High-Throughput Screening of the Alkoxide/Oxime-Based Library: An Alternative to Organotin Compounds for the Alkoxysilane Condensation in Adhesives and Sealants

Article abstract of DOI:10.1021/acscombsci.8b00161

In this work, a high-throughput screening (HTS) method was used to discover new efficient catalysts to substitute organotin compounds (DBTDL) for the cross-linking of silyl-modified polymers (SMPs). We report here on the use of our HTS method to investigate a library of alkoxide/oxime systems with different metal/ligand (M/L) ratios. Among the 156 candidates tested, 40 interesting hits were detected. Then, the cross-linking times for the better hits were measured on the SMP. Some of these seem to be more efficient than DBTDL and exhibit a good stability during storage in cartridges. Thereby, a high efficiency of alkoxide/oxime systems was established that shows great potential for the generation of new ligands to provide new tin-free catalysts for the cross-linking of adhesives and sealants.

Full text of DOI:10.1021/acscombsci.8b00161

Subscriber access provided by University of Winnipeg Library
Article
High-throughput screening of alkoxide/oxime-based
library: an alternative to organotin compounds for the
alkoxysilane condensation in adhesives and sealants
Boris Colin, Olivier LAVASTRE, Stephane Fouquay,
Guillaume Michaud, Frederic Simon, and Jean-Michel Brusson
ACS Comb. Sci., Just Accepted Manuscript • DOI: 10.1021/acscombsci.8b00161 • Publication Date (Web): 24 Jan 2019
Downloaded from http://pubs.acs.org on January 29, 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 30
ACS Combinatorial Science
1
2
3
4
5
6
7
8
9
High-throughput screening of alkoxide/oxime-based
library: an alternative to organotin compounds for the
alkoxysilane condensation in adhesives and sealants
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
Boris Colin†*, Olivier Lavastre‡, Stéphane Fouquay†, Guillaume Michaud†, Frédéric Simon† and
Jean-Michel Brusson§
† BOSTIK SMART TECHNOLOGY CENTER ZAC du bois de Plaisance 101, rue du Champ
Cailloux, 60280 Venette, France
‡ Laboratory of Functional Materials, IETR, UMR 6164 CNRS, University of RENNES 1, 263
Avenue du Général Leclerc, 35042 Rennes Cedex, France
§ TOTAL S.A., Tour Michelet A, 24, Cours Michelet, 92069 La Défense Cedex, France
*Corresponding author. Tel.: +33 344306114 ; E-mail address: boris.colin@bostik.com
KEYWORDS: silyl-modified polymers, tin-free catalysts, cross-linking, high-throughput
screening.
ABSTRACT: In this work, a high-throughput screening (HTS) method was used to discover new
efficient catalysts to substitute organotin compounds (DBTDL) for the cross-linking of silyl-
modified polymers (SMPs). We report here on the use of our HTS method to investigate a library
of alkoxide/oxime systems with different metal/ligand (M/L) ratios. Among the 156 candidates
tested, 40 interesting hits were detected. Then, the cross-linking times for the better hits were
measured on the SMP. Some of these seem to be more efficient than DBTDL and exhibit a good
ACS Paragon Plus Environment
1

ACS Combinatorial Science
Page 2 of 30
1
2
3
4
5
6
7
8
9
stability during storage in cartridges. Thereby, a high efficiency of alkoxide/oxime systems was
established that shows great potential for the generation of new ligands to provide new tin-free
catalysts for the cross-linking of adhesives and sealants.
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
GRAPHICAL ABSTRACT:
-
ACS Paragon Plus Environment
2

Page 3 of 30
ACS Combinatorial Science
1
2
3
4
INTRODUCTION
5
6
7
8
Silyl-modified polymers (SMPs) are organosilane-based materials containing organic and
inorganic moieties that can form polymers with the beneficial characteristics of both parts and
without some of their disadvantages. Presently, SMPs are used as key materials in adhesive1,
sealant2 and coating3 applications. These hybrid polymers are widely used in the do-it-yourself
(DIY) market as adhesives or sealants and are applied to a large variety of indoor and outdoor
applications for construction. They can be used in several other specific applications in the food
and beverage industry4 or as electrolytes5. The cross-linking of SMPs occurs with alkoxysilane
groups attached to the ends of the polymer chains by hydrolysis and polycondensation with
moisture (Fig. 1). Without the addition of any catalyst, the cross-linking of alkoxysilane slowly
occurs with moisture. Therefore, a catalysis of both a hydrolysis and condensation reaction is
necessary to accelerate the curing time and achieve a fully cross-linked material. Indeed, Brønsted
acids, such as HOTf6, -TSA7 or HCl8, and superbases, such as DBU and TBD9 or phosphazene10,
are very efficient at improving the cross-linking time of organosilanes. However, these are usually
avoided in sealants as they could result in some side effects, such as a reaction of acids with
calcium carbonate-based fillers, odor diffusion from the material, color change over time, or
migration of the catalyst from the 3D cross-linked polymer to another matrix in contact with the
polymer. As organometallic catalysts have a better compatibility, organotin compounds such as
dibutyltin dilaurate (DBTDL) or dioctyltin dilaurate (DOTL) represent the best references to
accelerate the cross-linking of SMP with moisture11. In this case, the true catalytic species is a
dibutyltin laurate hydroxide intermediate. However, in several years, the high toxicity of organotin
compounds has been confirmed and new regulatory restrictions on the use of tin catalysts have
been enacted. Therefore, there is a great need to find new efficient tin-free catalysts to anticipate
stronger restrictions in coming years or the ban of organotin compounds. Some alternatives were
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
3

ACS Combinatorial Science
Page 4 of 30
1
2
3
4
5
6
7
8
9
already used by DOW in silicone chemistry with the use of oxime compounds and titanium
catalysts12. More recently, titanium and zirconium alkoxides13 have been used as tin-free catalysts
in sealants containing SMP. This kind of alkoxide has the disadvantage of being unstable with
moisture and thus easily hydrolyzed with a longer curing time of the sealant after storage in
cartridges. To address this issue, alkoxides can be stabilized with a large variety of chelating
agents, such as acetylacetone14, acetic acid15 or diethanolamine16 to form M-O-C bonds.
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
OCH3
Si
OCH3
H3CO Si
OCH3
CH3
Polymer
R
R
H3C
Hydrolysis
H2O
OH
Si
OCH3
CH3
+ CH3OH
Polymer
Polymer
R
OR'
Si
Condensation
OCH3
OCH3
CH3
R
OCH3
Si
Si
H3C
+ R'OH
R
CH3
Polymer
Polymer
R
O
Polycondensation
CH3
Si
SiO
R
Polymer
CH3
O
Si
Si
SiO
R
R
Polymer
Polymer
O
OCH3
n
Figure 1. General scheme of hydrolysis and polycondensation reactions (R= CH2NHCOO or
C3H6O and R’= H or CH3).
ACS Paragon Plus Environment
4

