Deprotection of N-Boc Groups under Continuous-Flow High-Temperature Conditions

Article abstract of DOI:10.1021/acs.joc.8b02909

The scope of thermolytic, N-Boc deprotection was studied on 26 compounds from the Pfizer compound library, representing a diverse set of structural moieties. Among these compounds, 12 substrates resulted in clean (≥95% product) deprotection, and an additional three compounds gave ≥90% product. The thermal de-Boc conditions were found to be compatible with a large number of functional groups. A combination of computational modeling, statistical analysis, and kinetic model fitting was used to support an initial, slow, and concerted proton transfer with release of isobutylene, followed by a rapid decarboxylation. A strong correlation was found to exist between the electrophilicity of the N-Boc carbonyl group and the reaction rate.

Full text of DOI:10.1021/acs.joc.8b02909

Subscriber access provided by Iowa State University | Library
Article
Deprotection of N-Boc Groups Under
Continuous Flow High Temperature Conditions
Bryan Li, Ruizhi Li, Peter Dorff, James Christopher McWilliams,
Robert M Guinn, Steven M. Guinness, Lu Han, Ke Wang, and Shu Yu
J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.joc.8b02909 • Publication Date (Web): 08 Jan 2019
Downloaded from http://pubs.acs.org on January 9, 2019
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 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 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.
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 44
The Journal of Organic Chemistry
1
2
3
4
5
Deprotection of N-Boc Groups Under Continuous
6
7
8
Flow High Temperature Conditions
9
Bryan Li, Ruizhi Li,* Peter Dorff, J. Christopher McWilliams, Robert M. Guinn, Steven
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
M. Guinness, Lu Han, Ke Wang, and Shu Yu
Medicinal Science, Worldwide Research and Development, Pfizer Inc., Eastern Point
Road, Groton, Connecticut 06340, United States
Abstract: The scope of the thermolytic, N-Boc deprotection was studied on twenty six
compounds from the Pfizer compound library, representing a diverse set of structural
moieties. Among these compounds, 12 substrates resulted in clean (≥ 95% product)
deprotection, and an additional 3 compounds gave ≥ 90% product. The thermal de-
Boc conditions were found to be compatible with large number of functional groups. A
combination of computational modeling, statistical analysis, and kinetic model fitting,
was used to support an initial, slow and concerted proton transfer with release of
isobutylene, followed by a rapid decarboxylation. A strong correlation was found to
exist between the electrophilicity of the N-Boc carbonyl group and the reaction rate.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 44
1
2
3
Keywords: Green chemistry, amine, carbamate, de-Boc, continuous flow, and
4
5
6
7
modeling.
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
Table of Contents Graphic
H
H
H
O
O
Thermal deprotection
26 examples surveyed
-
H

