Development of the Commercial Route for the Manufacture of a 5-Lipoxygenase Inhibitor PF-04191834

Article abstract of DOI:10.1021/op500412a

A de novo three-step-one-pot process for the formation of PF-04191834 was developed. This methodology employed inexpensive, odorless, and readily available commodity chemical iso-octyl-3-mercaptopropionate as a sulfur source, which could be a general alternative to the popular TIPS-SH in the formation of diarylthioethers via Migita coupling. A kinetic study revealed that, at high temperature, reductive elimination could be the rate-limiting step in the catalytic cycle, which opens pathways for the generation of undesired impurities. By proper control of the reaction conditions, the desired API was synthesized in >70% crude yield and in 55% isolated yield after vigorous purifications. This process was successfully demonstrated on a 20 kg scale.

Full text of DOI:10.1021/op500412a

Page 1 of 26
Organic Process Research & Development
1
2
3
4
5
6
7
8
Development of the Commercial Route for the Manufacture of a 5-Lipoxygenase Inhibitor
PF-04191834
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
Brian Chekal, David Damon, Danny LaFrance, Kyle Leeman, Carlos Mojica, Andrew Palm, Michael St.
Pierre, Janice Sieser, Karen Sutherland, Rajappa Vaidyanathan, John Van Alsten, Brian Vanderplas,
Carrie Wager, Gerald Weisenburger, Gregory Withbroe, and Shu Yu*
Chemical Research and Development, Pfizer Worldwide R&DꢀGroton Laboratories,
558 Eastern Point Road, Groton, CT 06340, United States
Abstract:
A de novo threeꢀstepꢀoneꢀpot process for the formation of PF-04191834 was developed. This
methodology employed inexpensive, odorless and readily available commodity chemical isoꢀ
octylꢀ3ꢀmercaptopropionate as sulfur source, which could be a general alternative to the popular
TIPSꢀSH in the formation of diarylthioethers via Migita coupling. A kinetic study revealed that
at high temperature, reductive elimination could be the rateꢀlimiting step in the catalytic cycle,
which opens pathways for the generation of undesired impurities. By proper control of the
reaction conditions, the desired API was synthesized in >70% crude yield and in 55% isolated
yield after vigorous purifications. This process was successfully demonstrated on 20 kg scale.
Introduction
PF-04191834 (1), a potent competitive inhibitor of the 5ꢀlipoxygenase (5 LOX) enzyme, has
been explored as a treatment for mild to moderate asthma (Figure 1).1,2 Based on projections of a
high efficacious dose, it was expected that high volumes of API (1) would be required during
both the clinical phase as well as commercially. In order to meet the expected demand, a more
efficient and cost effective synthesis of 1 was required. This high volume posed an added
challenge to the development of the commercial manufacturing process. In addition to ensuring
the high quality of the API, considering the competitive nature of the asthma market, the
commercial viability of this inhibitor also hinged on the cost of goods. Thus, in order to support
ACS Paragon Plus Environment

Organic Process Research & Development
Page 2 of 26
1
2
3
4
Phase 2 clinical trials and beyond, we required a highly costꢀeffective process for the
5
6
manufacture of PF-04191834 (1).
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
Figure 1 Structure of 5ꢀlipoxygenase inhibitor PF-04191834 (1)
The medicinal chemistry team designed a highly convergent route to PF-04191834 (1) which
was later enabled into an effective fitꢀforꢀpurpose process for the manufacture of API for Phase 1
clinical trials (Scheme 1).3 This route offered several advantages: a convergent synthesis with
two raw materials of similar complexity (i.e., 2 and 5), crystalline lateꢀstage intermediates,
chemistry involving mostly readily available reagents while avoiding extreme temperatures (low
or high) and special equipment (such as high pressure and cryogenic conditions), and good to
excellent yields for synthetic steps. Importantly, the two advanced intermediates, aryl bromides 2
and 5, are both wellꢀbehaved, stable, highlyꢀcrystalline solids which could be sourced on multiꢀ
ton scale in reasonable time.
3
(TIPS-SH;
1.2 equiv)
Si
SH
O
O
TIPS
H2N
S
H2N
Br
PdCl2(dppf) (1 mol%)
NaOt-Bu (1.2 equiv)
PhMe, reflux
O
O
2
4
79%
N
N
Me
N
Br
N
O
5 (1.05 equiv)
Me
H2N
S
PdCl2(dppf) (2 mol%)
NaOMe (2.1 equiv)
2-MeTHF/MeOH, 70 oC
70%
O
1
Scheme 1 Firstꢀgeneration synthesis of PF-04191834 (1)
ACS Paragon Plus Environment

Page 3 of 26
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Further evaluation of the enabled chemistry revealed two major shortcomings that must be
addressed. The first one is the narrow optimum operating range of the second Migita coupling.
Under the reaction conditions, symmetrical sulfides 6 and 7 form readily (up to several percent,
see Scheme 2), of which, 7 could not be purged by crystallization and could only be removed by
chromatography. Since the impurities are the result of an unwanted reaction of the desired
product, their formation increased as the reaction proceeded. In order to manage the level of
symmetrical sulfides 6 and 7, the reaction required the charge of the catalyst at 70 °C and stop
within three hours. The fact that the API is not the most stable thermodynamic product in this
transformation posed high challenge and risk on robustness and the quality assurance of the
product.
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 other one is the availability of the sulfur source for the first of two Migita couplings.4
Triisopropylsilanethiol (3, TIPSꢀSH) is a versatile sulfur source that has been employed
extensively in academia as well as medicinal chemistry.5 However, its availability on scale is
limited and only a single vendor was able to provide it on multikilogram scale at time this work
was performed for a relatively high price. In addition, the reagent has a pungent rotten egg odor,
which is an environmental risk. Therefore we sought an alternative sulfur source.
Scheme 2 Migita coupling byꢀproducts
ACS Paragon Plus Environment

Organic Process Research & Development
Page 4 of 26
1
2
3
4
5
6
Discussion and Results
7
8
Evaluation of the sulfur source for initial Migita coupling
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
In the early stages of development, we considered both bromides 2 and 5 as substrates for the
initial Migita coupling. Based on factors including cost, availability, odor, and complexity of the
transformation, we began exploring alternatives to TIPSꢀSH for the sulfuration of bromide 5 to
thiophenol 8: S8,6 ClSSCl,7 CH3COSK,8 thiourea,9 and NaSCN/NBS.10 Among all the
candidates, molecular sulfur (S8) is the most desirable for its low cost, availability, and benign
smell. While metalation of aryl bromide 5 with nꢀBuLi at −78 °C generated an aryllithium,
alkylation of the resulting butyl bromide (byproduct of metalꢀhalogen exchange) proceeded to 9
prior to reaction with S8 (Scheme 3). Alternatively, the reaction of 5 with Bu3MgLi followed by
S8 successfully provided the thiophenoxide; however, this intermediate was alkylated by butyl
bromide to form 10 before it could be intercepted with an electrophile (e.g., Ac2O, AcOH).
Employing Rieke Mg successfully generated the arylmagnesium in the absence of BuBr;
however, reaction with S8 or ClSSCl afforded a mixture of sulfide, disulfide, and trisulfide
products 11–13. Potassium thioacetate was reported to couple with aryl halides at 150–160 °C
under microwave irradiation,8 but we were unable to adapt this chemistry to more processꢀ
friendly conditions. Thiourea coupled successfully with 5, but the resulting thioamidine
decomposed to sulfide 11 under the reaction conditions. Furthermore, attempts to introduce
sulfur onto desbromoꢀ5 via NaSCN/NBS10 provided 14 via chemoselectivity for the pyrazole
over the arene.
ACS Paragon Plus Environment

Page 5 of 26
Organic Process Research & Development
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
Scheme 3 Byproducts 9–14 from attempts to convert bromide 5 to thiophenol 8
These unsuccessful attempts led us to consider several commerciallyꢀavailable
3ꢀmercaptopropionates as reagents for an initial Pdꢀcatalyzed Migita coupling (Scheme 4).
Treatment of these coupling products with base would then liberate the free thiophenols (via
acrylate byproducts) for a second Migita coupling. Recognizing the alkyl acrylate side products
as potential Michael acceptors, a risk assessment was performed. It was concluded from the
review of the scientific literature,11 that although alkyl acrylates are clastogenic in mammalian
cells in vitro, based on the mutagenicity and carcinogenicity profile of structurally similar
acrylates, the weightꢀofꢀevidence is that isoꢀoctyl acrylate and nꢀpropyl acrylate, the
transesterification product between the initially formed isoꢀoctyl acrylate and nꢀpropanol in the
solvent, are not mutagenic or genotoxic carcinogens. Vogt et. al. reported that the primary route
of detoxification involves hydrolysis by carboxylesterases found in a number of tissues, which
result in the release of acrylic acid and the corresponding alcohol.11e
From an environmental standpoint, 3ꢀmercaptopropionates of higher alcohols tend to be less
odorous due to their large molecular weight. For instance, compared to the methyl and ethyl
analogs, propyl 3ꢀmercaptopropionate (15) has the advantage of a weaker odor. 2ꢀEthylhexyl 3ꢀ
mercaptopropionate (16)12, an excellent source of sulfur, has only a faint odor. Its isomer, isoꢀ
octyl 3ꢀmercaptopropionate (17), also has little odor. From availability and cost perspectives, 17
had the advantage of bulk availability as a commodity material used in the polymer industry.
Since all alkyl 3ꢀmercaptopropionates exhibit very similar reactivities, 17 was finally chosen for
development.
ACS Paragon Plus Environment

