Dehydrogenation of n -Alkanes by Solid-Phase Molecular Pincer-Iridium Catalysts. High Yields of α-Olefin Product

Article abstract of DOI:10.1021/jacs.5b05313

We report the transfer-dehydrogenation of gas-phase alkanes catalyzed by solid-phase, molecular, pincer-ligated iridium catalysts, using ethylene or propene as hydrogen acceptor. Iridium complexes of sterically unhindered pincer ligands such as ^iPr4PCP, in the solid phase, are found to give extremely high rates and turnover numbers for n-alkane dehydrogenation, and yields of terminal dehydrogenation product (α-olefin) that are much higher than those previously reported for solution-phase experiments. These results are explained by mechanistic studies and DFT calculations which jointly lead to the conclusion that olefin isomerization, which limits yields of α-olefin from pincer-Ir catalyzed alkane dehydrogenation, proceeds via two mechanistically distinct pathways in the case of (^iPr4PCP)Ir. The more conventional pathway involves 2,1-insertion of the α-olefin into an Ir-H bond of (^iPr4PCP)IrH₂, followed by 3,2-β-H elimination. The use of ethylene as hydrogen acceptor, or high pressures of propene, precludes this pathway by rapid hydrogenation of these small olefins by the dihydride. The second isomerization pathway proceeds via α-olefin C-H addition to (pincer)Ir to give an allyl intermediate as was previously reported for (^tBu4PCP)Ir. The improved understanding of the factors controlling rates and selectivity has led to solution-phase systems that afford improved yields of α-olefin, and provides a framework required for the future development of more active and selective catalytic systems. (Figure Presented).

Full text of DOI:10.1021/jacs.5b05313

Page 1 of 33
Journal of the American Chemical Society
Kumar et al. Solidꢀgas Phase Dehydrogenation
1
22ꢀJulꢀ15
1
2
3
Dehydrogenation of nꢀAlkanes by SolidꢀPhase Molecular PincerꢀIridium Catalysts. High
4
5
Yields of αꢀOlefin Product
Akshai Kumar,a Tian Zhou,a Thomas J. Emge,a Oleg Mironov,b Robert J. Saxton,b
Karsten KroghꢀJespersen,*a and Alan S. Goldman*a
6
7
8
9
a Department of Chemistry and Chemical Biology, Rutgers, The State University of New Jersey,
New Brunswick, New Jersey 08903, United States
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
b Chevron Energy Technology Company, 100 Chevron Way, Richmond, CA 94802
KEYWORDS: Pincer complexes, Iridium complexes, Gas–solid phase reactions, Alkane
dehydrogenation, CꢀH bond activation, Catalytic alkane conversion
ABSTRACT: We report the transferꢀdehydrogenation of gasꢀphase alkanes catalyzed by solidꢀphase,
molecular, pincerꢀligated iridium catalysts, using ethylene or propene as hydrogen acceptor. Iridium
complexes of sterically unhindered pincer ligands such as iPr4PCP, in the solid phase, are found to give
extremely high rates and turnover numbers for nꢀalkane dehydrogenation, and yields of terminal
dehydrogenation product (αꢀolefin) that are much higher than those previously reported for solutionꢀ
phase experiments. These results are explained by mechanistic studies and DFT calculations which
jointly lead to the conclusion that olefin isomerization, which limits yields of αꢀolefin from pincerꢀIr
catalyzed alkane dehydrogenation, proceeds via two mechanistically distinct pathways in the case of
(
(
iPr4PCP)Ir. The more conventional pathway involves 2,1ꢀinsertion of the αꢀolefin into an IrꢀH bond of
iPr4PCP)IrH2, followed by 3,2ꢀβꢀH elimination. The use of ethylene as hydrogen acceptor, or high
pressures of propene, precludes this pathway by rapid hydrogenation of these small olefins by the
dihydride. The second isomerization pathway proceeds via αꢀolefin CꢀH addition to (pincer)Ir to give an
allyl intermediate as was previously reported for (tBu4PCP)Ir. The improved understanding of the factors
controlling rates and selectivity has led to solutionꢀphase systems that afford improved yields of αꢀ
olefin, and provides a framework required for the future development of more active and selective
catalytic systems.
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Kumar et al. Solidꢀgas Phase Dehydrogenation
Page 2 of 33
2
22ꢀJulꢀ15
1
2
3
INTRODUCTION
4
5
6
7
8
9
Olefins are key intermediates in many, perhaps even most, processes in the fuel and commodity
chemical industries, and are also of great importance in the synthesis of fine chemicals. The
development of catalysts for the regioselective dehydrogenation of alkanes and alkyl groups to afford
olefins is therefore a goal of great interest to a broad range of chemists.
The most significant progress toward the goal of practical regioselective alkane dehydrogenation
catalysts has been realized with pincerꢀligated iridium complexes, beginning with the report by Kaska
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
and Jensen1 of alkane dehydrogenation by (tBu4PCP)IrHn (1ꢀHn; R4PCP = κ3ꢀC6H3ꢀ2,6ꢀ(CH2PR2)2; n = 2
or 4). Our group subsequently reported the synthesis and generally greater catalytic activity of the less
crowded iPr4PCP analogue (2)2 and soon discovered that both complexes showed kinetic selectivity for
dehydrogenation of nꢀalkanes at the terminal position to give the highly desirable corresponding αꢀ
olefins.3 Catalysts 1 and 2 were also found to be effective for the acceptorless dehydrogenation of
alkanes.2,4 Work with these complexes has been followed by reports of numerous catalytically active
variants with the (PCP)Ir motif,5ꢀ9 including other bisꢀphosphines,10ꢀ14 bisꢀphosphinites (POCOP),15ꢀ18
hybrid phosphine–phosphinites (PCOP),19,20 arsines (AsOCOAs),21 hybrid phosphineꢀthiophosphinites
(PSCOP)22 and hybrid amineꢀphosphinites (NCOP)23. In addition to simple alkane dehydrogenation,
these complexes have been employed for numerous other catalytic transformations of hydrocarbons,
including alkane metathesis,6,8,9,20,24ꢀ26 alkyl group metathesis,27 dehydroaromatization,19,28,29 alkane–
alkene coupling reactions,30ꢀ32 borylation of alkanes23 and the dehydrogenation of several nonꢀalkane
substrates.22,33,34 Several pincer motifs more recently explored, such as (CCC)Ir,35ꢀ38 (PCP)Ru39ꢀ41
,
(PCP)Os42, and (NCN)Ir43,44 have been found to show promise for alkane dehydrogenation but as of yet
none have proven to be competitive with the well investigated PCPꢀtype iridiumꢀbased systems.26
In early alkane dehydrogenation studies45 3,3ꢀdimethylꢀ1ꢀbutene (TBE) was found by Crabtree to be
a singularly effective hydrogen acceptor. In addition to being resistant to doubleꢀbond isomerization, the
bulky TBE is only weakly coordinating; in contrast, ethylene was found to completely inhibit catalytic
activity.45 TBE has thus become the most commonly used acceptor for alkane transfer
dehydrogenation.8,9 We have found that norbornene (NBE) is also very effective, presumably for similar
reasons.3,6 However, on a large scale, the use of smaller olefins, such as ethylene or propene, would
be much more practical. Ethylene, in particular, is efficiently dehydrogenated with heterogeneous
catalysts which could allow for recycling of ethane (without necessarily requiring the costly separation of
ethylene and ethane).46 We have earlier reported the use of propene as acceptor for
dehydroaromatization reactions.19 Very recently Brookhart and co–workers have demonstrated the role
of ethylene as both an acceptor and a dienophile in the synthesis of piperylene,47 toluene47 and p–
xylene.48
The dehydrogenation of lighter alkanes, e.g. butane and pentane,47,49 is of particular interest. Such
alkanes are generally undesirable as transportation fuel components, while the corresponding olefins
and dienes have many chemical applications and could potentially be dimerized (or crossꢀdimerized) to
give alkanes of molecular weight more suitable for fuel.32
Given these considerations we were led to study the transferꢀdehydrogenation of lighter alkanes
using gaseous olefins. At high temperatures, mixtures of these hydrocarbons are entirely in the gas
phase, while the catalyst is (at least primarily) in the solid phase.50 Much to our surprise the turnover
rates resulting from such dualꢀphase systems were found to be remarkably high. Although
heterogeneous solidꢀgas systems for alkane dehydrogenation are very well known,46 to our knowledge
these are the first examples of purely molecular solidꢀphase catalysts for alkane dehydrogenation.
Characteristic of their behavior in solution, and in contrast with nonꢀmolecular solidꢀphase
2
ACS Paragon Plus Environment

Page 3 of 33
Journal of the American Chemical Society
3
Kumar et al. Solidꢀgas Phase Dehydrogenation
22ꢀJulꢀ15
1
2
3
4
5
6
7
dehydrogenation catalysts, these systems are selective for the formation of αꢀolefins. Most remarkably,
the maximum yields of αꢀolefin from these heterogeneous systems are found to be much greater than
have been previously obtained from homogeneous solution phase systems (the highest previously
reported yield being 97 mM 1ꢀoctene3 from the transfer dehydrogenation of nꢀoctane with 0.5 M 1ꢀ
decene catalyzed by 1ꢀHn).
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
RESULTS AND DISCUSSION
A crystalline precursor of (iPr4PCP)Ir. As initial results (see below) indicated the particular
effectiveness of (iPr4PCP)Ir for our purposes, we explored several synthetic routes to viable precursors
of this catalyst (see Experimental Section). We successfully obtained crystalline (iPr4PCP)Ir(C2H4)
(2ꢀC2H4), which was characterized by Xꢀray diffraction (Figure 1). This is the first report of a crystal
structure of a direct precursor of the (iPr4PCP)Ir catalyst.
Figure 1. ORTEP diagram of (iPr4PCP)Ir(C2H4) with ellipsoids drawn at the 50% probability level. For
the sake of clarity only H atoms on the ethylene ligand are shown.
Transfer dehydrogenation by various pincerꢀiridium complexes: gasꢀsolid phase. In a typical
experimental setꢀup (Figure 2) 100 ꢁL of a stock nꢀpentane solution of catalyst (1 mM) was added to a
custom–made thickꢀwalled longꢀneck 1.5ꢀmL ampoule inside an argonꢀfilled glove box. The ampoule
was then connected to a Kontes adapter via Tygon tubing and degassed on a highꢀvacuum line.
Propene (1.0 atm) was then introduced to the system. The contents of the vials were frozen in liquid
nitrogen and the vials were flame sealed. (The total gas volume before sealing was 3 mL; thus, after
condensation, sealing, and warming, the pressure of propene is approximately 2 atm. The vials were
then placed in a preꢀheated aluminum block inside an oven maintained at 240 °C [note: extreme
caution must be exercised during this process, including the use of appropriate safety shields] and
subjected to interval free heating for a stipulated time. The oven was then cooled to room temperature,
the ampoules were removed, the contents were frozen in liquid nitrogen, the ampoules were broken
open, and the contents were analyzed by GC.
The vapor pressure of nꢀpentane at 200 °C and 240 °C is calculated to be 32 atm and 52 atm
respectively.51,52 100 ꢁl nꢀpentane in a 1.5 mL vial, upon converting fully to the gas phase, will generate
approximately 22 atm and 24 atm at 200 °C and 240 °C respectively. Thus, all hydrocarbons are
expected to be in the gas phase under these conditions.
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 33
4
Kumar et al. Solidꢀgas Phase Dehydrogenation
22ꢀJulꢀ15
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 2. Experimental setup for transfer dehydrogenation of n–pentane catalyzed by pincerꢀiridium
complexes using ethylene or propene as acceptor (values given for 2 atm acceptor at 240 °C).
nꢀPentane/propene transfer dehydrogenation was initially investigated with nine different pincerꢀIr
complexes (Scheme 1 and Table 1). We have previously reported that the relatively uncrowded mixed
methyl/tꢀbutyl substituted complexes (tBu3MePCP)IrHn (3) and (tBu2Me2PCP)IrHn (4) are catalytically more
active than 1 for the transfer dehydrogenation of n–octane using either TBE or NBE as acceptor.6 (Note
that under transferꢀdehydrogenation conditions, olefin, dihydride and tetrahydride complexes are
equivalent as precursors of the catalytically active (pincer)Ir fragments.) Likewise, complexes (pꢀOMeꢀ
iPr4PCP)IrH4 (7ꢀH4) and (iPr4PCP)IrH4 (2ꢀH4) were reported to be more active than 1 for transfer
dehydrogenation of nꢀalkanes using TBE.3,5,49
Scheme 1. PincerꢀIr catalysts investigated in this study
PtBu2
PiPr2
PtBuMe
PtBuMe
Ir
PtBu2
Ir
Ir
Ir
Ir
MeO
PtBu2
PiPr2
PtBu2
PMetBu
PtBu2
(
tBu4PCP)Ir (1)
(
iPr4PCP)Ir (2)
(
tBu3MePCP)Ir (3)
(
tBu2Me2PCP)Ir (4)
(pꢀMeOꢀtBu4PCP)Ir (5)
O
O
PtBu2
Ir
PiPr2
Ir
PiPr2
Ir
PiPr2
MeO
Ir
PtBu2
PiPr2
PiPr2
PiPr2
O
(
tBu4POCOP)Ir (6) (pꢀMeOꢀiPr4PCP)Ir (7)
(
iPr4PCOP)Ir (8)
(
iPr4Anthraphos)Ir (9)
4
ACS Paragon Plus Environment

