Kinetics and Mechanism of the Gold-Catalyzed Intermolecular Hydroalkoxylation of Allenes with Alcohols

Article abstract of DOI:10.1021/acscatal.8b02211

The mechanism of the hydroalkoxylation of 3-methyl-1,2-butadiene (1) with 1-phenylpropan-1-ol (2) catalyzed by (IPr)AuOTf in toluene has been evaluated through a combination of kinetic analysis, deuterium labeling studies, and in situ spectral analysis of catalytically active mixtures. These data are consistent with a mechanism involving endergonic conversion of (IPr)AuOTf and 1 to the cationic gold π-allene complex (IPr)Au[n²-H₂C=C=CMe₂][OTf] (5·OTf), which undergoes outer-sphere addition of 2 followed by rapid protodemetalation to form 1-(3-methyl-2-butenyloxy)propyl)benzene (3a) as the kinetic product. The microscopic rate constants associated with the formation and consumption of 5·OTf are of similar magnitude, such that the kinetic behavior of catalytic hydroalkoxylation changes as a function of the relative and absolute concentrations of allene and alcohol.

Full text of DOI:10.1021/acscatal.8b02211

Subscriber access provided by ST FRANCIS XAVIER UNIV
Article
Kinetics and Mechanism of the Gold-Catalyzed
Intermolecular Hydroalkoxylation of Allenes with Alcohols
Robert J. Harris, Robert G Carden, alethea duncan, and Ross A. Widenhoefer
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.8b02211 • Publication Date (Web): 10 Aug 2018
Downloaded from http://pubs.acs.org on August 10, 2018
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 41
ACS Catalysis
1
2
3
4
5
6
7
8
Kinetics and Mechanism of the GoldꢀCatalyzed Intermolecular
Hydroalkoxylation of Allenes with Alcohols
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Robert J. Harris, Robert G. Carden, Alethea N. Duncan, and Ross A. Widenhoefer*
Duke University, Department of Chemistry, French Family Science Center, Durham, North Carolina,
27708
rwidenho@chem.duke.edu
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
Abstract: The mechanism of the hydroalkoxylation of 3ꢀmethylꢀ1,2ꢀbutadiene (1) with 1ꢀphenylpropanꢀ
1ꢀol (2) catalyzed by (IPr)AuOTf in toluene has been evaluated through a combination of kinetic
analysis, deuterium labeling studies, and in situ spectral analysis of catalytically active mixtures. These
data are consistent with a mechanism involving endergonic conversion of (IPr)AuOTf and 1 to the
cationic gold πꢀallene complex (IPr)Au[η2ꢀH2C=C=CMe2][OTf] (5ꢀOTf), which undergoes outerꢀsphere
addition of 2 followed by rapid protodemetallation to form 1ꢀ(3ꢀmethylꢀ2ꢀbutenyloxy)propyl)benzene (3a)
as the kinetic product. The microscopic rate constants associated with the formation and consumption
of 5ꢀOTf are of similar magnitude, such that the kinetic behavior of catalytic hydroalkoxylation changes
as a function of the relative and absolute concentrations of allene and alcohol.
Keywords: hydrofunctionalization, C−O bond formation, mechanism, kinetics, gold, NHC, allene
1
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 41
1
2
3
4
Introduction
5
6
7
8
Cationic gold(I) complexes have emerged as a highly effective catalysts for the πꢀactivation of
C–C multiple bonds, most notably for the cycloisomerization of 1,nꢀenynes and for the
hydrofunctionalization of C–C multiple bonds with H–X nucleophiles (X = C, N, O).1ꢀ7 Within this latter
family of transformations, allenes have attracted considerable attention as substrates for gold(I)ꢀ
catalyzed hydrofunctionalization owing to the facility of the transformations,4 the potential for
stereospecific5ꢀ7 and enantioselective transformations,3 and the wide range of nucleophiles that
undergo allene hydrofunctionalization including water, alcohols, carbamates, esters, carboxamides,
ureas, thiols, and alkyl and aryl amines, electronꢀrich arenes, and activated methylene compounds.4
Similarly, gold(I)ꢀcatalyzed allene hydrofunctionalization has been applied to the total synthesis of a
number of natural products2 including (–)ꢀRhazinilam,8 flinderoles B and C,9 swainsonine,10 bejarol,11
jaspine B,12 indoxamycin,13 and (–)ꢀfunebrine.14
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
Along with the development of the synthetic aspects of gold(I)ꢀcatalyzed allene
hydrofunctionalization, there has been considerable interest in understanding the mechanisms of these
processes,15 although much of this effort has focused on computational analysis.16,17 Experimental
efforts in this area18 have focused primarily on the synthesis and interrogation of potential intermediates
in gold(I)ꢀcatalyzed allene hydrofunctionalization, including cationic gold(I) πꢀallene complexes,19,20
neutral gold vinyl complexes,6,18,21ꢀ24 and gemꢀdiaurated vinyl complexes,18,22,23 and on the
stereochemical analysis of goldꢀcatalyzed allene hydrofunctionalization.25ꢀ27 These latter studies have
established the net antiꢀaddition of the H–X bond of the nucleophile across a C=C of the allene,
consistent with mechanisms involving outerꢀsphere attack of the nucleophile on a gold πꢀallene
complex.25ꢀ27
Much less experimental information is available regarding the behavior of these potential
intermediates under catalytic conditions, including information regarding the catalyst resting state(s),
turnoverꢀlimiting and regioꢀ and/or stereochemical determining steps, and the nature of the reactive
2
ACS Paragon Plus Environment