Organic Process Research & Development
Page 6 of 26
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
Scheme 4 3ꢀMercaptopropionates for the initial Migita coupling
Selection of metal catalyst and coupling partners
As per the medicinal chemistry synthesis (Scheme 1), we planned to build the thioether 1
through a sequence of two Migita couplings. It has been well documented that bidentate ligands
are superior for Migita coupling and Pd2(dba)3/Xantphos is an excellent combination for this
transformation.12 Besides Pd, other base metal catalysts have been reported for Migita couplings
such as Cu,13 Ni,14 Fe,15 and Co;16 however, in our hands only Pd proved efficient for both C–S
bond formations en route to 1, and so we moved forward with Pd catalysts despite their higher
cost relative to other metals.
In addition to Pd2(dba)3/Xantphos, in recent years, Hartwig had demonstrated the high efficiency
of Pd2(dba)3/Josiphos in Migita couplings.5,17 Screening of palladium sources and ligands
quickly identified these two as the most efficient combinations. Since Josiphos is patented and
more expensive, Pd2(dba)3/Xantphos became the catalyst of choice for our process. Furthermore,
Pd2(dba)3/Xantphos appeared to minimize formation of the symmetrical sulfides, which greatly
reduced the burden on product purity upgrade.
ACS Paragon Plus Environment

Page 7 of 26
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
We investigated the order of C–S bond formation. Since bromide 2 coupled well with 3ꢀ
mercaptopropionates for the first C–S bond, we turned our attention to the second C–S bond
formation (Scheme 5). The coupling of bromide 5 and thiophenol 19 performed better than the
converse coupling of bromide 2 and thiophenol 8 using Pd2(dba)3 and Xantphos. Thus, we
decided to first prepare thiophenol 19 via sulfuration of 2 with a 3ꢀmercaptopropionate, and then
complete the API synthesis with a second Migita coupling of 19 and bromide 5.
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 Selection of the coupling partners for the second Migita coupling
Development of a one-pot, two-Migita coupling process
We set out to develop a oneꢀpot, telescoped process18 for the following threeꢀstep sequence: (1)
Migita coupling of bromide 2 and isoꢀoctyl 3ꢀmercaptopropionate (17); (2) unmasking of the
first Migita product 21 to thiophenol 19; (3) second Migita coupling of 19 and bromide 5. Two
main objectives were identified to reduce the cost of the process after a preliminary analysis.
Since the cost of the palladium/ligand represented the largest cost contribution, minimization of
the catalyst loading was paramount. Additionally, reduction of the number of units of operation
was pursued to reduce the operational cost of the process. In laboratory experiments (Scheme 6),
we found that 1 mol% Pd2(dba)3 and 2 mol% Xantphos performed quite well for the first Migita
coupling of bromide 2 and isoꢀoctyl 3ꢀmercaptopropionate (17). We were able to reduce the
ACS Paragon Plus Environment

Organic Process Research & Development
Page 8 of 26
1
2
3
4
5
6
7
8
9
Pd/Xantphos ratio from 1:2 to 1:1 without much change in performance; however, longer
reaction times were necessary when lowering the catalyst loading to 0.5 mol%. Thus, we heated
a mixture of bromide 2, thiol 17, Pd2(dba)3, Xantphos, and (iꢀPr)2NEt in toluene at reflux for 1 h
to afford thioether 21 in >97% (HPLC).
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
O
HS
OC8H17
O
O
O
17 (1.05 equiv)
H2N
Br
H2N
S
OC8H17
Pd2(dba)3 (1 mol%)
Xantphos (2 mol%)
(i-Pr)2NEt (2.0 equiv)
PhMe (10 vol)
O
O
21
2
reflux, 2 h
N
N
Me
Br
O
5 (0.96 equiv)
CO2(C8H17
)
H2N
SH
+
Pd2(dba)3 (0.5 mol%)
Xantphos (1 mol%)
NaOt-Amyl (2.2 equiv)
n-PrOH (10 vol)
19
22
O
second Migita coupling
94 ºC, 6 h
5
80%
N
N
Me
O
H2N
S
1
O
Scheme 6 Piloting a threeꢀstep, oneꢀpot process for synthesis of PF-04191834 (1)
We were gratified to discover we could telescope this reaction directly into thiophenol
deprotection and subsequent Migita coupling. The mixture of 21 (while at reflux) was treated
with a solution of NaOtꢀAmyl, bromide 5, and Pd2(dba)3/Xantphos in nꢀPrOH. These basic
conditions rapidly cleaved the 3ꢀmercaptopropionate to thiophenol 19 and acrylate 22 (a strong
base is required). The binary nꢀPrOH/toluene system was held at reflux for 6 h as the liberated
thiophenol coupled with 5 for the second C–S bond formation. The second charge of catalyst and
ligand, while not always required on laboratory scale, was added on large scale as a precaution to
facilitate complete conversion to PF-04191834 (1). Upon reaction completion (<3% of bromide
5 left by HPLC), the mixture was cooled to 70–80 °C and quenched with water. The phases were
allowed to separate and the aqueous layer was extracted once with iꢀPrOAc. The combined
ACS Paragon Plus Environment

Page 9 of 26
Organic Process Research & Development
1
2
organic layers were treated with 1,2ꢀdiaminopropane (1,2ꢀDAP)19 and Darco Gꢀ60 (for removal
of residual Pd) at 70–80 °C for 30–45 min, filtered while still heated, and concentrated for
crystallization upon MeOH addition and cooling at 0 ºC. The product was isolated by filtration
followed by drying under vacuum for more than 12 h. This overall process afforded the API in
80% yield over three steps on laboratory scale.
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
Impurity control
During the development of the chemistry for the synthesis of PF-04191834 (1), a number of
impurities had been observed, and their origin rationalized (see Scheme 7). Gratifyingly, the
employment of Xantphos as supporting ligand sequestered the formation of symmetrical sulfides
6 and 7, to very low levels below 0.1%. Formation of acids 23 and 29 was managed by
controlling the water content in formation of sulfide 21, by azeotropic distillation of toluene. The
formation of disulfide 24 was reduced by minimizing the exposure of the reaction mixture to air
to prevent the oxidation of thiolate 26. Impurity 28 was managed by maintaining moderate
reaction temperature (the 1,4ꢀaddition pathway was found to be very temperature dependent). In
addition, good purity upgrades were observed in the isolation of the crude API and ensuing
polymorph control step. With these measures in place, all impurities generated in the threeꢀstepꢀ
oneꢀpot process were driven to <0.1% level but one, which varied from 0.15–0.25%. Structure
elucidation work identified the impurity as 27. Extensive solubilityꢀbased purging studies were
perform but did not afford any conditions that consistently remove this impurity.
ACS Paragon Plus Environment

Organic Process Research & Development
Page 10 of 26
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
Scheme 7 Origin of impurities
Since the purge of 27 appeared difficult, its formation was studied in hope of minimizing its
formation during the reaction. It was postulated that a reasonable source of this impurity is from
the exchange between the palladiumꢀbound aryl group (derived from 5) and the phosphineꢀbound
phenyl group,20 a phenomenon also observed by Hartwig in his mechanistic study on C–S
reductive elimination,21 as well as researchers from Merck.22 A priori, the exchange between
phosphineꢀ and palladiumꢀbound aryl groups can occur either after oxidative addition23 (on
intermediate II; Figure 2) or after transmetalation (on intermediate III) via a C–P reductive
elimination (to form a phosphonium ion)/C–P oxidative addition mechanism24 prior to C–S
reductive elimination. Although both scenarios offer a plausible pathway to the impurity, some
lingering questions remained: (1) Xantphos has been employed by us and others for years in
various Pdꢀcatalyzed transformations without observations of aryl scrambling, so why does this
side reaction occur in Migita coupling?2222b (2) Is it possible to locate the point in the catalyst
Formatte
ACS Paragon Plus Environment

