A Comparative Study of Dibenzylideneacetone Palladium Complexes in Catalysis

Article abstract of DOI:10.1021/acs.oprd.9b00214

Commercial Pd_x(dba)_y from various suppliers was found to vary considerably in appearance, homogeneity, purity, and catalytic activity. The Buchwald-Hartwig amination of 4-bromoanisole (5) with aniline (6) was established as a sensitive test reaction to probe the efficiency of Pd_x(dba)_y batches in catalytic transformations. The yields obtained with 17 different Pd_x(dba)_y batches ranged from 10% to nearly quantitative and could not be predicted reliably on the basis of any physical or spectroscopic descriptor alone. The best results in the catalytic test reaction were consistently achieved with a self-made slowly crystallized Pd₂(dba)₃·toluene adduct. A protocol is disclosed that allows batches of Pd_x(dba)_y with unsatisfactory or inconsistent performance to be converted into this reliable precatalyst.

Full text of DOI:10.1021/acs.oprd.9b00214

Subscriber access provided by BUFFALO STATE
Communication
A comparative study of dibenzylideneacetone
palladium complexes in catalysis
Philip Weber, Agostino Biafora, Angelino Doppiu, Hans-Josef Bongard, Harald Kelm, and Lukas J. Gooßen
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.9b00214 • Publication Date (Web): 06 Jun 2019
Downloaded from http://pubs.acs.org on June 6, 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
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
A comparative study of dibenzylideneacetone
palladium complexes in catalysis
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
Philip Weber,†[+] Agostino Biafora,#[+] Angelino Doppiu,§ Hans-Josef Bongard,‡ Harald
Kelm,## and Lukas J. Gooßen†*
†Fakultät für Chemie und Biochemie, Ruhr Universität Bochum, Universitätsstr. 150,
44801 Bochum (Germany), lukas.goossen@rub.de, http://www.ruhr-uni-
bochum.de/oc1/goossen
#FB Chemie, Organische Chemie, Technische Universität Kaiserslautern, Erwin-
Schrödinger-Strasse, Geb. 54, 67663 Kaiserslautern (Germany)
##FB Chemie, Anorganische Chemie, Technische Universität Kaiserslautern, Erwin-
Schrödinger-Strasse, Geb. 54, 67663 Kaiserslautern (Germany)
§Umicore AG & Co. KG, Rodenbacher Chaussee 4, Building 895, Room 208, 63457
Hanau-Wolfgang (Germany)
ACS Paragon Plus Environment
1

Organic Process Research & Development
Page 2 of 44
1
2
3
4
5
‡Max-Planck-Institut für Kohlenforschung, Chemische Kristallographie und
6
7
8
Elektronenmikroskopie, Kaiser-Wilhelm-Platz 1, 45470 Mülheim an der Ruhr (Germany)
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
KEYWORDS: Palladium, Pd2(dba)3, Pd(dba)2, Pd catalysis, Pd nanoparticles, amination,
reproducibility.
ABSTRACT: Commercial Pdx(dba)y from various suppliers was found to vary
considerably in appearance, homogeneity, purity, and catalytic activity. The Buchwald-
Hartwig amination of 4-bromoanisole (5) with aniline (6) was established as a sensitive
test reaction to probe the efficiency of Pdx(dba)y batches in catalytic transformations. The
yields obtained with 17 different Pdx(dba)y batches range from 10% to near quantitative
yields, and cannot be predicted reliably based on any physical or spectroscopic descriptor
alone. The best results in the catalytic test reaction were consistently achieved with a self-
made, slowly crystallized Pd2dba3·toluene adduct. A protocol is disclosed that allows
converting batches of Pdx(dba)y with unsatisfactory or inconsistent performance into this
reliable pre-catalyst.
ACS Paragon Plus Environment
2

Page 3 of 44
Organic Process Research & Development
1
2
3
4
5
Introduction
6
7
8
9
Over the past decades, palladium-catalyzed coupling reactions have become
established as essential tools for C–C and C–heteroatom bond formation.1 The
pioneering contributions by Heck, Negishi, and Suzuki to this field of research were
acknowledged with the Nobel Prize in 2010.2 One of the key factors responsible for the
tremendous number of synthetic applications achieved with palladium catalysis was the
implementation of systematic catalyst screening. The optimal catalyst for a given
transformation can be rapidly determined by tuning a common palladium precursor using
a range of possible ligands and additives, and by optimizing further reaction parameters
such as solvents, bases, and temperatures. In this context, Pd0 adducts of
dibenzylideneacetone (dba) are among the most widely used Pd0 precursors.
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(dba)2 (1) was first synthesized by Takahashi et al. in 1970.3 The structural formula
suggests a defined stoichiometry, but it has meanwhile become clear that instead, it is a
rather complex mixture of species.4 Slow crystallization from coordinating solvents results
in the formation of defined solvent adducts of Pd2(dba)3 (2), for example, Pd2(dba)3·CHCl3
ACS Paragon Plus Environment
3