Page 5 of 33
Journal of the American Chemical Society
5
Kumar et al. Solidꢀgas Phase Dehydrogenation
22ꢀJulꢀ15
1
2
3
4
5
6
7
Table 1. Dehydrogenation of n–pentane “[8.7 M]”a by various pincerꢀIr catalysts, under 2 atm propene
“[1.2 M]”a at 240 °Cb
(pincer)Ir catalyst
(1.0 mM)
+
+
+
+
sealed vial aligned
vertically at 240 °C
8.7 M
2 atm
1ꢀpentene
(E+Z)ꢀ2ꢀpentene
1,3ꢀpentadiene
"1.2 M"
8
9
Total Olefinsc
/ mM
1ꢀPentene / mM
(% total monoenes)
% Propene
conversion (by GC)
Dienes
/ mM
Entry
1
Catalyst / 1 mM Time / min
10
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
8
0
ND
ND
0
–
180
4 (50%)
6
10
20
20
15 (75%)
17 (85%)
ND
ND
–
–
2
3
4
40
1
10
23
25
18 (75%)
20 (77%)
ND
ND
1
1
180
5
10
0
0
ND
ND
30
0
0
40
59
50 (85%)
105 (33%)
3
10
340
21
5
6
7
8
9
180
950
200 (24%)
98
150
4
10
630
140 (24%)
230 (24%)
63
90
40
180
1050
110
2
10
96
53 (56%)
10
52
2
180
630
170 (29%)
40
8
10
110
160
79 (72%)
ND
ND
2
4
180
100 (64%)
9
10
170
310
120 (70%)
180 (61%)
ND
ND
3
180
10
7
a Concentrations given represent concentrations obtained after the indicated time of heating, followed by cooling of the
reaction vessel to 25 °C so that all species are condensed or dissolved into liquid solution phase.
b Reaction vessels (glass ampoules) were oriented vertically.
c Most runs were repeated; results given are the average of two or more runs. Reproducibility averaged ± 3%.
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Kumar et al. Solidꢀgas Phase Dehydrogenation
Page 6 of 33
6
22ꢀJulꢀ15
1
2
3
4
5
6
7
8
Transfer dehydrogenation of gas phase nꢀpentane using propene (2 atm) was successfully
catalyzed under these conditions by the relatively crowded complexes (tBu4PCP)IrHn (1), the p–methoxy
derivative (pꢀOMeꢀtBu4PCP)IrHn (5) and the bisphosphinite complex (tBu4POCOP)IrHn (6). These
catalysts all gave relatively low TO numbers, less than 30 TO after 180 min at 240 °C (entries 1ꢀ3, Table
1). The apparent initial rates with catalysts 1 and 5 were moderately high, but were not maintained over
the course of the reaction (e.g. catalyst 1 gave 20 TO after 10 min, but the same TON was found after
40 min). Catalyst 3, in which one of the tBu groups of 1 is substituted by a methyl group, showed slightly
greater catalytic activity (entry 4, Table 1; 59 TO after 40 min). To our knowledge these are the first
examples of presumed molecular catalysts effecting heterogeneous (gasꢀsolid phase) dehydrogenation
of alkane.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In contrast with the moderate activity noted above, much higher rates and TONs were obtained with
the use of catalyst 4, in which methyl groups replace two of the tBu groups of 1 (entry 5, Table 1). After
10 min, 340 TO had been obtained, corresponding to consumption of ca. 30% of the propene in the
vessel, while after 180 min, the TON was 950, corresponding to hydrogenation of >90% of the propene.
Dehydrogenation catalyzed by (iPr4PCP)Ir(C2H4) (2) proceeded even more rapidly (entry 6, Table 1); 630
TO were obtained after 10 min and >1000 TO after 180 min. These rates and turnover numbers are
unprecedented even for solution phase alkane dehydrogenation systems.
The very high catalytic efficiency of (tBu2Me2PCP)IrH4 (4) compared to (tBu3MePCP)IrH4 (3) contrasts
with our earlier observations on n–octane transfer dehydrogenation with TBE or NBE as acceptors.6
Studies on 3 and 4 for n–octane transfer dehydrogenation using TBE or NBE indicated that 3 was the
more effective of the two catalysts (although both 3 and 4 provided higher activity than 1). However, as
4 showed a tendency to form dinuclear clusters, it was unclear whether this was responsible for its
lesser activity. Given the presumably much greater binding ability of propene vs. the bulkier TBE or
NBE, the formation of dimers or oligomers should be much less significant in the presence of propene;
the much greater reactivity of 4 vs. 3 when using propene thus lends support to this explanation for the
lesser activity of 4 obtained when NBE or TBE is used as acceptor. Another possible explanation, also
based on decreased dimerization of 4 under the present conditions, is that catalyst mobility is reduced
in the solid phase compared with the solution phase thereby inhibiting the kinetics of dimer formation.
The activity levels of hybrid phosphineꢀphosphinite catalyst (iPr4PCOP)Ir(C2H4) (8) (entry 6, Table 1),
and (iPr4Anthraphos)Ir(C2H4) (9) (entry 7, Table 1) were high, but less than those of either 4 or 2. The p–
methoxy derivative of 2, (p–OMeiPr4PCP)Ir(C2H4) (7) appeared to give a good initial rate (170 TO after
10 min) but much lower conversion than 2 after 180 min. We suspect this is due to intermolecular
reactions involving the methoxy groups, leading to catalytically inactive species.53,54
The great differences in catalytic activity among these various catalysts, and particularly the
disparity between the more crowded (three or more tꢀBu groups) and less crowded complexes was not
expected. Previous studies in our lab and others have indicated that less crowded complexes did
indeed tend to be more active, but the difference was much less dramatic. These studies generally
utilized NBE and TBE as hydrogen acceptors and, of course, were in the liquid phase. For example, the
difference between the activity of tBu4PCP and iPr4PCP catalysts was found to be a factor of ca. 3ꢀfold.6
The present results therefore raised the question as to whether the large differences between the
catalysts, most notably the tBu4PCP and iPr4PCP derivatives, were a result of the different acceptors
used in this study, or a result of the unusual conditions, particularly the solid vs. solution phase.
Accordingly, we conducted solutionꢀphase experiments (using nꢀoctane as dehydrogenation
substrate) under similar conditions, including the nature of the acceptor and the unusually high
temperature (240 °C). Results are shown in Table 2. A very pronounced difference in activity is
observed between the (tBu4PCP)Ir and (iPr4PCP)Ir precursors: a factor of ca. 11 in the initial data point, a
6
ACS Paragon Plus Environment

Page 7 of 33
Journal of the American Chemical Society
7
Kumar et al. Solidꢀgas Phase Dehydrogenation
22ꢀJulꢀ15
1
2
3
4
5
6
7
8
value which is of the same magnitude as the factor of ca. 30 observed in the gasꢀsolidꢀphase
experiments. Further, we find that in the case of catalyst 1 and propene as acceptor, in both solid and
liquid phases the rate decreases dramatically after an initial period of catalysis with a relatively slow
rate. We are not able to fully explain this behavior of catalyst 1, but our observations all seem
applicable to both solution and solid phase. Indeed, the fact that we observe, in both solution and solid
phase, both the dramatic difference between catalysts 1 and 2 and the particular temporal profile of
catalyst 1, strongly indicates that the catalysts are operating as discrete molecular species even in the
solid phase.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 2. Dehydrogenation of n–octane [6.2 M]a by various pincerꢀIr catalysts, under 6 atm propene
“[3.7 M]”a at 240 °Cb
+
+
(pincer)Ir catalyst
(1.0 mM)
1ꢀoctene
(E+Z)ꢀ2ꢀoctene
(E+Z)ꢀ3ꢀoctene
+
sealed vial aligned
vertically at 240 °C
6.2 M
6 atm
"3.7 M"
+
+ octadienes,
aromatics
(E+Z)ꢀ4ꢀoctene
Total Olefinsc
/ mM
1ꢀOctene / mM
(% total monoenes)
% Propene
conversion (by GC)
Dienes
/ mM
Catalyst / 1 mM
Time / min
Entry
10
40
170
190
56 (34%)
4
4
7
8
1
2
3
4
5
43 (23%)
75 (28%)
140 (25)
1
10
40
290
600
4
15
61
10
8
4
9
2
10
40
10
40
180
73 (31%)
140 (16%)
250 (24%)
270 (24%)
2
8
1100
1250
1310
15
15
20
260
230
190
10
40
1930
2430
160 (13%)
130 (10%)
44
65
670
1130
a Concentrations given represent concentrations obtained after the indicated time of heating, followed by cooling of the
reaction vessel to 25 °C so that all species are condensed or dissolved into liquid solution phase.
b Reaction vessels (glass ampoules) were oriented vertically.
c Most runs were repeated; results given are the average of two or more runs. Reproducibility averaged ± 2%.
In an effort to determine the physical distribution of catalyst during the gasꢀsolid phase experiments,
after several runs the GC oven temperature was cooled from 240 °C to 60 °C and slowly opened in the
range of a camcorder (see SI for images of one such experiment with (iPr4PCP)Ir(C2H4) (2)). At this point
the top portion of the vial was cool relative to the base, which was still hot as it was enclosed in the
aluminum block. The resulting images (Figure S7) clearly show bright red droplets formed along the
topmost portions of the vial as pentane condenses on the catalyst that had deposited on the glass. This
observation could be explained by vigorous splashing, when the pentane solution is heated to 240 °C
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 33
8
Kumar et al. Solidꢀgas Phase Dehydrogenation
22ꢀJulꢀ15
1
2
3
4
5
6
7
8
(followed by rapid solvent evaporation) or alternatively, by sublimation of the catalyst at this
temperature. To distinguish between these possibilities, an ampoule containing solid catalyst was
heated under the same conditions as the catalytic runs, including the presence of 2 atm propene but in
the absence of pentane or other liquid. Under such conditions, no significant migration of the iridium
complex within the ampoule was observed. Thus, rather than sublimation of catalyst, it seems likely that
when a pentane solution of catalyst is heated at 240 °C, the solution splashes and coats the glass
surface before the solvent is fully evaporated.
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
If it is assumed that the catalyst coats the glass surface, then, by having vials aligned horizontally
rather than vertically, the catalyst would have a greater surface area and should function more
efficiently. When the ampoules were positioned horizontally (using catalysts 2, 8 and 9 which proved
most effective in the experiments, cf. Table 1) even higher rates were achieved as shown in Table 3.
Remarkably, in the case of catalyst 2, the reaction had effectively proceeded to completion (≥ 97%
consumption of propene, > 1000 TO) after 10 min.
Table 3. Dehydrogenation of n–pentane “8.7 M”a under 2 atm propene “1.2 M”a at 240 °C with selected
catalysts and ampoules positioned horizontallyb
(pincer)Ir catalyst
(1.0 mM)
+
+
+
+
sealed vial aligned
horizontally at 240 °C
8.7 M
2 atm
1ꢀpentene
(E+Z)ꢀ2ꢀpentene
1,3ꢀpentadiene
"1.2 M"
Total Olefinsc
/ mM
1ꢀPentene / mM
(% total monoenes)
% Propene
conversion (by GC)
Dienes
/ mM
Entry
Time / min
10
Catalyst / 1 mM
220
920
520
680
120 (54%)
170 (20%)
240 (50%)
190 (31%)
21
74
48
67
3
1
2
3
180
10
82
32
64
8
9
2
180
10
1090
1200
200 (20%)
190 (17%)
97
97
110
114
180
a Concentrations given represent concentrations obtained after the indicated time of heating, followed by cooling of the
reaction vessel to 25 °C so that all species are condensed or dissolved into liquid solution phase.
b Reaction vessels (glass ampoules) were oriented horizontally.
c Most runs were repeated; results given are the average of two or more runs. Reproducibility averaged ± 1%.
nꢀButane was also investigated as a dehydrogenation substrate. Into an ampoule containing the
same quantity of catalyst 2 that was used in the experiments of Tables 1ꢀ3, a 1:1 butane/propene gas
mixture was condensed (see Figure 2) such that upon sealing and warming to room temperature the
pressures of butane and propene each reached 3 atm. High rates and turnover numbers were observed
(Table 4). In view of the much higher volatility of butane (b.p. = ꢀ1 °C) than pentane (b.p. = 36 °C), these
results may be interpreted as arguing against the possibility of a condensed amorphous catalyst/alkane
phase as opposed to a “true” solidꢀgas interaction. (It is well beyond the scope of this work, however, to
address in detail the question of the “phase” of any hydrocarbon adsorbed to the solid.)
8
ACS Paragon Plus Environment