R
R
+
O
Me
Me
N
O
N
O
H
H
R
+
N
O

H
CO2
R
NH2
1. Introduction
ACS Paragon Plus Environment

Page 3 of 44
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Deprotection of N-Boc groups (de-Boc) is one of the most often encountered reactions
in pharmaceutical research and development.1 Deprotection thermally with no added
reagents is advantageous over traditional acid-catalyzed techniques for compounds
that have functional group incompatibilities or workup complications, as the products
from the thermal deprotection are free amines that can be directly carried into the next
reaction step reaction without salt breaks, extractions or isolations (e.g. amidation or
SNAr).2 Recently published literature indicated continuous flow thermal conditions are
compatible with a variety of functional groups,3 and a kilogram scale continuous de-Boc
process at 270 °C demonstrated by Eli Lilly researchers marked a milestone for its
application in the pharmaceutical industry.4 In a survey of reactions executed in the
Pfizer Groton scale up facility in the past 30 months, we found N-Boc deprotections
and amidations ranked among the most often executed reactions. The results
coincided with Brown and Boström’s findings5 in which the top three most frequently
occurring reactions were amide bond formation, SNAr reactions, and N-Boc
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
protection/deprotection. As acid-mediated de-Boc typically follows either an extractive
aqueous workup and isolation or a salt break before the amidation step, it becomes
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 4 of 44
1
2
3
4
5
6
7
8
9
evident that continuous thermal deprotection conditions can enhance speed and
productivity by eliminating several unit operations in a large scale, production setting.
Thus, we set out to investigate continuous thermolytic deprotections of N-Boc over a
broad range of intermediates in our past and current portfolio projects.
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
2. Results and Discussion
Since the thermal de-Boc reaction involves high temperature and pressure, a PFR design was an
obvious choice as it can readily handle high temperature and pressure.4,6 Several PFR reaction
screen instruments are commercially available. As stable and homogenous feed and
effluent streams are critical for the development of continuous processes in plug flow
reactors (PFR),7 solubility screening was the first step for this study. In considering the
selection of solvents,8 alcohols were excluded as preliminary studies indicated
formation of carbamate impurities, RNHC(O)OR1 (R1 = Me, Et, or iPr from the alcoholic
solvents), under high temperature conditions. Lilly researchers found the use of
methanol led to significant impurity formation in thermal de-Boc under 270 °C. By
changing the solvent system to THF/MeOH mixture, the impurity was suppressed.4 In
our hands, we had also observed the use of t-amyl alcohol as solvent could lead a t-
ACS Paragon Plus Environment

Page 5 of 44
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
amyl carbamate impurity Likewise, acetate esters (EtOAc, iPrOAc) were not suitable
as they would react with amine products to give acetamide byproducts.9 Acetonitrile is
not preferred as it is known to have high vapor pressure; acetonitrile has a vapor
pressure of 4455 kPa (646 psi) at 540 K (266.8 °C), and its thermal decomposition
begins at ~536 K (262.6 °C).10 While the high vapor pressure of acetonitrile can be
handled in a flow reactor, the potential thermal decomposition under the high
temperature presents a process safety hazard. Toluene did not offer good solubility for
many substrates. Thus, the list of solvents was narrowed down to tetrahydrofuran,
anisole, and 2,2,2-trifluoroethanol (TFE11); TFE is known to facilitate de-Boc
reactions12 and also displayed unique solubilizing ability for many of the substrates
studied. Water was commonly used as an additive to increase the solubility and
accelerate the reaction.13 The substrates we selected encompassed both early and
late stage research portfolio intermediates in API syntheses. Thus, they were
representative of broad spectra of structural complexity and diversity.
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
2.1 Residence time vs. temperature.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 44
1
2
3
4
5
6
7
8
9
In the next phase, we sought to identify the temperature range of thermal de-Boc with
an objective of limiting the nominal residence time ( herein defined as reaction coil
volume/feed flow rate) to ≤60 min.14. We were hopeful that common tools used to profile
exotherms and endotherms when compounds are subjected to progressively
increasing heat, Differential Scan Calorimetry (DSC) and Thermal Screening Unit
(TSU), could be used as efficient predictive tools for thermolytic de-Boc, as one would
anticipate an exothermic event at the temperature where de-Boc begins. While
exotherms were typically observed at the DSC onset temperatures, as exemplified by
26 (Figure 1), wherein the DSC onset at 209.9 °C appeared to suggest the thermal de-
Boc event, the data did not offer a clear indication for many of the substrates. For
example, the DSC onset for 8 was at 267.7 °C, but the thermal de-Boc was observed
at the lower temperature of 225 °C (coinciding with an endotherm event spanning from
166.8 °C to 267.7 °C). It was also a surprise that Thermal Screening Unit did not
detect a significant exotherm for any of the compounds studied. Our initial screen in
the Uniqsis reactor suggested that thermolysis started to occur at 200 ~ 230 °C for
most of the substrates (except for two substrates vide infra), this temperature range is
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