Page 3 of 41
ACS Catalysis
1
2
3
4
5
6
7
8
9
gold allene intermediates. Furthermore, extant experimental information is restricted largely to
intramolecular hydrofunctionalization processes. For example, Gagné showed that a cationic bis(gold)
vinyl complex accumulated in solution during the gold(I)ꢀcatalyzed intramolecular hydroarylation of
allenes, which implicated a mechanism involving turnoverꢀlimiting protodeauration.22 A subsequent
study by Widenhoefer and Gagné of the hydroalkoxylation of 2,2ꢀdiphenylhexaꢀ4,5ꢀdienꢀ1ꢀol catalyzed
by [P(tꢀBu)2oꢀbiphenyl]AuOTs supported a mechanism involving reversible C–O bond formation
followed by turnoverꢀlimiting protonolysis of a mono(gold) vinyl complex that occurred competitively with
formation of an inactive bis(gold) vinyl complex.23,28 In comparison, Lalic's investigation of the
intramolecular hydroalkoxylation of γꢀhydroxy allenes catalyzed by a gold carboxylate complex
supported a mechanism involving irreversible C–O bond formation followed by turnoverꢀlimiting
protodeauration of a mono(gold) vinyl complex without competing formation of bis(gold) vinyl
complexes.6
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In what represents the only kinetic and mechanistic analysis of gold(I)ꢀcatalyzed intermolecular
allene hydrofunctionalization, Toste and Goddard have reported a combined kinetic/computational
analysis of the hydroamination of 1,7ꢀdiphenylheptaꢀ3,4ꢀdiene with methyl carbazate catalyzed by
(PPh3)AuNTf2. Here, authors invoked a "twoꢀstep, no intermediate" mechanism involving turnoverꢀ
limiting isomerization of a gold η2ꢀallene complex to a gold η1ꢀallylic cation transition state that is
trapped by methyl carbazate either prior to or after progression to the achiral η1ꢀallylic cation.7
Experimentally, this conclusion was supported by the ~zeroꢀorder dependence of the rate on methyl
carbazate concentration coupled with assignment of the gold πꢀallene complex [(Ph3P)Au(η2ꢀ
PhCH2CH2C=C=CCH2CH2Ph)]+ (A) as the catalyst resting state on the basis of in situ 31P NMR analysis
and independent synthesis. However, Widenhoefer and Brooner subsequently demonstrated that the
species assigned by Toste and Goddard as πꢀallene complex A is rather the catalyst decomposition
product [(PPh3)2Au]+,20 and without clear delineation of the catalyst resting state(s), zeroꢀorder rate
3
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 41
1
2
3
4
5
6
dependence on nucleophile concentration need not be attributed to turnoverꢀlimiting allene
isomerization.
7
8
9
Herein we report the mechanistic analysis of the gold(I)ꢀcatalyzed intermolecular
hydroalkoxylation of 3ꢀmethylꢀ1,2ꢀbutadiene (1) with 1ꢀphenylpropanꢀ1ꢀol (2), which forms primary
allylic ether 3a with high regioselectivity relative to tertiary allylic ether 4a (eq 1).25 This study includes
the kinetic and spectroscopic analysis of catalytic mixtures, deuteriumꢀlabeling studies, and
independent synthesis of potential intermediates. We targeted goldꢀcatalyzed hydroalkoxylation for
this study owing to the high efficiency of these processes25 and because aliphatic alcohols are
representative of the weakly basic nucleophiles typically employed in gold(I)ꢀcatalyzed allene
hydrofunctionalization.4 The results of these studies are in accord with a mechanism involving
reversible, endergonic formation of a cationic gold πꢀallene complex that undergoes irreversible, outerꢀ
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
sphere attack by alcohol on an η2ꢀallene complex followed by rapid protodeauration.
RESULTS AND DISCUSSION
Kinetics of the Gold(I)ꢀCatalyzed Hydroalkoxylation of excess 1 with 2. Toward an
understanding of the mechanism of gold(I)ꢀcatalyzed intermolecular allene hydroalkoxylation, we
investigated the kinetics of the reaction of 1 with 2 catalyzed by (IPr)AuOTf in toluene under conditions
of excess allene 1. In these studies, the singleꢀcomponent catalyst (IPr)AuOTf was employed in
preference to a mixture of (IPr)AuCl and AgOTf to avoid potential complications associated with the
4
ACS Paragon Plus Environment

Page 5 of 41
ACS Catalysis
1
2
3
4
5
6
7
8
9
presence of silver salts either as coꢀcatalysts or byproducts of catalyst activation.29 In an initial
experiment, a toluene solution of 1 (1.6 M), 2 (0.16 M), hexadecane (internal standard), and a catalytic
amount of (IPr)AuOTf (15 mM) was stirred at 30 °C and analyzed periodically by GC. A plot of ln[2]
versus time was linear to >3 halfꢀlives with a pseudoꢀfirstꢀorder rate constant of 9.8 ± 0.4 × 10–4 s–1
(Figure 1, Table 1, entry 1), which established firstꢀorder dependence of the rate on alcohol
concentration under these conditions. Importantly, no significant (≤2%) formation of the regioisomeric
allenyl alcohol 4a was detected throughout complete conversion of 1 and 2 to 3a.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0.2
ꢀ0.3
ꢀ0.8
[2]t
ln ––––
ꢀ1.3
ꢀ1.8
ꢀ2.3
ꢀ2.8
[2]0
0
1000
2000
time (s)
3000
Figure 1. Pseudoꢀfirstꢀorder plot for the hydroalkoxylation of 1 (1.6 M) with 2 (0.16 M) catalyzed by
(IPr)AuOTf (16 mM) in toluene at 30 °C.
5
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 41
1
2
3
4
Table 1. Pseudoꢀfirstꢀorder rate constants for the hydroalkoxylation of 1 with 2 catalyzed by (IPr)AuOTf
5
6
in toluene at 30 °C.
7
8
9
entry
[1] (M)
1.6
[2] (M)
0.16
0.16
0.16
0.16
0.16
0.16
0.16
0.16
[cat] (mM)
additive (mM)
(104)kobs (s–1)
9.8 ± 0.4
1
2
3
4
5
6
7
8
15
15
15
15
15
8.2
30
16
––
––
––
––
––
––
––
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0.78
1.2
5.4 ± 0.2
7.1 ± 0.4
11.9 ± 0.3
12.8 ± 0.5
5.4 ± 0.4
17.1 ± 0.6
4.0 ± 0.2
2.3
3.1
1.6
1.6
1.6
Bu NOTf (16)
4
9
1.6
1.6
0.16
0.16
16
15
Bu NOTf (24)
4
2.39 ± 0.08
1.9 ± 0.2
10
Bu NOTf (77)
4
11
12
13
14a
15
16
17
18
19
20
21
22a
1.6
0.16
0.16
0.16
0.90
1.8
15
15
15
18
18
18
6.0
13
15
20
45
15
HOTf (8.6)
4.4 ± 0.3
1.6
3a (75)
––
10.1 ± 0.4
11 ± 1
1.6
0.18
0.18
0.18
0.14
0.14
0.14
0.14
0.14
0.14
––
3.39 ± 0.01
3.06 ± 0.02
2.71 ± 0.01
0.834 ± 0.007
2.49 ± 0.03
3.00 ± 0.07
4.06 ± 0.03
8.35 ± 0.15
––
2.7
––
1.0
––
1.0
––
1.0
––
1.0
––
1.0
––
1.0
––
3.03 ± 0.02
a1ꢀPhenylpropanꢀ1ꢀolꢀOꢀd (2ꢀOꢀd, ~90% d) was used in this reaction.
To determine the dependence of the rate of goldꢀcatalyzed hydroalkoxylation on allene
concentration, pseudoꢀfirstꢀorder rate constants for the goldꢀcatalyzed reaction of 1 with 2 (0.16 M)
were determined as a function of allene concentration from 0.78 to 3.1 M (Table 1, entries 1ꢀ5). A plot
of kobs versus [1] revealed positive, nonꢀlinear dependence of the rate on allene concentration,
consistent with a kinetic order between zero and one (Figure 2). To determine the dependence of the
6
ACS Paragon Plus Environment