Page 11 of 26
Organic Process Research & Development
1
2
cycle that aryl exchange occurs,25 and hence manipulate the reaction conditions to disfavor the
3
4
5
6
pathway leading to this 27?
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
Figure 2 Catalytic cycle for Migita coupling
Kinetic Study
In order to gain insights into the Migita coupling reactions, a kinetic study was performed. Since
Novak and Grushin have established that the aryl exchange occurs in the resting state of the
catalyst,23 the goal of the study was to identify the resting state of the catalyst under our
conditions of Migita coupling. Based on the catalytic cycle of Migita coupling, as depicted in
Figure 2, the rate of the reaction is defined by equation (1).26
∗
ꢈ
ꢋ
ꢌꢍꢌꢎꢏ
ꢉꢊ
ꢈ
ꢋ
ꢀꢁꢂꢃ = ꢄ
ꢠꢡꢢ ∙ꢈꢣꢤꢥꢦꢖꢋ ∙ ꢩꢀꢪꢀ
(1)
ꢅꢆꢇ
ꢈ
ꢛꢟ
ꢋ
ꢒ
ꢈꢖꢗꢘꢗꢋ
ꢞ
ꢓꢔꢕ
ꢒ
∙
∙
ꢚꢗꢛꢜꢝ
ꢧ
ꢈꢣꢤꢥꢦꢖꢢ ꢋ
ꢈ
ꢋ
ꢞ
ꢠꢡꢢ ∙ꢈꢣꢤꢥꢦꢖꢋ
ꢛꢟ
ꢒ
∙
ꢚꢗꢛꢜꢝ
ꢧ
ꢈꢣꢤꢥꢦꢖꢢ
ꢋ
ꢒ
∙
ꢗꢙꢕ
ꢒ
ꢈꢖꢗꢘꢗꢋ
ꢠꢡꢢ ∙ꢈꢣꢤꢥꢦꢖꢋ
ꢈꢣꢤꢥꢦꢖꢢ
ꢒ
ꢧꢒ
ꢈꢘꢗꢋ
ꢚꢗꢛꢜꢝ
ꢨ
ꢨ
ꢓꢔꢕ
ꢞ
ꢗꢙꢕ
ꢒ
ꢐꢑ
ꢑ
ꢈ
ꢋ
ꢧꢒ
ꢈꢘꢗꢋ
ꢚꢗꢛꢜꢝ
ꢛꢟ
ꢗꢙꢕ
ꢒ
∙
∙
ꢚꢗꢛꢜꢝ
ꢧ
ꢋ
ꢒ
ꢗꢙꢕ
ꢒ
ꢗꢙꢕ
ꢧꢒ
ꢈꢘꢗꢋ
ꢚꢗꢛꢜꢝ
ꢨ
ACS Paragon Plus Environment

Organic Process Research & Development
Page 12 of 26
1
2
3
4
Note that during the deduction of the rate law RSH was not specified as alkylꢀ or arylꢀthiol,
5
6
equation (1) is applicable to both the first and the second Migita coupling. By inspecting
7
8
equation (1), one very important conclusion one can draw is that the rate is first order in [Pd]total.
In order to keep the kinetic study within reasonable complexity, it is important to maintain
[Pd]total constant as a known value. It was felt that the first step is a better candidate for this
study. We reasoned that for the second Migita coupling, to keep the concentration of [Pd]total
constant could be hard to achieve, as there are two sources of palladium/ligand in this step, i.e.
the recycled catalyst from the first step and the freshly charged. Experimentally, it is very
difficult to measure the concentration of the ligated palladium as against colloidal palladium. On
the other hand, all reagents are used fresh in the first Migita, it is much more practical to control
the concentration of the catalyst. For above considerations, the kinetic study was conducted on
the first step. However, as pointed out already, the conclusion should be illustrative for both
steps, considering that the catalytic cycles are the same and the differences between coupling
partners are incremental.
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 first goal of this study was to establish the rate law of the reaction. Reaction Progression
Kinetic Analysis (RPKA) protocol employed by Blackmond27 was selected for this study, and
the iC Kinetics software was used to process the data. Considering the reaction was quite fast (on
the time scale of minutes), the inꢀsitu PAT monitoring approach was utilized to serve our
purposes better than the offꢀline (HPLC) analysis, (offꢀline samples were indeed taken and
analyzed to confirm the reaction completion).
Scheme 8 Formation of thioether 21.
The formation of thioether 21 was hence carried out with simultaneous inꢀsitu IR/Raman
monitoring. Gratifyingly, the disappearance of 2 (ArBr) and isoꢀoctylꢀ3ꢀmercaptopropionate (17,
RSH) could be monitored by Raman and the formation of 21 could be monitored by IR
ACS Paragon Plus Environment

Page 13 of 26
Organic Process Research & Development
1
2
3
4
simultaneously in real time, a typical graph of concentration change against time is shown in
5
6
Figure 3.
7
8
9
0.04
0.035
0.03
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
0.025
Bromide 2
0.02
Product 21
0.015
Thiol 17
0.01
0.005
0
0:00:00 0:07:12 0:14:24 0:21:36 0:28:48 0:36:00
Reaction conditions: a mixture of bromide 2 (4.00 g, 14.1 mmol), Pd2(dba)3 (194 mg, 0.212 mmol, 1.5 mol%),
Xantphos (244 mg, 0.422 mmol, 3.0 mol%), Hunig’s base (4.92 mL, 28.3 mmol, 2.0 eq.) in 36.0 mL of toluene was
degassed and heated to a vigorous reflux, isoꢀoctylꢀ3ꢀmercaptopropionate (3.60 mL, 16.1 mmol, 1.14 eq.) was added
bolus through a septum. The progression of the reaction was monitored simultaneously by both an IR probe and a
Raman probe: Raman trend of bromide 2 (blue) and isoꢀoctylꢀ3ꢀmercaptopropionate (green); IR trend of thioether 21
(red). The disappearance of the bromide, the thiol, as well as the formation of the thioether all exhibited linear
behavior against time. (Note: in the developed process, lower catalyst loading was employed, as it was found that
such high catalyst loading was unnecessary.)
Figure 3. Real time monitoring for the formation of thioether 21.
From the same excess experiments, the overlay (see Figure 4) of the two curves indicated that at
this time scale, no extensive catalyst deactivation or product inhibition occurred, and deemed this
transformation “wellꢀbehaved”.
ACS Paragon Plus Environment

Organic Process Research & Development
Page 14 of 26
1
2
3
4
5
6
7
8
9
0.004
0.0035
0.003
0.0025
0.002
0.0015
0.001
0.0005
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
Reaction A
Reaction B
0
500
1000
1500
2000
Initial concentration of Reaction A: [Pd]0= 0.0086 M, [ArBr]0= 0.288 M, [RSH]0= 0.329 M,[Excess]= 0.041 M;
Initial concentration of Reaction B: [Pd]0= 0.0086 M, [ArBr]0= 0.202 M, [RSH]0= 0.243 M,[Excess]= 0.041 M;
Reaction condition: a mixture of bromide 2, Pd2(dba)3, Xantphos, Hunig’s base in toluene was degassed and heated
to a vigorous reflux, isoꢀoctylꢀ3ꢀmercaptopropionate was added bolus through a septum. The formation of thioether
21 was monitored by IR probe.
Figure 4 Same excess experiment for the formation of thioether 21
Performing the same reaction at different initial concentrations of 2 (ArBr) but same isoꢀoctylꢀ3ꢀ
mercaptopropionate (17, RSH) concentration (different excess), the kinetic behavior of the
reactants was analyzed with iC Kinetics software and the power law was deduced as
Rate=0.00083[RSH]0.00[ArBr]0.13 (T=110 °C), see Figure 5. Within the sensitivity of IR and
Raman, no sign of change of mechanism from low conversion to high conversion was observed.
ACS Paragon Plus Environment