Page 5 of 30
ACS Combinatorial Science
1
2
3
4
5
6
7
8
9
Oxime ligands can also present the potential to modify titanium and zirconium alkoxides to obtain
complexes, which are more resistant to air and moisture17. To our knowledge, the use of these
ligands combined with titanium or zirconium alkoxides to provide M-O-N bonds has not been
described for the cross-linking of SMPs in adhesives and sealants. Thereby, a library of N-tripodal
dioxime ligands was established from a previous work by our team to form copper-, nickel- and
cobalt-based catalysts with dioxime ligands for the copolymerization of ethylene/-olefins18.
Several mono-oxime ligands were added to this library (Fig. 2) to compare the effects of di- or
mono-oximes and their roles in coordination with alkoxides to cross-link SMPs (Fig. 3)19.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
N
N OH
N
N OH
N
N OH
N
N
N
1
2
3
OH
OH
OH
N
N OH
5
N
N OH
N
O
N
OH
N
OH
N
4
6
OH
OH
N
8
OH
N
OH
N
N
7
OH
9
10
OH
H
N
OH
N
OH
N
OH
11
13
12
Figure 2. Library of dioxime and mono-oxime ligands.
ACS Paragon Plus Environment
5

ACS Combinatorial Science
Page 6 of 30
1
2
3
4
5
6
7
8
9
H
R'
OH
N
R
O
N
O
N
N
N OH
N
N
N OH
M
M
R'
M
M
M
= without
R'
R'
R'
R'
R'
N
heteroatom
OH
= with
heteroatom
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 M
R
N
N OH
OH
OH
N
R'
N
N
M
N O M
R'
M
R'
R'
Figure 3. Possible mode of coordination by using simple mono-oxime (a), N-
tripodal mono-oximes (b) and N-tripodal dioximes (c).
It has already been established that high-throughput screening can play an important role in
discovering new efficient catalysts20. Therefore, we report here on using our HTS method, which
was previously described to detect efficient catalysts for the fast cross-linking of SMPs21. This
method is based on a multiwell plate containing SMPs and with 96 different catalysts (one per
well, Fig. 4a). Then, for a given time, the surface of each well was touched with a 24 multichannel
tip system to visualize the physical state (viscous or skin effect, Fig. 4b). Finally, the multichannel
tip system is applied to paper sheet to evaluate the cross-linking time. Thus, when the surface is
viscous, there is a drop on the tip system, meaning that the catalyst is inefficient (no cross-linking).
When a skin effect occurs, there is no drop on the tip system, meaning that the catalyst is efficient
(Fig. 4c). This method is applied here to screen a large number of alkoxide/oxime systems over
different periods of time and to quickly identify new and efficient catalysts able to replace
organotin compounds.
ACS Paragon Plus Environment
6

Page 7 of 30
ACS Combinatorial Science
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
Figure 4. Scheme of HTS method representing a multiwell plate containing SMP and a
maximum of 96 catalysts (a), the 24 multichannel tip system (b) and an example of the
results obtained after the screening of 24 catalysts at a given time (c).
ACS Paragon Plus Environment
7

ACS Combinatorial Science
Page 8 of 30
1
2
3
4
EXPERIMENTAL SECTION
5
6
General data. Commercially available reagents were used without further purification, and
7
8
solvents were dried by standard procedures. NMR spectra were recorded using a Bruker 300 MHz
9
1
13
WB proton Advance I 300 instrument operating at 300 MHz for H spectra and 50 MHz for C
spectra. Mass spectrometry analyses were performed using a JEOL JMS-T100CS (AccuTOF CS)
orthogonal time-of-flight (TOF) mass spectrometer (Peabody, MA) equipped with an IonSense
DART Source (Danvers, MA). Elemental analyses were performed with a microanalysor Flash
EA1112 instrument.
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
Adhesive materials. The silyl-modified polymer used in this study is named SMP1 and was sent
by BOSTIK. This polymer is a dimethoxysilane-terminated polyether (Fig. 5). As a reminder, rates
of RTV cure as follow RCOO-Si > R-C=N-O-Si > RO-Si meaning that the reactivity of the
alkoxysilanes used in the current study is low.
CH3
CH3
Polyether
H3CO Si CH2 HNCO
OCNH CH2 Si OCH3
OCH3
O
O
OCH3
Figure 5. Structure of SMP1.
Chemical products. Hydroxylamine hydrochloride (NH2OH.HCl, 99%), Chloroacetone (95%),
Potassium carbonate (K2CO3, ≥99%), Sodium hydroxide (pellets, ≥97%), Triethylamine (TEA,
≥99%), Cyclohexylamine (99%), Benzylamine (99%), Isobutylamine (99%), Dibenzylamine
(97%), Furfurylamine (≥99%), Amylamine (≥99%), Cyclohexanone oxime (97%), Acetone oxime
(98%), 2-Butanone oxime (99%), Metal-based catalysts: Dibutyltin dilaurate (DBTDL, 95%wt),
Zirconium(IV) propoxide (Zr(IV)nOPr, 70%wt), Zirconium(IV) isopropoxide (Zr(IV)iOPr,
99,9%wt), Zirconium(IV) butoxide (Zr(IV)nOBu, 80%wt), Titanium(IV) propoxide (Ti(IV)nOPr,
98%wt), Titanium(IV) isopropoxide (Ti(IV)iOPr, 97%wt), Titanium(IV) butoxide (Ti(IV)nOBu,
ACS Paragon Plus Environment
8

Page 9 of 30
ACS Combinatorial Science
1
2
3
4
5
6
7
8
9
97%wt), Solvents: Tetrahydrofuran (≥99%), Ethyl ether (≥99%), Acetonitrile (≥99,5%) and Ethyl
acetate (≥99,5%) were purchased from Sigma Aldrich and used without further purification.
Synthesis of Triethylamine Adduct of Chloroacetone Oxime (TACO reagent)22.
Hydroxylamine hydrochloride (182 mmol, 12.65 g) is dissolved in distilled water (80 mL) and a
solution of chloroacetone (165.4 mmol, 13.63 mL) in ethyl ether (200 mL) is added. The mixture
is cooled to 0°C and potassium carbonate (91 mmol, 12.57 g) is slowly added under stirring
(magnetic stir bar), leading to a gas release. After 2 hours, the layers are separated and the aqueous
phase is extracted with ethyl ether (60 mL). Triethylamine (172 mmol, 24 mL) diluted in
acetonitrile (60 mL) is added dropwise onto the organic phase and the mixture is stirred for 30
minutes to obtain a white precipitate. The solid is filtered and washed with cold acetonitrile (120
mL). A second precipitation can occur in the filtrate, which is in that case filtered and the solid is
also washed with cold acetonitrile (3x40 mL). The combined solids are dried under vacuum during
1 night to provide TACO (34.42 g, 99,7%) as a white powder. 1H NMR (300 MHz, DMSO-d6) δ:
11.81 (s, 1H), 4.05 (s, 2H), 3.29 (m, 6H), 1.95 (s, 3H), 1.24 (t, J=6.7Hz, 9H); 13C NMR (50 MHz,
DMSO-d6) δ: 147.58, 59.43, 53.31, 15.76, 7.86. Anal. Calcd C, 51.79; H, 10.14; N, 13.42. Found:
C, 52.05; H, 10.23; N, 13.71.
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
General procedure for the preparation of oxime ligands 1 to 623. Commercially available
primary or secondary amine (6 mmol) is added into a flask. TACO reagent (from 6.6 to 13.2
mmol) and acetonitrile (80 mL) are successively added and the mixture is stirred (magnetic stir
bar), at 85°C during 6 hours. After cooling, acetonitrile is evaporated and ethyl acetate is added
(30 mL) to precipitate residual TACO and formed ammonium salt. The mixture is filtered and
washed with ethyl acetate (2 x 14 mL). The combined organic layers are evaporated to provide the
oxime ligand 1 to 6.
ACS Paragon Plus Environment
9