Page 7 of 41
ACS Catalysis
1
2
3
4
5
6
7
8
9
rate of allene hydroalkoxylation on catalyst concentration, pseudoꢀfirstꢀorder rate constants for the
hydroalkoxylation of 1 (1.6 M) with 2 (0.16 M) were determined as a function of catalyst concentration
from 8.2 to 30 mM (Table 1, entries 1, 6, and 7). A plot of kobs versus [(IPr)AuOTf] was linear (Figure
3), which established the firstꢀorder dependence of the rate on catalyst concentration.
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
14.0
12.0
10.0
8.0
6.0
4.0
2.0
0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
[1] (M)
Figure 2. Allene concentration dependence of the rate of hydroalkoxylation of 1 (0.78 ꢀ 3.1 M) with 2
(0.16 M) catalyzed by (IPr)AuOTf (15 mM) in toluene at 30 °C.
7
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 41
1
2
3
4
5
6
7
8
9
20.0
16.0
12.0
8.0
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4.0
0.0
0
5
10 15 20 25 30 35
[(IPr)AuOTf] (mM)
Figure 3. Plot of pseudoꢀfirstꢀorder rate constants versus catalyst concentration for the
hydroalkoxylation of 1 (1.6 M) with 2 (0.16 M) catalyzed by (IPr)AuOTf (0ꢀ30 mM) in toluene at 30 °C.
To determine the dependence of the rate of goldꢀcatalyzed hydroalkoxylation on exogenous
triflate ion concentration, pseudoꢀfirstꢀorder rate constants for the reaction of 1 (1.6 M) with 2 (0.16 M)
catalyzed by (IPr)AuOTf (16 mM) were determined as a function of tetrabutylammonium triflate
concentration from 16 to 77 mM (Table 1, entries 8ꢀ10). The corresponding plot of pseudoꢀfirstꢀorder
rate constants versus the concentration of tetrabutylammonium triflate established inhibition of the rate
of hydroalkoxylation by exogenous triflate (Figure 4). Similarly, goldꢀcatalyzed reaction of 2 with excess
1 that contained triflic acid (8.6 mM) was ~50% slower than in the absence triflic acid (Table 1, entries 1
and 11). The rate of goldꢀcatalyzed reaction of 2 with excess 1 that contained allylic ether 3a (75 mM)
in toluene was not significantly different from the rate of hydroalkoxylation in the absence of excess 3a
(Table 1, entries 1 and 12). Goldꢀcatalyzed deuterioalkoxylation of 2 with 1ꢀphenylpropanꢀ1ꢀolꢀOꢀd (2ꢀ
d1; ~90% d) formed 3aꢀd1 with 77% deuterium incorporation exclusively at the internal vinylic position of
3aꢀd1 and with no significant deuterium kinetic isotope effect (KIE): kH/kD = 1.1 ± 0.1 (eq 2; Table 1,
entries 1 and 13).
8
ACS Paragon Plus Environment

Page 9 of 41
ACS Catalysis
1
2
3
4
5
6
10
8
7
8
9
6
10
11
12
13
14
15
16
17
18
19
20
21
4
2
0
0
20
40
60
80
[(nꢀBu4)NOTf] (mM)
Figure 4. Plot of the concentration of tetrabutylammonium triflate versus kobs for the reaction of
M) with (0.16 M) catalyzed by (IPr)AuOTf (~15 mM).
1 (1.6
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
Kinetics of the Gold(I)ꢀCatalyzed Hydroalkoxylation of 1 with excess 2. In a second set of
experiments, we analyzed the kinetics of the goldꢀcatalyzed hydroalkoxylation of allene with alcohol
1
2
under conditions of excess alcohol. In one experiment, a solution of
1 (0.18 M), 2 (0.90 M), CHCl3
(internal standard) and (IPr)AuOTf (18 mM) in tolueneꢀ
d
8 at 30 °C was analyzed periodically by 1H NMR
spectroscopy. A plot of ln[
1
] versus time was linear to >3 halfꢀlives with a pseudoꢀfirstꢀorder rate
constant of kobs = 3.39 ± 0.01
×
10–4 s–1 (Figure 5, Table 1, entry 14), which established firstꢀorder
dependence of the rate on allene concentration under these conditions. As was the case with
hydroalkoxylation of under conditions of excess allene, no significant (≤2%) formation of the
1
regioisomeric allenyl alcohol 4a was observed during the conversion of
1 and 2 to 3a.
9
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 41
1
2
3
4
5
6
7
8
To determine the dependence of the rate of hydroalkoxylation on alcohol concentration, pseudoꢀ
firstꢀorder rate constants for the disappearance of were determined as a function of [ ] from 0.9 to 2.7
M (Table 1, entries 14ꢀ16). A plot of the corresponding pseudoꢀfirstꢀorder rate constants versus [2 0
1
2
]
9
revealed that the rate of hydroalkoxylation displayed near zeroꢀorder dependence on alcohol
concentration over this range (Figure 6), in sharp contrast to the firstꢀorder rate dependence on alcohol
concentration under conditions of excess allene. To determine the dependence of the rate on 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
concentration, pseudoꢀfirstꢀorder rate constants for the hydroalkoxylation of
were determined as a function of catalyst concentration from 8.2 to 30 mM (Table 1, entries 12, 17ꢀ21).
A plot of obs versus [(IPr)AuOTf] was linear (Figure 7), which established the firstꢀorder dependence of
1 (0.15 M) with 2 (1.0 M)
k
the rate of hydroalkoxylation on catalyst concentration and overall the secondꢀorder rate law under
10–2 M–1 s–1 at [
2]0 = 0.9 M. Goldꢀ
these conditions: rate =
k'[
1][(IPr)AuOTf] where
k
' = 1.88 ± 0.01
×
catalyzed deuterioalkoxylation of
1 with excess 1ꢀphenylpropanꢀ1ꢀolꢀOꢀd (2ꢀd1; 90% d) occurred with
no detectable KIE (kH
/
k
D = 1.00 ± 0.03; Table 1, entries 19 and 22).
0.2
ꢀ0.3
ꢀ0.8
[1]t
ln ––––
[1]0
ꢀ1.3
ꢀ1.8
ꢀ2.3
0
2000
4000
time (s)
6000
Figure 5. Pseudoꢀfirstꢀorder plot for the hydroalkoxylation of
(IPr)AuOTf (18 mM) in tolueneꢀ 8 at 30 °C.
1 (0.18 M) with 2 (0.90 M) catalyzed by
d
10
ACS Paragon Plus Environment

Page 11 of 41
ACS Catalysis
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
4.0
3.0
2.0
1.0
0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0
[2] (M)
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6. Plot of pseudoꢀfirstꢀorder rate constants versus alcohol concentration for the
hydroalkoxylation of
30 °C.
1 (0.18 M) with 2 (0.90 ꢀ 2.7 M) catalyzed by (IPr)AuOTf (18 mM) in tolueneꢀd8 at
9
8
7
6
5
4
3
2
1
0
0
10
20
30
40
50
[(IPr)AuOTf] (mM)
Figure 7. Plot of pseudoꢀfirstꢀorder constants versus catalyst concentration for the hydroalkoxylation of
1
(0.15 M) with
2
(1.0 M) catalyzed by (IPr)AuOTf (0ꢀ45 mM) in toluene at 30 °C.
11
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 41
1
2
3
4
Spectroscopic Analysis of Catalytic Mixtures. A series of experiments were performed to
5
6
gain insight into the resting state(s) of the goldꢀcatalyzed hydroalkoxylation of
1
with
2. Unfortunately,
7
8
31
the absence of a phosphine supporting ligand precluded analysis of catalytically active mixtures by P
9
1
NMR spectroscopy and H NMR spectroscopy alone proved unsuitable for this task due to excessive
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
broadening and poor dispersion of chemical shifts. We therefore employed the 13Cꢀlabeled gold
complex (IPr*)AuOTf (IPr* = IPrꢀ1ꢀ13C1) which exploits the diagnostic and generally well dispersed
carbene C1 resonances of potential reaction intermediates. For example, the carbene C1 resonance of
(IPr*)AuOTf appears at
δ
163 in the 13C NMR spectrum as compared to
δ
~180 for cationic gold
πꢀ
allene and alkene complexes,19,30 and ~200 for neutral IPr gold
complexes.32
σ
ꢀvinyl31 and cationic IPr bis(gold) vinyl
In one experiment, allene
1
was added incrementally (0.50, 1.0, 1.5, 2.0 M) to a solution of
13
(IPr*)AuOTf (~30 mM) in tolueneꢀd8 and analyzed after each addition by C NMR spectroscopy at 30
13
°C (eq 3). As the concentration of
1 increased, the carbene C resonance of (IPr*)AuOTf broadened
and shifted slightly downfield and when the concentration of
1 reached ~2 M, a broad resonance at δ
~180, assigned to gold
πꢀallene complex (IPr*)Au(η
2ꢀH2C=C=CMe2)]+ OTf– (5ꢀ13C1), was observed
along with the resonance of (IPr*)AuOTf at
δ 165, although the excessive broadening of these
resonances precluded accurate determination of the relative concentrations (eq 3, Figure 8).
Nevertheless, these observations are consistent with the (1) endergonic conversion of (IPr*)AuOTf and
1
to 5 and (2) the formation of kineticallyꢀrelevant concentrations of 5 in the presence of excess 1. The
resulting solution of (IPr*)AuOTf (30 mM) and excess
1
(~2 M) was then treated with alcohol
2
(0.14 M)
165 and
and monitored periodically by 13C NMR spectroscopy at 30 °C. The broad resonances at
δ
180 persisted throughout complete conversion of
1 and 2 to 3a (t1/2 = 20 min) without the appearance of
any additional carbene resonances. These observations established the gold triflate complex
(IPr)AuOTf and the gold
conditions.
πꢀallene complex 5 as the predominant gold complexes present under catalytic
12
ACS Paragon Plus Environment