Organic Process Research & Development
Page 4 of 44
1
2
3
4
5
(3), ·CH2Cl2 or ·C6H6, in which the solvent molecules are part of elemental crystal cells
6
7
8
(Table 1).5
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 1. Known Pdx(dba)y-species
molecular structure of Pd2(dba)3
species
appearance
(2)
brown, crystalline3
red-violet, crystalline4a
orange-brown, crystalline4b
purple, crystalline6
purple needles5
a
Pd(dba)2 (1)
Pd(dba)3∙C6H6
a
Pd
Pd
Pd2(dba)3 (2)
O
a
Pd2(dba)3∙CHCl3 (3)
Pd2(dba)3∙CH2Cl2
Pd2(dba)3∙C6H6
wine-red, crystalline4d
purple needles5
3
Pd2(dba)3∙PhMe
purple needles5
a commercially available.
It was initially assumed that the reaction of Pd0-dba adducts with stronger donor ligands,
e.g. phosphines or N-heterocyclic carbenes, would give rise to homoleptic PdLn
complexes in which the number of ligands would depend on the palladium-to-ligand
stoichiometry.7 However, it has repeatedly been observed that dba coordinates rather
ACS Paragon Plus Environment
4

Page 5 of 44
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
strongly and remains partially bound to the metal, thus influencing the overall catalyst
activity.6,8 Fairlamb et al. have performed comparative studies with substituted dba
derivatives and found a clear influence of the electron-donating nature of the aryl residue
on the catalytic activity of palladium-dba-type species.9 The more strongly donating the
aryl residue, the less stable is the coordination of dba and the easier it is to fully replace
it by a phosphine ligand. However, weakening the dba interaction also leads to lower
stability of the palladium intermediates and thus, a shorter catalyst lifetimes. The wealth
of applications seem to suggest that Pd complexes with dba itself strike the ideal balance
between activity and stability that is required for a reliable in situ-conversion to active
catalysts bearing customized ligands.
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 gives an overview of representative reactions in which dba adducts of
palladium are used as catalyst precursors. They include Heck, Suzuki, Negishi,
Buchwald-Hartwig, Tsuji-Trost, Sonogashira, Stille, and our own decarboxylative
couplings.10 In some of these cases, dba acts as the stabilizing agent itself, but in most
reactions, phosphines or other ligands are added.
ACS Paragon Plus Environment
5

Organic Process Research & Development
Page 6 of 44
1
2
3
4
5
6
Ar-Br, PCy
Ar'-Cl, PCy
3
3
Heck
Ar
R
R
Suzuki-Miyaura
BY2
Ar Ar'
Ar
7
8
9
Ar-X, PCyp
3
Negishi
Ar R
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 Zn Y
X = I, Br, Cl, OTs
Ar-X, DTBNpP
X = Br, Cl
Buchwald-Hartwig
Tsuji-Trost
Ar NR'R''
Nu
H NR'R''
X
Pd (dba)
x
y
t
NuH , Bu-PHOX
Ar-X, TBAF
X = I, Br
H
R
Ar
R
Sonogashira
Stille
R''-X
X = I, Br, Cl, OTf
SnR'3
CO2H
R
R
R R''
R R'
Decarboxylative
Cross-Couplings
R'-X
X = Br, Cl, OTf
Scheme 1. Pdx(dba)y-catalyzed reactions.
As reaction developers, we are frequently contacted by synthetic chemists from
academia or industry who experience substantial deviations in their experimental
outcomes from published reactions when using Pdx(dba)y precursors, despite all efforts
to avoid known pitfalls, such as varying solvent quality or different handling of reaction
substrates.11 In our experience, this is often a consequence of different traditions in the
groups of catalyst developers and synthetic chemists: Whereas the former prefer to
prepare their own Pdx(dba)y complexes using optimized protocols handed down within
ACS Paragon Plus Environment
6