Page 9 of 33
Journal of the American Chemical Society
9
Kumar et al. Solidꢀgas Phase Dehydrogenation
22ꢀJulꢀ15
1
2
Table 4. Dehydrogenation of nꢀbutane (3 atm) “[6.1 M]”a with propene (3 atm) “[6.1 M]”a at 240 °C catalyzed by 2. b
3
4
5
6
(pincer)Ir catalyst
(1.0 mM)
+
+
+
+
sealed vial aligned
horizontally at 240 °C
3 atm
"6.1 M"
3 atm
"6.1 M"
1ꢀbutene
(E+Z)ꢀ2ꢀbutene
butadiene
7
Total Olefinc
/ TON
Butadiene
/ TON
1ꢀButene
/ TON
1ꢀButene Fraction
/ %
8
9
Catalyst / 1 mM
Time / min
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10
40
335
590
680
40
40
65
185
370
280
65
65
45
180
a Concentrations given represent concentrations obtained after the indicated time of heating, followed by cooling of the
reaction vessel to ꢀ15 °C so that all species are condensed or dissolved into liquid solution phase.
b Reaction vessels (glass ampoules) were oriented horizontally.
Ethylene would be even more attractive as a hydrogen acceptor than propene (in addition to the
abundance of ethylene derived from shale gas in North America, ethane could be more easily recycled
via separation from the alkane substrate and conventional dehydrogenation methods46). In this context,
experiments with ethylene gave highly encouraging results (Table 5), although rates were roughly a
factor of ten slower than when propene was used as acceptor.
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 33
10
Kumar et al. Solidꢀgas Phase Dehydrogenation
22ꢀJulꢀ15
1
2
3
4
5
6
7
Table 5. Dehydrogenation of nꢀpentane “8.7 M”a with ethylene (2 atm, “1.2 M”)a at 240 °Cb by various
pincerꢀIr catalysts
(pincer)Ir catalyst
(1.0 mM)
+
+
+
+
C2H6
sealed vial aligned
vertically at 240 °C
8.7 M
2 atm
1ꢀpentene
(E+Z)ꢀ2ꢀpentene
1,3ꢀpentadiene
"1.2 M"
8
9
Catalyst /
1 mM
Total Olefinsc
/ mM
1ꢀPentene / mM
(% total monoenes)
% Ethylene
conversion (by GC)
Dienes
/ mM
Entry
Time / min
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10
40
17
45
12 (71%)
34 (76%)
ND
0
0
1
ND
8
4
9
10
180
10
98
254
41
70 (71%)
135 (55%)
36 (88%)
120 (78%)
ND
ND
ND
ND
1
7
0
1
2
3
4
180
150
10
41
36 (88%)
ND
ND
0
1
180
156
120 (78%)
7
10
40
72
60 (88%)
250 (79%)
430 (65%)
420 (61%)
4
2
5
320
660
720
ND
ND
44
5
100
180
28
41
2
a Concentrations given represent concentrations obtained after the indicated time of heating, followed by cooling of the
reaction vessel to 25 °C so that all species are condensed or dissolved into liquid solution phase.
b Reaction vessels (glass ampoules) were oriented vertically.
c Most runs were repeated; results given are the average of two or more runs. Reproducibility averaged ± 2%.
Selectivity for production of αꢀolefins. Regioselective functionalization of the terminal position of
nꢀalkanes (or nꢀalkyl groups) has long been one of the major goals of research in catalytic hydrocarbon
conversion. Ever since the earliest examples of organometallic CꢀH bond activation revealed selectivity
for oxidative addition at 1° vs. 2° positions55 – and thus a remarkable preference for cleaving stronger
CꢀH bonds – this selectivity has been viewed as perhaps the most important potential advantage of
“homogeneous” vs. “heterogeneous” catalysts for the functionalization of alkanes or alkyl groups. Thus,
we were quite surprised to observe that the heterogeneous systems described above appeared to show
greater selectivity for the formation of αꢀolefins than the homogeneous (solutionꢀphase) systems based
on the same catalysts. For example, 2ꢀcatalyzed pentaneꢀethylene transfer dehydrogenation yielded
430 mM αꢀolefin (upon condensation to the liquid phase; entry 5, Table 5) and formation of 660 mM
total olefin. This is an unprecedented yield of αꢀolefin from nꢀalkane dehydrogenation; as noted above,
to our knowledge the highest yield of αꢀolefin previously reported from any catalytic alkane
dehydrogenation system was 97 mM (out of a total conversion to olefin of 143 mM).56
The high selectivity, in even a qualitative sense, for αꢀolefin formation resulting from a
heterogeneous system is certainly noteworthy; for example, after 10 min at 240 °C, 88% selectivity with
total conversion to 72 mM (upon condensation) is obtained (entry 5, Table 5). This observed
10
ACS Paragon Plus Environment

Page 11 of 33
Journal of the American Chemical Society
Kumar et al. Solidꢀgas Phase Dehydrogenation
11
22ꢀJulꢀ15
1
2
3
4
5
6
7
regioselectivity certainly supports the argument that the catalyst, although not in solution, is still
operating as a discrete molecular species. But even more remarkable is the appearance of even
greater selectivity for αꢀolefin formation from the heterogeneous system as compared with the same
catalyst in solution. Accordingly, further experiments were conducted in large part with an aim toward
explaining this phenomenon.
8
9
Table 6. Dehydrogenation of n–pentane [8.7 M] under 2 atm and 6 atm propene at 200 °C and 240 °C
catalyzed by 2ꢀC2H4
10
11
12
13
14
15
16
(
iPr4PCP)Ir(C2H4)
(1.0 mM)
+
+
+
+
sealed vial aligned
vertically
200 °C or 240 °C
8.7 M
1ꢀpentene
(E+Z)ꢀ2ꢀpentene
1,3ꢀpentadiene
Total Olefinsc
/ mM
1ꢀPentene / mM
(% total monoenes)
% Propene
conversion (by GC)
Dienes
/ mM
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
Entry
1
Conditions
Time / min
240 °C
2 atm
10
40
630
850
140 (24%)
170 (22%)
230 (24%)
63
77
90
40
70
“1.2 M”
180
1050
110
240 °C
4 atm
10
40
1370
1450
1590
420 (37%)
440 (36%)
430 (33%)
56
66
73
210
230
290
2
3
4
5
6
7
“2.5 M”
180
240 °C
6 atm
10
40
700
870
300 (48%)
400 (51%)
440 (46%)
20
20
30
76
87
“3.7 M”
180
1060
116
240 °C
6 atm
“7.4 M”d
10
40
80
695
930
380 (58%)
485 (56%)
575 (46%)
8
40
65
10
16
1420
170
200 °C
2 atm
10
40
410
690
720
150 (41%)
190 (32%)
190 (30%)
36
60
67
34
75
74
“1.2M”
180
200 °C
4 atm
10
40
370
510
800
155 (46%)
210 (45%)
260 (37%)
15
20
39
30
40
94
“2.5M”
180
200 °C
6 atm
10
40
270
470
950
130 (50%)
220 (50%)
320 (40%)
ND
14
17
30
“3.7M”
180
35%
110
a Concentrations given represent concentrations obtained after the indicated time of heating, followed by cooling of the
reaction vessel to 25 °C so that all species are condensed or dissolved into liquid solution phase.
b Reaction vessels (glass ampoules) were oriented vertically.
c Most runs were repeated; results given are the average of two or more runs. Reproducibility averaged ± 3%.
d Volume of pentane in each vial was reduced from 100 ꢂL to 50 ꢂL; thus the propene/pentane ratio was doubled.
When propene pressure is varied from 2 atm to 4 atm at 240 °C (Table 6) the overall rate of
dehydrogenation increases by ca. 2ꢀfold. Further increase in propene pressure to 6 atm results in a
decreased rate, which is comparable to the rate observed with 2 atm of propene. The yield of αꢀolefin,
however, depends significantly upon propene pressure; for example, after 40 min at 240 °C, conversion
was very similar at 2 atm and 6 atm propene (850 – 870 mM), but yields of αꢀolefin were quite different,
11
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 33
12
Kumar et al. Solidꢀgas Phase Dehydrogenation
22ꢀJulꢀ15
1
2
3
4
5
6
7
8
170 mM (22%) and 400 mM (51%), respectively. Reducing the amount of pentane to 50 ꢂL allowed us
to have approximately “7.4 M” propene while working under 6 atm (entry 4, Table 6). This reaction
system, when heated at 240 °C for 80 min, gave the highest yield of 1ꢀpentene yet reported from
transferꢀdehydrogenation, ca. 575 mM, which is about 5.9 times greater than the αꢀolefin yields
obtained in previous reports.3 At lower temperatures, a similar dependence of propene pressure on
yields of 1ꢀpentene was observed (entries 5, 6 and 7, Table 6) as illustrated in Figure 3.
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
(a)
Figure 3. Plot of 1ꢀpentene as fraction of pentenes formed in the dehydrogenation of nꢀpentane
catalyzed by 2ꢀC2H4 under 2 atm, 4 atm, and 6 atm propene at 200 °C
Origin of the high αꢀolefin yields. As discussed above, high yields of αꢀolefin are obtained when
using ethylene as acceptor and under high pressures of propene in particular. Both olefins, but ethylene
in particular, are expected to bind strongly to the pincerꢀiridium complex. This strong binding explains
the relatively low rate of dehydrogenation obtained in the presence of ethylene, given that
(pincer)Ir(ethylene) is presumably not catalytically active. We have recently shown that the
dehydrogenation catalyst (tBu4PCP)Ir (1) catalyzes olefin isomerization via an η3–allyl pathway which, in
turn, proceeds via CꢀH addition prior to olefin coordination.57 Thus olefins like ethylene that bind very
strongly, or high pressures of propene which binds fairly strongly, would be expected to inhibit
isomerization of the αꢀolefin primary product. However, such binding of the acceptor olefins would also
be expected to inhibit (equally) the rate of dehydrogenation; if the rate of isomerization relative to
dehydrogenation were unchanged, the maximum concentration of αꢀolefin would also be unchanged
(although it would of course take more time to reach that maximum).
The conclusion, in our earlier study, of an η3–allyl isomerization pathway being operative for catalyst
1 was based on several lines of evidence.57 In particular, we gave very strong consideration to the most
commonly proposed pathway for olefin isomerization: insertion into an MꢀH bond (e.g. 2,1ꢀaddition of 1ꢀ
alkene), followed by βꢀH migration at C3 to give the more stable doubleꢀbond isomer, 2ꢀalkene); we will
refer to this as a “hydride addition pathway”. Under the conditions of our studies the only observable
resting state was always (tBu4PCP)Ir(1ꢀalkene). Thus, a hydride addition pathway would proceed via a
small, if unobservable, concentration of a catalytically active hydride, most likely (tBu4PCP)IrH2, which
should be present according to eq 1.
(pincer)Ir(1ꢀalkene) + alkane = (pincer)IrH2 + 1ꢀalkene + alkene’
(1)
The concentration of the hydride species would be much greater in alkane than in arene solvent
due to the steady state of eq 1; therefore, isomerization rates would be commensurately much greater if
12
ACS Paragon Plus Environment