Page 13 of 41
ACS Catalysis
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
Figure 8. Portion of the 13C NMR spectrum of a mixture of
1 (2.0 M) and (IPr*)AuOTf (~30 mM) in
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
tolueneꢀ 8 at 30 °C. The resonance at ~208 corresponds to the sp carbon atom of 1.
d
δ
In a second experiment, alcohol
2 was added incrementally to a tolueneꢀd8 solution of
(IPr*)AuOTf (~30 mM) at 30 °C and analyzed after each addition by 13C NMR spectroscopy. Addition of
13
a large excess of
to (IPr*)AuOTf (
2 (~1.5 M) produced no detectable change in the C NMR resonance corresponding
δ
163) nor were any other carbene resonances observed (eq 4). These observations
indicate that the equilibrium constant for the conversion of (IPr*)AuOTf to gold alcohol complex
CH(Et)Ph)]+ ( ) is exceedingly small.33,34 The resulting solution of (IPr*)AuOTf and excess
[(IPr)Au(
6
2
was then treated with allene
1
(0.14 M) and the solution was monitored periodically by 13C NMR
13
ACS Paragon Plus Environment

ACS Catalysis
Page 14 of 41
1
2
3
4
5
6
spectroscopy at 30 °C. The sharp resonance of (IPr*)AuOTf at
δ
163 broadened slightly upon addition
and to without
of allene but otherwise remained unchanged throughout complete conversion of
1
2
3
7
8
9
the appearance of any additional carbene resonances, which established (IPr*)AuOTf as the
predominant goldꢀcontaining species present during catalysis under conditions of excess alcohol.
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
Kinetic Regioselectivity of Hydroalkoxylation
.
Paton and Maseras have previously
with methanol catalyzed by (IPr)Au+
investigated the mechanism of the hydroalkoxylation of
1
employing DFT calculations.17 This analysis supported a mechanism involving outerꢀsphere addition
of methanol to the more substituted terminus of to form tertiary allylic ether 4b followed by secondary
1
isomerization of 4b to primary allylic alcohol 3b via outerꢀsphere addition of methanol to the
γꢀcarbon
atom of the πꢀ(allylic ether) complex I
followed by elimination of methanol (Scheme 1).17 The goldꢀ
catalyzed transposition of allylic ethers in the presence of alcohol has been validated experimentally by
us and others.35 Furthermore, Lee has shown that gold(I)ꢀcatalyzed hydroalkoxylation of allene
1
with
excess primary alcohol in DMF at 0 °C leads to selective formation of the corresponding tertiary allylic
ether (eq 5).36
14
ACS Paragon Plus Environment

Page 15 of 41
ACS Catalysis
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Scheme 1. Proposed mechanism for the goldꢀcatalyzed isomerization of tertiary allylic ether 4b to
primary allylic alcohol 3b in the presence of methanol.
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
For the reasons outline in the preceding paragraph, we sought to evaluate the possibility that 3a
was formed via secondary isomerization of 4a in the goldꢀcatalyzed hydroalkoxylation of
noted above, periodic analysis of the gold(I)ꢀcatalyzed conversion of and to 3a under conditions of
either excess allene or alcohol revealed no detectable accumulation of tertiary allylic alcohol 4a
Therefore, if 4a were an intermediate in the gold(I)ꢀcatalyzed conversion of and to 3a, conversion of
1 with 2. As
1
2
.
1
2
4a to 3a under reaction conditions would have to be much faster than the formation of 3a. Importantly,
a pair of experiments employing methanol as nucleophile argue strongly against this possibility and,
rather point to formation of 3a as the kinetic product of hydroalkoxylation. Specifically, reaction of
1
(0.8 M) with methanol (0.8 M) and a catalytic amount of (IPr)AuOTf (~30 mM) at 23 °C reached >95%
conversion after 2 h to form 3b as the exclusive product (Scheme 2). In comparison, treatment of a 1:1
mixture of 4b (0.8 M) and methanol (~0.8 M) with a catalytic amount of (IPr)AuOTf (~30 mM) at 23 °C
15
ACS Paragon Plus Environment

ACS Catalysis
Page 16 of 41
1
2
3
4
for 2 h led to no detectable consumption of 4b or formation of 3b as determined by 1H NMR analysis of
5
6
the crude reaction mixture.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2. Goldꢀcatalyzed reactions of
1 and 4b with methanol.
Kinetic Model for the Hydroalkoxylation of 1 with 2. Although our analysis of the gold(I)ꢀ
catalyzed reaction of with provides no direct insight into the stereochemistry of C–O bond formation,
1
2
we have previously established the net antiꢀaddition of the O–H bond of an alcohol across the C=C
bond of an allene for both intraꢀ and intermolecular goldꢀcatalyzed hydroalkoxylation, and these
observations are consistent with outerꢀsphere mechanisms for C–O bond formation.25ꢀ27,37 We
therefore interpreted the kinetics of the hydroalkoxylation of
1 with 2 catalyzed by (IPr)AuOTf in the
context of the mechanism depicted in Scheme 3 involving reversible reaction of (IPr)AuOTf with
1 to
form gold
π
ꢀallene complex 5ꢀOTf,19 outerꢀsphere addition of alcohol
2 to 5ꢀOTf to form the Oꢀ
protonated gold
IIIꢀOTf that collapses to release
derivation of rate laws were that (1) protodeauration (II → IIIꢀOTf), and hence hydroalkoxylation, is
irreversible, (2) gold triflate complex (IPr)AuOTf and gold ꢀallene complex 5ꢀOTf are the only
complexes that accumulate under catalytic conditions ([Au]tot = [(IPr)AuOTf] + [5ꢀOTf]), and (3) cationic
σꢀvinyl intermediate II, and rapid protodeauration to form gold
πꢀallylic ether complex
3
and close the catalytic cycle. Key assumptions made in the
π
πꢀcomplex
5
exists as the tight ion pair 5ꢀOTf in the nonꢀpolar reaction medium (K
A[OTf–] >> 1; see
below).
16
ACS Paragon Plus Environment