Page 7 of 44
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
their research groups, the latter usually rely on commercially available sources. The
characteristics of the Pdx(dba)y precursor employed will substantially affect the yields
whenever the reaction under investigation is close to the performance limit of the
palladium catalyst and/or the reaction outcome critically depends on the palladium-to-
ligand stoichiometry. In contrast, the quality of the Pdx(dba)y complexes will not
immediately influence the reaction outcome if catalyst activity is so high that comparable
results are achieved even if only a fraction of the palladium precursor is converted to the
actual catalyst. The outcome will also be unaffected if the actual catalytic species is
colloidal palladium rather than defined PdLn complexes. Ananikov and Zalesskiy found
that commercially available Pdx(dba)y samples contain up to 40% Pd nanoparticles and
propose that many reactions that are assumed to depend on homogeneous catalysts are
in fact catalyzed by such particles.12 Colacot intensively studied palladium acetate against
the two major impurities found in commercial samples: Pd3(OAc)5(NO2) and polymeric
[Pd(OAc)2]n in a variety of cross-coupling reactions.13 Whereas the impurities can be as
active as the pure palladium acetate in certain reactions, pure palladium acetate turned
out to be superior in precatalyst formation, in this case a cyclopalladation.
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
7

Organic Process Research & Development
Page 8 of 44
1
2
3
4
5
6
7
8
9
The aim of the present study was to correlate the activity of Pdx(dba)y samples with their
physical and spectroscopic properties and explore whether descriptors exist to predict
their catalytic activity.
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
Results & Discussion
In order to quantify and compare the catalytic activity of a set of commercially available
and self-made Pdx(dba)y batches, we searched our laboratory database for an easy-to-
perform reaction that requires a molecularly defined catalyst and that had proven to be
particularly sensitive to the choice of the catalyst precursor. We evaluated a series of
reactions, including Suzuki couplings, Heck reaction and amination reactions. All these
reactions showed varying yields depending on the choice of the palladium precursor, but
the Buchwald-Hartwig amination of aryl bromides was most strongly affected. We finally
chose the coupling of 4-bromoanisole (5) with aniline (6, Scheme 2) as a test reaction,
because it is easy to conduct and gave yields that varied between 10% all the way up to
quantitative yields, depending on the choice of the palladium pre-catalyst.
Using this test reaction, we then investigated the performance of seventeen batches of
Pd(dba)2 (1), Pd2(dba)3 (2) and Pd2(dba)3·chloroform (3) newly purchased from six
ACS Paragon Plus Environment
8

Page 9 of 44
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
different suppliers A-F (Table 2). The catalytic reaction was performed three times for
every batch, using material from different layers of each container. All commercial
materials were used as shipped.
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
t
NaOBu (1.5 eq.)
H
N
[Pd] (0.5 mol%)
Br
H2N
t
PBu (0.4 mol%)
3
+
PhMe, rt, 1h
MeO
MeO
7
5
6
Scheme 2. Test reaction of 4-bromoanisole (5) with aniline (6).
As a reference material, we used a self-made Pd2(dba)3∙toluene (4) adduct. In the
original synthetic procedure by Ishii, 4 was obtained by reaction of PdCl2, dba, and sodium
acetate followed by crystallization of the resulting dba-palladium adduct from a toluene
solution. This way, purple needles with a melting point of 140-141 °C were obtained in
36% yield.5 We used an analogous method to convert commercial Pd(dba)2 (1) into a
well-defined toluene adduct by slowly crystallizing it from a toluene solution layered with
excess pentane. Only the first crop of crystals, which usually corresponded to no more
than 50% of the initial amount, was used. This way, 4 was obtained as dark purple crystals
ACS Paragon Plus Environment
9

Organic Process Research & Development
Page 10 of 44
1
2
3
4
5
6
7
8
of cubic or octahedral shape that were around 0.2 mm in size and had a melting point of
185 °C.
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 palladium loadings for the reaction were calculated based on the Pd content
specified by the supplier14 or, if not available, based on the molecular formula provided
(Table 1). The palladium content of each sample was double-checked by atomic
absorption spectroscopy, revealing no major inconsistencies between our own
measurements and the data provided by the suppliers. The elemental analyses of all
samples of Pd2(dba)3 (2), i.e. 3D and 3E, showed strong discrepancies between
experimentally found and calculated molecular formula, suggesting that some samples
contain excess dba (Supporting Information, Table S1). This is not unexpected since Pd
dba adducts are often non-stoichiometric in nature, and are known to incorporate solvent
molecules during crystallization.
Table 2 lists catalyst specifications for all samples investigated and shows the yields
obtained in the test reaction along with visible light microscopy and SEM images. The
results were strongly variable, with yields ranging from 10% to nearly 100%. Some
catalyst sources consistently gave similar yields, whereas the results obtained with others
ACS Paragon Plus Environment
10