Page 15 of 26
Organic Process Research & Development
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
Initial concentration of Reaction A: [Pd]0= 0.0086 M, [ArBr]0= 0.288 M, [RSH]0= 0.329 M,[Excess]= 0.041 M;
Initial concentration of Reaction B: [Pd]0= 0.0086 M, [ArBr]0= 0.202 M, [RSH]0= 0.329 M,[Excess]= 0.127 M;
Reaction condition: a mixture of bromide 2, Pd2(dba)3, Xantphos, Hunig’s base in toluene was degassed and heated
to a vigorous reflux, isoꢀoctylꢀ3ꢀmercaptopropionate was added bolus through a septum. The formation of thioether
21 was monitored by IR.
Figure 5. Different excess experiment for the formation of thioether 21
Based on the observation that the concentration vs. time plot is linear (within experimental error,
vide supra) and the output of iC Kinetics, it appeared that the Migita coupling is zeroth order in
both 2 (ArBr) and isoꢀoctylꢀ3ꢀmercaptopropionate (17, RSH). As depicted in the mechanism of
the Migita coupling summarized in Figure 2, the catalytic cycle is composed of three steps:
oxidative addition, transmetallation, and reductive elimination. Since the rate law indicated that
the reaction is zeroth order in both the aryl bromide and the thiol, it meant that those two
components are not involved in the rate limiting step, which suggested that the reductive
elimination step is the rate limiting step, a kinetic model that closely resembled palladiumꢀ
catalyzed C(sp)ꢀC(sp) crossꢀcoupling.28 And this implied that the resting state of the catalyst is
intermediate III in Figure 2, where both the aryl group and the thiol are bound to the palladium
center.29
ACS Paragon Plus Environment

Organic Process Research & Development
Page 16 of 26
1
2
3
4
5
6
7
8
9
This observation offered a good explanation for the generation of byproduct 27. With
intermediate III as the resting state of the catalyst, slow30 C–S reductive elimination presented an
opportunity for competitive ligand dissociation/C–P reductive elimination from 3123b and
subsequent oxidative addition to the Ph–P bond en route to 27 (see Scheme 9). It should be
pointed out that for most of the Pd(Xantphos)ꢀmediated transformations, oxidative addition is the
rate limiting step.4 This suggests that typically Pd(Xantphos) I is the resting state of the catalyst
rather than intermediate III type, which is perhaps the reason for the scarce citation of this side
reaction.
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
O
O
O
H2N
SNa
N
Ph2P
PPh2
ligand
dissociation
Ph2P
PPh2
Pd(0)
Xantphos
26
O
Pd
Pd
+
ArS
ArS
30
N
Me
31
Br
5
N
Me
N
Me
N
N
C–P
reductive
elimination
O
Ph2P
ArS
C–S
reductive
elimination
Ph
Ph
oxidative
addition
P+
O
O
Pd
Ph
Ph2P
P
H2N
S
Pd
33
27
O
32
Me
ArS
N
N
Me
N
N
Scheme 9 Proposed pathway to impurity 27
This kinetic analysis also offered a rationale for the empirically discovered conditions that
minimized the formation of 27 (i.e. performing the reaction at lower temperature and reducing
the nꢀPrOH to toluene ratio). Presumably, at lower temperature, dissociation of one phosphine
ACS Paragon Plus Environment

Page 17 of 26
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
from 30 became much more difficult, hence suppressing CꢀP reductive elimination; also, reduced
nꢀPrOH to toluene ratio lowered the polarity of the reaction medium, which further hampered C–
P reductive elimination/oxidative addition pathway, consistent with Novak’s observation of the
effect of lower solvent polarity on aryl exchange.2323a Under the modified conditions, the second
Migita coupling became substantially slower and required >10 h to reach completion (still an
acceptable reaction time to be practical); however, under such conditions, 27 did not form in any
appreciable amount (vide supra).
Formatte
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
Crystallization of PF-04191834 (1) with color removal and polymorph control
Several polymorphs have been identified for PFꢀ04191834 (1), and we need to develop a workup
and crystallization to ensure isolation of API with the desired form (designated Form A) and
purity. Solid form studies had identified three anhydrous polymorphs: Forms A, B, and C. In
addition, solvated forms with acetone and CH2Cl2 were discovered during early polymorph
screening studies. Competitive slurry studies demonstrated that Form A was the most stable of
these three anhydrous forms at room temperature. Additional solid form characterization studies
determined that Forms A and B are enantiotropicallyꢀrelated polymorphs, with Form B being the
more stable above a transition temperature determined to be in the range of 63–100 °C. When the
API is already present in high purity, it crystallizes out of THF as Form A; however, this process
offered very limited purity upgrade when less pure material was employed. With this
understanding of the polymorph landscape for 1, a recrystallization process was developed to
purge impurities and ensure isolation of the desired form.
As shown in Scheme 6, we prepared 1 in a binary solvent system of nꢀPrOH and toluene, and we
explored mixtures of these two solvents for API recrystallization. Kinetic solubility and
metastable
zone
limit
(MSZL)
data
of
the
API
in
23.7%
nꢀPrOH/toluene (a specific binary system that was found empirically to have the highest
dissolving power for the API at 90 °C) was measured as a function of temperature. These data
illustrate that 1 has a rather wide metastable zone width of approximately 30 °C (Figure 6,
cooling rate=0.2 °C/min.). Based on this measured solubility data, a process was developed in
which the crude API was dissolved in approximately 35 mL/g of the nꢀPrOH/toluene
ACS Paragon Plus Environment

Organic Process Research & Development
Page 18 of 26
1
2
3
4
5
6
7
8
9
combination at 80 °C. The resulting mixture was a deep red color, and close inspection showed
the batch contained a thin haze of colored particulate. This nonꢀdissolved material (origin of dark
color) was subsequently removed by filtering the batch through Celite, which had been preheated
to prevent the sudden crystallization of API during the transfer. Inspection of the Celite bed after
the filtration showed a covering of dark orange material on the surface with substantially less
color deeper into the bed.
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
6
Kinetic solubility and metastable zone limit data for
1
in 23.7%
nꢀPrOH/toluene mixture as a function of temperature
The resulting pale yellow filtered solution was cooled below 40 °C to induce supersaturation,
and seeded with 0.2 wt% of Form A crystals to induce nucleation of the desired form. It was
observed that the system may be slow to crystallize at 40 °C, and reducing the temperature to
20 °C promoted the crystallization of the API. Partial vacuum distillation of solvent further
promoted crystallization of 1 from the mother liquor. At this point in the process, the resulting
slurry typically contained a mixture of the metastable Form B along with the desired Form A.
Extended stirring between 20–45 °C allowed for polymorph conversion to the desired Form A.
The upperꢀend of the temperature range for the polymorph conversion remained below the
ACS Paragon Plus Environment

Page 19 of 26
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
transition temperature between Form A and Form B, which may be as low as 63 °C. Testing by
powder xꢀray diffraction (PXRD) of an inꢀprocess sample was used to determine that complete
conversion to the desired form had been achieved. Once the PXRD analysis was complete, the
batch was held between 0–5 °C to maximize yield. After filtration, the batch was washed by
toluene followed by nꢀPrOH.
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
Final commercial process on multikilogram scale
With the developmental issues resolved, the process was demonstrated on multikilogram scale
(Scheme 10). The threeꢀstep, oneꢀpot process successfully converted 28.0 kg of bromide 2 and
21.0 kg of bromide 5 to 25.0 kg (72% yield) of the crude API as a MeOHꢀwet cake with 99.1%
purity. (A decision was made not to perform the carbon treatment during the workup.) Assay of
the mother liquor suggested the loss of ~10% API which was not pursued. Next, the polymorph
control step was executed to demonstrate control of impurities. Thus, the crude API was charged
to the reactor as a MeOHꢀwet cake from the previous step. This residual MeOH had no impact
on the recrystallization process as it was removed by atmospheric distillation prior to nꢀPrOH
addition. Due to the challenge of confirming API dissolution because of insoluble colored
particulate and limited sight in the pilotꢀplant reactor, an additional 7 volumes of toluene were
added to the batch to ensure dissolution. Special care was taken to ensure that the Nutsche filter
containing the Celite bed and the subsequent cartridge filter remained at 75 °C to prevent the
crystallization of the API during the filtration step. The batch was concentrated via partial
vacuum distillation and cooled to <20 °C to promote API crystallization. PXRD analysis of an
inꢀprocess sample after 10 h of stirring at 45 °C indicated the presence of Form A alone. With
the desired polymorph confirmed, the batch was cooled to <5 °C, filtered and washed with
toluene followed by nꢀPrOH.
After the purity upgrade and polymorph control step, 19.9 kg of PF-04191834 (1) as Form A
was isolated with 99.6% HPLC purity but contained 300 ppm residual Pd without additional
treatment to purge palladium (Scheme 10). The carbon treatment was executed to purge residual
palladium (Thiolꢀ3 functionalized silica was equally effective; carbon was selected for lower
cost). Accordingly, the API was dissolved in 45 volumes of THF at 45 °C and treated with
ACS Paragon Plus Environment