Page 17 of 41
ACS Catalysis
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 3. Proposed mechanism for the goldꢀcatalyzed hydroalkoxylation of
1 with 2.
Within the mechanism depicted in Scheme 3, kinetic scenarios involving turnoverꢀlimiting
protodeauration or deprotonation of II can be safely discounted on the basis of our experimental
observations. For example, mechanisms involving reversible C–O bond formation followed by
turnoverꢀlimiting protodeauration or deprotonation would be expected to display significant deuterium
KIEs for deuterioalkoxylation and firstꢀorder dependence on [
2] under all conditions, neither of which
was observed experimentally. Importantly, both we23 and Lalic6 have observed large KIEs (kH
/kD > 5)
for goldꢀcatalyzed intramolecular hydroalkoxylation/deuterioalkoxylation of allenes under conditions of
turnoverꢀlimiting protodeauration. Similarly, mechanisms involving irreversible C–O bond formation
followed by turnoverꢀlimiting protodeauration would be expected to display a significant KIE for
deuterioalkoxylation, zeroꢀorder rate dependence on both [1] and [2] under all conditions, and the
accumulation of mono(gold) or bis(gold) vinyl complexes under catalytic conditions, none of which were
observed experimentally.
17
ACS Paragon Plus Environment

ACS Catalysis
Page 18 of 41
1
2
3
4
The observed change in the kinetic order of the goldꢀcatalyzed hydroalkoxylation of
1
with
2
as
5
6
a function of the
1:2 ratio points to a change in the resting state catalyst composition and/or the
7
8
9
turnoverꢀlimiting step of catalysis, the latter of which requires that two or more microscopic steps within
the catalytic cycle occur at similar rates. As such, application of the preꢀequilibrium assumption to 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
reversible formation of gold
hydrofunctionalization processes, appears overly restrictive. Rather, application of the Bodenstein
(steady state) approximation to gold ꢀallene complex 5ꢀOTf with no additional restrictions generates
πꢀallene complex 5ꢀOTf, which is often assumed valid for goldꢀcatalyzed
π
the rate law depicted in equation 6, which is of the same form as the BriggsꢀHaldane equation for
enzyme kinetics.38
From the general rate equation for hydroalkoxylation of
limiting kinetic scenarios based on the relationship between the rate of C–O bond formation (k2
the rates of interconversion of (IPr)AuOTf and 5ꢀOTf (k1 ] and –1). In the case where C–O bond
formation is much slower than is the interconversion of (IPr)AuOTf and 5ꢀOTf (k2 ] << k1 ] + –1), the
rate equation simplifies to the twoꢀterm rate equation depicted in eq 7, which is of the same form as the
Michaelis–Menten equation and which predicts firstꢀorder rate dependence on [ ] and [Au]tot, and
between zeroꢀ and firstꢀorder dependence on [ ]. This rate equation is of the same form as the
experimental rate law determined for the goldꢀcatalyzed reaction of and under conditions of excess
allene , which displayed firstꢀorder dependence on [ ] and a positive, nonꢀinteger dependence on
1
with
2
(eq 6), we considered two
[
2]) and
[
1
k
[
2
[
1
k
2
1
1
2
1
2
1.
18
ACS Paragon Plus Environment

Page 19 of 41
ACS Catalysis
1
2
3
4
5
6
7
8
9
In the case where C–O bond formation is much faster than is the interconversion (IPr)AuOTf and 5ꢀOTf
k2 ] >> k1 ] + –1), the rate equation depicted in eq 6 simplifies to the secondꢀorder rate equation
depicted in eq 8 predicting firstꢀorder dependence of the rate on [ ] and [Au]tot and zeroꢀorder
dependence on [ ]. This rate equation is of the same form as the experimental rate law determined for
the goldꢀcatalyzed reaction of and under conditions of excess alcohol where rate =
'[ ][(IPr)AuOTf], where the macroscopic rate constant ' corresponds directly to the microscopic rate
constant k1
(
[
2
[
1
k
1
2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
2
2
k
1
k
.
Rate equation 7, which describes the rate behavior of catalytic hydroalkoxylation under
conditions of excess allene, can alternatively be represented at eq 9. Taking the reciprocal of the
equation describing
under conditions of excess allene at constant catalyst concentration with slope = k–1
intercept = 1/ 2[Au]tot. Indeed a plot of the reciprocal of the experimentallyꢀdetermined pseudoꢀfirstꢀ
k
obs gives equation 10, which predicts a linear relationship between 1/
kobs and 1/[1]
/
k1k2[Au]tot and with
k
order rate constants for the hydroalkoxylation of
1
under conditions of excess allene at constant
103 M s and
102 s (Figure 9), from which values for the microscopic rate constant
0.02 s–1 M–1 and equilibrium constant k1 K1 = 0.34 ± 0.02 M–1 were derived. Furthermore, using
catalyst concentration ([cat] = 16 mM) versus 1/[1] was linear with a slope = 1.15 ± 0.05 ×
intercept = 3.9 ± 0.4
×
k2 = 0.17 ±
/k–1 =
the value for
k
1, which corresponds to the secondꢀorder rate constant for hydroalkoxylation under
10–2 s–1 M–1), we likewise derived the value for rate
conditions of excess alcohol (
k' = k1 = 1.88 ± 0.01 ×
constant k–1 = 5.6 ± 0.5 ×
10–2 s–1.
19
ACS Paragon Plus Environment

ACS Catalysis
Page 20 of 41
1
2
3
4
5
6
7
8
9
2000
1500
1000
500
0
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0.0
0.5
1.0
1.5
1/[1] (M)
Figure 9. Plot of 1/kobs versus 1/[
1
] for the hydroalkoxylation of 1 with 2 catalyzed by (IPr)AuOTf (15
mM) in toluene at 30 °C.
The equilibrium constant K1 = 0.34 M–1 determined from the above analysis predicts an
equilibrium ratio of (IPr)AuOTf:5•OTf ≈ 1.5:1 at [ ] = 2 M and [Au]tot = 30 mM according to the
1
relationship K1 = [5•OTf]/[(IPr)AuOTf][
1
]. This value is consistent with our experimental observations
regarding the formation of 5•OTf from (IPr)AuOTf and
1
, considering the low sensitivity of our 13C NMR
measurements. Less clear is that the values determined for the microscopic rate constants k1
2 validate the limiting conditions k2 ] << k–1 k1 ] (eq 7) and k2 ] >> k–1 k1 ] (eq 8) corresponding
to the experimental conditions of excess allene and excess alcohol, respectively. For example, the
, k–1, and
k
[
2
+
[1
[
2
+
[1
calculated values for k1
0.16 M) are 0.034, 0.056, and 0.027 s–1, respectively, and the calculated values for k1
] = 1.8 M) are 0.0034, 0.056, and 0.31 s–1,
[1], k–1, and k2[2] under typical conditions of excess allene ([1] = 1.6 M, [2] =
[1], k–1, and k2[2]
under general conditions of excess alcohol ([1] = 0.18 M, [2
respectfully. However, here it should be noted that the value for k1 employed in this analysis was
determined under conditions of excess alcohol, whereas the values for k2 and K1 were determined
under conditions of excess allene, under the assumption that the magnitudes of these microscopic rate
20
ACS Paragon Plus Environment