ACS Combinatorial Science
Page 10 of 30
1
2
3
4
5
6
7
8
9
Ligand 1: 1-[Cyclohexyl-(2-hydroxyimino-propyl)-amino]-propan-2-one oxime (91%); 1H NMR
(300 MHz, CDCl3) δ: 8.97 (s, 2H), 3.14 (s, 4H), 2.45 (m, 1H), 1.89 (s, 6H), 1.81 (m, 4H), 1.62 (m,
1H), 1.28-1.19 (m, 5H); 13C NMR (50 MHz, CDCl3) δ: 157.51, 60.02, 54.19, 28.59, 26.21, 26.02,
12.11. MS found for [M+H+] = 242.18133.Anal. Calcd C, 59.72; H, 9.61; N, 17.41. Found: C,
59.32; H, 9.67; N, 17.15.
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
1
Ligand 2: 1-[Benzyl-(2-hydroxyimino-propyl)-amino]-propan-2-one oxime (80%); H NMR
(300 MHz, CDCl3) δ: 9.16 (s, 2H), 7.34-7.27 (m, 5H), 3.53 (s, 2H), 3.06 (s, 4H), 1.92 (s, 6H); 13C
NMR (50 MHz, CDCl3) δ: 156.89, 129.04, 128.35, 127.24, 58.50, 57.68, 12.45. MS found for
[M+H+] = 250.16327. Anal. Calcd C, 62.63; H, 7.68; N, 16.85. Found: C, 62.34; H, 7.62; N, 17.09.
1
Ligand 3: 1-[isobutyl-(2-hydroxyimino-propyl)-amino]-propan-2-one oxime (84%); H NMR
(300 MHz, CDCl3) δ: 8,28 (s,2H), 3.14-3.03 (m, 4H), 2.17 (m, 2H), 2.05 (m,1H), 1.93 (s, 6H),
0.89 (m, 6H); 13C NMR (50 MHz, CDCl3) δ: 157.17, 62.92, 58.73, 26.22, 20.68, 12.44. MS found
for [M+H+] = 216.15743. Anal. Calcd C, 55.79; H, 9.83; N, 19.52. Found: C, 55.57; H, 9.75; N,
19.69.
1
Ligand 4: 1-[Dibenzyl-(2-hydroxyimino-propyl)-amino]-propan-2-one oxime (92%); H NMR
(300 MHz, CDCl3) δ: 8.53 (s, 1H), 7.41-7.27 (m, 10H), 3.60 (m, 4H), 3.10 (s, 2H), 1.95 (s, 3H);
13C NMR (50 MHz, CDCl3) δ: 139.02, 128.92, 128.34, 127.16, 58.11, 57.54, 12.31. MS found for
[M+H+] = 269.17766. Anal. Calcd C, 76.09; H, 7.51; N, 10.44. Found: C, 75.53; H, 7.57; N,
10.55.
1
Ligand 5: 1-[Furanyl-(2-hydroxyimino-propyl)-amino]-propan-2-one oxime (98%); H NMR
(300 MHz, CDCl3) δ: 9.09 (s, 2H), 7.79-7.39 (m, 3H), 3.58 (m, 2H), 3.12 (s, 4H), 1.92 (s, 6H); 13C
NMR (50 MHz, CDCl3) δ: 154.31, 152.17, 142.9, 110.77, 109.41, 57.36, 49.21, 12.47. MS found
ACS Paragon Plus Environment
10

Page 11 of 30
ACS Combinatorial Science
1
2
3
4
5
for [M+H+] = 240.14754. Anal. Calcd C, 55.22; H, 7.16; N, 17.56. Found: C, 55.07; H, 7.15; N,
17.29.
6
7
8
9
1
Ligand 6: 1-[amyl-(2-hydroxyimino-propyl)-amino]-propan-2-one oxime (89%); H NMR
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
(300 MHz, CDCl3) δ: 8.07 (s, 2H), 3.96 (s, 1H), 3.64 (m, 8H), 3.12 (s, 4H), 2.03 (s,2H), 1.93 (s,
6H); 13C NMR (50 MHz, CDCl3) δ: 130.48, 129.32, 129.07, 128.95, 128.42, 57.99, 48.98, 12.36.
MS found for [M+H+] = 230.18952. Anal. Calcd C, 57.61; H, 10.11; N, 18.32. Found: C, 55.46;
H, 10.05; N, 17.37.
General procedure for the preparation of oxime ligands with ketone 10 to 1324: Sodium
hydroxide (100 mmol, 4 g) is dissolved in distilled water (10 mL) and a solution of ketone (80
mmol) is added. The mixture is cooled to 0°C and a solution of hydroxylamine hydrochloride (100
mmol, 6.95 g) in distilled water (12 mL) is slowly added under stirring (magnetic stir bar). After
1 night, the layers are separated and the organic phase is washed with distilled water (2x20 mL).
Water residues are evaporated and the organic layer is dried under vacuum during 1 night to
provide ketoxime ligand 10 to 13.
Ligand 10: Methylisobutylketoxime (96%); 1H NMR (300 MHz, CDCl3) δ: 8.68 (s, 1H), 2.08 (s,
2H), 1.95-1.88 (m, 4H), 0.94 (s, 3H), 0.92 (s, 3H); 13C NMR (50 MHz, CDCl3) δ: 158.11, 44.70,
25.83, 22.34, 13.54. MS found for [M + H]+ 116.1012. Anal. Calcd C, 62.57; H, 11.38; N, 12.16.
Found: C, 61.88; H, 11.19; N, 11.78.
Ligand 11: 2-Pentanone oxime (94%); 1H NMR (300 MHz, CDCl3) δ: 8.87 (s, 1H), 2.37 (m, 2H),
13
1.90 (s, 3H), 1.57 (m, 2H), 0.94 (m, 3H); C NMR (50 MHz, CDCl3) δ: 158.58, 37.71, 19.76,
19.55, 13.61. MS found for [M + H]+ 102.0589. Anal. Calcd C, 59.37; H, 10.96; N, 13.85. Found:
C, 58.39; H, 10.41; N, 13.36.
ACS Paragon Plus Environment
11