Organic Process Research & Development
Page 20 of 26
1
2
3
4
5
6
7
8
9
20 wt% of activated carbon for 18 h. The mixture was filtered, concentrated, and charged with
heptane as antisolvent to precipitate the API in high recovery. After filtration and drying, 19.3 kg
of API was isolated in 97% yield with 99.7% purity (HPLC) and <1 ppm Pd. The overall yield of
the entire process was 55%.
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
O
HS
OC8H17
O
O
O
17 (1.05 equiv)
NaOt-Amyl
H2N
Br
H2N
S
OR
(R = C8H17
n-PrOH
PhMe
80 ºC
Pd2(dba)3 (1 mol%)
Xantphos (2 mol%)
(i-Pr)2NEt (2 equiv)
PhMe (10 vol)
)
O
O
2
21
not isolated
28.0 kg
reflux, 2 h
N
N
Me
Br
N
N
Me
O
5 (0.90 equiv)
O
H2N
SNa
Pd2(dba)3 (0.5 mol%)
Xantphos (1 mol%)
NaOt-Amyl (2.2 equiv)
n-PrOH (10 vol)
H2N
S
O
O
26
not isolated
1
80 ºC, 18 h
API in reaction mixture
i.
suspend in PhMe
distill off MeOH
add n-PrOH, heat
filter speck-free
N
N
Me
ii.
iii.
iv.
i.
ii.
H2O wash (2x), 70 ºC
(DARCO KBB)/1,2 DAP
filter through Celite
O
H2N
S
v. cool to 45-50 ºC
vi. seed with Form A,
granulate
iii.
add MeOH
1
iv. concentrate to 10 vol
O
filter, MeOH rinse
v.
distill off n-PrOH
filter, PhMe wash,
n-PrOH rinse
vii.
isolated as MeOH-wet cake
71%
viii.
25.0 kg
99.1% (HPLC)
80%
N
dissolve in THF, 45 ºC
add Darco KBB (20 wt%),
agitate at 45 ºC, 18 h
i.
ii.
N
O
Me
1
H2N
S
iii.
iv.
filter speck-free
concentrate
19.3 kg (97%)
99.7% (HPLC)
<1 ppm Pd
1
O
v. add heptane
19.9 kg
filter, heptane rinse
vi.
99.6% (HPLC),
300 ppm Pd
97%
Scheme 10 Multikilogramꢀscale production of PF-04191834 (1)
ACS Paragon Plus Environment

Page 21 of 26
Organic Process Research & Development
1
2
3
4
Conclusion
5
6
7
8
9
We developed a threeꢀstep, oneꢀpot process for the synthesis of PF-04191834 (1) via a sequence
of two Migita couplings. This strategy employed cheap, odorless, and readilyꢀavailable isoꢀoctyl
3ꢀmercaptopropionate as the sulfur source for the initial Migita coupling as a general alternative
to the popular TIPSꢀSH for the formation of diaryl thioethers. A kinetic study revealed that at
high temperature, reductive elimination could be the rate limiting step in the catalytic cycle,
which opens pathways for the generation of undesired impurities. By properly controlling the
reaction conditions, the desired API was synthesized in 72% crude yield and 55% isolated yield
after purification. This process was successfully demonstrated on 20ꢀkg scale.
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
General Procedures Intermediates were analyzed by reverse phase LC–MS on an Agilent 1100
series instrument, coupled to a Waters Micromass ZQ mass spectrometer according to the
following conditions: column ExtendꢀC18 3.0 mm × 50 mm i.d., 1.8 ꢁm; eluent A, 5% v/v
acetonitrile in 10 mM aqueous ammonium acetate; eluent B, acetonitrile; flow rate 1.2 mL/min;
wavelength, diode array (190–400 ꢁm); column temperature, 50 °C; injection volume, 10 ꢁL; at t
= 0 min, 5% eluent B; at t = 3.5 min, 100% eluent B; at t = 4.5 min, 100% eluent B; at t = 4.6
1
min, 5% eluent B. H and 13C NMR spectra were recorded on a Bruker Ultrashield 400 Plus
spectrometer at 400 and 100.6 MHz, respectively. The quoted melting points for all materials are
the onset temperatures observed by DSC.
Formation of the crude API (1): A 300 gal reactor was charged with 28.0 kg (98.6 mol) of
bromide 2 and 292 kg (336 L) of toluene. Under nitrogen protection at atmospheric pressure, ~10
gal of toluene was distilled out to remove stray and surface water. After the contents had been
cooled to 30 °C, the reactor was charged with 0.90 kg of Pd2(dba)3, 1.14 kg of Xantphos, 25.47
kg of diisopropylethylamine, and 22.6 kg of isoꢀoctylꢀ3ꢀmercaptopropionate (17). The vessel was
thoroughly degassed by alternately connecting the head space to house vacuum and ultra pure
nitrogen. The contents in the vessel were heated at reflux for 90 min, at which time, an offꢀline
sample indicated all bromide 2 was consumed and the desired thioether 21 had formed. While
maintaining the internal temperature at 80 °C, the vessel was charged with 134.4 kg (168 L) of 1ꢀ
propanol, a warm (50 °C) solution of 25.7 kg of NaOꢀtꢀAm in 134 kg (154 L) of toluene,
ACS Paragon Plus Environment

Organic Process Research & Development
Page 22 of 26
1
2
3
4
5
6
followed by a 26 kg (33 L) 1ꢀpropanol chase. The contents were then allowed to heat at 80 °C for
1 h. The vessel contained thiolate 26 at this point.
7
8
9
While maintaining the internal temperature at 80 °C, the vessel was charged a solution of 21.03
kg of aryl bromide 5 in 36.6 kg (42 L) of toluene, followed by a mixture of 0.57 kg of Xantphos
and 0.45 kg of Pd2(dba)3 in 14.1 kg (16.6 L) of toluene. The contents in the tank were allowed to
heat at 80 °C for 18 h, at which time, offꢀline HPLC assay indicated the consumption of aryl
bromide 5 and the formation of the desired API.
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
Isolation of the crude API. The temperature of the contents was brought to 70 °C and 84 kg of
water was added. After agitation for 30 min at 70 °C, the phases were allowed to split and the
lower aqueous layer was removed. The organic phase was washed one more time with 84 kg of
water at 70 °C. To the organic phase was charged 27.76 kg of 1,2ꢀdiaminopropane,31 and the
contents in the vessel was heated for one hour at 70 °C, filtered through a pad of Celite in a preꢀ
heated Nutsche filter, and concentrated from a total volume of ~250 gal to ~105 gal at 35 °C
under vacuum. To the mixture was charged 66.3 kg (84 L) of methanol, and the mixture was
allowed to heat at 55 °C for 30 min. After cooling to 5 °C and granulating for 12 h, the mixture
was filtered, and the cake was washed with 66.4 kg (84 L) of cold methanol. After pulling dry,
the methanolꢀwet cake weighed 43.7 kg.
KarlꢀFischer assay and GC head space analysis indicated that the material contained 1.36% water
and 41.3% methanol, which translate to 25.0 kg (72% over three steps) of the crude API on dry
basis. Evaluation of the mother liquor indicated that ~10% of the API was lost to the mother
liquor, no efforts were made to recover this material.
Purity upgrade. The wet cake was charge to a 300 gal vessel, followed by 337.5 kg (389 L) of
toluene. Methanol was distilled out under atmospheric pressure until the pot temperature reached
80 °C. To the vessel was charged 150 kg of 1ꢀpropanol, and the resulting mixture was heated at
80 °C for 2 h, until a clear solution formed. The temperature of the solution was brought to 45
°C, seeded with form B crystals, and aged for 18 h, at which time, a thick slurry formed. The
slurry was chilled to 5 °C, aged for 2 h, filtered, washed with 32.62 kg of toluene to afford 19.9
1
kg of pure API after drying under vacuum. Mp 173 °C. H NMR (400 MHz, DMSOꢀd6) 7.52
ACS Paragon Plus Environment