Page 21 of 41
ACS Catalysis
1
2
3
4
5
constants are invariant of the reaction conditions. One observation that suggests this may not be the
case is the observed decrease in obs, where obs/[Au]tot 1, with increasing [ ] under conditions of
k
k
=
k
2
6
7
8
excess alcohol (Figure 5), which suggests that k1 may be larger under conditions of excess allene,
9
although the origin of this rate/medium effect remains unclear.
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
Ion Pairing and Role of Exogenous Triflate
coordinating anions and cationic transition metal complexes,39 including cationic gold
104 in CH2Cl2.41,42 Furthermore,
.
The formation of ion pairs between nonꢀ
π
ꢀcomplexes,40 is
well established, with association constants that typically exceed 1
×
because log KA for ion pair association typically scales with the reciprocal of the dielectric constant of
),43 ion pairing of with OTf– is anticipated to be exceptionally strong in the nonꢀ
(toluene) = 2.38].44ꢀ46
the medium (i.e. 1/
ε
5
polar reaction medium employed in catalytic hydroalkoxylation [ε(CH2Cl2) = 8.93; ε
Importantly, ligand substitution of cationic transition metal complexes with strongly associated anionic
ligands occurs through an interchange mechanism and displays zeroꢀorder rate dependence on the
concentration of the anionic ligand.39,41,42,47 In the context of gold(I)ꢀcatalyzed allene hydroalkoxylation,
the presence of strong ion pairing between
5 and OTf is consistent with the absence of any significant
curvature of the plot of kobs versus [(IPr)AuOTf] at constant, excess [
1] (Figure 3). For example, under
conditions of negligible ion pairing between
5 and OTf (K
A[OTf–] << 1), equation 7 is replaced by eq 11
containing the term k–1KA[OTf] in the denominator.48 Because [OTf–] would increase with increasing
[Au]tot under conditions of negligible ion pairing, the rate dependence on catalyst concentration under
such conditions would approach halfꢀorder, which was not observed (Figure 3).49
Under conditions of strong ion pairing between
5
and OTf–, the observed inhibition of the rate of
hydroalkoxylation by tetrabutylammonium triflate (Figure 4) points to the presence of an additional
triflateꢀdependent pathway for the reversion of 5•OTf to (IPr)AuOTf. Here, two points are worth noting.
21
ACS Paragon Plus Environment

ACS Catalysis
Page 22 of 41
1
2
3
4
Firstly, simply relaxing the condition of strong ion pairing between
5 and OTf (KA[OTf] >> 1) does not
5
6
7
8
account for the observed rate dependence on both catalyst and tetrabutylammonium triflate
concentration. In particular, in the case of a modest association constant for ion pairing between and
5
9
OTf–, the rate dependence on triflate concentration would be most pronounced at the lowest triflate
concentrations, such as in the determination of the rate dependence on (IPr)AuOTf concentration.
However, as was noted above, no significant deviation from linearity was observed for a plot of kobs
versus [(IPr)AuOTf] from 0 ꢀ 30 mM (Figure 3). Secondly, owing to the strong ion pairing between
tetrabutylammonium and triflate in the nonꢀpolar reaction medium50 and the higher concentrations of
triflate employed in these experiments relative to catalytic conditions (16 ꢀ 77 mM versus ≤30 mM,
respectively), it is unlikely that this triflateꢀdependent pathway is relevant under conditions of catalytic
hydroalkoxylation.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Because both 5•OTf and tetrabutylammonium triflate are both strongly ion paired in the nonꢀ
polar reaction medium,50 a mechanism for the triflateꢀdependent reversion of 5•OTf to (IPr)AuOTf
involving direct attack of free triflate at the gold center of 5•OTf followed by expulsion of allene
1
appears unlikely. Rather, allene for triflate ligand exchange presumably occurs via association of the
[
nꢀBu4N][OTf] ion pair with 5ꢀOTf to form an ion aggregate such as {[
5][OTf]2[(nꢀBu4N]} followed by
ligand interchange.51 This mechanism is analogous to that proposed by Romeo to rationalize the
kinetics of the displacement of dimethyl sulfoxide from [Pt(Me4En)(Me2SO)Cl]Cl (Me4En = (
tetramethyldiaminoethane) by halide ions in CH2Cl2.42
N,N,N',N'ꢀ
Control of Regioselectivity in the Hydroalkoxylation of 1. The selective formation of the
linear allylic ether 3a in the goldꢀcatalyzed hydroalkoxylation of with is intriguing owing both to the
selective formation of tertiary allylic ether in the goldꢀcatalyzed hydroalkoxylation of in DMF (eq 5) and
1
2
1
also the formation of the tertiary allylic carbamate in the closely related gold(I)ꢀcatalyzed
with benzyl carbamate (eq 12).36,52 Our experimental data argue strongly against
hydroamination of
1
isomerization of 4a to 3a under reaction conditions and we can also safely rule out regiochemicallyꢀ
22
ACS Paragon Plus Environment

Page 23 of 41
ACS Catalysis
1
2
3
determining formation of gold
π
ꢀallene complex 5ꢀOTf owing to the facile (ꢀG‡ ≤ 11 kcal/mol)
4
5
6
7
8
unimolecular interconversion of gold
(Scheme 4).19 The simplest rationale for the observed regioselectivity of the goldꢀcatalyzed
πꢀallene complexes 5 and 5' via η
1ꢀallene intermediate IV
9
hydroalkoxylation of
1
with
2
is to invoke selective, irreversible attack of
2
at the less substituted allene
on
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
terminus of . However, we cannot rule out a mechanism involving rapid and reversible attack of
5
2
5' that is superimposed on slower, irreversible attack of
2 on 5 and indeed, this mechanism might better
account all the observations made in the context of goldꢀcatalyzed allene hydrofunctionalization.4.36.52
Supporting the feasibility of such a mechanism, gold vinyl complex II' generated via attack of
2 on 5' is
expected to be less stable than II owing to the diminished olefinic substitution and likewise,
experimental and computational analysis of protodeauration supports the more facile protodeauration of
II relative to II' owing to the presence of the electronꢀreleasing olefinic methyl groups of II 53
.
Scheme 4. Interconversion of gold
πꢀallene complexes
5
and 5' via
η
1ꢀallene intermediate IV and
potential mechanism for regiocontrol in catalytic hydroalkoxylation.
23
ACS Paragon Plus Environment

ACS Catalysis
Page 24 of 41
1
2
3
4
5
6
7
8
Role of activated η1ꢀallene species in hydroalkoxylation. There has been considerable
1ꢀallene species in goldꢀ
catalyzed allene functionalization processes,16 most notably from the work of Toste and Goddard who
1ꢀallylic cation in
the goldꢀcatalyzed hydroamination of 1,7ꢀdiphenylheptaꢀ3,4ꢀdiene.7 Our analysis of the fluxional
speculation regarding the potential attack of nucleophile on an activated
η
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
invoked barrierless attack of methyl carbazate on the transition state leading to the
η
behavior of gold
πꢀcomplexes containing aliphatic 1,1ꢀ and 1,3ꢀdisubstituted allenes established two
distinct gold
η
1ꢀallene species: a lower energy (ꢀG‡ ≤11 kcal/mol) staggered
η
1ꢀallene intermediate or
transition state (e.g
and a higher energy (ꢀG‡ ≥17.5 kcal/mol) planar
ꢀface exchange and allene stereomutation.19,54,55
.
IV, Schemes 4 and 5) that leads to
πꢀface exchange without allene stereomutation
η
1ꢀallylic cation intermediate (e.g
. V, Scheme 5) that
leads that leads to both
π
Scheme 5. η2ꢀAllene complex
5 and potential η
1ꢀallene isomers.
In the context of the gold(I)ꢀcatalyzed hydroalkoxylation of
1 with 2, a mechanism involving
attack on the staggered
η
1ꢀallene intermediate IV is fully consistent with our kinetic data and is
reminiscent of the slippage that is proposed to occur along the reaction coordinate for nucleophilic
addition to a transition metal
π
ꢀcomplex.56 However, it appears unlikely that formation of IV
(
ꢀG‡ ≤11
24
ACS Paragon Plus Environment