ACS Combinatorial Science
Page 12 of 30
1
2
3
4
5
6
7
8
9
Ligand 12: Cyclopentanone oxime (85%); 1H NMR (300 MHz, CDCl3) δ: 7.18 (s, 1H), 2.48 (m,
2H), 2.38 (m, 2H), 1.77 (m, 4H); 13C NMR (50 MHz, CDCl3) δ: 167.56, 30.85, 27.16, 25.18, 24.56.
MS found for [M + H]+ 100.0408. Anal. Calcd C, 60.58; H, 9.15; N, 14.13. Found: C, 59.86; H,
8.99; N, 13.80.
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
Ligand 13: Salicylaldoxime (91%); 1H NMR (300 MHz, CDCl3) δ: 9.74 (s, 1H), 8.17 (s, 1H), 7.21
(dd, J=7.83Hz, 2H), 6.95 (dt, J=7.46Hz, 2H), 4.94 (s, 1H); 13C NMR (50 MHz, CDCl3) δ: 158.07,
152.96, 131.30, 130.78, 119.87, 116.71, 116.47. MS found for [M + H]+ 138.0738. Anal. Calcd C,
61.31; H, 5.14; N, 10.21. Found: C, 60.75; H, 4.99; N, 10.11.
Preparation of the library containing complexes M-OR/dioximes: 1 mL tubes (from VWR) are
placed in a 96 well plate in glove box protected from oxygen and moisture. Metal precursors
(2,25.10-5 mol) are disposed as follows: columns 2 (Zr(nOPr)4, 10,5 mg, 10,1 µL), columns 3
(Zr(iOPr)4, 8,7 mg), columns 4 (Zr(nOBu)4, 10,8 mg, 10,3 µL), columns 5 (Ti(nOPr)4, 6,5 mg, 6,3
µL), columns 6 (Ti(iOPr)4, 6,5 mg, 6,8 µL) and columns 7 (Ti(nOBu)4, 7,9 mg, 7,9 µL). Then, the
dioxime ligands 1, 2, 3, 5 and 6 (2,25.10-5 (1:1) to 4,5.10-5 mol (1:2)) are added in rows B to F
respectively. Each tube in each column is charged with THF (50 L) and stirred with a magnetic
stir bar during 4 hours.
Preparation of the library containing complexes M-OR/mono-oximes: 1 mL tubes (from
VWR) are placed in a 96 well plate in glove box protected from oxygen and moisture. Metal
precursors (2,25.10-5 mol) are disposed as follows: columns 2 (Zr(nOPr)4, 10,5 mg, 10,1 µL),
columns 3 (Zr(iOPr)4, 8,7 mg), columns 4 (Zr(nOBu)4, 10,8 mg, 10,3 µL), columns 5 (Ti(nOPr)4,
6,5 mg, 6,3 µL), columns 6 (Ti(iOPr)4, 6,5 mg, 6,8 µL) and columns 7 (Ti(nOBu)4, 7,9 mg, 7,9
µL). Then, the mono-oxime ligands 4, 7, 8, 9, 10, 11, 12 and 13 (4,5.10-5 (1:2) to 9.10-5 mol (1:4))
ACS Paragon Plus Environment
12

Page 13 of 30
ACS Combinatorial Science
1
2
3
4
5
6
are added in rows B to I respectively. Each tube in each column is charged with THF (50 L) and
stirred with a magnetic stir bar during 4 hours.
7
8
9
General procedure for parallel HTS method: SMP1 (55 g) was introduced in a large plastic
vessel (185*95*15 mm). Another 96 well plate B was slowly disposed on the plastic tank to allow
the viscous liquid to fill to 96 wells by capillarity. By this way, 96 identical samples of prepolymer
are generated without any transfer by multichannel pipette or individual weighting. Each reactor
has an 8 mm diameter and 8 mm height. Catalytic systems were then transferred from the plate A
into the 96 wells of plate B with a multichannel pipette to form a film at the surface of each reactor.
Then, the plate was removed of glove box and was disposed into laboratory in contact with ambient
moisture. The cross-linking time was followed every 5 minutes during 1 hour at room temperature
(classical conditions in the laboratory are 55%H2O and 23°C).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
General procedure for the cross-linking of SMP1 in bulk: Catalysts (4,5.10-4 mol) and THF
(100 L) were added in 1 mL tubes (from VWR). Mixtures were stirred with a magnetic stir bar
during 1 hour at room temperature. SMP1 prepolymer (10 g) was introduced in a plastic cup
(diameter 50mm and height 30mm). Catalytic systems were introduced in the plastic cup and
stirred with SMP1 during 1 minute. Cross-linking time was evaluated by touching the surface
every 5 minutes during 1 hour, then every 30 minutes until 8 hours and every day until 1 week.
General procedure for the cross-linking of sealant containing SMP1 in bulk: Catalysts (4,5.10-
4 mol) and THF (100 L) were added in 1 mL tubes (from VWR). Mixtures were stirred with a
magnetic stir bar during 1 h at room temperature. SMP1 prepolymer (5 g) and Calcium carbonate
(Calofort SV, 5 g) were introduced in a plastic cup (diameter 50mm and height 30mm) and stirred
with a mechanical stirrer during 10 minutes. Catalytic systems were introduced in the plastic cup
and stirred with prepolymer during 1 minute. Cross-linking time was evaluated by touching the
ACS Paragon Plus Environment
13

ACS Combinatorial Science
Page 14 of 30
1
2
3
4
5
6
surface every 5 minutes during 1 hour, then every 30 minutes until 8 hours and every day until 1
week.
7
8
9
General procedure to determine stability of the sealant: same procedure that for the cross-
linking of sealant containing SMP1 but this mixture is introduced in a cartridge and put in an oven
at 40°C during 1 month. After 1 month, the cartridge was removed and the cross-linking time was
evaluated by classical method in room condition (classical conditions in the laboratory are
55%H2O and 23°C).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
14