Page 11 of 44
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
varied greatly, which indicates inhomogeneity of the material within the batch. The toluene
adduct 4 used as a reference material performed consistently well throughout, confirming
that the deviations are mostly caused by factors associated with the catalyst precursors
rather than by experimental errors.
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
We also investigated the palladium precatalysts in the Suzuki-Miyaura coupling of 4-
bromo-6-methyl-2H-pyrone 8 and 4-fluorophenylboronic acid 9 (Scheme 3).
F
(HO)2B
Br
[Pd] (0.5 mol%)
+
O
O
THF, 1 M Na CO , rt, 10 or 60 min
2
3
F
O
O
10
8
9
Scheme 3. Test reaction of 4-bromo-6-methyl-2H-pyrone 8 and 4-fluorophenylboronic
acid 9.
Fairlamb et al. had established this reaction as a test to investigate the influence of
substituents at the dba ligands.9a However, 8 is not commercially available, which
complicated the use of this test reaction.15 Moreover, the yields became too high and too
similar to each other already after short reaction time, which precluded a reliable
ACS Paragon Plus Environment
11

Organic Process Research & Development
Page 12 of 44
1
2
3
4
5
6
7
8
9
comparison of pre-catalysts. Less reactive substrate combinations would be required to
test this reaction type. Still, as can be seen from the SI (Table S6), the reactivity trends
found in this alternative test reaction do not contradict those found for the amination.
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
12

Page 13 of 44
Organic Process Research & Development
Species
Sample
Pd(dba)2
1C2
Pd2(dba)3
Pd2(dba)3·CHCl3
Pd2(dba)3·PhMe
1
2
3
4
1A
1B1
1B2
1C1
1D
1E
2A
2B1
2B2
2C
2D
2E
2F
3A
3D
3E
4
16.00- 18.40 18.67 20.00
21.00
20.33
18.90
18.10
23.24c
23.03
23.03 20.00
15.00- 22.10
22.00
15.50
20.56b
20.56b
20.56b
21.12b
Pd(supp)
/ %
5
6
7
8
9
19.57
n.d.
18.37 19.90
20.22
18.50
16.97
23.24
22.85
22.91 19.89
21.39
20.97
15.50
19.14
n.d.
20.51
Pd(AAS)
/ %
0.41
n.d.
n.d.
11.26 0.36
5.70
n.d.
n.d.
2.90
3.02
7.47
6.71
0.62
5.26
2.32
n.d.
n.d.
2.28
n.d.
n.d.
2.20
AiC / %
AsP / %
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
19.16
7.11
19.54
14.52
14.07
20.22
15.38
16.20 19.27
16.13
18.65
16.86
18.31
100
90
80
70
60
50
40
30
20
10
0
1A
1B₁
1B₂
1C₁
1C₂
1D
1E
2A
2B₁
2B₂
2C
2D
2E
2F
3A
3D
3E
4
Table 2. Analytical data, catalytic performance, microscopy at 200x magnification, and SEM for all samples under investigation.a
43
44
45
46
47
48
ACS Paragon Plus Environment
13

Organic Process Research & Development
Page 14 of 44
aReaction conditions: 5 (1.0 mmol), 6 (1.0 mmol), NaOtBu (1.5 mmol), Pdx(dba)y (0.5 mol% Pd, calculated from the Pd content reported by the supplier) and PtBu3 (0.4 mol%) in toluene (2 mL),
b
1
2
3
4
5
6
7
8
9
room temperature, 1h. Yields were determined by GC analysis using n-dodecane as internal standard. Theoretical amount of palladium. Pd(supp.) – palladium content reported by the supplier.
Pd(AAS) – palladium content determined by atomic absorption spectroscopy. AiC – Amount of insoluble components. AsP – Amount of soluble palladium (equal to Pd(AAS) – AiC).
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
14