Page 23 of 26
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
(2H, m), 7.48 (2H, m), 7.42 (2H, m), 7.35 (2H, m), 7.29 (2H, m), 7.07 (1H, br. s), 6.42 (1H, d, J
= 1.8 Hz), 3.85 (3H, s), 3.74 (2H, dt, J = 11.7, 3.7 Hz), 3.47 (2H, br. t, J = 11.7 Hz), 2.41 (2H, br.
d, J = 13.3 Hz), 1.80 (2H, m). 13C NMR (100.6 MHz, DMSOꢀd6) 174.6, 146.0, 141.9, 137.9,
136.0, 133.2, 130.1, 129.7, 129.4, 129.3, 128.6, 125.6, 105.9, 64.6, 47.8, 37.6, 33.9. LCMS:
found m/z 394.17 [M + H]+. Anal. Calcd for C22H23N3O2S: C, 67.15; H, 5.89; N, 10.68; S, 8.15.
Found: C, 67.09; H, 5.93; N, 10.69; S, 8.16.
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
Pd removal. A glassꢀlined vessel was charged with 796 kg of THF, followed by 19.9 kg of PF-
04191834 (1). The contents were heated to 45 °C, at which time, a clear solution formed. To the
solution was charged 3.98 kg of DARCO KBB and the resulting mixture was agitated at 45 °C
for 14 h. The activated carbon was filteredꢀoff and the THF solution was further filtered speckꢀ
free. The total volume of the solution was reduced to 50 gal under reduced pressure at 45 °C,
yielding a suspension of the API in Form A. After cooling to 20 °C, 140.5 kg of heptane was
added as antiꢀsolvent, and the resulting suspension was allowed to agitate for an additional 30
min. The mixture was filtered and the cake was washed with 33 kg of heptane. Upon drying
under vacuum, 19.3 kg of the API was isolated, which met all the preꢀset specifications and was
1
released for clinical uses. Mp 173 °C. H NMR (400 MHz, DMSOꢀd6) 7.52 (2H, m), 7.48 (2H,
m), 7.42 (2H, m), 7.35 (2H, m), 7.29 (2H, m), 7.07 (1H, br. s), 6.42 (1H, d, J = 1.8 Hz), 3.85
(3H, s), 3.74 (2H, dt, J = 11.7, 3.7 Hz), 3.47 (2H, br. t, J = 11.7 Hz), 2.41 (2H, br. d, J = 13.3
13
Hz), 1.80 (2H, m). C NMR (100.6 MHz, DMSOꢀd6) 174.6, 146.0, 141.9, 137.9, 136.0, 133.2,
130.1, 129.7, 129.4, 129.3, 128.6, 125.6, 105.9, 64.6, 47.8, 37.6, 33.9. LCMS: found m/z 394.17
[M + H]+. Anal. Calcd for C22H23N3O2S: C, 67.15; H, 5.89; N, 10.68; S, 8.15. Found: C, 67.09;
H, 5.93; N, 10.69; S, 8.16.
Acknowledgments
The authors would like to thank Drs. Stephane Caron, Juan Colberg, Joel Hawkins, Robert
Singer, Sebastien Monfette, John Ragan, Nathan Ide, Zhihui Peng, Jason Mustakis, Jade Nelson,
Joshua Dunetz, Michelle Kenyon, Trish Ellis as well as Javier Magano and Professors Steven
Ley, David Collum, Justin Du Bois, and Donald Watson for helpful discussion. Special thanks to
Drs. Stephane Caron, Robert Singer and Sebastien Monfette for proof reading the manuscript.
ACS Paragon Plus Environment

Organic Process Research & Development
Page 24 of 26
1
2
3
4
5
6
7
8
We thank Weiling Cai for performing important experiments in this work. We are grateful to Dr.
Dominique Hebrault from Mettler for providing the initial trial version of the iC Kinetics
software.
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
References
1
Alvira, E.; Graneto, M. J.; Grapperhaus, M. L.; Iyanar, K.; Maddux, T. M.; Mahoney, M. W.; Massa, M. A.; Sample, R. K.;
Schmidt, M. A.; Seidel, R. E.; Selbo, J. G.; Tollefson, M. B.; Vonder Embse, R. A.; Wagner, G. M.; Woodard, S. S. WO
2009069044 A1, 2009.
2 Graneto, M. J.; Maddux, T. M.; Masferrer, J. L. US 20080125474 A1, 2008.
3 (a) de Koning, P. D.; Murtagh, L.; Lawson, J. P.; Vonder Embse, R. A.; Kunda, S. A.; Kong, W. Org. Process Res. Dev. 2011,
15, 1046. Also see: (b) Mano, T.; Stevens, R. W.; Ando, K.; Kawai, M.; Kawamura, K.; Nakao, K.; Okumura, Y.; Okumura, T.;
Sakakibara, M.; Miyamoto, K.; Tamura, T. Chem. Pharm. Bull. 2005, 53, 965. (c) Mano, T.; Stevens, R. W.; Okumura, Y.;
Kawai, M.; Okumura, T.; Sakakibara, M. Bioorg. Med. Chem. Lett. 2005, 15, 2611. Note: In the second Migita coupling, NaOMe
was employed instead of NaOꢀtꢀBu, as NaOMe was found experimentally to remove the TIPS group more efficiently, see ref.
3(a).
4 For a recent review on Migita coupling see, Beletskaya, I. P.; Ananikov, V. P. Chem. Rev. 2011, 111, 1596.
5 Fernández–Rodríguez, M. A.; Shen, Q.; Hartwig, J. F. J. Am. Chem. Soc. 2006, 128, 2180.
6 Brown, A. D.; Davis, R. D.; Fitzgerald, R. N.; Glover, B. N.; Harvey, K. A.; Jones, L. A.; Liu, B.; Patterson, D. E.; Sharp, M. J.
Org. Process Res. Dev. 2009, 13, 297.
7 Korn, T. J.; Knochel, P. Synlett 2005, 1185.
8 Lai, C.; Backes, B. J. Tetrahedron Lett. 2007, 48, 3033.
9 (a) Takagi, K. Chem. Lett. 1985, 1307. (b) Takagi, K. Chem. Lett. 1986, 1379.
10 Moseley, J. D.; Brown, D.; Firkin, C. R.; Jenkin, S. L.; Patel, B.; Snape, E. W. Org. Process Res. Dev. 2008, 12, 1044. Note:
the role of NBS here is to convert NaSCN into a electrophile.
11 (a) Dobo, K.; Greene, N.; Cyr, M.; Caron, S.; Ku, W. Regul. Toxicol. Pharmacol. 2006, 44; 282; (b) Seifried, H.; Seifried, R.;
Clarke, J.; Junghans, T.; San, R. Chem. Res. Toxicol. 2006, 19; 627; (c) Andrews, L. S.; Clary, J. J. J. Toxicol. Environ. Health
1986 19, 149; (d) Gordon, S. C., Zimmerman, D. D., Griffith, F. D. J. Toxicol. Environ. Health 1991, 34. 297; (e) Johannsen, F.
R.; Vogt, B.; Waite, M.; Deskin, R. Regul. Toxicol. Pharmacol. 2008 50, 322.
12 (a) Itoh, T.; Mase, T. Org. Lett. 2004, 6, 4587. (b) Itoh, T.; Mase, T. J. Org. Chem. 2006, 71, 2203.
13 Huang, Y.ꢀB.; Yang, C.ꢀT.; Yi, J.; Deng, X.ꢀJ.; Fu, Y.; Liu, L. J. Org. Chem. 2011, 76, 800.
14 Zhang, J.; Medley, C. M.; Krause, J. A.; Guan, H. Organometallics 2010, 29, 6393.
15 Correa, A.; Carril, M.; Bolm, C. Angew. Chem., Int. Ed. 2008, 47, 2880.
16 Wong, Y.ꢀC.; Jayanth, T. T.; Cheng, C.ꢀH. Org. Lett. 2006, 8, 5613.
17 Hartwig, J. F. Acc. Chem. Res. 2008, 41, 1534.
18
While our work was onꢀgoing, a TIPSꢀSH based oneꢀpot protocol was published by Hartwig: Fernández–Rodríguez, M. A.;
Hartwig, J. F. Chem. Eur. J. 2010, 16, 2355.
19 Flahive, E. J.; Ewanicki, B. L.; Sach, N. W.; O’Neill–Slawecki, S. A.; Stankovic, N. S.; Yu, S.; Guinness, S. M.; Dunn, J. Org.
Process Res. Dev. 2008, 12, 637.
20 For reports on exchange of the phosphineꢀ and palladiumꢀbound aryl groups, see: (a) Kong, K. C.; Cheng, C. H. J. Am. Chem.
Soc. 1991, 113, 6313. (b) Morita, D. K.; Stille, J. K.; Norton, J. R. J. Am. Chem. Soc. 1995, 117, 8576.
21 Barañano, D.; Hartwig, J. F. J. Am. Chem. Soc. 1995, 117, 2937.
22 (a) Zheng, N.; McWilliams, J. C.; Fleitz, F. J; Armstrong, J. D., III; Volante, R. P. J. Org. Chem. 1998, 63, 9606. (b) Wager, K.
M.; Daniels, M. H. Org. Lett. 2011, 13, 4052.
23
(a) Goodson, F. E.; Wallow, T. I.; Novak, B. M. J. Am. Chem. Soc. 1997, 119, 12441. (b) Grushin, V. V. Organometallics
2000, 19, 1888; [Note: Grushin established that CꢀP reductive elimination occurred by dissociation of one phosphine ligand first
to give a coordinately unsaturated tricoordinated Pd(II) intermediate, which is competent to reductively eliminate phosphonium
ion; the presence of added phosphine ligand retarded the formation of tricoordinated Pd(II) intermediate, hence disfavored CꢀP
reductive elimination; the same observation was also made by Hartwig, see Ref. 20.]
ACS Paragon Plus Environment