Page 13 of 33
Journal of the American Chemical Society
13
Kumar et al. Solidꢀgas Phase Dehydrogenation
22ꢀJulꢀ15
1
2
3
4
such a species were largely responsible for isomerization. In fact we found that rates of 1ꢀalkene
isomerization by 1 were identical in nꢀoctane and pꢀxylene solvent57 (and this result was reproduced
during the present study).
5
6
7
8
In the case of (iPr4PCP)Ir (2), as with 1, the major resting state in the presence of any appreciable
concentration of 1ꢀalkene is the 1ꢀalkene complex (or the propene or ethylene complex in the presence
of these olefins). However, the effect of alkane vs. arene solvent on the rate of isomerization proved to
be very different in the case of 2. Addition of 1ꢀoctene [100 mM] to 2ꢀC2H4 (1 mM) in either nꢀoctane or
pꢀxylene solvent resulted in complete conversion to (iPr4PCP)Ir(1ꢀoctene), without any hydride species
observable in either solvent. However, in contrast with the catalytic behavior of 1, 1ꢀoctene
isomerization is indeed significantly faster in nꢀoctane than in pꢀxylene, by ca. twoꢀfold (Figure 4). This
indicates that dihydride 2ꢀH2 is a much more active catalyst (on a per mol basis) than 2ꢀ(1ꢀoctene).
Nevertheless, given that the small concentration of dihydride would be many times greater in alkane
than in arene (eq 1), the fact that there is only a ca. 2ꢀfold difference indicates that isomerization by 2
does not proceed exclusively via the hydride pathway. Instead, it can be concluded that the observed
isomerization in pꢀxylene solvent is not due to a hydride pathway, but is presumably due to an allyl
pathway; the rate of this pathway might be considered a “baseline”, while the presence of any 2ꢀH2
could add to this baseline rate.
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
PiPr2
Ir
PiPr2
H
+
Ir
R
+ 2
R
H
1ꢀoctene
3ꢀoctenes
4ꢀoctenes
R
R = C6H13
+
PiPr2
PiPr2
pꢀxylene or nꢀoctane solvent, 125 °C
120 mM
2ꢀoctenes
Figure 4. Isomerization of 1ꢀoctene in nꢀoctane and pꢀxylene at 125 °C catalyzed by 2
Ethylene is presumably a better hydrogen acceptor than propene, given that the thermodynamics of
its insertion and hydrogenation are more favorable, and that it is sterically less demanding. Thus any
contribution to isomerization from a hydride pathway would be expected to be minimized by the
presence of ethylene, particularly at high pressure. Indeed, under 4 atm ethylene (and with the reaction
ampoule oriented horizontally so as to maximize surface area and minimize diffusion limitations) the
yield of 1ꢀpentene was the highest we have obtained to date, 520 mM (after condensation) after 180
min (entry 2, Table 7). Figure 5a illustrates that at a given conversion level, the fraction of 1ꢀpentenes is
significantly higher when ethylene is the acceptor instead of propene. The apparent suppression of
isomerization via the hydride pathway is apparently maximized even at only 2 atm ethylene; thus at a
higher ethylene pressure (4 atm) the 1ꢀpentene fraction is not significantly greater, at a given
conversion level, than under 2 atm ethylene (Figure 5b).
13
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 33
14
Kumar et al. Solidꢀgas Phase Dehydrogenation
22ꢀJulꢀ15
1
2
Table 7. Dehydrogenation of n–pentane [8.7 M] with ethylene at 240 °C catalyzed by 2.a,b
3
4
5
6
(
iPr4PCP)Ir(C2H4)
(1.0 mM)
+
+
+
+
sealed vial aligned
horizontally
8.7 M
1ꢀpentene
(E+Z)ꢀ2ꢀpentene
1,3ꢀpentadiene
200 °C or 240 °C
7
8
9
Time /
min
Total Olefinsc
/ mM
1ꢀPentene / mM
(% total monoenes)
% Ethylene
conversion (by GC)
Dienes
/ mM
Entry
1
Conditions
240 °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
10
560
350 (64%)
63
24
2 atm
“1.2 M”
40
1290
115 (10%)
100
82
2
3
240 °C
4 atm
10
40
130
310
114 (88%)
240 (79%)
520 (52%)
8
0
4
16
60
“2.4 M”
180
1100
90
200 °C
2 atm
10
40
26
22 (85%)
90 (80%)
190 (72%)
250 (61%)
85%
80%
24
0
0
114
260
420
“1.2M”
80
3
180
61%
14
4
200 °C
4 atm
“2.4”
10
80
3
3 (>97%)
26 (93%)
170 (78%)
280 (64%)
ND
1
0
0
28
600
1200
220
460
8
3
15
11
a Concentrations given represent concentrations obtained after the indicated time of heating, followed by cooling of the
reaction vessel to 25 °C so that all species are condensed or dissolved into liquid solution phase.
b Reaction vessels (glass ampoules) were oriented horizontally.
c Most runs were repeated; results given are the average of two or more runs. Reproducibility averaged ± 3%.
(a)
(b)
Figure 5. Plot of 1ꢀpentene as percentage fraction of pentenes formed in the dehydrogenation of nꢀ
pentane catalyzed by 2ꢀC2H4 at 200 °C under (a) 2 atm propene and ethylene, (b) 2 atm and 4 atm
ethylene.
With 2 atm instead of 4 atm ethylene (entry 1, Table 7), the reaction rate is expected to be
somewhat (up to 2ꢀfold) faster due to decreased inhibition. Surprisingly, however, under these
conditions the reaction was found to be ca. 4ꢀfold faster. We suspect this result is due to a diffusion
limitation which lowers the local ethylene concentration and thus (somewhat counterꢀintuitively)
14
ACS Paragon Plus Environment

Page 15 of 33
Journal of the American Chemical Society
Kumar et al. Solidꢀgas Phase Dehydrogenation
15
22ꢀJulꢀ15
1
2
3
produces a rate even faster than would be predicted. With respect to mechanistic study, the value of
this experiment is thus doubtful.
4
5
6
Overall, the picture that emerges from these studies is illustrated in Scheme 2.
7
Scheme 2
8
9
R
R
(
iPr4PCP)Ir(1ꢀalkene)
R
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(
iPr4PCP)IrH2
(
iPr4PCP)Ir
(
iPr4PCP)Ir(CH2=CHA)
R
fast
olefin
isomztn
slow
olefin
isomztn
R
= ethylene or
propene
A
A
A
R
R
Implicit in the above explanation of selectivity is a model of reactivity that is at least qualitatively not
different from the behavior of the catalyst in solution. We therefore further explored the liquid phase
reactivity with the same acceptors and nꢀalkanes as an obvious test of this model. Moreover, the
solution phase does not present the issue of irreproducible surface area and physical distribution of the
catalyst, thus allowing a more rigorously quantitative study of the reaction kinetics.
2ꢀcatalyzed, solutionꢀphase, transfer dehydrogenation of nꢀoctane was conducted with ethylene and
with propene as acceptor. As with the gasꢀsolid phase reactions, a glass ampoule was charged with
catalyst, alkane, and olefin acceptor, and then sealed. The ampoule was rotated in the oven to promote
gasꢀliquid mixing. Transferꢀdehydrogenation was run with 2 atm, 4 atm, and 6 atm propene pressure.
Higher propene pressures resulted in somewhat lower rates, indicating that a significant fraction of the
catalyst was present as the outꢀofꢀcycle species 2ꢀpropene. The effect on the rate from a 3ꢀfold
increase in Ppropene, however, was less than a factor of 3 (<2ꢀfold), suggesting that 2ꢀ propene is not the
only major species present.
If we assume that 2ꢀpropene is catalytically inactive with respect to both octane dehydrogenation
and 1ꢀalkene isomerization, then an increase in [2ꢀpropene] (effectuated by increasing the propene
pressure) is expected to lower the rates of both processes equally. Assuming that fragment 2 can react
with either nꢀalkane (leading to dehydrogenation) or with αꢀolefin (leading to isomerization), an increase
in Ppropene would not be expected to have any direct effect on the ratio of dehydrogenation to
isomerization if these were the only paths leading to dehydrogenation and isomerization, respectively.
In that case, at a given level of conversion (i.e. after dehydrogenation had proceeded to a given extent),
the degree of isomerization would be proportional, and the fraction of unisomerized 1ꢀalkene product
formed would be independent of propene pressure. However, as seen in Table 8 and Figure 6a, at any
given level of conversion the fraction of αꢀolefin is in fact higher under the higher propene pressure[s].
This is explained in terms of the left side of Scheme 2: if a small concentration of 2ꢀH2 is responsible for
a significant fraction of the isomerization, then increasing propene concentration will decrease the
steadyꢀstate concentration of 2ꢀH2, thus resulting in decreased isomerization and higher αꢀolefin yields.
When ethylene is used as the hydrogen acceptor, the yields of 1ꢀoctene show a weak dependence
on ethylene pressure. At high ethylene pressures, the yields of 1ꢀoctene are somewhat higher than with
15
ACS Paragon Plus Environment

Journal of the American Chemical Society
Kumar et al. Solidꢀgas Phase Dehydrogenation
Page 16 of 33
16
22ꢀJulꢀ15
1
2
3
propene but the difference between the two acceptors is not large. We interpret these results as
approaching a regime where the concentration of 2ꢀH2 is too low to contribute significantly to
isomerization; in such a regime, the ratio of dehydrogenation to isomerization should be independent of
ethylene or propene concentration.
4
5
6
7
8
9
Table 8. Dehydrogenation of nꢀoctane [6.2 M]a catalyzed by 2 (1 mM) under varying propene or
ethylene pressures at 180 °C (spinning reaction vials).b
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
Acceptor,
Pressure
1ꢀOctene / mM
(Fraction %)
% Acceptor
conversion (by GC)
Time / min
Total olefinsc / mM
Dienes /mM
propene
2 atm
5
340
590
93 (29%)
120 (22%)
120 (14%)
80 (8%)
18
32
60
82
17
49
10
20
30
“1.2 M”
990
150
280
1240
5
270
490
92 (35%)
165 (30%)
146 (20%)
120 (11%)
< 5
10
15
10
25
propene
4 atm
10
20
50
90
840
30
90
“2.4 M”
1455
1980
60
345
750
100
5
200
500
70 (37%)
160 (33%)
190 (24%)
190 (16%)
170 (12%)
4
10
25
propene
6 atm
10
30
60
120
9
870
18
30
46
88
“3.6 M”
1410
1860
250
480
10
40
23
87
9 (40%)
35 (40%)
76 (37%)
130 (32%)
160 (26%)
160 (18%)
2
1
2
ethylene
2 atm
8
110
250
480
840
210
430
690
1040
17
38
55
75
8
“1.2 M”
25
64
150
10
40
12
24
8 (66%)
15 (63%)
62 (67%)
85 (52%)
150 (37%)
210 (24%)
200 (21%)
ꢀ
0
0
ethylene
4 atm
1
80
92
5
1
“2.4 M”
200
480
960
1440
170
430
960
1100
7
3
17
40
44
18
100
140
40
<4
33
3 (83%)
20 (67%)
78 (56%)
127 (29%)
130 (15%)
ND
3
0
0
(
tBu4PCP)Ir (1)
ethylene
2 atm
180
960
156
455
1000
15
35
75
2
2160
5400
20
135
“1.2 M”
a Concentrations given represent concentrations obtained after the indicated time of heating, followed by cooling of the
reaction vessel to 25 °C so that all species are condensed or dissolved into liquid solution phase.
b Reaction vessels (glass ampoules) were oriented horizontally.
c Most runs were repeated; results given are the average of two or more runs. Reproducibility averaged ± 1%.
16
ACS Paragon Plus Environment

Page 17 of 33
Journal of the American Chemical Society
17
Kumar et al. Solidꢀgas Phase Dehydrogenation
22ꢀJulꢀ15
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
(a)
(b)
(c)
(d)
Figure 6. Plot of 1ꢀoctene as percentage fraction of octenes formed in the dehydrogenation of nꢀoctane
catalyzed by 2ꢀC2H4 at 180 °C under (a) 2 atm, 4 atm, and 6 atm propene, (b) 2 atm and 4 atm
ethylene, (c) 2 atm ethylene and propene, (d) data from all plots combined.
The reaction of nꢀoctene with propene is probably too fast at 180 °C to obtain good kinetics data.
For this reason, and also to investigate the effect of temperature further, we also conducted runs under
propene at 160 °C. We also varied temperature with ethylene as acceptor, but in this case we raised
the temperature to 200 °C, since the kinetics with ethylene were quite slow at 180 °C. Generally
speaking, higher selectivity is of course associated with lower temperature – and this is particularly true
in the case of formation of thermodynamically less favorable products. Inspection of Tables 8 and 9,
however, reveals that at any given level of conversion, with any given pressure of either propene or
ethylene, higher temperatures are found to give greater fractions of αꢀolefin. Accordingly, at 200 °C and
at the highest pressure of ethylene used (4 atm), an αꢀolefin yield as high as 250 mM (250 TO) is
obtained, a factor of 2.5 greater than any previously reported value in solution.
17
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 33
18
Kumar et al. Solidꢀgas Phase Dehydrogenation
22ꢀJulꢀ15
1
2
3
Table 9. Dehydrogenation of nꢀoctane (6.2 M) catalyzed by 2 (1.0 mM) under propene at 160 °C, and under
ethylene at 200 °C
4
5
% Acceptor
conversion (by GC)
Dienes
/ mM
1ꢀOctene / mM
(Selectivity %)
Acceptor
Conditions
Time /min
Total Olefins / mM
6
7
8
9
10
20
96
190
450
720
1310
55 (57%)
76 (41%)
100 (24%)
105(16%)
52 (5%)
5
0
3
160 °C
2 atm
9
propene
40
23
40
90
21
66
340
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 M”
90
180
10
20
55
37 (67%)
55 (47%)
1
2
0
2
160 °C
6 atm
118
270
540
690
850
propene
40
110 (40%)
120 (25%)
140 (22%)
140 (19%)
4
7
“3.6 M”
80
11
12
20
40
50
110
120
180
10
20
170
360
90 (56%)
140 (40%)
200 (40%)
195 (24%)
2(<1%)
16%
23%
45%
75%
100%
2
200 °C
2 atm
14±1
28±1
96±2
206±6
ethylene
ethylene
40
580
“1.227 M
90
950
120
1190
10
40
40
28 (74%)
96 (56%)
2%
7%
1
200 °C
4 atm”
180
390
560
770
910
4±1
80
170 (44%)
210 (40%)
250 (34%)
230 (27%)
17%
25%
32%
38%
13±1
30±4
55±1
55±1
“2.45 M”
120
180
280
a Concentrations given represent concentrations obtained after the indicated time of heating, followed by cooling of the
reaction vessel to 25 °C so that all species are condensed or dissolved into liquid solution phase.
b Reaction vessels (glass ampoules) were oriented horizontally.
c Most runs were repeated; results given are the average of two or more runs. Reproducibility averaged ± 1%.
Figure 7. Plot of 1ꢀalkene as percentage fraction of alkenes formed in the dehydrogenation of nꢀoctane
and nꢀpentane catalyzed by 2ꢀC2H4 at 200 °C under 2 atm ethylene.
18
ACS Paragon Plus Environment