Page 15 of 44
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Influence of the Pd content. The results summarized in Table 2 reveal no clear
correlation between Pd content and catalytic performance. This is an important finding,
because we are aware from requests by synthetic chemists that it is often assumed that
batches with a high Pd content have a higher purity and will perform better in catalytic
reactions than those with a Pd content that is lower than expected from the molecular
formula. Besides the normal Pd content, the relative amount of soluble Palladium (AsP in
%) versus the amount of insoluble components (AiC in %, consisting mostly of
agglomerated Pd black) were also determined.15
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
Especially when working with catalyst stock solutions, it is important to know how much
of the material actually dissolves in organic solvents, and how much of the bulk material
is insoluble metallic palladium. This correlation was verified in a control experiment. The
yield obtained in our test reaction dropped from 60% to 28% when the Pd-dba precursor
was mixed with Pd black in a 1:1 ratio based on the overall Pd content (Supporting
information, Table S7).
ACS Paragon Plus Environment
15

Organic Process Research & Development
Page 16 of 44
1
2
3
4
5
6
7
8
9
However, the results listed in Table 2 do not reveal any such correlation between AsP
and catalyst activity. Hence, the AsP is no reliable predictor for the performance of a batch
of palladium-dba material in sensitive catalytic 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
Influence of the molecular formula provided. When comparing the activities of
commercial samples of 1, 2, and 3, it seems as though the better control of the
stoichiometry in the Pd2(dba)3∙CHCl3 adduct (3) causes smaller variations with regard to
catalyst performance than the other two compounds. Similar yields were obtained both
within each batch and when using materials obtained from different suppliers. However,
the overall performance of all Pd2(dba)3∙CHCl3 samples (3) is mediocre, which sustains
the prejudice catalyst developers have against adducts that contain non-innocent
components such as chloroform. The activities of Pd(dba)2 (1) and Pd2(dba)3 (2)
complexes are subject to greater variability both within and between batches of 1 and 2
than for Pd2(dba)3∙CHCl3 (3). Thus, at least for this test reaction, the catalyst performance
cannot be predicted based on the dba/Pd ratio. For a comparative study of of Pd(dba)2
vs. Pd2(dba)3, see also the protocol of Peng.16
ACS Paragon Plus Environment
16

Page 17 of 44
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Influence of particle shape and size. Next, all samples were investigated by microscopy
at 200x magnification. Table 2 shows tremendous differences between the samples (see
the Supporting Information for more detailed pictures). In the literature, Pd(dba)2 (1)
crystals are described as brown to red-violet needles.3,4a Such needles were seen for
samples 1A, 1C, 2A, 2C, 2E and 3E. In sample 1A, silver-colored octahedrons, indicative
of elemental Pd,17 are found alongside the needles. Samples 1D, 2B and 2D consist of
hexagonal crystals, which has not previously been described for Pdx(dba)y. In contrast,
samples 1B1 and 1B2 are black, amorphous powders. 1E and 2F consist of cubic crystals.
This crystal shape has been reported for microstructures of Pd2(dba)3 (2) crystallized from
THF/H2O mixtures.18 3A and 3D are violet needles, which is in agreement with literature
reports for 3.5 These examples demonstrate how strongly the manufacturing process
affects crystal shape and appearance of Pdx(dba)y materials.
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 only conclusion one might possibly draw from the microscopic investigations in
combination with the catalytic test reaction is that amorphous powders should be avoided,
because both Pd(dba)2 (1) samples by supplier B gave low yields in the test reaction.
ACS Paragon Plus Environment
17

Organic Process Research & Development
Page 18 of 44
1
2
3
4
5
6
7
8
Other factors detectable by microscopy, such as crystal shape or color do not lead to
reliable predictions with regard to the catalytic performance of the 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
49
50
51
52
53
54
55
56
57
58
59
60
Scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDX).
We next investigated the solid materials by SEM and EDX to obtain further information
on the elementary composition and homogeneity of the samples, and to determine the
presence, form, and size of palladium nanoparticles (Table 2). All samples were sprinkled
on Cu grids and embedded in resin (for more information about methods and sample
preparation see the Experimental Section).
The brightness of the material correlates with the Pd content. At the brightest spots in
the picture of 1B, a palladium content of 85% was measured by SEM, the darker areas
correspond to a Pd content of around 20%. This is indicative of a mixture of elementary
palladium particles with Pd-dba complexes. The palladium particles have sizes of up to
3800 nm, well beyond those of soluble nanoparticles. In contrast, 1C displays a high level
of homogeneity with a Pd content of at most 18.6%. A magnification of 250,000x was
necessary to detect bright spots indicative of elemental palladium. Most of these
nanoparticles have a size of 2 nm, whereas very few measuring up to 20 nm seem to be
ACS Paragon Plus Environment
18