Page 25 of 26
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
24 (a) Segelstein, B. E.; Butler, T. W.; Chenard, B. L. J. Org. Chem. 1995, 60, 12. (b) Kwong, F. Y.; Lai, C. W.; Yu, M.; Chan, K.
S. Tetrahedron 2004, 60, 5635. (c) Marcoux, D.; Charette, A. B. J. Org. Chem. 2008, 73, 590. (d) Wakioka, M.; Nakajima, Y.;
Ozawa, F. Organometallics 2009, 28, 2527.
In Suzuki coupling, Novak suggests that exchange between phosphineꢀ and palladiumꢀbound aryl moieties occurs after
oxidative addition before transmetalation (ref. 23). In ref. 22b, the authors postulate that the exchange could occur either before
25
or after transmetalation.
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
26 Based on the catalytic cycle of Migita coupling, as depicted in Figure 2, the rate of the reaction is defined by:
∗
ꢈ
ꢀꢁꢂꢃ = ꢊ ꢫꢫꢫ ꢋ = ꢄ
ꢬꢬꢬ
(2)
(3)
∗
ꢈ
ꢋ
ꢀꢃꢇ
ꢊꢌ
According to BriggsꢀHaldane approximation:
∗
∗
−
ꢈ
ꢯꢰ
ꢂꢀꢁꢭꢮ
∗
ꢋ ꢈ ꢋ
∙ ꢬꢬ
ꢈ
ꢋ
ꢈ
ꢋ ꢈ
ꢋ
ꢄ
ꢀꢃꢇ
ꢬꢬꢬ + ꢄ
ꢬꢬꢬ ∙ ꢪꢀ = ꢄ
ꢂꢀꢁꢭꢮ
−
And:
∗
∗
ꢋꢈ
ꢩꢀꢪꢀ
ꢈ
ꢋ
ꢈ
ꢋ
ꢄ
ꢀꢃꢇ
ꢬꢬꢬ = ꢄ
ꢬ
(4)
(5)
ꢅꢆꢇ
Where:
ꢈ
ꢋ
ꢱ
ꢳꢴꢵ ∙ꢈꢶꢫꢉꢷꢸꢋ
ꢑ
ꢎꢲ
−
ꢈ
ꢯꢰ
ꢋ
=
ꢈꢶꢫꢉꢷꢸꢵ
ꢋ
Substitute equation (4) into (2) and solve for [III*]:
ꢈ
ꢋ
ꢞ
ꢠꢡꢢ ∙ꢈꢣꢤꢥꢦꢖꢋ
ꢛꢟ
∗
∙ꢈꢫꢫ ꢋ∙
ꢹ
ꢌꢺꢎꢻꢼ
ꢧ
ꢈꢣꢤꢥꢦꢖꢢ
ꢋ
∗
ꢈ
ꢬꢬꢬ
ꢋ
=
(6)
(7)
ꢹ
ꢑꢹ
ꢈꢿꢺꢋ
ꢌꢺꢎꢻꢼ
ꢺꢽꢊ
ꢾ
Substitute equations (4) and (5) into (3):
ꢈ
ꢋ
ꢞ
ꢠꢡꢢ ∙ꢈꢣꢤꢥꢦꢖꢋ
ꢛꢟ
∗
∙ꢈꢫꢫ ꢋ∙
ꢹ
ꢌꢺꢎꢻꢼ
ꢧ
ꢈꢣꢤꢥꢦꢖꢢ
ꢋ
∗
ꢈ
= ꢄ ꢬ
ꢅꢆꢇ
ꢋꢈ ꢋ
ꢩꢀꢪꢀ
ꢄ
ꢀꢃꢇ
∙
ꢹ
ꢑꢹ
ꢈꢿꢺꢋ
ꢌꢺꢎꢻꢼ
ꢺꢽꢊ
ꢾ
Rearrange and solve for [II]:
∗
ꢈ
ꢋ
ꢹ
ꢫ
ꢈꢸꢺꢿꢺꢋ
ꢍꣀꢊ
∗
ꢋ
=
ꢈ
ꢬꢬ
(8)
ꢈ
ꢋ
ꢞ
ꢠꢡꢢ ∙ꢈꢣꢤꢥꢦꢖꢋ
ꢛꢟ
ꢒ
∙
ꢧ
ꢚꢗꢛꢜꢝ
ꢈꢣꢤꢥꢦꢖꢢ
ꢧꢒ ꢈꢘꢗꢋ
ꢚꢗꢛꢜꢝ
ꢨ
ꢋ
ꢹ
∙
ꢺꢽꢊ
ꢒ
ꢗꢙꢕ
Substitute (7) into (5):
∗
ꢈ
ꢞ
ꢋ
ꢈ
ꢞ
ꢛꢟ
ꢋ
ꢒ
ꢤ
ꢈꢖꢗꢘꢗꢋ
ꢈ
ꢠꢡꢢ ∙ꢈꢣꢤꢥꢦꢖꢋ
ꢧ
ꢈꢣꢤꢥꢦꢖꢢ ꢋ
ꢓꢔꢕ
ꢹ
∙
∙
ꢠꢡꢢ ∙ꢈꢣꢤꢥꢦꢖꢋ
ꢌꢺꢎꢻꢼ
ꢋ
ꢛꢟ
∙
ꢒ
ꢧ
ꢚꢗꢛꢜꢝ
∙
ꢒ
ꢈꢣꢤꢥꢦꢖꢢ
ꢧꢒ
ꢋ
ꢒ
ꢗꢙꢕ
ꢈꢘꢗꢋ
∗
ꢗꢙꢕ
ꢚꢗꢛꢜꢝ
ꢨ
ꢈ
ꢬꢬꢬ
ꢋ
=
(9)
ꢹ
ꢑꢹ
ꢈꢿꢺꢋ
ꢌꢺꢎꢻꢼ
ꢺꢽꢊ
ꢾ
Based on material balance:
∗
∗
∗
ꢋ
∗
ꢈ
ꣁꢇ
ꢋ
ꢈ
ꢋ
ꢈ
= ꢬ + ꢬꢬ + ꢈꢬꢬꢬ ꢋ
(10)
ꢂꢅꢂꢁꣂ
ACS Paragon Plus Environment