Page 19 of 33
Journal of the American Chemical Society
Kumar et al. Solidꢀgas Phase Dehydrogenation
19
22ꢀJulꢀ15
1
2
Computational results, discussion, and overview
3
4
5
6
7
8
9
Quantum mechanical calculations (DFT, see Computational for details) modeling (in vacuo) the
iPr4PCP (2) and tBu4PCP (1) systems offer significant insight into the surprisingly high yields of αꢀolefin
obtained in this work, both in solution and solidꢀgas phase experiments, and enable us to put the
mechanistic hypotheses advanced above on a much firmer footing. Moreover, while we have previously
reported that 2 is a more effective catalyst than 1 for several alkane dehydrogenation reactions,2,6,19 the
difference was not as pronounced as in much of the present experimental work; the present set of DFT
calculations help explain this observation as well.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In the following, we use hexane/1ꢀhexene as our representative nꢀalkane/1ꢀalkene pair. If we
consider the simple case of an nꢀalkane/1ꢀalkene transferꢀdehydrogenation cycle, the resting state of
either catalyst 1 or 2 is the corresponding 1ꢀalkene complex, while the rateꢀdetermining step is βꢀH
elimination of the 1ꢀalkyl iridium hydride CꢀH bond addition product (Schemes 3 and 4). For (tBu4PCP)Ir,
1, the difference in free energy between the 1ꢀalkene complex resting state and the rateꢀdetermining TS
(RDTS) is calculated to be 34.1 kcal/mol (38.9 kcal/mol ꢀ 4.8 kcal/mol) at 220 °C (ꢀG220; most of the
experiments in this work were conducted at 200 °C or 240 °C and for convenience, free energies are
given at the intermediate temperature, 220 °C; Scheme 3). For catalyst 2 (Scheme 4), the difference in
free energy between the RDTS for dehydrogenation and the 1ꢀalkene complex resting state is ꢀG220
33.1 kcal/mol (25.4 kcal/mol ꢀ (ꢀ7.7) kcal/mol); the computed difference between the two catalysts,
ꢀꢀG220 = 1.0 kcal/mol, is consistent with the experimentally observed significant but not extreme
difference in catalytic activity of 2 vs. 1 for nꢀalkane/1ꢀalkene transferꢀdehydrogenation.6
=
Scheme 3. Calculated pathway with relative free energies (220 °C) for transferꢀdehydrogenation and αꢀ
olefin isomerization catalyzed by 1. Free energies of key resting states and rateꢀlimiting (determining)
transition states (RDTS) shown in red.
P
P
nꢀBu
(
tBu4PCP)Ir (1) G (220 °C)
Ir
Ir
5.3
kcal/mol
P
P
4.8
H
P
kcal/mol
‡
‡
allyl isomerization path
P
Ir
± 1ꢀhexene
nꢀBu
nꢀBu
Ir
nꢀPr
nꢀPr
P
H
P
nꢀPr
P
P
P
PtBu2
P
34.7 kcal/mol
Ir
32.5 kcal/mol
± ethylene
Ir
Ir
Ir
Ir
RDTS
H
H
nꢀPr
P
P
P
PtBu2
P
14.7
kcal/mol
17.2
kcal/mol
ꢀ1.9
kcal/mol
ꢀ3.6 kcal/mol
(0.0)
‡
P
Ir
nꢀBu
nꢀBu
P
A
‡
Ir
C2H4: 25.2 kcal/mol
C3H8: 33.8 kcal/mol
H
H
30.3
kcal/mol
P
P
Functional: PBE
Basis Set: LANL2TZ/6ꢀ311G(d,p)
P
A
P
Ir
C2H4: 18.6 kcal/mol
C3H8: 29.6 kcal/mol
Free energies relative to free (tBu4PCP)Ir
‡
H
H
A
Ir
P
P
H
P
Ir
26.3 kcal/mol
H
P
C2H4: 26.0
C3H8: 36.4
TBE: 47.0
P
‡
H
P
H
nꢀBu
‡
‡
H
P
Ir
P
Ir
Ir
P
Ir
P
H
H
nꢀPr
P
H
nꢀBu
P
P
H
P
Ir
nꢀPr
H
H
H
44.0 kcal/mol
RDTS
37.8 kcal/mol
Ir
38.9 kcal/mol
RDTS for
H
P
P
3.0 kcal/mol
ꢀ0.7 kcal/mol
32.0 kcal/mol
dehydrogenation
+
hydride isomerization path
19
ACS Paragon Plus Environment

Journal of the American Chemical Society
Kumar et al. Solidꢀgas Phase Dehydrogenation
Page 20 of 33
20
22ꢀJulꢀ15
1
2
3
4
Scheme 4. Calculated pathway with relative free energies (220 °C) for transferꢀdehydrogenation and αꢀ
olefin isomerization catalyzed by 2. Free energies of key resting states and rateꢀlimiting (determining)
transition states (RDTS) shown in red.
5
6
P
P
nꢀBu
7
8
9
Ir
Ir
(
iPr4PCP)Ir (2) G (220 °C)
allyl isomerization path
ꢀ6.4
kcal/mol
P
P
ꢀ7.7
kcal/mol
H
P
‡
‡
P
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
± 1ꢀhexene
nꢀBu
nꢀBu
Ir
Ir
nꢀPr
nꢀPr
P
H
nꢀPr
P
P
P
P
P
PiPr2
± ethylene
19.2 kcal/mol
RDTS
Ir
Ir
19.2 kcal/mol
Ir
Ir
Ir
H
nꢀPr
H
P
P
P
P
PiPr2
ꢀ1.8
kcal/mol
0.9
kcal/mol
ꢀ3.7
kcal/mol
ꢀ11.6 kcal/mol
(0.0)
‡
P
Ir
nꢀBu
nꢀBu
P
A
‡
Ir
C2H4: 16.6 kcal/mol
C3H8: 25.3 kcal/mol
H
H
22.0
P
P
kcal/mol
Functional: PBE
Basis Set: LANL2TZ/6ꢀ311G(d,p)
Free energies relative to free (iPr4PCP)Ir
P
A
P
Ir
C2H4: 11.1 kcal/mol
C3H8: 19.2 kcal/mol
‡
P
H
A
Ir
P
H
H
P
Ir
18.7 kcal/mol
H
P
C2H4: 18.4
C3H8: 25.2
TBE: 30.0
‡
P
P
P
H
‡
nꢀBu
‡
H
P
Ir
P
Ir
H
Ir
P
H
H
nꢀPr
P
H
nꢀBu
H
P
Ir
P
nꢀPr
Ir
H
H
H
26.2 kcal/mol
RDTS
25.1 kcal/mol
Ir
P
25.4 kcal/mol
RDTS for
dehydrogenation
H
P
P
1.3 kcal/mol
ꢀ2.4 kcal/mol
20.6 kcal/mol
+
hydride isomerization path
However, catalyst 1 binds much more strongly to ethylene than to αꢀolefin (ꢀꢀG220 = 8.4 kcal/mol,
Scheme 3). The overall barrier for dehydrogenation in the ethylene reaction (GTSꢀbetaꢀelim – G(1ꢀethene))
is therefore very high, ꢀG220 = 42.5 kcal/mol. In the case of the less crowded catalyst 2, ethylene binds
only 3.9 kcal/mol more strongly than does 1ꢀhexene, and ꢀG220 for the ethylene complex vs. the βꢀH
elimination RDTS is 37.0 kcal/mol. Thus, in comparing catalysis by 2 vs. 1, ꢀꢀG220 = 5.5 kcal/mol in the
case when ethylene is the acceptor; this corresponds to a very large difference in reaction rate between
the two catalysts, in accord with observations (see Table 8, for example).
Both catalysts 1 and 2 are kinetically highly regioselective for the dehydrogenation of nꢀalkanes to
give αꢀolefins;8 however, due to subsequent doubleꢀbond isomerization, with neither catalyst (nor with
any other alkane dehydrogenation catalyst) have αꢀolefin yields previously been reported above 100
mM.8 The following computational results may explain why the conditions in the present work afford
much higher αꢀolefin yields, in solution and especially in the solidꢀgas system.
As discussed above, we have previously determined that the major pathway for olefin isomerization
by 157 proceeds via formation of an iridium allyl complex (involving addition of the allylic sp3 CꢀH bond
to 14ꢀelectron fragment 1). While allylꢀbased olefin isomerization pathways have long been known,58ꢀ64
the more commonly proposed “hydride isomerization pathway” involves insertion into a metalꢀH bond,
followed by βꢀH elimination at the adjacent position (e.g. 2,1ꢀaddition of MꢀH to an αꢀolefin), and then
2,3ꢀelimination.63ꢀ71 This mechanism would seem to be a particularly likely path for a transferꢀ
dehydrogenation system in which insertion of olefin (the acceptor) into MꢀH bonds is a necessary part
20
ACS Paragon Plus Environment

Page 21 of 33
Journal of the American Chemical Society
Kumar et al. Solidꢀgas Phase Dehydrogenation
21
22ꢀJulꢀ15
1
2
3
of the catalytic cycle. The computational results shown in Scheme 3, however, explain why the “allyl
isomerization pathway” predominates for isomerization catalyzed by 1.
4
5
6
7
8
9
Dehydrogenation of nꢀalkane substrate yields (pincer)IrH2. The predicted importance of the hydride
isomerization pathway can be expressed in terms of the three possible reactions of the dihydride
(Scheme 3). The RDTS for the hydride isomerization pathway by (tBu4PCP)IrH2 (1ꢀH2) (3,2ꢀβꢀHꢀ
elimination) has a free energy of 44.0 kcal/mol (all energies are expressed relative to the free (pincer)Ir
complex plus appropriate substrates unless noted otherwise). Alternatively, 1ꢀH2 can hydrogenate
acceptor to complete one catalytic cycle. The respective RDTS free energies for hydrogenation are
much lower: 26.0 kcal/mol and 36.4 kcal/mol for ethylene or propene, and 38.9 kcal/mol for higher αꢀ
olefin acceptors (the reverse of the βꢀH elimination step in dehydrogenation), respectively. Thus, the
calculations predict that isomerization by 1ꢀH2 would be negligible in the presence of these acceptors.
Even if the barrier to hydrogenation of the acceptor were much higher, as is calculated in the case of
TBE at 47.0 kcal/mol, back reaction with the αꢀolefin product (GRDTS = 38.9 kcal/mol) followed by
isomerization via the allyl path (GRDTS = 34.7 kcal/mol) would be much more rapid than isomerization by
the hydride path.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In the case of catalyst 2 the free energy of the RDTS for the hydride isomerization path is much
lower (26.2 kcal/mol, Scheme 4) than that of 1 and, importantly, in contrast with 1, much lower relative
to the competitive hydrogenations of acceptor or αꢀolefin. As with 1ꢀH2, hydrogenation of ethylene still
has a RDTS of much lower energy (18.4 kcal/mol); but for the much less crowded dihydride 2ꢀH2, the
RDTS for hydrogenation of propene (25.3 kcal/mol) and for the back reaction with αꢀolefin (25.4
kcal/mol) are quite comparable to the hydride isomerization RDTS. The calculations thus indicate that
for catalyst 2 in the presence of ethylene, or in the limit of very high propene concentration or pressure,
isomerization via the hydride path will not play a large role. In the case of low propene concentration or
pressure, however, or in the case of an acceptor with a higher barrier to hydrogenation (e.g. TBE), the
hydride isomerization path can be significant. Moreover, at lower concentrations of propene or in the
case of a poor acceptor, the back reaction of αꢀolefin product will predominate over the forward
hydrogenation of acceptor. This backꢀreaction lowers the net rate of hydrogenation, while isomerization
can still proceed via the allyl pathway.
The relative rates of isomerization and dehydrogenation (which determine the ultimate buildꢀup of
αꢀolefin) are expressed algebraically in eq 2, based on the rate constants indicated in Scheme 5.
Scheme 5. Simplified scheme illustrating relative rates of isomerization and dehydrogenation as well as
corresponding rate constants. (Terms in the numerator of equation 2 are colorꢀcoded to indicate their
origin in the corresponding isomerization pathways depicted in the scheme.)
[Ir]
+
R
[Ir](olefin*)
k2[αꢀolefin]
allyl isomzt'n path
allylꢀpathway isomerization: k2[αꢀolefin][Ir]
hydrideꢀpathway isomerization: k3[αꢀolefin][IrH2]
net dehydrogenation = k1[RH][Ir] ꢀ kꢀ1[αꢀolefin][IrH2]
[Ir]
k1[RH]
AH2
k4[A]
k1[Ir][RH]
[IrH2] =
kꢀ1[αꢀolefin]
{kꢀ1[αꢀolefin] + k4[A]}
acceptor (A)
hydride isomzt'n path
[Ir]H2
R
[Ir]H2
(αꢀolefin)
k3[αꢀolefin]
[Ir] = (pincer)Ir
+
R
(olefin* = acceptor or product
depending on their concentration
and binding ability)
21
ACS Paragon Plus Environment