Page 25 of 41
ACS Catalysis
1
2
3
4
5
6
7
8
kcal/mol) could become kineticallyꢀcontrolling under any conditions. A mechanism involving outerꢀ
1ꢀallylic cation
1ꢀallylic cation formation, it is plausible that isomerization of
5 to V could become
sphere attack of
2
on
η
V is likewise consistent with our kinetic data, and given the higher
energy barrier for
η
9
turnoverꢀlimiting under conditions of excess alcohol, as was observed experimentally. However, a
1ꢀallylic cation
V, generated either reversibly or
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
mechanism involving outerꢀsphere attack of
2
on
η
irreversibly, provides no pathway for chirality transfer, which is common for the gold(I)ꢀcatalyzed
hydroalkoxylation of axially chiral allenes.5ꢀ7 Finally, a mechanism akin to that invoked by Toste and
Goddard involving turnoverꢀlimiting generation and trapping of an η )
1ꢀallylic cation transition state (VꢀTS
is inconsistent with the observed firstꢀorder rate dependence on alcohol concentration under conditions
of excess allene. Furthermore, such a pathway can be firmly discounted for the closely related goldꢀ
catalyzed intramolecular hydroalkoxylation of 2,2ꢀdiphenylhexaꢀ4,5ꢀdienꢀ1ꢀol, where the energy barrier
for C–O bond formation (ꢀG‡ ≤ 13 kcal/mol) occurred well below the threshold for allylic cation
formation (ꢀG‡ ≥ 17.5 kcal/mol).23
Conclusions
In summary, we have investigated the kinetics and mechanism of the intramolecular
hydroalkoxylation of 3ꢀmethylꢀ1,2ꢀbutadiene (
toluene. All of our data are consistent with the mechanism depicted in Scheme 3 involving endergonic
formation of the cationic gold ꢀallene complex , which exists as the tight ion pair 5ꢀOTf in the nonꢀ
polar reaction medium. Outerꢀsphere addition of
collapse of the resulting gold ꢀalkene complex
1) with 1ꢀphenylpropanꢀ1ꢀol (2) catalyzed by (IPr)AuOTf in
π
5
2
to 5ꢀOTf followed by rapid protodemetallation and
π
6 releases the primary allylic ether 3a as the kinetic
product and regenerates (IPr)AuOTf. The microscopic rate constants for the conversion of (IPr)AuOTf
and to 5ꢀOTf (k1 ]), the collapse of 5ꢀOTf to (IPr)AuOTf and –1), and the rate of attack of on
5ꢀOTf (k2 ]) are similar enough such that (1) application of the preꢀequilibrium assumption to the
1
[
1
1
(
k
2
[
2
formation of 5ꢀOTf is not valid and (2) the rate behavior of catalytic hydroalkoxylation changes as a
25
ACS Paragon Plus Environment

ACS Catalysis
Page 26 of 41
1
2
3
4
5
6
7
8
9
function of the relative and absolute concentrations of
the reaction rate displayed firstꢀorder dependence on [(IPr)AuOTf] and [
order dependence on [ ], while under conditions of excess alcohol , the reaction rate displayed firstꢀ
order dependence on [(IPr)AuOTf] and [ ] and zeroꢀorder dependence on [ ].
1
and
2
. Under conditions of excess allene
1,
2] and between zeroꢀ and firstꢀ
1
2
1
2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
There has been a considerable speculation and debate concerning the role of counterion in
gold(I)ꢀcatalyzed hydrofunctionalization reactions, in particular that outerꢀsphere C–X (X = N, O) bond
formation is assisted by hydrogen bonding between the incoming nucleophile and the counterion.57
Indeed, we have obtained evidence for counterionꢀassisted C–O bond formation in our investigation of
goldꢀcatalyzed intramolecular allene hydroalkoxylation.58 The present investigation of the mechanism
of the gold(I)ꢀcatalyzed hydroalkoxylation of
1 with 2 provides little additional insight in this regard.
However, our data do show that potential triflate/alcohol hydrogen bonding cannot lead to significant O–
H bond cleavage in the transition state for C–O bond formation, which would have been revealed by the
presence of a KIE for the deuterioalkoxylation of
1 with 2ꢀOꢀd, which was not observed.
In comparison to counterion effects, there has been much less discussion regarding the effect of
reaction medium on the kinetic behavior and mechanism of gold(I)ꢀcatalyzed hydrofunctionalization,
and the present investigation points to the potential importance of such medium effects. In particular,
the kinetics of catalytic hydroalkoxylation of
1 in toluene are dominated by the slow and endergonic
conversion of (IPr)AuOTf to gold
π
ꢀallene complex 5ꢀOTf, which was corroborated by in situ 13C NMR
analysis of the reaction of
1
with (IPr*)AuOTf. This behavior stands in sharp contrast to the reaction of
1
with (IPr)AuOTf in CD2Cl2, which forms 5
quantitatively with <2 equiv of allene.19 Therefore, it is quite
reasonable to assume that very different kinetic behavior of the goldꢀcatalyzed hydroalkoxylation of
1
with would be observed in more polar solvents, although predictions beyond this point are
2
unwarranted owing to the absence of information regarding medium effects on the rate of C–O bond
formation and/or on the equilibrium constants for the potentially competing formation of goldꢀ
nucleophile complexes.
26
ACS Paragon Plus Environment

Page 27 of 41
ACS Catalysis
1
2
3
■ ASSOCIATED CONTENT
4
5
6
7
8
■ AUTHOR INFORMATION
Corresponding Author
9
*Eꢀmail: rwidenho@chem.duke.edu.
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
ORCID
Notes
Ross A. Widenhoefer: 0000ꢀ0002ꢀ5349ꢀ8477
The authors declare no competing financial interest.
Supporting Information
The Supporting Information is available free of charge on the ACS Publications website at DOI:xxxxxxx
Experimental procedures and derivation of differential rate equations (PDF)
■ ACKNOWLEDGMENTS We acknowledge the NSF (CHEꢀ1465209) for support of this
research. RGC was supported through a GAANN fellowship (P200A150114).
References
(1) a) Li, Y.; Li, W.; Zhang, J. Gold‐Catalyzed Enantioselective Annulations. Chem. Eur. J. 2017, 23,
467 – 512; b) Quintavalla, A.; Bandini, M. Gold
‐Catalyzed Allylation Reactions. ChemCatChem 2016
,
8, 1437ꢀ1453; c) Miróand, J.; del Pozo, C. Fluorine and Gold: A Fruitful Partnership. Chem. Rev. 2016
,
116, 11924−11966; d) Hopkinson, M. N.; TlahuextꢀAca, A.; Glorius, F. Merging Visible Light
Photoredox and Gold Catalysis. Acc. Chem. Res 2016 49, 2261−2272; e) Paioti, P. H. S.; Aponick,
A. GoldꢀCatalyzed Transformation of Unsaturated Alcohols. Top. Curr. Chem 2015 357, 63–94; f)
Dorel, R.; Echavarren, A. M. Gold(I)ꢀCatalyzed Activation of Alkynes for the Construction of Molecular
.
,
.
,
Complexity. Chem. Rev
.
2015
,
115, 9028−9072; g) Asiria, A. M.; Hashmi, A. S. K. GoldꢀCatalysed
27
ACS Paragon Plus Environment