Page 7 of 44
The Journal of Organic Chemistry
1
2
3
4
5
6
7
substantially lower than in previous reports under continuous flow,2,4 and is considered
operationally more achievable in our current scale-up PFR platform.15
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
Figure 1. Differential Scan Calorimetry (DSC) for 8 and 26 (Table 1)
The selected compounds were subjected to thermal de-Boc screens and the results
are summarized in Table 1. A nominal residence time of 10 min at 225 °C was initially
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 8 of 44
1
2
3
4
5
6
7
8
9
tested for thermal de-Boc feasibility assessment.16 Five compounds (entries 1, 2, 3, 17
and 19, Table 1) were immediately found not suitable for thermolytic de-Boc as
decomposition occurred at the temperature required to effect de-Boc. Those showing
reasonably clean reaction profiles but with incomplete conversion (≤ 98%) were further
examined after a 60 min residence time at 225 °C. Out of 26 compounds studied, 12 of
them gave reasonably clean (≥ 95% product) deprotection, and an additional three
substrates gave ≥ 90% product. While some compounds gave fairly clean conversion
at shorter residence times, longer reaction times led to formation of impurities. For
instance, compound 19 gave 44% conversion under 225 °C for 10 min, it gave a UPLC
area percent reaction profile of 97% product (excluding the starting material). However,
when the reaction was carried out for 60 min at the same temperature, the conversion
increased to 97%, but the reaction profile dropped to 85% product with multiple
impurities. This degradation was more pronounced for compounds 6, 9, and 17 with
longer residence times. Increased reaction temperatures did not seem to improve the
reaction profiles for at least two of the substrates (4 and 26) investigated. This was
particularly obvious for 26, a very clean reaction was observed under 200 °C in 10 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
ACS Paragon Plus Environment

Page 9 of 44
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
though only with 15% conversion. As the reaction temperature was increased, the
conversion rate grew accordingly to 100%, however, undesired reactions started to
dominate and the product was fully decomposed at 265 °C. After a further reaction
screen, it was determined that the optimal temperature was 210 °C at a longer
residence time (80 min). These observations suggest that a simple temperature vs.
residence time screen is insufficient to determine the optimal reaction conditions for
thermal de-Boc reactions.
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
2.2 Functional group compatibility
A number of functionalities were found to be compatible under thermal de-Boc
conditions, including ketones, amides, esters, enol ethers, ketals, nitriles, alkyl halides,
aryl halides, aryl- and akyl trifluoromethyl, benzyl carbamate and phenol. The results
are summarized in Table 2. Aryl and heteroaryl systems were well-tolerated, including
pyridines, pyrimidines, pyrazoles, pyrazolones, imidazoles, thiazoles, and pyrroles.
Pinacol boronate esters and the thiadiazole (3) gave complete decomposition. While
aryl fluorides and chlorides are well-tolerated, aryl bromides gave partial
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 10 of 44
1
2
3
4
5
6
7
8
9
dehalogenation byproducts (5 ~ 20%). We hypothesized that this might be related to
the formation of aryl metallic bromides in the metallic reaction coil under the high
temperature.17 This was subsequently confirmed by heating substrates in glass
capillary tubes (solid state without any solvent) using a melting point device (Buchi
Model B-545). In both cases (5 and 6, Table 1), the solid state thermal deprotection in
the glass capillary tubes was fairly clean with no detection of the dehalogenated
byproduct.18 The results suggest that N-Boc compounds containing aryl bromide
moieties can be thermally deprotected under continuous flow conditions in a PFR with
proper construction of material.19 Ethers are expected to survive thermal conditions;
nevertheless, some aryl alkyl ethers were observed to give ether cleavage byproducts.
This seemed to be correlated with the leaving group (ArO) stability; 6-fluoroquinoline
(9, Table 1) gave the highest level of ether cleavage byproduct, which suggested the
aryl alkyl ether cleavage proceeded via a SNAr solvolysis20 mechanism with the ArO
being a 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
ACS Paragon Plus Environment