Journal of the American Chemical Society
Kumar et al. Solidꢀgas Phase Dehydrogenation
Page 22 of 33
22
22ꢀJulꢀ15
1
2
3
4
5
6
7
k3[αꢀolefin]k1[RH]
k2[αꢀolefin] +
{kꢀ1[αꢀolefin] + k4[A}]
isomerization
=
net dehydrogent'n
kꢀ1[αꢀolefin]
k1[RH] 1 ꢀ
{kꢀ1[αꢀolefin] + k4[A]}
(2)
8
9
From eq 2 it can be seen that in the limit of k4[A] >> kꢀ1[αꢀolefin] and k4[A] >> k3[αꢀolefin] (i.e.
conditions of fast hydrogenation of the sacrificial acceptor) the isomerization/dehydrogenation ratio
reduces to k2[αꢀolefin]/k1[RH] (eq 3). Eq 3 reflects the competing reactions of (pincer)Ir (present in a
very small steadyꢀstate concentration) with αꢀolefin (isomerization) vs. alkane (dehydrogenation).
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
k2[αꢀolefin]
If k4[A] >> kꢀ1[αꢀolefin] and k4[A] >> k3[αꢀolefin]: isomerization/dehydrogenation =
k1[RH]
(3)
When ethylene is the acceptor, the difference in free energies of the respective RDTS’s is very large
for 1, ca. 13 kcal/mol, and large even for 2 (ca. 7 kcal/mol); thus, either catalyst should be in the fastꢀ
acceptor hydrogenation limit and eq 3 is expected to be applicable, with no significant contribution to
isomerization from a hydride pathway.
In the case of propene acceptor and (tBu4PCP)Ir catalyst (1) the difference in free energies for the
RDTS’s of propene hydrogenation vs. isomerization via the hydride pathway is also very large (7.6
kcal/mol). The difference between propene hydrogenation vs. backꢀreaction with αꢀolefin is 2.5
kcal/mol. Hence, as long as propene concentration is comparable to αꢀolefin concentration the backꢀ
reaction rate will be small, and the system will still be in the fastꢀacceptor hydrogenation limit described
by eq 3. Similarly, when αꢀolefin is used as acceptor, the relative rate of back reaction will be small as
long as the acceptor is present in excess.
In the case of propene acceptor and the less bulky (iPr4PCP)Ir catalyst (2), however, the calculated
difference in free energies for the RDTS’s of propene hydrogenation vs. isomerization via the hydride
pathway is small (1.0 kcal/mol), and the difference between propene hydrogenation vs. backꢀreaction
with higher αꢀolefin is negligible (0.2 kcal/mol). In both cases this represents a competition between the
reaction of 2ꢀH2 with propene vs. αꢀolefin. Thus, only in the limit of very high propene concentration or
pressure (relative to concentration or pressure of α−olefin) will the isomerization/dehydrogenation ratio
approach the lower limit of eq 3. As the reaction progresses and reaches the limit of complete
consumption of propene the isomerization/dehydrogenation ratio will rapidly increase.
Regarding ethylene or propene acceptors, it should be noted that as the overall rate of
dehydrogenation is inhibited by their binding to the iridium center, as the acceptor concentration or local
pressure is lowered, its rate of consumption is increased. Thus, slow diffusion could result in a (counterꢀ
intuitive) fasterꢀthanꢀexpected rate (either in the gas or solution phase) and a selfꢀpropagating cycle
which in turn could further lower concentrations of acceptor; consecutively, this would result in a higherꢀ
thanꢀexpected rate of isomerization (eq 2) as well as a fast but diffusionꢀlimited rate of hydrogenation.
The use of a higher αꢀolefin as sacrificial acceptor (one with chain length different from the nꢀalkane
substrate so that the reaction is nonꢀdegenerate) allows simplification of eq 2, since we can then
assume k4 = kꢀ1. If we consider the point at which the acceptor concentration is equal to the product (αꢀ
olefin) concentration, eq 2 can be simplified to eq 4:
22
ACS Paragon Plus Environment

Page 23 of 33
Journal of the American Chemical Society
Kumar et al. Solidꢀgas Phase Dehydrogenation
23
22ꢀJulꢀ15
1
2
3
4
isomerization
2k2[αꢀolefin]
k3
If [αꢀolefin acceptor] = [αꢀolefin product]:
=
(4)
+
k1[RH]
kꢀ1
net dehydrogent'n
5
6
7
8
In the case of catalyst 1, the term k3/kꢀ1 is predicted to be negligible due to the RDTS difference of 5.1
kcal/mol for the respective steps. For 2 the calculated difference (0.8 kcal/mol) is small, even within the
range of error of the calculations, consistent with a significant contribution to isomerization from the
hydrideꢀisomerization pathway under this (fairly typical) set of conditions. This is also consistent with
the results of the isomerization experiment in which only αꢀolefin is present (in that case αꢀolefin of only
one chain length is present but that will not affect the rates of isomerization resulting from the different
isomerization pathways.)
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
Note that in the case of catalyst 2 the use of highly reactive acceptors suppresses the dihydride
path, which could otherwise play a significant role. But even independent of suppressing the hydride
path, in the case of 1 or 2, higher concentrations of more active acceptors (i.e. high values of k4[A]) will
favor a higher αꢀolefin fraction of total olefin produced by disfavoring the backꢀreaction of dihydride with
αꢀolefin product. This is reflected in eq 2 in that, even in the limit of k3 = 0, the ratio of isomerization to
net dehydrogenation (eq 2) is still inversely dependent on k4[A].
(Pincer)Irꢀcatalyzed alkane dehydrogenation has been of particular interest in the context of alkane
metathesis in which (pincer)Ir catalysts operate in tandem with olefin metathesis catalysts.26 As noted
above, the hydride isomerization pathway is suppressed by the presence of effective hydrogenꢀ
acceptors in high concentration. In the course of an alkane metathesis reaction, however, the steadyꢀ
state concentration of olefin is quite low, and the conditions favor the buildꢀup of dihydride complex.
Given that both catalyst 1 and 2 dehydrogenate alkanes with high regioselectivity for the terminal
position, these results may well explain why catalyst 1 gives much better yields of C2nꢀ2 product in
alkane metathesis (e.g. nꢀdecane from nꢀhexane) than does 2.6 Indeed, although 2 and several other
catalysts have shown high regioselectivity for dehydrogenation, catalyst 1 has proven nearly unique
with respect to good selectivity in alkane metathesis;6,20,26 a possible explanation is that the dihydride
isomerization path in particular is anomalously unfavorable for the highly crowded catalyst 1.
The effect of the nature and concentration of acceptor on the αꢀolefin fraction as elucidated above
may offer insight into the higher yields of αꢀolefin obtained in the gas phase vs. liquid. In a gasꢀphase
experiment the ratio of total acceptor to αꢀolefin present in the reaction vessel equals the relative
concentrations of these species to which the catalyst is exposed. In the solution phase experiments,
however, while essentially all catalyst and αꢀolefin are in the solution phase, a large fraction of the
ethylene or propene acceptor is in the gas phase, thus biasing the system toward isomerization vs.
hydrogenation.
We note one additional effect which the DFT calculations suggest would contribute to the high
yields of αꢀolefins reported in this work. High temperature is generally associated with a lack of
selectivity. However, the RDTS (βꢀH elimination) for the dehydrogenation of nꢀalkane by 2 has an
enthalpy barrier calculated to be 4.4 kcal/mol greater than that of the RDTS for αꢀolefin isomerization
catalyzed by 2. Higher temperatures should thus favor dehydrogenation vs. isomerization, and
ultimately the apparent “selectivity” for αꢀolefin production. The error in the calculated difference in
enthalpy between these very different TSs (allylꢀisomerization vs. βꢀHꢀelimination) is surely large
relative to the small difference itself, but taking the calculated value of 4.4 kcal/mol as a “best guess”,
we can consider the effect of conducting the reaction at 220 °C, for example, compared with a more
typical reaction temperature of 150 °C. The predicted change in the ratio of dehydrogenation to
isomerization via the allyl pathway is significant, viz. a factor of exp[(4.4 kcal•molꢀ1/R)(1/T1 – 1/T2)] = 2.1
23
ACS Paragon Plus Environment

Journal of the American Chemical Society
Kumar et al. Solidꢀgas Phase Dehydrogenation
Page 24 of 33
24
22ꢀJulꢀ15
1
2
3
4
5
6
(T1 = 423 K; T2 = 493 K). Note, however, that the RDTS for the hydride isomerization pathway has a
slightly higher calculated enthalpy than the RDTSs for the competitive hydrogenation reactions; thus,
the hydride pathway will be favored by higher temperature, highlighting further the importance of the
use of highly effective hydrogen acceptors in obtaining high yields of αꢀolefin.
7
8
CONCLUSIONS
9
We report that pure solid phase (pincer)Ir catalysts are highly effective for the dehydrogenation of
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
nꢀalkanes in the gas phase. (iPr4PCP)Ir (2) was found to be particularly effective for this purpose, while
commonly used bulkier catalysts such as (tBu4PCP)Ir (1) or (tBu4POCOP)Ir are much less effective. High
selectivity for αꢀolefin (the thermodynamically least stable doubleꢀbond isomer) is obtained,
demonstrating that the solid catalyst is operating as the molecular species. Remarkably, the fractional
yields of αꢀolefin obtained from the heterogeneous systems are actually much greater than have been
previously reported with homogeneous solutionꢀphase systems.
In an effort to elucidate the origin of the unusually high αꢀolefin fraction, as well as the much greater
reactivity of the less crowded catalysts, we conducted solutionꢀphase studies and DFT calculations on
complexes 1 and 2. With ethylene as hydrogen acceptor the much greater reactivity of complex 2 is
well explained by the DFT calculations. The difference in energy between the RDTS for
dehydrogenation and the ethyleneꢀbound resting state is calculated to be much greater for the bulky
complex 1 than for 2; in the case of higher olefins the difference is much smaller. This effect is
applicable to propene also, but to a much lesser extent. However, decomposition of catalyst 1 seems to
be promoted by propene via a mechanism that we do not yet understand.
While both 1 and 2 are known to be regioselective for dehydrogenation of nꢀalkanes to give αꢀ
olefins, yields of αꢀolefin are limited by doubleꢀbond isomerization; the highest yield of αꢀolefin
previously reported in solution experiments was 97 mM (obtained with the use of 2). We have reported
that the mechanism of isomerization in the case of catalyst 1 proceeds entirely by reaction of the 14ꢀ
electron fragment 1 with olefin via an allyl intermediate and not via the more typical hydride class of
mechanism. DFT calculations show that the hydride pathway is much more competitive in the case of 2;
this is supported by experiments showing that 1ꢀoctene is isomerized ca. 2ꢀfold more rapidly by 2 in nꢀ
octane vs. pꢀxylene solvent (whereas, in the case of complex 1, the rate of isomerization in these two
solvents is identical).
The contribution of the hydride pathway to isomerization is dependent upon a competition for the
dihydride complex between hydrogenation of acceptor, hydrogenation of αꢀolefin (i.e. backꢀreaction),
and 2,1ꢀinsertion of the αꢀolefin leading to isomerization. DFT calculations indicate that the reaction of
either 1ꢀH2 or 2ꢀH2 with ethylene is much more rapid than either backꢀreaction with αꢀolefin or
isomerization of αꢀolefin. Hence the use of ethylene as hydrogen acceptor gives the highest yields of αꢀ
olefin in either solutionꢀphase or solidꢀphase experiments; at high pressures, propene gives αꢀolefin
yields that are only slightly lower. As the calculated activation enthalpy for dehydrogenation is higher
than that for isomerization via the allyl pathway, higher temperatures favor the
dehydrogenation/isomerization ratio and therefore higher αꢀolefin yields. Thus the high αꢀolefin yields
obtained from the gasꢀsolid systems with catalyst 2 are in large part simply a result of the conditions
that lead to the gas/solidꢀphase state: use of highly volatile hydrogen acceptors and the high
temperature, both of which mitigate the hydride isomerization pathway. A further advantage of the gas
phase is that the catalyst is exposed to the same ratio of acceptor to αꢀolefin product that is present in
the reaction vessel. In contrast, in the solution runs, the volatile acceptors propene and ethylene are
partitioned largely into the gas phase while the αꢀolefin product remains in solution, thus favoring
24
ACS Paragon Plus Environment

Page 25 of 33
Journal of the American Chemical Society
Kumar et al. Solidꢀgas Phase Dehydrogenation
25
22ꢀJulꢀ15
1
2
3
isomerization via the hydride path, as well as hydrogenation of αꢀolefin (back reaction); both effects
contribute to an increase in the ratio of isomerization to dehydrogenation.
4
5
6
7
8
9
Thus we report a novel molecular gasꢀsolid system which shows high kinetic selectivity for
dehydrogenation of nꢀalkanes at the terminal position. The solid phase itself has little if any effect on the
intrinsic selectivity and reactivity; indeed DFT calculation modeling the system in vacuo capture the key
properties of the catalyst in the gasꢀsolid system as well as in solution. Experiment and calculation have
led to greater insight into the factors that determine the yields of the desirable terminal dehydrogenation
products. We find that it is possible to effectively eliminate one of two pathways for olefin isomerization
with the appropriate conditions and hydrogen acceptor. A focus of further work will be on the design of
catalysts for which the remaining η3ꢀallyl isomerization pathway is less active relative to
dehydrogenation.
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
25
ACS Paragon Plus Environment