Page 15 of 30
ACS Combinatorial Science
1
2
3
4
RESULTS AND DISCUSSION
5
6
Screening of alkoxide/dioxime systems for SMP1
7
8
9
Zirconium- and titanium-based alkoxides are unstable in the presence of moisture. To improve
their behavior during storage, some chelating agents, such as acetylacetone can be used. Buyuktas
et al.25 reported on the synthesis and stability of systems that provided a reaction between
zirconium or titanium alkoxides with dimethylglyoxime. In this work, only two molecules of
dioxime can participate in the coordination of one molecule of Ti(nOBu)4 or Zr(nOBu)4 to form M-
O-N bonds. In our case, several oximes were evaluated instead of the classical chelating agents to
test their potential for cross-linking SMPs. Thus, the curing times of alkoxide/dioxime systems at
two different M/L ratios (1/1 and 1/2) were measured by the HTS method. For the results obtained
with the 1/1 ratio (Fig. 6a), all the systems can be organized into four categories (i.e., fast, medium
and slow curing times or inefficient after 1 hour). According to this classification, it is logical to
expect that fast systems are better than the others. In fact, we looked for new ligands able to
stabilize alkoxides, meaning the better hits are systems able to modify the curing time of alkoxides
alone. Indeed, interesting hits will be alkoxide/oxime couples capable of stabilizing alkoxides
while maintaining a relatively good curing time. Therefore, plots including the curing time (CT)
were transformed into new ones with a ∆ parameter (Fig. 6b) to identify more interesting hits. The
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
∆ parameter is described by the equation:
∆ = CT(alkoxide/oxime system) – CT(corresponding alkoxide)
When ∆ is equal to 0, there is no modification of the curing time between a new system and
alkoxides used alone (shown in red). At this stage, there are two hypotheses: a) there is no reaction
between the ligand and the alkoxide, meaning the alkoxide reacts by itself, or b) there is a
complexation that is not able to control the reaction. In contrast, when ∆ is greater than 0, this
ACS Paragon Plus Environment
15

ACS Combinatorial Science
Page 16 of 30
1
2
3
4
5
6
means that alkoxide/oxime systems (shown in blue) are able to control the curing time of the initial
alkoxide used.
7
8
9
 The use of the dioxime alone (without metal) as catalyst does not allow the cross-linking
of SMP1 after 1 hour (column 1, row B to F).
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
 All akoxides are quite active (column 2 to 7, row A), but it is well known that they are
unstable in the presence of moisture. The curing time of many systems is equal to the curing
time of the corresponding alkoxide (Zr(nOPr)4 in column 2, Zr(iOPr)4 in column 3 and
Ti(iOPr)4 in column 6).
 Using an M/L ratio equal to 1/1, only 4 systems among the 36 have a ∆ value different
from 0, meaning there are only 4 interesting hits for our application: Zr(nOBu)4 + 2,
Ti(nOPr)4 + 2, Ti(nOBu)4 + 2 and Ti(nOBu)4 + 5. We also report the minimum differences
in curing times between the system and alkoxide alone for these interesting hits, here 30
minutes.
Figure 6. The curing times of alkoxide/dioxime systems with an M/L ratio of 1/1 obtained by
HTS method (a) and the transformation of curing times in ∆ parameter for each systems (b)
ACS Paragon Plus Environment
16

Page 17 of 30
ACS Combinatorial Science
1
2
3
4
5
6
7
8
9
HTS was used also to investigate alkoxide/dioxime systems with an M/L ratio equal to 1/2 to
observe the impact of ligands. Similarly, the curing times were also transformed into ∆ values (Fig.
7).
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
 Using an M/L ratio equal to 1/2, the curing time of 6 systems increased (higher than 1
hour), making the determination of the ∆ value impossible (cross in the graph). This means
that the new zirconium- and titanium-based complexes are inactive for Zr(nOBu)4 + 2, 5
and 6, Ti(nOPr)4 + 2 and Ti(iOPr)4 + 2 and 5. The creation of catalytic species is due to the
reaction of these systems with moisture to form a hydroxyalkoxide as well as for the tin
catalyst. It is suggested that the hindrance of these 6 systems is more important when the
amount of oxime increases and involves a lower reactivity with moisture, leading to an
excessive curing time.
x
x
x
x
x
x
Figure 7. The curing times of alkoxide/dioxime systems with an M/L ratio of 1/2
transformed into ∆ parameter
ACS Paragon Plus Environment
17

ACS Combinatorial Science
Page 18 of 30
1
2
3
4
5
6
7
8
9
 The curing time for some systems containing Zr(iOPr)4 or Ti(iOPr)4 decreased using an
M/L ratio equal to 1/2, whereas there was no modification of curing times at 1/1.
 Considering only efficient systems with a 1/2 ratio, such as Zr(iOPr)4 + 1, Zr(iOPr)4 + 2,
Zr(iOPr)4 + 5, Ti(iOPr)4 + 6 and Zr(nOPr)4 + 3, the curing time increases. The efficiency of
Ti(nOBu)4 + 5 is the same at 1/1 and 1/2. Therefore, this system will be further studied only
at a 1/1 ratio, which is enough to increase the stability and maintain a relatively good curing
time.
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
 There are no interesting systems containing Zr(nOBu)4 at a 1/2 ratio, as all of these systems,
which have different curing time than with alkoxides alone, are inefficient after 1 hour of
reaction.
 Ti(nOPr)4 + 3 and Ti(nOPr)4 + 5 are interesting only at the 1/2 ratio.
 The reactivity of the Ti(iOPr)4 + 6, Ti(nOBu)4 + 1 and Ti(nOBu)4 + 2 systems decreases
compared to alkoxides alone using a 1/2 ratio.
Thus, all of these systems are studied in a second step to determine their real cross-linking time.
In summary, among the 72 alkoxide/dioxime systems tested by the HTS method, only 12 were
detected as promising hits. Indeed, they induce the cross-linking of SMP1 by generating a steric
hindrance around the alkoxide to control the hydrolysis rate. Thereby, it is logical to hope that
these systems will be more stable in cartridges than alkoxides alone. Despite the good results
obtained with Zr(iOPr)4 + 1, Zr(iOPr)4 + 2, Zr(iOPr)4 + 5 and Ti(iOPr)4 + 6, it is not possible to go
further to determine their real cross-linking times, as Zr(iOPr)4 and Ti(iOPr)4 are not available from
industrial suppliers with stable compositions between the different samples. Therefore, only 8
alkoxide/dioxime systems will be studied.
ACS Paragon Plus Environment
18

Page 19 of 30
ACS Combinatorial Science
1
2
3
4
Screening of alkoxide/mono-oxime systems for SMP1
5
6
7
8
The reaction between mono-oximes and alkoxides consists of the nucleophilic substitution of
alcohol groups to form metallic oximate complexes26. Theoretically, one to four molecules of
mono-oximes can react with one molecule of titanium- or zirconium-based alkoxides. Therefore,
the HTS method was used and the corresponding curing times were transformed into ∆ values for
two M/L ratios (1/2 and 1/4). For a 1/2 ratio (Fig. 8) for the alkoxide/dioxime systems, we look to
detect the mono-oximes able to modify the curing time of alkoxides alone. Surprisingly, in several
cases, the ∆ parameter is lower than 0, meaning that the alkoxide/mono-oxime systems (shown in
yellow) decrease the curing time of the initial alkoxide used and accelerate their reactivity. In fact,
by using mono-oximes, several systems are found to have faster curing times than alkoxides and
may be a potential option to replace organotin compounds. For example, five systems, namely,
Zr(nOPr)4 + 4, Zr(nOPr)4 + 7, Zr(nOPr)4 + 8, Zr(nOPr)4 + 9 and Zr(nOPr)4 + 11 are more efficient
than Zr(nOPr)4 alone.
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
x
x
x
x
x
Figure 8. The curing times of alkoxide/mono-oxime systems with an M/L ratio of 1/2
transformed into ∆ parameter
ACS Paragon Plus Environment
19