Page 11 of 44
The Journal of Organic Chemistry
1
2
3
4
5
Table 1. Thermal de-Boc screen results of 26 substrates
6
7
8
9
225 °C 10 min 
225 °C 60 min 
φ
Computational Indices of C=O
Solvent
Entry/
compo
und
(unless noted otherwise)
(unless noted otherwise)
Structure
Reaction Profile
Product
Product
Nucleophi-
Electrophi-
Conversion
Conversion
Charge (O)
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
¥
§
(%)
(%)
licity (O)
licity(C=O)
(%)
(%)
N
N
Cl
N
N
Dehalogenation as the
major impurity
N
121
O
0.0315
-0.579
0.0069
THF
7.61%
17%
N/A
N/A
N
N
Boc
O
Boc
N
Pinacol boronate ester
hydrolysis as the major
byproduct
0.48%
(200 °C)
222
N
F
0.2809
0.0868
-0.535
-0.488
0.014
anisole
2%
0%
N/A
N/A
N/A
N
O
B
O
O Cl
F
Multiple impurities;
major impurity loss of
SO2
anisole
(20 mL/g)
0%
(150 °C)
323
N
O
0.0379
N/A
88%
N
S
Boc
O
F
33%
97%
97%
97%
95%
O
anisole
90%
(240 °C 30 min)
92%
(240 °C)
Water as additive to
suppress urea formation
Ph
N
424
0.1681
-0.552
0.013
O
NH
THF(1%
water)
Boc
27%
91%
90%
94%
N/A
N/A
O
TFE
anisole
44%
37%
95%
95%
93%
N/A
93%
Single impurity (~5%
dehalogenated)
N
N
THF
525
0.3297
-0.540
-0.551
0.0136
0.00
N
N
Boc
Clean de-Boc
(borosilicate glass
capillary tube, solid
without solvent)
Br
54%
(225 °C 10 min)
None
100%
94%
N/A
80%
Br
dehalogenated as
major impurity
626
0.1
TFE
78%
80%
Boc
N
N
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 12 of 44
1
2
3
4
5
6
anisole
None
33%
83%
99%
N/A
N/A
Clean de-Boc
(borosilicate glass
capillary tube, solid
without solvent)
11%
(225 °C 10 min)
7
8
9
Boc
O
N
anisole
(15 mL/g)
24%
(200 °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
727
0.21
0.04
-0.509
-0.548
0.06
0.01
95%
50%
95%
50%
95%
Multiple impurities
N
F
F
O
anisole
THF
44%
39%
75%
93%
94%
83%
~2% methyl ketone
~50% methyl ketone
O
O
N
N
N
828
N
N
N
N
nBu
Boc
TFE
F
O
aryl ether cleavage as
major impurity
929
0.13
0.17
-0.533
-0.553
0.01
0.00
anisole
anisole
44%
91%
65%
65%
N
Boc
N
H
N
Boc
1030
N
51%
77%
97%
88%
88%
99%
clean
clean
anisole
THF
100%
100%
O
N
CF3
N
N
1131
56%
70%
100%
97%
0.13
-0.512
0.04
N
Boc
THF (1%
H2O)
N
1232
1333
0.15
0.31
-0.548
-0.553
0.02
0.04
anisole
48%
89%
96%
96%
98%
clean
clean
H
Boc
N
O
CO2Et
Boc
anisole
THF
72%
64%
95%
95%
99%
N/A
N
N
N
CF3
ACS Paragon Plus Environment