Page 19 of 44
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
embedded within the crystal matrix and could not be detected on the sample surface,
even when incorporating the sample into a resin and preparing resin slices (see
Supporting Information and Experimental Section).
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
Since this highly homogeneous material performs comparatively well in the catalytic
test reaction, one might conclude that SEM analysis allows to predict catalytic
performance based on the homogeneity of the sample and the absence of large, insoluble
Pd particles. Indeed, many homogeneous samples including all samples by manufacturer
C and 2A and 2E performed well, whereas the catalytic activity was lower for many
inhomogeneous samples, e.g. 1E and all samples by supplier B. However, this approach
has its limitations, considering that both 1A and 1D look similar by SEM but have a very
different catalytic activity.
The chloroform adducts were also investigated by EDX, a technique which allows
visualizing the distribution of different atom types. EDX showed that the distribution of
chlorine atoms was remarkably inhomogeneous in sample 3A (Figure 1, EDX 3A Cl K).
The highest chlorine concentration was found in areas that also have a high palladium
ACS Paragon Plus Environment
19

Organic Process Research & Development
Page 20 of 44
1
2
3
4
5
6
7
8
concentration (EDX 3A Pd L). This points towards either chloroform-coated palladium
particles or dissociation of chloroform with formation of palladium chloride particles.
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. SEM and EDX measurements of sample 3A, 3D and 3E. Visualizing chlorine (yellow)
and palladium content (red).
This is interesting, since one would assume from published crystal structures that the
chloroform is embedded within the crystal matrix and, thus, homogeneously distributed.5
The same observation applies to sample 3D. However, in this sample, the chloroform is
not only adsorbed onto the Pd surface, but is also present inside the crystal matrix (EDX
3D Pd L and EDX 3D Cl K). In our experience, homogeneous solvent adducts of
Pd2(dba)3 only form if the sample crystallizes very slowly. We thus assume that the
ACS Paragon Plus Environment
20

Page 21 of 44
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
commercial materials were rapidly precipitated from a chloroform-containing solution,
resulting in nonstoichiometric mixtures of Pdx(dba)y·(CHCl3)z with chloroform-coated
nanoparticles and/or palladium chloride. This may be a reason why the catalytic activity
of the Pd2(dba)3∙CHCl3 samples in our test reaction was relatively low. Interestingly, no
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
chlorine was detected in sample 3E even though H-NMR in solution shows that it is
present in the sample (EDX 3E Cl). We believe that either the concentration of chloroform
is too low for detection, or that chloroform-containing particles have been removed in the
separation step that is part of sample preparation (for more information see the
Experimental Section). In any case, a prediction of catalytic activity based on EDX is not
possible, since despite their differences, all tested samples of 3 have a very similar
catalytic activity.
Other spectroscopic studies. Several additional investigations were performed with the
goal of establishing a correlation between physical or spectroscopic descriptors and the
catalytic activity of dba-palladium adducts.
Any solvation process changes the composition of the dba adducts. This is the main
1
problem when investigating the samples in homogeneous phase, e.g. by H-NMR
ACS Paragon Plus Environment
21

Organic Process Research & Development
Page 22 of 44
1
2
3
4
5
6
7
8
9
spectroscopy. Solution analysis can, thus, be used only to detect organic impurities. In
this context, we found peaks in all tested samples that are indicative of traces of water,
pentane, acetone, alcohols, or acetic acid (see Supporting Information, Figure S18-S23).
However, these small contaminants did not seem to decisively influence the catalytic
activity. The samples were also investigated by MAS (magic-angle spinning, Figure S24)
13C-NMR, IR (infrared, Table S3) and XRD (X-ray diffraction, Figure S25) spectroscopy.
However, none of the observations made in these investigations could be correlated to
the performance of the samples in the test reaction (Supporting 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
We next followed up on the observation that two different batches of Pd2(dba)3 (2)
obtained from the same supplier had a markedly different catalytic activity (Table 2,
sample 2B1 versus 2B2). A possible explanation is that the material loses activity during
storage, and that newer samples are more active than older ones. A series of control
experiments revealed that the sample 2C, which had performed rather well in the first set
of experiments (94, 86, and 81% yield), gave only 71, 40, and 68% yield after it had been
stored under air for 14 months. In sharp contrast, the crystalline Pd2(dba)3∙toluene adduct
(4) still gave near-quantitative conversion (99, 99, and 99% yield) after this time. We
ACS Paragon Plus Environment
22