ACS Combinatorial Science
Page 20 of 30
1
2
3
4
5
6
7
8
9
Furthermore, the systems provided by the combination of salicylaldoxime 13 and Zr(nOPr)4,
Zr(nOBu)4, Ti(nOPr)4 and Ti(nOBu)4 are inefficient after 1 hour of reaction. This behavior may be
explained by Buyukthas et al., indeed, the structures of the complexes formed by the reaction
between the salicylaldoxime 13 and acetone oxime 9 with Ti(nOBu)4 are completely different. In
the first case, 13 can be coordinated on titanium by the oxygen of the phenol group, but also by
the nitrogen of the oxime group (Fig. 9a). In the second case, ligand 9 involves the nucleophilic
substitution of alcohol groups to form Ti-O-N bonds (Fig. 9b). Infrared analyses were performed
on these complexes, indicating that there is a shift of N-O stretching to a higher frequency from 9
to Ti(nOBu)2(9)2, which can be explained by the direct interaction between titanium and oxygen
atoms, shifting the electron density from nitrogen to oxygen. In our case, this effect has a great
impact on the catalysis of SMP, as the titanium becomes more sensitive to moisture and thus
creates more easily the active species. Therefore, systems with 13 which does not contain the M-
O-N bonds, are less efficient than alkoxides alone, whereas systems with other mono-oximes are
more efficient than alkoxides alone or at least have the same efficiency. However, it is worth
nothing that a longer curing time (shown in green) can be obtained, such as for Ti(nOPr)4 + 9,
Ti(nOBu)4 + 9 and Ti(nOBu)4 + 12. These three exceptions are probably due to the structure of the
complexes formed, which can be dimer, trimer or others27. Indeed, the formation of complexes is
difficult to predict, as it depends on the alkoxide/oxime couple used, but also on the solvent,
temperature and moisture rate. It appears that the formation of M-O-N bonds is essential to a fast
curing time. Finally, this screening allows for the detection of 16 systems for an M/L ratio of 1/2,
that have better curing times than when alkoxides are used alone.
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
20

Page 21 of 30
ACS Combinatorial Science
1
2
3
4
5
6
7
8
9
O
H
(a)
(b)
(BuO)3 Ti O N
OBu
O
N
O
Ti
OBu
H O
N
N O Ti (OBu)3
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 9. Possible structures of (a) Ti(nOBu)2(13)2 and (b) Ti(nOBu)3(9)
The HTS method was also used to screen alkoxide/mono-oximes systems with an M/L ratio equal
to 1/4 (Fig. 10) to compare the substitution effect for the cross-linking of SMP1. However, there
were no differences between the 1/2 and 1/4 ratios excepted for Ti(nOBu)4 + 12, which becomes
inefficient after 1 hour at a M/L ratio of 1/4. In fact, this may be due to some differences between
interesting hits (shown in yellow), but our HTS method does not allow detection of the differences
between several systems with curing times between 5 and 10 minutes.
x
x
x
x
x
x
Figure 10. The curing times of alkoxide/mono-oxime systems with an M/L ratio of 1/4
transformed into ∆ parameter
ACS Paragon Plus Environment
21

ACS Combinatorial Science
Page 22 of 30
1
2
3
4
5
6
7
8
9
In summary, among the 96 alkoxide/mono-oxime systems tested using the HTS method, 32 were
detected as faster than alkoxides alone. More specifically, 16 systems are very interesting at an
M/L ratio of 1/2 but also at a ratio of 1/4. All these hits will be further studied to measure their real
cross-linking time and check if the same order of curing time is obtained under standard conditions.
Cross-linking time of interesting hits for SMP1
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 HTS method allowed the detection of several interesting hits based on alkoxide/oxime systems
(i.e., 8 systems based on dioximes and 32 based on mono-oximes). The second step is to measure
the real cross-linking times compared to DBTDL and zirconium- or titanium-based alkoxides.
Using the same conditions (temperature, humidity and molar ratio of the catalytic system), the
reactions were performed in bulk conditions for 10 g of SMP1 prepolymer and fast mixing of the
mixture. Then, the cross-linking times were determined by touching the surface with a needle
several times (standard method) by using alkoxide/dioxime systems (Table 1) and alkoxide/mono-
oxime systems (Table 2).
Table 1 Cross-linking times of efficient systems based on alkoxide/dioxime by standard method
at room temperature on SMP1.
Cross-linking time
Catalysts
M/L ratio
(min)
150
30
DBTDL
-
Zr(nOPr)4
-
Zr(nOBu)4
-
90
Ti(nOPr)4
-
30
Ti(nOBu)4
-
60
Zr(nOPr)4 + 3
Zr(nOBu)4 + 2
Ti(nOPr)4 + 2
Ti(nOPr)4 + 3
Ti(nOPr)4 + 5
Ti(nOBu)4 + 5
Ti(nOBu)4 + 1
Ti(nOBu)4 + 2
1/2
1/1
1/1
1/2
1/2
1/1
1/2
1/1
180
480
360
480
1440
480
300
120
The cross-linking times of zirconium- and titanium-based alkoxides are faster than the tin catalyst
(DBTDL, from 30 to 90 minutes compared to 150 minutes with DBTDL). The use of dioxime
ACS Paragon Plus Environment
22

Page 23 of 30
ACS Combinatorial Science
1
2
3
4
5
6
7
8
9
ligands increases the cross-linking times for zirconium- and titanium-based systems, allowing a
longer open time, which is important for a specific target where the adhesive has to be
repositionable. The combinations of Zr(nOPr)4 + 3 and Ti(nOBu)4 + 2 are very close to DBTDL
and may be interesting due to their potential high stability against moisture in cartridges compared
to alkoxides alone.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 2 Cross-linking times of efficient systems based on alkoxide/mono-oxime by standard
method at room temperature on SMP1.
Cross-linking time
Catalysts
M/L ratio
(min)
20
10
15
45
10
25
10
10
15
15
10
10
20
20
10
15
20
20
10
15
40
30
15
10
20
30
20
15
15
15
15
15
Zr(nOPr)4 + 4
Zr(nOPr)4 + 7
Zr(nOPr)4 + 8
Zr(nOPr)4 + 9
Zr(nOPr)4 + 11
Zr(nOPr)4 + 4
Zr(nOPr)4 + 7
Zr(nOPr)4 + 8
Zr(nOPr)4 + 9
Zr(nOPr)4 + 11
Zr(nOBu)4 + 7
Zr(nOBu)4 + 8
Zr(nOBu)4 + 9
Zr(nOBu)4 + 10
Zr(nOBu)4 + 7
Zr(nOBu)4 + 8
Zr(nOBu)4 + 9
Zr(nOBu)4 + 10
Ti(nOPr)4 + 4
Ti(nOPr)4 + 7
Ti(nOPr)4 + 10
Ti(nOPr)4 + 11
Ti(nOPr)4 + 4
Ti(nOPr)4 + 7
Ti(nOPr)4 + 10
Ti(nOPr)4 + 11
Ti(nOBu)4 + 7
Ti(nOBu)4 + 10
Ti(nOBu)4 + 11
Ti(nOBu)4 + 7
Ti(nOBu)4 + 10
Ti(nOBu)4 + 11
1/2
1/2
1/2
1/2
1/2
1/4
1/4
1/4
1/4
1/4
1/2
1/2
1/2
1/2
1/4
1/4
1/4
1/4
1/2
1/2
1/2
1/2
1/4
1/4
1/4
1/4
1/2
1/2
1/2
1/4
1/4
1/4
Contrary to dioxime, the use of mono-oxime allows a shorter cross-linking time in all cases.
According to the systems used, the cross-linking times vary from 10 to 45 minutes compared to
ACS Paragon Plus Environment
23