Page 13 of 44
The Journal of Organic Chemistry
1
2
3
4
5
6
THF (1%
H2O)
68%
97%
98%
N/A
N/A
TFE
93%
7
8
9
N
N
N
69%
(200°C)
1434
1535
1636
1737
0.17
0.15
0.1
-0.541
-0.529
-0.534
-0.540
0.01
0.2
TFE
98%
100%
98%
94%
98%
96%
N/A
94%
98%
96%
Clean
Clean
N
Boc
Cl
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
N
F
NHBoc
anisole
anisole
84%
94%
17%
F
N
Boc
0.18
0.02
Clean
EtO2C
N
S
H
Boc
N
N
S
N
0.09
anisole
TFE
24%
Multiple impurities
N
N
EtO2C
N
H
93%
56% (20 min)
13%
95%
98%
82%
N/A
N/A
N/A
N
Single impurity, ~5%
demethylated
THF(1%
H2O)
1838
1939
0.11
0.05
-0.523
-0.539
0.1
Boc
O
N
H
anisole
N
0
anisole
44%
97%
85%
85%
Multiple impurities
BocHN
CO2Bn
O
N
Boc
O
N
N
THF
(15 mL/g)
70%
(200 °C)
2040
2141
0.04
0.06
-0.47
0.03
0.02
100%
100%
95%
clean
Si
O
100%
(220 °C)
Clean; Solvent must be
degassed to avoid N-
-0.540
N/A
100%
N
TFE
H2N
Boc
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 14 of 44
1
2
3
4
5
6
7
8
9
oxidation
clean
45%
(220 °C)
THF
N/A
55%
100%
(220 °C)
H2N
H
N
-0.57
-0.57
TFE
THF
TFE
N/A
N/A
100%
45%
2242
2343
0.05
0.08
0.01
0.01
Boc
N
N
45%
98%
(220 °C , 15 min)
100%
100%
100%
100%
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
Boc
MeO
N
20%
(220 °C, 22 min)
toluene
THF
N/A
45%
clean
N
H
N
N/A
45%
N/A
100%
(210 °C)
TFE
THF
100%
90%
N/A
Boc
N
O
2444
0.04
-0.57
0.01
N
N
5%
H
(220 °C, 22 min)
O
N
N
N
F
0.03/0.
03
THF
(13 mL/g)
2545
0.14/0.14
-0.48/-0.46
99%
99%
N/A
N/A
Clean
O
Boc
N
N
N
Boc
57%
(200 °C, 80
min)
74%
(210 °C, 80
min)
89%
(220 °C, 60
min)
15%
(200 °C, 10 min)
¥
¥
99%
92%
27%
(210°C /10mi)
¥
¥
97%
90%
O
NH2
O
40%
(220°C /10min)
¥
¥
F
95%
87%
Ether cleavage was
predominant byproduct
at higher temperature.
2646
0.08
-0.55
0.01
THF
58%
(235°C /10min)
N
¥
86%
Boc
OH
96%
(250°C /10min)
¥
6%
99%
(265°C /10min)
0%
ACS Paragon Plus Environment

Page 15 of 44
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
φ 10 ml per gram of substrate was used unless otherwise noted.
47
¥
Defined as product area% in the reaction mixture excluding the unconverted starting material.
47
§
Defined as product area% in the reaction mixture including the unconverted starting material.
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
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 16 of 44
1
2
3
4
5
6
7
8
Table 2. Functional group compatibility under thermal de-Boc conditions
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
Compatibility
Functionalities
Entries in Table 1
Comments
(Y/N)
ketone
7
Y
amide
enol ether
ketal
4, 7, 25, 26
Y
Including primary amide
methyl ketone formed as a minor impurity
presumably due to inadvertent presence of
water
8
4
Y
Y
nitrile
5, 20, 25
9, 15, 25
13
Y
Ar-F
Y
Ar-CF3
Y
Alk-CF3
Alk-F
11
Y
7, 26
1,14
5, 6
20
Y
Limited to 2o examples
Dehalogenation impurity was observed for
Entry 1 substrate.
Ar-Cl
See Comments
Up to 90% desired product observed,
dehalogenation was observed as an
impurity
Ar-Br
See comments
TBDMS ether
Boronate ester
ArNH-SO2-NHAr
Y
N
N
2
3
Loss of SO2 as the major byproduct
ACS Paragon Plus Environment