Journal of the American Chemical Society
Kumar et al. Solidꢀgas Phase Dehydrogenation
Page 26 of 33
26
22ꢀJulꢀ15
1
2
3
4
5
6
7
8
9
EXPERIMENTAL AND COMPUTATIONAL DETAILS
General. All manipulations were carried out under an inert atmosphere of dry argon either in a glove
box or using a standard double manifold. (tBu4PCP)IrHn,1 (pꢀOMeꢀtBu4PCP)IrHn,72 tBu4POCOP)IrHn,15ꢀ18
tBu3MePCP)IrHn,6 (tBu2Me2PCP)IrHn,6 (iPr4PCOP)Ir(C2H4),19 (pꢀOMeꢀiPr4PCP)Ir(C2H4)6 and
iPr4Anthraphos)Ir(C2H4)73 were synthesized according to literature methods. Anhydrous toluene,
(
(
(
anhydrous benzene, anhydrous acetone, anhydrous triethylamine and anhydrous pentane were
purchased from Sigma Aldrich and used as such. Diꢀisopropylphosphine, [(COD)IrCl]2 and LiBEt3H
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
were purchased from STREM Chemicals and dibromoꢀm–xylene was purchased from Fluka. Ethylene
used for iPr4PCPIr(C2H4) synthesis was 99.5% pure and was supplied by Matheson. Ultra High Purity
Hydrogen from Airgas was used for synthesis of iPr4PCPIrH4. For catalytic studies, anhydrous pentane
and octane obtained from Sigma Aldrich was distilled over NaK alloy and stored over 4Å molecular
sieves. Research grade purity (99.999 %) ethylene and propylene supplied by Matheson were used for
catalytic studies.
Physical Measurements. 1H NMR, 13C(H), 31P(H) NMR, were recorded on Bruker AMX 400 operating
at 400 MHz for 1H NMR, 100 MHz for 13C NMR, and 161.9 MHz for 31P NMR. GC analyses (FID
detection) were performed on a Varian 430– GC instrument fitted with Agilent J&W GSꢀGasPro column
(60 m length x 0.32 mm ID).
Synthesis of C6H4[CH2(PiPr2)]2 (iPr4PCPꢀH) Dibromoꢀmꢀxylene (5.00 g, 19 mmol), diꢀ
isopropylphosphine (4.5 g, 38 mmol) and triethylamine (10.5 ml, 75.5 mmol) were dissolved in 40.0 ml
THF in the glove box. The mixture was refluxed under an argon atmosphere for 24 hours. The reaction
mixture was cooled in an ice bath and the dense white precipitate was separated by cannula filtration.
The solvent from the filtrate was removed under reduced pressure to yield 4.5 g (70%) of colorless oil.
1
31P(H) NMR (C6D6, 161.9 MHz): δ 10.19 (s) H NMR (C6D6, 400 MHz ): δ 6.83‒6.92 (m, 4H, Arene H),
2.73 (s, 4H, CH2P)1.32 (d of sept, JHH= 5.6Hz, JPH= 1.6 Hz, 4H, PCH(CH3)2), 0.73 (d, 5.6 Hz, 12H,
PCH(CH3)2), 0.70 (dd, JHH= 5.6 Hz, JPH= 1.2 Hz, 12H, PCH(CH3)2)
Synthesis of (iPr4PCP)Ir(HCl/Br). Ultra high purity H2 was bubbled into a 35.0 ml toluene solution of
[(COD)IrCl]2 (3 g, 4.3 mmol) and C6H4[CH2(PiPr2)]2 (3 g, 8.8 mmol) for about 15 minutes. The reaction
mixture was then stirred at 80 °C under an H2 atmosphere for 15 h. Solvent was removed from the
resulting red solution to yield a red solid. The red solid was then stirred with 100.0 ml pentane for 30
minutes. The solution was cannula filtered and filtrate was collected. The extraction was repeated five
more times and filtrate was collected. Then the residue was further stirred with 100 ml pentane
overnight and extracted. All filtrate were mixed and the solvent was evaporated to obtain a red solid in
51 % yield (2.6 g). NMR analysis indicated the red solid to be 5:1 mixture of iPr4PCPIr(HCl) and
iPr4PCPIr(HBr)
31P NMR (C6D6, 161.9 MHz): δ 58.40 (HBr complex, d, 12.3 Hz), δ 58.44 (HCl complex, d, 12.3 Hz)
1H NMR (C6D6, 400 MHz ): δ 6.95 (s, 3H, Arene H), 2.84 (d of vt, JPH= 4.0 Hz, JHH=17.6 Hz, 2H, CH2P),
2.73 (d of vt, JPH= 4.4 Hz, JHH=17.6 Hz, 2H, CH2P), 2.71 (m, 2H, PCH(CH3)2), 1.96 (m, 2H, PCH(CH3)2)
1.19 (app. Sext (dqt), 7.7 Hz, 12H, PCH(CH3)2), 0.86‒0.93 (m, 12H, PCH(CH3)2),
[‒36.25 (HCl complex) and ‒38.25 (HBr complex)] (t, JPH= 13.2 Hz, 1H, IrH)
Synthesis of (iPr4PCP)Ir(C2H4). Ethylene was bubbled for about 15 minutes into a reddish 60.0 ml
pentane solution of iPr4PCPIr(HCl/Br) (0.21 g, 0.4 mmol) whereupon the solution turns colorless. To the
26
ACS Paragon Plus Environment

Page 27 of 33
Journal of the American Chemical Society
Kumar et al. Solidꢀgas Phase Dehydrogenation
27
22ꢀJulꢀ15
1
2
3
4
5
6
7
8
above solution 1M LiBEt3H (0.37 ml, 0.4 mmol) was added drop wise under ethylene atmosphere. The
colorless solution then gradually turns brownish. The reaction mixture was stirred overnight under
ethylene atmosphere. Removal of pentane from the filtrate obtained from cannula filtration yields 0.20 g
of a brown solid in quantitative yield. NMR and elemental analysis indicate the formation of expected
compound in >99% purity. Crystals suitable for X–ray analysis were grown by slow evaporation of a 10
mg solution of (iPr4PCP)Ir(C2H4) in 1.0 ml pentane.
9
31P NMR (C6D6, 161.9 MHz): 51.45
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
1H NMR (C6D6, 400 MHz ): 7.45‒7.32 (m, 3H, Arene H), 3.18 (t, J = 3.8 Hz, 4H, CH2P), 3.11 (t, J = 3.0
Hz, 4H, Ir(C2H4)), 2.15 (m, 4H, PCH(CH3)2), 1.15 (dd, J = 14.7, 7.2 Hz, 12H, PCH(CH3)2), 1.00 (dd, J =
13.2, 6.7 Hz, 12H, PCH(CH3)2).
Calcd for C22H39IrP2: C, 47.38; H, 7.05; Found: C, 48.23; H, 7.23
Synthesis of (iPr4PCP)IrH4. Introducing H2 to a pentane solution of (iPr4PCP)Ir(C2H4) into atmosphere
results in formation of (iPr4PCP)Ir(H4) in quantitative yields as an orange solid.
31P NMR (p–xylene–d10, 161.9 MHz) : δ 54.70
1H NMR (p–xylene–d10, 400 MHz ): δ 7.02 (s, 3H, Arene H), 3.17 (vt, JPH= 4.0 Hz, 4H, CH2P), 1.55 (m,
4H, PCH(CH3)2), 1.03 (app. qt, 7.5 Hz, 12H, PCH(CH3)2) 0.97 (app. qt, 7.1 Hz, 12H, PCH(CH3)2), –9.40
(t, 10.2 Hz, IrH4)
Computational. All electronic structure calculations employed the DFT method74 and the PBE75
exchange correlation functional. A relativistic, smallꢀcore ECP and corresponding basis set were used
for the Ir atom (LANL2TZ model)76,77 allꢀelectron 6ꢀ311G(d,p) basis sets were applied to all P, C, and H
atoms.78 The (R4PCP)Ir species was modeled with R = tꢀBu and iꢀPr, the phosphine substituents
actually used in the experiments. Reactant, transition state and product geometries were fully
optimized, and the stationary points were characterized further by normal mode analysis. Expanded
integration grid sizes (pruned (99,590) atomic grids invoked using the integral=ultrafine keyword) were
applied to increase numerical accuracy and stability in both geometry optimizations and normal mode
analysis.79 The (unscaled) vibrational frequencies formed the basis for the calculation of vibrational
zeroꢀpoint energy (ZPE) corrections; standard thermodynamic corrections (based on the harmonic
oscillator/rigid rotor approximations and ideal gas behavior) were made to convert from purely
electronic (reaction or activation) energies (E) to (standard) enthalpies (H) and Gibbs free energies (G;
P = 1 atm).80 H, entropy (S), and G were evaluated at two temperatures, T = 25 °C (= 298 K) and T =
220 °C (= 493 K). The latter T corresponds approximately to the temperature used in the experiments,
and all energy values quoted in the principal text refer to T = 220 °C unless noted otherwise. In
Supporting Information, we tabulate E, H, S, and G at T = 298 K (P = 1 atm) as well as G at T = 220 °C
(P = 1 atm). All calculations were executed using the GAUSSIAN 09 series of computer programs.81
27
ACS Paragon Plus Environment

Journal of the American Chemical Society
Kumar et al. Solidꢀgas Phase Dehydrogenation
Page 28 of 33
28
22ꢀJulꢀ15
1
2
3
4
ASSOCIATED CONTENT
Supporting Information.
Experimental details including gas chromatographic analysis methods, GC profiles, images of the
reaction vessel during the gasꢀphase reaction, NMR spectra of (iPr4PCP)Ir complexes, Xꢀray analysis
data including cif files, calculated molecular energies and geometries relevant to Schemes 3 and 4.
This material is available free of charge via the Internet at http://pubs.acs.org
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
AUTHOR INFORMATION
Corresponding Authors
*kroghjes@rutgers.edu
*alan.goldman@rutgers.edu
Notes
The authors declare no competing financial interests.
ACKNOWLEDGMENT
This work was supported by NSF through the CCI Center for Enabling New Technologies through
Catalysis (CENTC), grant CHEꢀ1205189, and by Chevron Energy Technology Company. We gratefully
thank Jason Hackenberg, Gao Zhuo and Andrew Steffens for donations of catalyst.
REFERENCES
(1) Gupta, M.; Hagen, C.; Flesher, R. J.; Kaska, W. C.; Jensen, C. M. Chem. Commun. 1996, 2083ꢀ2084.
(2) Liu, F.; Goldman, A. S. Chem. Commun. 1999, 655ꢀ656.
(3) Liu, F.; Pak, E. B.; Singh, B.; Jensen, C. M.; Goldman, A. S. J. Am. Chem. Soc. 1999, 121, 4086ꢀ4087.
(4) Xu, W.; Rosini, G. P.; Gupta, M.; Jensen, C. M.; Kaska, W. C.; KroghꢀJespersen, K.; Goldman, A. S.
Chem. Commun. 1997, 2273ꢀ2274.
(5) Zhu, K.; Achord, P. D.; Zhang, X.; KroghꢀJespersen, K.; Goldman, A. S. J. Am. Chem. Soc. 2004, 126,
13044ꢀ13053.
(6) Kundu, S.; Choliy, Y.; Zhuo, G.; Ahuja, R.; Emge, T. J.; Warmuth, R.; Brookhart, M.; KroghꢀJespersen,
K.; Goldman, A. S. Organometallics 2009, 28, 5432ꢀ5444.
(7) Punji, B.; Emge, T. J.; Goldman, A. S. Organometallics 2010, 29, 2702ꢀ2709.
(8) Choi, J.; MacArthur, A. H. R.; Brookhart, M.; Goldman, A. S. Chem. Rev. 2011, 111, 1761ꢀ1779.
(9) Findlater, M.; Choi, J.; Goldman, A. S.; Brookhart, M. In Alkane CꢀH Activation by SingleꢀSite Metal
Catalysis; Pérez, P. J., Ed.; Springer: New York, 2012; Vol. 38.
(10) Haenel, M. W.; Oevers, S.; Angermund, K.; Kaska, W. C.; Fan, H.ꢀJ.; Hall, M. B. Angew. Chem., Intl.
Ed. 2001, 40, 3596ꢀ3600.
(11) Bezier, D.; Brookhart, M. ACS Catal. 2014, 4, 3411ꢀ3420.
(12) Adams, J. J.; Arulsamy, N.; Roddick, D. M. Organometallics 2012, 31, 1439ꢀ1447.
(13) Adams, J. J.; Lau, A.; Arulsamy, N.; Roddick, D. M. Organometallics 2011, 30, 689ꢀ696.
(14) Shi, Y.; Suguri, T.; Dohi, C.; Yamada, H.; Kojima, S.; Yamamoto, Y. Chem.ꢀEur. J. 2013, 19, 10672ꢀ
10689.
(15) GöttkerꢀSchnetmann, I.; Brookhart, M. J. Am. Chem. Soc. 2004, 126, 9330ꢀ9338.
(16) GöttkerꢀSchnetmann, I.; White, P.; Brookhart, M. J. Am. Chem. Soc. 2004, 126, 1804 ꢀ1811.
(17) GöttkerꢀSchnetmann, I.; White, P. S.; Brookhart, M. Organometallics 2004, 23, 1766ꢀ1776.
(18) MoralesꢀMorales, D.; Redon, R.; Yung, C.; Jensen, C. M. Inorg. Chim. Acta. 2004, 357, 2953ꢀ2956.
(19) Ahuja, R.; Punji, B.; Findlater, M.; Supplee, C.; Schinski, W.; Brookhart, M.; Goldman, A. S. Nature
Chem. 2011, 3, 167ꢀ171.
(20) NawaraꢀHultzsch, A. J.; Hackenberg, J. D.; Punji, B.; Supplee, C.; Emge, T. J.; Bailey, B. C.; Schrock,
R. R.; Brookhart, M.; Goldman, A. S. ACS Catalysis 2013, 3, 2505ꢀ2514.
28
ACS Paragon Plus Environment