ACS Combinatorial Science
Page 24 of 30
1
2
3
4
5
6
7
8
9
150 minutes with DBTDL. Moreover, in most cases, the alkoxide/mono-oxime systems are better
than zirconium or titanium alkoxides alone. The M/L ratio (1/2 or 1/4) slightly altered the cross-
linking, meaning that the complexes formed are similar or that the cross-linking times are too fast
to observe any differences. In fact, there are only two systems with shorter cross-linking times
when the ratio moves from 1/2 to 1/4. Indeed, the cross-linking time moves from 45 to 15 minutes
for Zr(nOPr)4 + 9 and from 40 to 20 minutes for Ti(nOPr)4 + 10. In these two cases, the structure
of the complex formed has to be different and likely contains more M-O-N bonds, increasing the
electrophilic characteristic of the metal atoms, which should be more reactive against a
nucleophilic attack by moisture. Finally, many alkoxide/mono-oxime systems may be interesting
when a fast cross-linking time is required, particularly for grab adhesives.
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
Cross-linking time of some efficient catalysts for SMP1-based sealants
The best catalysts were used to evaluate the effect of storage on the cross-linking time for SMP1-
based sealants. We studied the following two alkoxide/dioxime systems: Zr(nOPr)4 + 3 and
Ti(nOBu)4 + 2, and only 10 alkoxide/mono-oxime systems were used to determine an initial idea
of the stability. Thus, the cross-linking times for fresh product and after storage were determined
(Table 3). The employment of zirconium and titanium alkoxides alone induces a longer cross-
linking time after storage due to their high moisture-sensitive behavior involving condensation
into themselves. The cross-linking times of the two alkoxide/dioxime systems slightly increased
after the storage: 75 to 90 and 45 to 55 minutes, respectively. This is also the case for the Zr(nOPr)4
+ 4 system, which is composed of a mono-oxime as the ligand. These three catalysts possess N-
tripodal oxime as the ligand (2, 3 and 4). It is suggested that this hindered the process and that a
stable structure formed during storage, likely due to the interaction of the central nitrogen atom
with other metals or by a reaction between a free oxime group and another metal. Then, the
ACS Paragon Plus Environment
24

Page 25 of 30
ACS Combinatorial Science
1
2
3
4
5
6
7
8
9
Ti(nOPr)4 + 7 system induces instantaneous cross-linking that is difficult to control during mixing.
However, this system is not very useful for adhesive applications, as a minimum setting time is
necessary to generate a homogeneous mixture for bulk reactions or to apply the adhesive material
onto a surface. Lastly, the cross-linking times of the six systems provided by the reaction between
Zr(nOPr)4, Zr(nOBu)4 and Ti(nOBu)4 with ligand 7 or 8 are faster than DBTDL and were not
modified during storage. Therefore, these systems have a high potential for use as the substitute to
the tin catalyst in this sealant.
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 3 Cross-linking times of fresh product and after storage of sealant during 1 month in an
oven at 40°C using standard method. *=longer cross-linking time after storage, **=cross-linked
sealant directly after storage
Cross-linking time Cross-linking time
Catalysts
M/L ratio
fresh (min)
stored (min)
DBTDL
-
120
110
Zr(nOPr)4
-
25
*
Zr(nOBu)4
-
50
*
Ti(nOPr)4
-
20
*
Ti(nOBu)4
-
30
*
Zr(nOPr)4 + 3
Ti(nOBu)4 + 2
Zr(nOPr)4 + 4
Zr(nOPr)4 + 4
Zr(nOPr)4 + 7
Zr(nOBu)4 + 7
Zr(nOPr)4 + 8
Zr(nOBu)4 + 8
Ti(nOBu)4 + 7
Ti(nOBu)4 + 7
Ti(nOPr)4 + 7
Ti(nOPr)4 + 7
1/2
1/1
1/2
1/4
1/2
1/2
1/2
1/2
1/2
1/4
1/2
1/4
75
90
55
20
30
< 5
< 5
< 5
< 5
< 5
< 5
**
**
45
15
20
< 5
< 5
< 5
< 5
< 5
< 5
Instantaneous
Instantaneous
CONCLUSIONS
The use of oxime as a chelating agent instead of acetylacetone allowed us to identify a new library
of catalysts to cross-link silyloxane polymer. An HTS method was used to perform a fast
comparison of 156 alkoxide/oxime-based systems for the cross-linking of SMP. This rapid method
allowed the detection of approximately 40 efficient systems and extraction of the better hits, which
ACS Paragon Plus Environment
25

ACS Combinatorial Science
Page 26 of 30
1
2
3
4
5
6
7
8
9
were analyzed in a second step to determine their cross-linking times with the SMP1 polymer. The
study of alkoxide/dioxime systems allowed the detection of two new systems: Zr(nOPr)4 + 3 and
Ti(nOBu)4 + 2, which are closer to the performance of DBTDL. For alkoxide/mono-oxime
systems, many are better than DBTDL, mainly the six systems operating via the reaction between
Zr(nOPr)4, Zr(nOBu)4 and Ti(nOBu)4 with ligand 7 or 8. These catalysts could be used in two types
of applications: repositionable adhesives to bond two surfaces (dioxime-based systems) and fast-
setting adhesives (mono-oxime based systems). Finally, the high efficiency and stability of
alkoxide/oxime systems were established here and show the great potential of this library as an
alternative to tin catalysts for the cross-linking of an adhesive and a sealant.
ACKNOWLEDGEMENT
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 authors gratefully acknowledge CNRS, ANRT, BOSTIK and TOTAL for their financial
support.
ASSOCIATED CONTENTS
Supporting information. preparation procedures, chemical schemes, NMR spectra of oxime-
based compounds.
ACS Paragon Plus Environment
26