Page 23 of 44
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
suspected that the relatively high palladium content of 2C was at the origin of these
stability issues, because there may have been insufficient dba to shield the Pd from the
environment. We thus purchased a fresh sample of Pd2(dba)3 (2) from supplier F, which
had a Pd content of only 15.31%, and contained excess dba. The new sample had roughly
the same initial catalytic performance as 2C (83, 89, and 86% yield). We divided this
batch into several samples which we stored under various conditions, and then
reinvestigated the catalytic activity after 6 months. These investigations revealed that
regardless of whether the material was stored under argon or air, in a glass or air-
permeable polyethylene vessel, at room temperature, 40°C or 4°C, the catalytic activity
remained almost unchanged (Figure 2). Although additional systematic stability studies
are clearly required, this comparative experiments suggest that excess dba leads to
enhanced storability of the Pd catalysts.
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
23

Organic Process Research & Development
Page 24 of 44
1
2
3
4
5
6
7
Reference
Air, PE, light, rt
Air, glass, dark, 40 °C
Air, glass, dark, 4 °C
Argon, PE, light, rt
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
Argon, glass, dark, 40 °C
Argon, glass, dark, 4 °C
0 10 20 30 40 50 60 70 80 90 100
Yield / %
Figure 2. Reaction of 4-bromoanisole (5) with aniline (6) under standard conditions before and
after storing the Pd2(dba)3 (2) sample under the given conditions for six months.a
Improving the quality of Pd(dba)2 batches. The above investigations show how difficult
it is for synthetic chemists to be sure that the material they have in stock will perform well
in a catalytic transformation. From our experience, it is advisable to convert stock
palladium dba into a crystalline adduct with an inert solvent such as toluene.
The efficiency of this strategy was probed for sample 1D, which had performed
particularly poorly in the test reaction (Table 2), and had then been stored under air at
room temperature for more than a year. 50 mg (0.09 mmol) of this material were
suspended in 15 mL of toluene (p.a. grade, Acros Organics) and stirred it for 30 minutes
at room temperature. The resulting deep red solution was then separated from the
insoluble material by filtration through a pad of Celite. The clear filtrate was frozen by
ACS Paragon Plus Environment
24

Page 25 of 44
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
cooling the Schlenk flask with liquid nitrogen while applying a slight vacuum. The solidified
product solution was then carefully layered with 5 mL of diethyl ether and 30 mL of
pentane. After warming to room temperature, the slow diffusion of the initially sharply
separated solvent layers caused the product to crystallize within a week (Figure 3). This
yielded 16.5 mg (18%) large, dark purple crystals of toluene adduct (4) with a melting
point of 185 °C. The material thus obtained gave near-quantitative conversion in the test
reaction in three independent runs.
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. Crystallization of the toluene adduct (4) shortly after removing the cold bath.
The fact that this crystallization is time-consuming and low-yielding explains why it is
not a valid large-scale manufacturing process for this material. However, in our hands,
ACS Paragon Plus Environment
25

Organic Process Research & Development
Page 26 of 44
1
2
3
4
5
6
7
8
9
any attempts to obtain the toluene adduct (4) faster and in a higher yield led to non-
stoichiometric materials with unreliable catalytic activity (See Experimental Section). As
can be seen from the crystal structure (Figure 4), even after slow crystallization, the
material still has a relatively high level of disorder (50%). However, it is a well-defined
adduct of two palladium atoms, three dba molecules and one toluene molecule.
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 4. X-ray structure of the Pd2(dba)3∙toluene adduct (4, 50% disorder).
Conclusion
This systematic evaluation of physical and spectroscopic properties of various
commercial dba-Pd adducts has revealed marked differences between the samples with
regard to homogeneity, crystal shape and size, color, and palladium content. As to be
ACS Paragon Plus Environment
26