Organic Process Research & Development
Page 26 of 26
1
2
3
∗
4
5
6
7
ꢈ
ꢞ
ꢋ
ꢈ
ꢋ
ꢒ
ꢤ
ꢈꢖꢗꢘꢗꢋ
ꢞ
ꢠꢡꢢ ∙ꢈꢣꢤꢥꢦꢖꢋ
ꢓꢔꢕ
ꢛꢟ
ꢹ
∙
∙
ꢠꢡꢢ ∙ꢈꢣꢤꢥꢦꢖꢋ
ꢌꢺꢎꢻꢼ
ꢧ
ꢈ
ꢋ
ꢛꢟ
∙
ꢈꢣꢤꢥꢦꢖꢢ
ꢋ
ꢒ
ꢧ
ꢚꢗꢛꢜꢝ
∙
ꢗꢙꢕ
ꢒ
ꢈꢣꢤꢥꢦꢖꢢ
ꢋ
∗
ꢒ
ꢈ
∙
ꢋ
ꢹ
ꢫ
ꢈꢸꢺꢿꢺꢋ
ꢧꢒ
ꢈꢘꢗꢋ
ꢍꣀꢊ
ꢗꢙꢕ
ꢚꢗꢛꢜꢝ
ꢨ
∗
ꢋ
∗
ꢋ
+
ꢈ
ꣁꢇ
ꢈ
= ꢬ
+
ꢋ
ꢠꢡꢢ ∙ꢈꢣꢤꢥꢦꢖꢋ
(11)
(12)
ꢂꢅꢂꢁꣂ
ꢈ
ꢞ
ꢛꢟ
ꢹ
ꢑꢹ
ꢈꢿꢺꢋ
ꢌꢺꢎꢻꢼ
ꢒ
ꢧ
ꢺꢽꢊ
ꢚꢗꢛꢜꢝ
ꢾ
ꢈꢣꢤꢥꢦꢖꢢ
ꢧꢒ ꢈꢘꢗꢋ
ꢚꢗꢛꢜꢝ
ꢨ
ꢋ
ꢹ
∙
ꢺꢽꢊ
ꢒ
ꢗꢙꢕ
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
So:
∗
ꢈ
ꢋ
ꢉꢊ
∙
ꢌꢍꢌꢎꢏ
ꢒ
∗
ꢈ
ꢬ
ꢋ
=
ꢈ ꢋ
ꢠꢡꢢ ∙ꢈꢣꢤꢥꢦꢖꢋ
ꢛꢟ
ꢒ
ꢈꢖꢗꢘꢗꢋ
ꢞ
ꢓꢔꢕ
ꢒ
∙
ꢋ
ꢠꢡꢢ ∙ꢈꢣꢤꢥꢦꢖꢋ
ꢧ
ꢈꢣꢤꢥꢦꢖꢢ ꢋ
ꢚꢗꢛꢜꢝ
ꢈ
ꢞ
ꢛꢟ
∙
ꢧ
ꢚꢗꢛꢜꢝ
ꢈꢣꢤꢥꢦꢖꢢ
ꢋ
ꢒ
∙
ꢗꢙꢕ
ꢒ
ꢈꢖꢗꢘꢗꢋ
ꢒ
ꢧꢒ ꢈꢘꢗꢋ
ꢓꢔꢕ
ꢗꢙꢕ
ꢒ
ꢚꢗꢛꢜꢝ
ꢨ
ꢐꢑ
ꢠꢡꢢ ∙ꢈꢣꢤꢥꢦꢖꢋꢑ
ꢈ
ꢋ
ꢞ
ꢛꢟ
ꢧꢒ ꢈꢘꢗꢋ
ꢚꢗꢛꢜꢝ
ꢨ
ꢗꢙꢕ
ꢒ
∙
∙
ꢧ
ꢚꢗꢛꢜꢝ
ꢈꢣꢤꢥꢦꢖꢢ
ꢋ
ꢒ
ꢗꢙꢕ
ꢒ
ꢧꢒ
ꢚꢗꢛꢜꢝ
ꢨ
ꢈꢘꢗꢋ
ꢗꢙꢕ
Consequently the unꢀbridged steady state rate equation for the overall reaction should be:
∗
∗
ꢈ
ꢋ
ꢈ
ꢋꢈ
ꢋ
ꢀꢁꢂꢃ = ꢄ
ꢀꢃꢇ
= ꢄ
ꢅꢆꢇ
ꢬꢬꢬ = ꢄ
ꢅꢆꢇ
ꢬ
ꢩꢀꢪꢀ
∗
ꢈ
ꢋ
ꢉꢊ
ꢌꢍꢌꢎꢏ
ꢒ
ꢈ
ꢋ
ꢠꢡꢢ ∙ꢈꢣꢤꢥꢦꢖꢋ ∙ ꢩꢀꢪꢀ
ꢧ
ꢈꢣꢤꢥꢦꢖꢢ ꢋ
(1)
ꢈ
ꢋ
ꢒ
ꢈꢖꢗꢘꢗꢋ
ꢓꢔꢕ
ꢞ
ꢛꢟ
ꢒ
∙
∙
ꢠꢡꢢ ∙ꢈꢣꢤꢥꢦꢖꢋ
ꢈꢣꢤꢥꢦꢖꢢ
ꢚꢗꢛꢜꢝ
ꢈ
ꢋ
ꢞ
ꢛꢟ
∙
ꢧ
ꢚꢗꢛꢜꢝ
ꢋ
ꢒ
∙
ꢗꢙꢕ
ꢒ
ꢈꢖꢗꢘꢗꢋ
ꢒ
ꢧꢒ
ꢚꢗꢛꢜꢝ
ꢨ
ꢈꢘꢗꢋ
ꢓꢔꢕ
ꢞ
ꢗꢙꢕ
ꢒ
ꢐꢑ
ꢠꢡꢢ ∙ꢈꢣꢤꢥꢦꢖꢋꢑ
ꢈ
ꢋ
ꢛꢟ
ꢧꢒ
ꢈꢘꢗꢋ
ꢚꢗꢛꢜꢝ
ꢗꢙꢕ
ꢒ
∙
∙
ꢨ
ꢧ
ꢚꢗꢛꢜꢝ
ꢈꢣꢤꢥꢦꢖꢢ
ꢋ
ꢒ
ꢗꢙꢕ
ꢒ
ꢧꢒ
ꢈꢘꢗꢋ
ꢗꢙꢕ
ꢚꢗꢛꢜꢝ
ꢨ
27 Blackmond, D. G. Angew. Chem. Int. Ed. 2005, 44: 4302. For inductrial application of RPKA, see: Haddad, N.; Qu, B.; Lee,
H.; Lorenz, J.; Varsolona, R.; Kapadia, S.; Sarvestani, M.; Feng, X.; Busacca, C.A.; Hebrault, D.; Rea, S.; Schellekens, L.;
Senanayake, C.H. Org. Process Res. Dev. 2015, 19, 132.
28 Shi, W.; Luo, Y.; Luo, X.; Chao, L.; Zhang, H.; Wang, J.; Lei, A. J. Am. Chem. Soc. 2008, 130, 14713.
29 Inspection of the rateꢀlaw:
∗
∗
ꢋ
ꢌꢍꢌꢎꢏ
ꢀꢁꢂꢃ = ꢊ ꢫꢫꢫ ꢋ = ꢄ
ꢈꢖꢗꢘꢗꢋ ꢈꢩꢀꢪꢀꢋ
ꢗꢙꢕ
(1’)
ꢈ
ꢈ
ꢉꢊ
ꢅꢆꢇ ꢐꢑ
ꢒ
∙
∙ꢈꢖꢗꢘꢗꢋ
ꢒ
ꢓꢔꢕ
ꢓꢔꢕ
ꢈ
ꢊꢌ
ꢠꢡꢢ ∙ꢈꢣꢤꢥꢦꢖꢋꢑ
ꢋ
ꢞ
ꢛꢟ
ꢒ
ꢒ
ꢧ
ꢚꢗꢛꢜꢝ
ꢈꢣꢤꢥꢦꢖꢢ
ꢋ
Reveals that when ꢄ
the equation further reduced to:
ꢄ
ꢄ
ꢀꢃꢇ is far far smaller than ꢂꢀꢁꢭꢮ and ꢅꢆꢇ, the third term in the denominator became the dominant term,
∗
ꢋ
∗
ꢋ
ꢌꢍꢌꢎꢏ
ꢈ
ꢈ
ꢒ
ꢉꢊ
ꢉꢊ
ꢈꢖꢗꢘꢗꢋ
ꢓꢔꢕ
ꢒ
ꢗꢙꢕ
ꢌꢍꢌꢎꢏ
∗
ꢈ ꢋ
∙ ꣁꢇ
ꢂꢅꢂꢁꣂ
ꢈ
ꢋ
ꢈ
ꢋ
ꢀꢁꢂꢃ = ꢄ
ꢅꢆꢇ ꢐꢑ
∙ ꢩꢀꢪꢀ = ꢄ
ꢅꢆꢇ
∙ ꢩꢀꢪꢀ = ꢄ
ꢀꢃꢇ
(13)
ꢒ
∙
∙ꢈꢖꢗꢘꢗꢋ
ꢈ
ꢒ
ꢈꢖꢗꢘꢗꢋ
ꢓꢔꢕ
ꢗꢙꢕ
ꢓꢔꢕ
ꢛꢟ
ꢠꢡꢢ ∙ꢈꢣꢤꢥꢦꢖꢋꢑ
ꢋ
ꢞ
ꢒ
ꢒ
ꢧ
ꢚꢗꢛꢜꢝ
ꢈꢣꢤꢥꢦꢖꢢ
ꢋ
In other words, when reductive elimination is the rateꢀdetermining step, the overall reaction is zeroth order in both the bromide
and the thiolate.
30
The word “slow” is used in relative terms, compared to oxidative addition and transmetalation under current conditions. In
absolute terms, the C–S reductive elimination is actually very fast, as all the reactions were completed within 10 min in the
kinetic studies; Hartwig’s kinetics work on C–S reductive elimination demonstrated that such process proceeded readily at
temperature as low as 50 °C, see ref. 21.
31
Novak and Grushin have established that Pd and P bond aryl exchange is highly dependent on the halide on Pd. For instance,
the exchange in much more pronounced when Iꢀ is on Pd than Brꢀ. Considering that preꢀdissociation of one phosphine is required
prior to exchange, this could be the result of trans effect. Since Brꢀ ranked was lower that sulfur in the trans effect hierarchy (Brꢀ
<Iꢀ<S), intermediate II in Figure 2 is not expected to be prone to undergo aryl exchange. Also note that aryl exchange is most
severe when monodentate ligands like PPh3 are used, see ref 20; such exchanges are suppressed by the employment of bidentate
ligands.
ACS Paragon Plus Environment