Page 27 of 30
ACS Combinatorial Science
1
2
3
4
REFERENCES
5
6
7
8
9
1
(a) Maudgal, S.; St Clair, TL. Preparation and properties of silane-endcapped polyimide
adhesives. Int. J. Adhes. Adhes. 1984, 4, 129-132. (b) US 8,012,302 B2. Moisture-curing
compositions containing silane-functional polymers and aminosilane adducts with good adhesive
properties. Sika Technology AG 2011.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
(a) Lauriks, L.; Wouters, I.; Belis, J. Compressive and lap shear tests on traditional putty and
polymer sealants. Int. J. Adhes. Adhes. 2016, 64, 109-115. (b) Foster, V.R. Polymers in caulking
and sealant materials. J. Chem. Educ. 1987, 64, 861-865. (c) US 8,501,903 B2. Urea-bonded
alkoxysilanes for use in sealants and adhesives. Henkel AG 2013.
3
(a) Meera, K.M.S.; Sankar, R.M.; Jaisankar, S.N. Physicochemical studies on
polyurethane/siloxane cross-linked films for hydrophobic surfaces by the sol-gel process. J. Phys.
Chem. B 2013, 117, 2682-2694. (b) Alyamac, E.; Gu, H.; Soucek, M.D., Qiu, S.; Buchheit, R.G.
Alkoxysilane oligomer modified epoxide primers. Prog. Org. Coat. 2012, 74, 67-81.
4
Kregiel, D. Advances in biofilm control for food and beverage industry using organo-silane
technology: A review. Food Control 2014, 40, 32-40.
5
Saikia, D.; Wu, C.G.; Fang, J.; Tsai, L.D; Kao, H.M. Organic-inorganic hybrid polymer
electrolytes based on polyether diamine, alkoxysilane, and trichlorotriazine: Synthesis,
characterization, and electrochemical applications. J. of Power Sources 2014, 269, 651-660.
6
Kaddami, H.; Surivet, F.; Gérard, J.F.; Lam, T.M.; Pascault, J.P. Hybrid organic-inorganic
materials synthesized by reaction with alkoxysilanes: Effect of the acid-to-alkoxide ratio on
morphology. J. Inorg. Organomet. Polym. 1994, 4, 183-198.
ACS Paragon Plus Environment
27

ACS Combinatorial Science
Page 28 of 30
1
2
3
4
5
6
7
Ni, H.; Skaja, A.D.; Soucek, M.D. Acid-catalyzed moisture-curing polyurea/polysiloxane
7
8
9
ceramer coatings. Prog. In Organic Coatings 2000, 40, 175-184.
8 Brinker, C.J.; Keefer, K.D.; Schaefer, D.W.; Ashley, C.S. Sol-gel transition in simple silicates.
J. of Non-Cryst. Solids 1982, 48, 47-64.
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
9
US 8,124,690 B2. Moisture curable polymer having SiF group, and curable composition
containing the same. Kaneka Corporation 2012.
10 EP 2813528 A1. Curable composition on the basis of polyethers containing alkoxysilane groups
and a phosphazene base. Sika Technology AG 2014.
11 Burkhalter, R.S.; Hogue, C.L.; Smith, D.L.; Sonner, S.M.; Winningham, M.J.; Yougman, R.E.
Controlled tin catalyzed hydrolysis of 3-acryloxypropyltrimethoxysilane with mono- and multi-
functional mercaptans. J. Organomet. Chem. 2013, 724, 213-224.
12 US 4,956,435 Neutral cure silicone sealants. Dow Corning Corporation 1990.
13
(a) Huber, M.P.; Kelch, S.; Berke, H. FTIR investigations on hydrolysis and condensation
reactions of alkoxysilane terminated polymers for use in adhesives and sealants. Int. J. Adhes.
Adhes. 2016, 64, 153-162. (b) US 9,321,878 B2. Process for the preparation of silylated
polyurethane polymers using titanium-containing and zirconium-containing catalysts. Momentive
Performance Materials 2016.
14 (a) Guglielmi, M.; Carturan, G.; Precursors for sol-gel preparations. J. of Non-Cryst. Solids 1988,
100, 16-30. (b) Sanchez, C.; Livage, J.; Henry, M.; Babonneau, F. Chemical modification of
alkoxide precursors. J. of Non-Cryst. Solids 1988, 100, 65-76. (c) US 9,376,525 B2. Curable
composition and cured synthetic resin using same. MIE University 2015.
15
Mendez-Vivar, J.; Brinker, C.J. Development of Two New Approaches for Preparation of
Zirconium Silicate Systems at Low Temperature. J. Sol-Gel Technol. 1994, 2, 393-397.
ACS Paragon Plus Environment
28

Page 29 of 30
ACS Combinatorial Science
1
2
3
4
5
6
7
8
16 Verma, A.; Kar, M.; Singh, D.P. Aging effect of diethanolamine derived precursor sol on TiO2
films deposited at different annealing temperature. J. Sol-Gel Technol. 2010, 54, 129-138.
17 Mehrotra, R.C.; Rai, A.K.; Singh, A.; Bohra, R. Oxime and hydroxylamine derivatives of metals
and metalloids. Inorganica Chimica Acta 1975, 13, 91-103.
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
18
WO2007/009977 A1. Polymerization catalyst system based on dioxime ligands. Total
Petrochemicals/CNRS 2007.
19 (a) Nicholson, G.A.; Lazarus, C.R.; McCormick, B.J. Copper(II) complexes derived from 2,6-
Diacetylpyridine dioxime. Inorg. Chem. 1980, 19, 192-195. (b) El-Dissouky, A.; Mohamed, G.B.;
El-Sonbati, A.Z. Metal Chelates of heterocyclic nitrogen-containing ketones. XI. Copper (II)
complexes of syn-2-picolyl phenyl and syn-2-picolyl methyl ketoximes. Inorg. Chimica Acta.
1983, 74, 279-284. (c) Holynska, M. Structural variety and magnetic properties of oxime-bridged
copper(II) complexes. J. of Mol. Struct. 2015, 1098, 175-180.
20
(a) Jandeleit, B.; Schaefer, D.J.; Powers, T.S.; Turner, H.W.; Weinberg, W.H. Combinatorial
materials science and catalysis. Angew. Chem. Int. Ed. 1999, 38, 2494-2532. (b) Reetz, M.T.
Combinatorial and evolution-based methods in the creation of enantioselective catalysts. Angew.
Chem. Int. Ed. 2001, 40, 284-310. (c) Garbacia, S.; Touzani, R.; Lavastre, O. Image analysis as a
quantitative screening test in combinatorial catalysis: discovery of an unexpected ruthenium-based
catalyst for the sonogashira reaction. J. Comb. Chem. 2004, 6, 297-300. (d) Lavastre, O.; Bonette,
F.; Gallard, L. Parallel and combinatorial approaches for synthesis of ligands. Current opinion in
Chemical Biology 2004, 8, 311-318.
ACS Paragon Plus Environment
29