ACS Catalysis
Page 28 of 41
1
2
3
4
5
6
7
8
9
Reactions of Diynes. Chem. Soc. Rev
Transformations of ꢀDiazocarbonyl Compounds: Selectivity and Diversity. Chem. Soc. Rev. 2016
506ꢀ516; i) Shin, S. GoldꢀCatalyzed Carbene Transfer Reactions. Top. Curr. Chem 2015 357, 25–62;
j) Michelet, V. GoldꢀCatalyzed Domino Reactions. Top. Curr. Chem 2015 357, 95–132.
(2) a) Rudolph, M.; Hashmi, A. S. K. Gold Catalysis in Total Synthesis ꢀ an Update. Chem. Soc. Rev
2012
Achievements. Chem. Soc. Rev
(3) a) Zi, W.; Toste, F. D. Recent Advances in Enantioselective Gold Catalysis Chem. Soc. Rev
.
2016
,
45, 4471ꢀ4503; h) Liu, L.; Zhang, J. GoldꢀCatalyzed
α
,
45,
.
,
.
,
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
.
,
,
41, 2448–2462; b) Pflästerer, D.; Hashmi, A. S. K. Gold Catalysis in Total Synthesis – Recent
.
2016, 45, 1331ꢀ1367.
.
2016
45, 4567ꢀ4589; b) Pradal, A.; Toullec, P. Y.; Michelet, V. Recent Developments in Asymmetric
Catalysis in the Presence of Chiral Gold Complexes. Synthesis 2011, 1501ꢀ1514; c) Sengupta, S.; Shi,
X. Recent Advances in Asymmetric Gold Catalysis. ChemCatChem 2010
R. A. Recent Developments in Enantioselective Gold(I) Catalysis. Chem. Eur. J
(4) a) Krause, N.; Winter, C. GoldꢀCatalyzed Nucleophilic Cyclization of Functionalized Allenes: A
Powerful Access to Carboꢀ and Heterocycles. Chem. Rev 2011 111, 1994ꢀ2009; b) Shen, H. C.
,
2, 609ꢀ619; d) Widenhoefer,
.
2008 14, 5382ꢀ5391.
,
.
,
Recent Advances in Syntheses of Heterocycles and arbocycles via Homogeneous Gold Catalysis. Part
1: Heteroatom Addition and Hydroarylation Reactions of Alkynes, Allenes, and Alkenes. Tetrahedron
2008, 64, 3885ꢀ3903.
(5) a) Patil, N. T. Chirality Transfer and Memory of Chirality in Gold‐Catalyzed Reactions. Chem.
Asian J
.
2012
,
7, 2186 – 2194; b) Campolo, D.; Gastaldi, S.; Roussel, C.; Bertrand, M. P.; Nechab, M.
AxialꢀtoꢀCentral Chirality Transfer in Cyclization Processes. Chem. Soc. Rev
.
2013 42, 8434ꢀꢀ8466; c)
,
Teller, H.; Corbet, M.; Mantilli, L.; Gopakumar, G.; Goddard, R.; Thiel W.; Fürstner, A. OneꢀPoint
Binding Ligands for Asymmetric Gold Catalysis: Phosphoramidites with a TADDOLꢀRelated but Acyclic
Backbone. J. Am. Chem. Soc
.
2012
,
134, 15331ꢀ15342; d) Miles, D. H.; Veguillas, M.; Toste, F. D.
2013 , 3427ꢀ
Gold(I)ꢀCatalyzed Enantioselective Bromocyclization Reactions of Allenes. Chem. Sci
.
, 4
3431; e) Handa, S.; Lippincott, D. J.; Aue, D. H.; Lipshutz, B. H. Asymmetric GoldꢀCatalyzed
28
ACS Paragon Plus Environment

Page 29 of 41
ACS Catalysis
1
2
3
4
5
Lactonizations in Water at Room Temperature. Angew. Chem., Int. Ed
.
2014
,
53, 10658ꢀ10662; f) Zi,
Diols Through
W.; Toste, F. D. Gold(I) Catalyzed Enantioselective Desymmetrization of 1,3
‐
‐
6
7
8
9
Intramolecular Hydroalkoxylation of Allenes. Angew. Chem., Int. Ed. 2015, 54, 14447ꢀ14451; g) Shu,
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.ꢀZ.; Nguyen, S. C.; He, Y.; Oba, F.; Zhang, Q.; Canlas, C.; Somorjai, G. A.; Alivisatos A. P.; Toste, F.
D. SilicaꢀSupported Cationic Gold(I) Complexes as Heterogeneous Catalysts for Regioꢀ and
Enantioselective Lactonization Reactions. J. Am. Chem. Soc
(6) Cox, N.; Uehling, M. R.; Haelsig, K. T.; Lalic, G. Catalytic Asymmetric Synthesis of Cyclic Ethers
Containing an α‐Tetrasubstituted Stereocenter. Angew. Chem. Int. Ed 2013 52, 4878ꢀ4882.
. 2015, 137, 7083ꢀ7086.
.
,
(7) Wang, Z. J.; Benitez, D.; Tkatchouk, E.; Goddard, W. A.; Toste, F. D. Mechanistic Study of Gold(I)ꢀ
Catalyzed Intermolecular Hydroamination of Allenes. J. Am. Chem. Soc 2010 132, 13064ꢀ13071.
(8) Liu, Z.; Wasmuth, A. S.; Nelson, S. G. Au(I)ꢀCatalyzed Annulation of Enantioenriched Allenes in
the Enantioselective Total Synthesis of (−)ꢀRhazinilam. J. Am. Chem. Soc 2006 128, 10352ꢀ10353.
(9) Zeldin, R. M.; Toste, F. D. Synthesis of Flinderoles B and C by a GoldꢀCatalyzed Allene
Hydroarylation. Chem. Sci 2011 , 1706ꢀ1709.
(10) Bates, R. W.; Dewey, M. R. A Formal Synthesis of Swainsonine by GoldꢀCatalyzed Allene
Cyclization. Org. Lett 2009 11, 3706ꢀ3708.
(11) Sawama, Y.; Sawama, Y.; Krause, N. First Total Synthesis of (R,R,R)ꢀ and (3
GoldꢀCatalyzed Allene Cycloisomerization and Determination of Absolute Configuration of the Natural
Product. Org. Biomol. Chem. 2008 , 3573ꢀ3579.
.
,
.
,
.
, 2
.
,
R,5S,9R)ꢀBejarol by
,
6
(12) Schmiedel, V. M.; Stefani, S.; Reissig, H.ꢀU. Stereodivergent Synthesis of Jaspine B and its
Isomers Using a CarbohydrateꢀDerived Alkoxyallene as C3ꢀbuilding Block. Beilstein J. Org. Chem.
2013, 9, 2564ꢀ2569.
29
ACS Paragon Plus Environment