Page 27 of 44
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
expected for a non-stoichiometric adduct, the molecular formula did not always correlate
with the palladium content of the samples. Many of the dba adducts contain impurities
and particulate Pd. However, no reliable correlation seems to exist between the
spectroscopic or physical data and the catalytic activity. We have found the Buchwald-
Hartwig amination of 4-bromoanisole (5) and aniline (6) to be a sensitive and easy-to-
perform test reaction that allows to rapidly determine the catalytic activity of Pdx(dba)y
samples. It is recommended to first investigate the catalytic activity of a new batch of
Pdx(dba)y with this known reaction before using it for new catalytic transformations. The
best results in the test reactions were consistently obtained with a self-made
Pd2(dba)3∙toluene adduct (4), which had been slowly crystallized over several days. It can
be stored over years under ambient conditions without losing its catalytic performance,
which is superior to most newly purchased commercial samples. Even older and partially
decomposed dba-Pd batches with unsatisfactory catalytic performance can easily be
converted into this reliable pre-catalyst.
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
ACS Paragon Plus Environment
27

Organic Process Research & Development
Page 28 of 44
1
2
3
4
5
6
7
8
9
General Information. All solvents were purchased from Acros Organics in p.a. grade
and purified by standard procedures prior to use.19 4-Bromoanisole (≥99% purity), Tri-
tert-butylphosphine (98% purity), disodium tetrachloropalladate (99.99% purity) and
sodium acetate (≥99% purity) were purchased from Sigma Aldrich and used without
further purification. Aniline (≥99% purity) was purchased from Sigma Aldrich and purified
by standard procedures prior to use.19 4-Bromo-6-methyl-2H-pyrone was synthesized
according to literature.Error! Bookmark not defined. 4-Fluorophenylboronic acid (97% purity) and
dibenzylideneacetone (≥99% purity) were purchased from TCI and used without further
purification. Sodium tert-butoxide (97% purity) and sodium carbonate (99.5% purity) were
purchased from Alfa Aesar and used without further purification. All Solvents and liquids
were degassed with Argon prior to use. Reactions were performed in oven-dried
glassware under a nitrogen atmosphere containing a Teflon-coated stirring bar and dry
septum. All reactions were monitored by GC using n-Dodecane as an internal standard. Response
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
factors of the products with regard to n-Dodecane were obtained experimentally by analyzing
known quantities of the substances. GC analyzes were carried out using an HP-5 capillary column
(Phenyl methyl siloxane, 30 m x 320 x 0.25, 100/2.3-30-300/3, 2 min at 60 °C, heating rate 30 °C
min-1, 3 min at 300 °C). The elementary analyses were measured on a Vario Micro Cube by
ACS Paragon Plus Environment
28

Page 29 of 44
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Elementar. The melting points were measured on a Digimelt MPA161 of Stanford Research
Systems. The microscope pictures were taken with the digital microscope VHX-500F of Keyence.
The sample was placed on a glass slide and illuminated from above (left side) or below (right side).
1
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 IR spectra have been measured on a Spectrum 100 ATR-FTIR of Perkin Elmer. H-NMR
spectra were measured on a FT-NMR spectrometer DPX 400 of Bruker and recorded at ambient
temperature using CD2Cl2 as solvent, with proton resonances at 400 MHz, respectively. All NMR
13
data are reported in ppm relative to the solvent signal. C-MAS spectra were measured on an
AVANCE III 500 Solid State NMR by Bruker. The powder diffractograms were measured on a
Siemens X-Ray powder diffractometer D 5005 using CuKα-radiation (wavelenght 0.15406 nm,
acceleration voltage 40 kV, amperage 30 mA), recorded in a measuring range of 1° < 2Θ < 100°.
The X-ray analysis of Pd2(dba)3·PhMe was done on a diffractometer Oxford Diffraction Gemini
S Ultra at 150 K by using MoKα- or CuKα-radiation.
General procedure for the Buchwald-Hartwig amination. An oven-dried 20 mL vessel
was charged with Pdx(dba)y (0.5 mol% of Palladium), NaOtBu (149 mg, 1.5 mmol) and
PtBu3 (0.4 mol%). Toluene (2 mL), 4-bromoanisole 5 (1 mmol) and n-dodecane (50 µL)
was added as the standard. The reaction mixture was allowed to stir for 10 minutes before
aniline 6 (1 mmol) was added. The resulting mixture was stirred at room temperature
under a nitrogen atmosphere. After 60 minutes, the mixture was diluted with ethyl acetate
(4 mL), washed with water (2 mL) and extracted with ethyl acetate (3 mL). The organic
layer was dried over MgSO4, filtered and directly analyzed by GC.
ACS Paragon Plus Environment
29