compounds 16a and 19a were prepared following the procedure described for the synthesis
of 1b, starting from 17a (66 mg, 0.35 mmol), NaOH (45 mg, 1.12 mmol) and a solution of
propargylbromide in toluene (80 wt.%, 39 µL, 0.36 mmol). Amines 16a (15 mg, 19%) and
19a (37 mg, 56%) were obtained after purification by chromatography on silica gel
20
1
(heptane/AcOEt, 80/20) as colorless oils. 16a: [α]D ꢀ22.9 (c 0.76, CHCl3); H NMR (400
4
MHz, CDCl3, TMS) δ 7.32ꢀ7.20 (m, 5H), 4.66ꢀ4.35 (m, 2H), 3.70 (d, JHH = 2.4 Hz, 4H),
3.33ꢀ3.20 (m, 1H), 3.07ꢀ3.03 (m, 1H), 2.83 (dd, JHH = 13.2 Hz, 3JHH = 10.0 Hz, 1H), 2.28 (t,
2
4JHH = 2.4 Hz, 2H); 13C NMR (100.6 MHz, CDCl3, TMS) δ 139.1, 129.6, 128.9, 126.7, 83.2
(d, 1JCF = 170.6 Hz), 80.4, 73.3, 63.2 (d, 2JCF = 17.9 Hz), 40.3 (d, 4JCF = 2.6 Hz), 33.7 (d, 3JCF
19
= 5.9 Hz); F NMR (376.5 MHz, CDCl3) δ ꢀ227.2; IR (ATR) νmax: 3293, 3027, 3026, 1603,
1455, 1125, 999, 918, 742, 700, 638; HRMS (ESIꢀTOF) m/z: [M+H]+ Calcd. for C15H17NF
20
230.1345; Found 230.1350; HPLC purity: eluent MeOH/H2O 70/30, tr = 11.4 min. 19a: [α]D
1
+4.5 (c 1.00, CHCl3); H NMR (400 MHz, CDCl3) δ 7.36ꢀ7.24 (m, 5H), 4.50ꢀ4.27 (m, 2H),
3.53 (d, 4JHH = 2.4 Hz, 2H), 3.38ꢀ3.30 (m, 1H), 2.81 (d, 3JHH = 7.2 Hz, 2H), 2.24 (t, 4JHH = 2.4
Hz, 1H), 1.54 (bs, 1H); 13C NMR (100.6 MHz, CDCl3, TMS) δ 138.0, 129.6, 128.9, 127.0,
1
2
3
84.7 (d, JCF = 170.3 Hz), 82.0, 72.0, 57.4 (d, JCF = 18.7 Hz), 37.1 (d, JCF = 6.1 Hz), 36.4;
19F NMR (376.5 MHz, CDCl3) δ ꢀ229.0; IR (ATR) νmax: 3294, 2925, 1454, 1007; 744, 699,
636; HRMS (ESIꢀTOF) m/z: [M+H]+ Calcd. for C12H15NF 192.1189; Found 192.1192.
ACS Paragon Plus Environment

Page 29 of 39
The Journal of Organic Chemistry
1
2
(S)-N
,
N
-Methylpropargyl-1-fluoro-3-phenylpropan-2-amine (8a)
3
4
5
6
7
8
Compound 8a was prepared following the procedure described for the synthesis of 10,
starting from 19a (30 mg, 0.16 mmol), formic acid (60 µL, 1.59 mmol) and formaldehyde
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(37% in H2O) (71 µL, 0.95 mmol). Compound 8a was obtained after purification by
20
chromatography on silica gel (heptane/AcOEt, 80/20) as colorless oil (3.6 mg, 11%). [α]D
ꢀ
24.7 (c 0.36, CHCl3); 1H NMR (400 MHz, CDCl3, TMS) δ 7.35ꢀ7.22 (m, 5H), 4.62ꢀ4.33 (m,
2
2H), 3.55 (s, 2H), 3.16ꢀ3.00 (m, 2H), 2.77 (dd, JHH = 13.2 Hz, 3JHH = 10.0 Hz, 1H), 2.55 (s,
3H), 2.30 (t, 4JHH = 2.4 Hz, 1H); 13C NMR (100.6 MHz, CDCl3, TMS) δ 139.4, 129.6, 128.9,
2
4
126.7, 82.6 (d, 1JCF = 170.9 Hz), 80.5, 73.2, 64.2 (d, JCF = 17.5 Hz), 44.3 (d, JCF = 2.0 Hz),
4
3
38.5 (d, JCF = 1.7 Hz), 33.3 (d, JCF = 6.2 Hz); 19F NMR (376.5 MHz, CDCl3) δ ꢀ227.5; IR
(ATR) νmax: 3295, 2360, 1454, 1016, 740, 700, 648; HRMS (ESIꢀTOF) m/z: [M+H]+ Calcd.
for C13H17NF 206.1345; Found 206.1341; HPLC purity: eluent MeOH/H2O/TFA 30/70/0.1, tr
= 10.1 min.
(R)-2-Fluoro-3-phenylpropan-1-ammonium chloride (17b)
Following the procedure for the preparation of 17a starting from 13b (286 mg, 1.23 mmol),
compound 17b (140 mg, 60%) was obtained as a white solid. m.p.: 236ꢀ237 °C (lit. 129ꢀ132
20
°C)44; [α]D + 6.3 (c 1.00, MeOH). Characterization data are in accordance to the published
data.44
(R)-N,N-Dimethyl-2-fluoro-3-phenylpropan-1-amine (18b)
Compound 18b was prepared following the procedure described for the synthesis of 10,
starting from 17b (20 mg, 0.1 mmol), formic acid (0.071 mL, 9.5 mmol) and formaldehyde
(37% in H2O) (0.06 mL, 9.0 mmol). Compound 18b was obtained without further purification
20
1
as colorless oil (16 mg, 84%). [α]D ꢀ 21.3 (c 0.83, MeOH); H NMR (400 MHz, CD3OD) δ
2
3
7.37ꢀ7.27 (m, 5H), 5.32ꢀ5.14 (m, 1H), 3.44ꢀ3.33 (m, 3H), 3.05 (dd, 1H, JHH = 23.4 Hz, JHH
ACS Paragon Plus Environment