Page 17 of 44
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
ArCO2Et
Ar-O-Alk ether
Ar-O-Ar ether
ArNH2
12, 16
Y
Mostly tolerated, desired products were
observed as the major product, but some
aryl alkyl ethers were partially cleaved
9, 18, 23, 24, 25
See comments
12
Y
Y
Y
Y
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, 22, 23, 24, 27
The impurity formation for the substrate
was unrelated to CBZ.
CBz
19
26
Phenol
2.3Solvent effects
Among the solvents (TFE, THF, anisole, and toluene) studied, TFE was observed to
have the highest rate of reaction. This was clearly illustrated with results from entry 4
(Table 1), the conversion rates at 10 min under 225 °C were 91% in TFE, 33% in
anisole and 27% in THF (1 wt% water). In entry 8 (Table 1) the conversion rates were
75% in TFE, 44% in anisole and 39% in THF; and entry 23 (Table 1) the conversion rate
were 100% in TFE and 20% in toluene. The reaction rates in THF (with or without 1
wt% water as an additive)48 and anisole were comparable as shown in entries 11 and
13. We attempted to add catalytic amounts of acids49 (aq HCl or TsOH), and
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 18 of 44
1
2
3
4
5
6
7
8
9
hexafluoroisopropanol (HFIPA), but did not observe accelerated reaction rates. In
general, solvents were not dried or degassed before use, and were expected to contain
adventitious amount of water, especially for TFE, a highly polar protic solvent. However,
for one particular substrate (21, Table 1), the solvent (THF or TFE) needed to be
nitrogen-sparged thoroughly to avoid the formation of an N-oxide impurity. Urea (27,
Figure 2) was observed as a major byproduct in the thermal de-Boc of the compound 4.
With the addition of 1 wt% water, the impurity was completely suppressed. The rate of
the thermal de-Boc reaction appeared to have no correlation with the substrate
concentration (vide infra).
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
Figure 2 Byproduct from deprotection of 4
O
O
O
O
Ph
N
N
H
N
NH
O
O
O
Ph
27
2.4 Statistical analysis
ACS Paragon Plus Environment

Page 19 of 44
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Reactive indices of substrates were calculated using an internally development
computational model. 50 The carbonyl electrophilicity, oxygen nucleophilicity and charge
(of C=O) in N-Boc carbonyl groups of each substrate were tabulated in Table 1 with the
hope that in silico data can help predict the proclivity towards thermal de-Boc. Linear
regression was used to explore the functional relationship between de-Boc conversion
and reaction parameters, including reaction time, temperature, solvent, concentration
and reactive indices. As displayed in Figure 3, reaction time, solvent and electrophilicity
(C=O) were significantly associated with the product conversion. TFE offers faster rate
of reaction compared to anisole and THF, and higher C=O electrophilicity leads to more
facile thermal de-Boc. The observed parameter effects become less evident as the
reaction time is extended.
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
Figure 3. Contour plots of predicted conversion to product vs electrophilicity (C=O)
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 20 of 44
1
2
3
2.5 Computational modeling and Kinetics
4
5
6
Acid catalyzed N-Boc deprotection has been well studied,3b the reaction typically proceeds at
7
8
9
ambient temperature when strong acids are employed. The reaction rate exhibited a second-order
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
kinetics with a dependency on the acid concentration51 (Scheme 1). Water mediated N-Boc
deprotections were also reported at 100 °C or higher temperature (under pressure). It was
attributed to increased self-ionization of water, i.e more abundant H+ and OH− at higher temperature
( −log Kw = 12 at 100 °C vs 14 at ambient temperature), and appeared to follow a second order
reaction kinetics (Scheme 2).52 For the study of thermal de-Boc reaction kinetics, two
substrates (4 and 5, Table 1) were chosen as they both offered clean deprotections and
displayed distinctively different electronics. The experiments were performed with a
range of residence times (10-60 min), while the temperature and initial substrate
concentration were fixed at 225 °C and 10 mL of solvent/g of substrate. As a first-level
approximation, the de-Boc reaction was postulated to follow first order kinetics. The
kinetic rate constant for each substrate was fitted to the experimentally-derived
conversions from UPLC results. The model was implemented in Matlab (R2017b) and
solver lsqnonlin was applied to perform the parameter estimation. The results from the
model fitting are presented in Figure 4 and the estimated rate constants are shown in
the figure caption. Based on the good agreement between model and experimental
ACS Paragon Plus Environment

Page 21 of 44
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
data, this result suggests that thermal de-Boc reactions follow a first order kinetics
(Scheme 3), with the rate determination step being the formation of the carbamic acid.
As adventitious amount of water was always present in the thermal deBoc reactions
(and at times purposely introduced), it is possible the reaction may follow both
mechanisms shown in Schemes 2 and 3, with the latter being the predominant one.
This is supported by computational modeling using a Gaussian 0953 as it seems to
proceed initially through a concerted mechanism.54 When TFE is employed in thermal
de-Boc, the reaction could possible follow a hybrid of all three mechanisms as it is a
weakly acidic solvent (pKa 12.5 at 25 °C). The acidity of TFE is expected to be further
increased at higher temperature.55 For most acids, disassociation is an endothermic
process. Hence according to Le Chatelier’s Principle, adding heat to an endothermic
process means a shift in equilibrium to the right. A similar phenomena is also exhibited
with water.56 Thermal de-Boc without added catalysis has a considerably higher
activation energy as illustrated in Figure 5. The activation energy for the thermal de-
Boc of 16 and 13 was calculated to be ~34.0 and ~34.8 kcal/mol, respectively, which
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 22 of 44
1
2
3
4
5
6
7
was consistent with the experimental results. The conversion rate for 16 and 13 in 10
min (nominal ) at 225 °C was 94% and 72%, respectively (Table 1).
8
9
Figure 4. Thermal de-Boc reaction kinetics (4 and 5) at 225 °C in Anisole or THFa
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
a The estimated kinetic rates are the following: 0.0394 min-1 for 4 and 0.0525 min-1 for 5.
Scheme 1. Acid catalyzed de-Boc reaction mechanism
H
OH+
O
O
O
H+
+
R
+
R
R
N
O
slow
fast
N
N
H
O
H
H
H+
fast
fast
CO2
RNH2
Scheme 2 Water participation in thermal de-Boc reactions
ACS Paragon Plus Environment

Page 23 of 44
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
O
O+H
HO
O
OH
H3O+
pathway B
R
R
R
RNH2 +
R
O
N
O
N
H
O
N
H
O
H
OH
pathway A
OH
O
H
H
-OH
CO2 + tBuOH
+
9
N
H
O
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
CO2
RNH2
Scheme 3. Thermally driven, concerted release of isobutylene, followed by
decarboxylation
H
-
H
O

H
H
H
O
O
H

R
R
+
+
N 
H
N
H
O
slow
H
H
H
H
fast
CO2
RNH2
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 24 of 44
1
2
3
4
5
6
7
Figure 5. Thermolytic de-Boc Transition State and Activation Energy
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

Page 25 of 44
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
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 26 of 44
1
2
3
4
5
6
7
2.6 Scale up considerations
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
From the thermal de-Boc studies, it appears there is not a simple protocol that fits all N-
Boc substrates, especially pharmaceutical intermediates of high complexity. A thorough
reaction screening of solvents, additives (water/TFE), temperature, and residence time
will be needed to determine the optimal conditions. As TFE offers great solubities for
almost all substrates we have surveyed, it can be used a good starting point to use a
TFE solution of the substrate by performing a constant temperature (225 °C) screen
with various residence time (10 – 60 min) and a constant time (30 min) screen with
various temperature (180 – 250 °C). Other solvent systems, such as water/THF or
water/THF/TFE mixtures, may be studied thereafter for scale up considerations to
address both cost and low level impurities that may arise from the use of TFE. (TFE
solvolysis was noted for some amide substrates to give RC(O)OCH2CF3 impurities).
While the focus of this study was on reaction screening in lab-based flow chemistry
equipment, here we provide a brief primer on process engineering and scale-up
ACS Paragon Plus Environment

Page 27 of 44
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
considerations. Special attention should be paid to the change in density from both
thermal expansion of solvents below their critical temperature (Tc) and supercritical fluid
behavior beyond Tc.4 The actual residence time can be significantly shorter than the
nominal residence time in a PFR system when solvent density changes are significant
within the system. The volume expansion (%)57 of common solvents at elevated
temperatures can be found in the Supplemental Information.
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
Furthermore, hydrodynamics and heat transfer phenomena should be considered when
scaling from laboratory PFRs (10-40 ml) to manufacturing scale PFRs (>1000 ml). In
brief, ideal plug flow conditions have perfect mixing in the radial direction but no mixing
in the axial direction. In practice, the degree of axial dispersion will depend on the flow
regime (laminar vs. turbulent) and in the case of thermal deprotection also the
temperature.4 Axial dispersion will determine the true residence time distribution of the
substrate in the PFR. This directly impacts product quality as substrate exposed to
reaction conditions shorter or longer than deemed necessary may result in incomplete
reaction, impurity formation, or thermal degradation inside the PFR. Furthermore, heat
transfer in the radial direction as a function of flow rate should also be considered as the
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 28 of 44
1
2
3
4
5
6
7
8
9
PFR coil diameter increases. Overall, the thermal expansion of the solvent, axial
dispersion and its associated residence time distribution, and heat transfer are
important factors to be aware of when scaling-up a thermal deprotection reaction. May
and colleagues from Eli Lilly have elegantly covered deprotection of an N-Boc-protected
imidazole, its scale-up, and the associated engineering principles.4
3. Conclusion
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
Out of 26 compounds surveyed, 12 gave clean (≥ 95% product) deprotection, and a
broad functional group compatibility. This suggests the thermal de-Boc can be a viable
method in developing a continuous end to end process, in particular, when the
immediate next step is on the resulting amine. Among the compounds giving clean
reactions, 3 compounds (11, 12, and 13) were chosen for multi-gram scale up, and all
gave > 92% isolated yields. In addition, the conversion of 25 to lorlatinib on 10 g scale
proceeded in essentially quantitative yield. Kinetic analysis and computational modeling
suggests the thermolytic de-Boc follows an initial concerted fragmentation as the
predominant mechanism, releasing isobutylene, followed by rapid decarboxylation.
ACS Paragon Plus Environment

Page 29 of 44
The Journal of Organic Chemistry
1
2
3
Statistical analysis shows a strong correlation of thermal de-Boc reaction rate with a
4
5
6
7
computationally determined electrophilicity coefficient of the N-Boc carbonyl group.
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
Experimental Section
General Methods. Flow experiments were performed using a Uniqsis FlowSyn reactor
equipped with a 22 ml stainless steel coil and a 750 psi back pressure regulator. The
Uniqsis FlowSyn reactor was flushed with the solvent the tested substrate was
dissolved in at 1 ml/min at room temperature. The flow reactor was then set to the
desired temperature and allowed to equilibrate for 15 mins with the said solvent being
continuously perfused through the reactor. One gram of each substrate was dissolved in
the specified solvents (anisole, THF, and/or TFE) at a desired volume (ml/g) in two
dram vials with pressure relief caps (Chemglass Life Sciences). Samples were then
loaded into an injection loop. The flow rate was set to achieve the intended nominal
residence time, and samples were injected into the Uniqsis Flow Syn reactor. Reaction
screening for compound 25 was carried out using Propel, a segmented flow reactor
system.58 For the multi-gram scale demo on compound 25, Vapourtec was used
ACS Paragon Plus Environment