Page 29 of 33
Journal of the American Chemical Society
Kumar et al. Solidꢀgas Phase Dehydrogenation
29
22ꢀJulꢀ15
1
2
3
4
5
6
7
8
(21) Brayton, D. F.; Beaumont, P. R.; Fukushima, E. Y.; Sartain, H. T.; MoralesꢀMorales, D.; Jensen, C. M.
Organometallics 2014, 33, 5198ꢀ5202.
(22) Yao, W.; Zhang, Y.; Jia, X.; Huang, Z. Angew. Chem., Intl. Ed. 2014, 53, 1390ꢀ1394.
(23) Jia, X.; Zhang, L.; Qin, C.; Leng, X.; Huang, Z. Chem. Commun. (Cambridge, U. K.) 2014, 50, 11056ꢀ
11059.
(24) Goldman, A. S.; Roy, A. H.; Huang, Z.; Ahuja, R.; Schinski, W.; Brookhart, M. Science 2006, 312, 257ꢀ
261.
(25) Ahuja, R.; Kundu, S.; Goldman, A. S.; Brookhart, M.; Vicente, B. C.; Scott, S. L. Chem. Commun.
2008, 253ꢀ255.
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
(26) Haibach, M. C.; Kundu, S.; Brookhart, M.; Goldman, A. S. Acc. Chem. Res. 2012, 45, 947ꢀ958.
(27) Dobereiner, G. E.; Yuan, J.; Schrock, R. R.; Goldman, A. S.; Hackenberg, J. D. J Am Chem Soc 2013,
135, 12572ꢀ5.
(28) Goldman, A.; Ahuja, R.; Schinski, W. Rutgers, The State University of New Jersey, USA . 2013
US20130123552A1
(29) Steffens, A. M.; Goldman, A. S. "Catalytic dehydroaromatization of nꢀalkanes: Toward a mechanistic
understanding and control of product distribution" Abstracts of Papers, 245th ACS National Meeting &
Exposition, New Orleans, LA, United States, April 7ꢀ11, 2013, INORꢀ1328.
(30) Goldman, A. S.; Stibrany, R. T.; Schinski, W. L. Chevron U.S.A. Inc., USA; Rutgers, The State
University of New Jersey . 2013 US20130090503A1
(31) Leitch, D. C.; Labinger, J. A.; Bercaw, J. E. Organometallics 2014, 33, 3353ꢀ3365.
(32) Leitch, D. C.; Lam, Y. C.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc. 2013, 135, 10302ꢀ10305.
(33) Gupta, M.; C. Kaska, W.; M. Jensen, C. Chem. Commun. 1997, 461ꢀ462.
(34) Zhang, X.; Fried, A.; Knapp, S.; Goldman, A. S. Chem. Commun. 2003, 2060ꢀ2061.
(35) Chianese, A. R.; Drance, M. J.; Jensen, K. H.; McCollom, S. P.; Yusufova, N.; Shaner, S. E.; Shopov,
D. Y.; Tendler, J. A. Organometallics 2014, 33, 457ꢀ464.
(36) Chianese, A. R.; Shaner, S. E.; Tendler, J. A.; Pudalov, D. M.; Shopov, D. Y.; Kim, D.; Rogers, S. L.;
Mo, A. Organometallics 2012, 31, 7359ꢀ7367.
(37) Chianese, A. R.; Mo, A.; Lampland, N. L.; Swartz, R. L.; Bremer, P. T. Organometallics 2010, 29,
3019ꢀ3026.
(38) Knapp, S. M. M.; Shaner, S. E.; Kim, D.; Shopov, D. Y.; Tendler, J. A.; Pudalov, D. M.; Chianese, A. R.
Organometallics 2014, 33, 473ꢀ484.
(39) Gruver, B. C.; Adams, J. J.; Warner, S. J.; Arulsamy, N.; Roddick, D. M. Organometallics 2011, 30,
5133–5140.
(40) Adams, J. J.; Gruver, B. C.; Donohoue, R.; Arulsamy, N.; Roddick, D. M. Dalton Trans. 2012, 41,
12601ꢀ12611.
(41) Roddick, D. M. Top. Organomet. Chem. 2013, 40, 49ꢀ88.
(42) Gruver, B. C.; Adams, J. J.; Arulsamy, N.; Roddick, D. M. Organometallics 2013, 32, 6468ꢀ6475.
(43) Allen, K. E.; Heinekey, D. M.; Goldman, A. S.; Goldberg, K. I. Organometallics 2013, 32, 1579ꢀ1582.
(44) Allen, K. E.; Heinekey, D. M.; Goldman, A. S.; Goldberg, K. I. Organometallics 2014, 33, 1337ꢀ1340.
(45) (a) Crabtree, R. H.; Mihelcic, J. M.; Quirk, J. M. J. Am. Chem. Soc. 1979, 101, 7738ꢀ7740. (b)
Crabtree, R. H.; Mellea, M. F.; Mihelcic, J. M.; Quirk, J. M. J. Am. Chem. Soc. 1982, 104, 107ꢀ113. (c)
Burk, M. J.; Crabtree, R. H.; McGrath, D. V. J. Chem. Soc., Chem. Commun. 1985, 1829ꢀ30. (d)
Baudry, D.; Ephritikine, M.; Felkin, H. J. Chem. Soc., Chem. Commun. 1980, 1243–1244. (d) Baudry,
D.; Ephritikine, M.; Felkin, H.; HolmesꢀSmith, R. J. Chem. Soc., Chem. Commun. 1983, 788ꢀ789. (e)
Burk, M. J.; Crabtree, R. H.; Parnell, C. P.; Uriarte, R. J. Organometallics 1984, 3, 816ꢀ817.
(46) (a) Galvita, V.; Siddiqi, G.; Sun, P.; Bell, A. T. J. Catal. 2010, 271, 209ꢀ219. (b) Siddiqi, G.; Sun, P.;
Galvita, V.; Bell, A. T. J. Catal. 2010, 274, 200ꢀ206. (c) Sun, P.; Siddiqi, G.; Vining, W. C.; Chi, M.; Bell,
A. T. J. Catal. 2011, 282, 165ꢀ174. (d) Lobera, M. P.; Escolastico, S.; Serra, J. M. ChemCatChem
2011, 3, 1503ꢀ1508. (e) Solsona, B.; Concepcion, P.; Hernandez, S.; Demicol, B.; Nieto, J. M. L. Catal.
Today 2012, 180, 51ꢀ58. (f) Ausavasukhi, A.; Sooknoi, T. Catal. Commun. 2014, 45, 63ꢀ68. (g) Wu, J.;
Peng, Z.; Bell, A. T. J. Catal. 2014, 311, 161ꢀ168. (h) Avila, A. M.; Yu, Z.; Fazli, S.; Sawada, J. A.;
Kuznicki, S. M. Microporous Mesoporous Mater. 2014, 190, 301ꢀ308. (j) Sui, K.; Yang, M.; Zhou, T.;
Guan, Z.; Sun, R.; Han, D. Integr. Ferroelectr. 2014, 154, 57ꢀ63. (j) Shin, H. H.; McIntosh, S. ACS
Catal. 2015, 5, 95ꢀ103.
(47) Kundu, S.; Lyons, T. W.; Brookhart, M. ACS Catal. 2013, 3, 1768ꢀ1773.
(48) Lyons, T. W.; Guironnet, D.; Findlater, M.; Brookhart, M. J. Am. Chem. Soc. 2012, 134, 15708ꢀ15711.
29
ACS Paragon Plus Environment

Journal of the American Chemical Society
Kumar et al. Solidꢀgas Phase Dehydrogenation
Page 30 of 33
30
22ꢀJulꢀ15
1
2
3
4
5
6
7
8
(49) Kumar, A.; Goldman, A. S. "Pincer based iridium catalysts for light alkane dehydrogenation" Abstracts
of Papers, 245th ACS National Meeting & Exposition, New Orleans, LA, United States, April 7ꢀ11,
2013, INORꢀ1202.
(50) Kumar, A.; Mironov, O.; Saxton, R. J.; Goldman, A. S. "Pincer iridium catalyzed gasꢀsolid phase and
liquidꢀphase alkane dehydrogenation using acceptors that inhibit olefin isomerization" Abstracts of
Papers, 248th ACS National Meeting & Exposition, San Francisco, CA, United States, August 10ꢀ14,
2014, INORꢀ1041.
(51) Ohe, S. Computer Aided Data Book of Vapor Pressure; Data Book Publishing Company, Tokyo, 1976.
(52) Ohe, S. In Vapor Pressure Calculation; http://eꢀdata.jp/vpcal1/e/ ; Vol. 2015.
(53) Choi, J.; Choliy, Y.; Zhang, X.; Emge, T. J.; KroghꢀJespersen, K.; Goldman, A. S. J. Am. Chem. Soc.
2009, 131, 15627ꢀ15629.
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
(54) Kundu, S.; Choi, J.; Wang, D. Y.; Choliy, Y.; Emge, T. J.; KroghꢀJespersen, K.; Goldman, A. S. J. Am.
Chem. Soc. 2013, 135, 5127ꢀ5143.
(55) Some lead references to organometallic CꢀH addition, with particular emphasis on selectivity: (a)
Bennett, J. L.; Vaid, T. P.; Wolczanski, P. T. Inorg. Chim. Acta. 1998, 270(1ꢀ2), 414ꢀ423 (b) Wick, D.
D.; Jones, W. D. Organometallics 1999, 18, 495ꢀ505. (c) Asbury, J. B.; Hang, K.; Yeston, J. S.;
Cordaro, J. G.; Bergman, R. G.; Lian, T. J. Am. Chem. Soc. 2000, 122, 12870 ꢀ12871 and references
4ꢀ11 therein. (d) Vetter, A. J.; Flaschenriem, C.; Jones, W. D. J. Am. Chem. Soc. 2005, 127, 12315ꢀ
12322. (e) Balcells, D.; Clot, E.; Eisenstein, O. Chem. Rev. 2010, 110, 749ꢀ823.
(56) Liu, F.; Pak, E. B.; Singh, B.; Jensen, C. M.; Goldman, A. S. J. Am. Chem. Soc. 1999, 121, 4086ꢀ4087.
(57) Biswas, S.; Huang, Z.; Choliy, Y.; Wang, D. Y.; Brookhart, M.; KroghꢀJespersen, K.; Goldman, A. S. J.
Am. Chem. Soc. 2012, 134, 13276ꢀ13295.
(58) Manuel, T. A. J. Org. Chem. 1962, 27, 3941ꢀ3945.
(59) Alper, H.; LePort, P. C.; Wolfe, S. J. Am. Chem. Soc. 1969, 91, 7553ꢀ7554.
(60) Cowherd, F. G.; Von Rosenberg, J. L. J. Am. Chem. Soc. 1969, 91, 2157ꢀ2158.
(61) Misono, M.; Grabowski, W.; Yoneda, Y. J. Catal. 1977, 49, 363ꢀ8.
(62) Casey, C. P.; Cyr, C. R. J. Am. Chem. Soc. 1973, 95, 2248ꢀ53.
(63) Herrmann, W. A.; Prinz, M. In Applied Homogeneous Catalysis with Organometallic Compounds (2nd
Edition); Cornils, B., Herrmann, W. A., Eds.; WileyꢀVCH Verlag GmbH: Weinheim, Germany, 2002;
Vol. 3, p 1119ꢀ1130.
(64) Crabtree, R. H. In The Organometallic Chemistry of the Transition Metals; 5th ed.; John Wiley & Sons:
Hoboken, NJ, 2009, p 229ꢀ231.
(65) Mol, J. C. J. Mol. Catal. A Chem. 2004, 213, 39ꢀ45.
(66) van Leeuwen, P. W. N. M. In Homogeneous Catalysis: Understanding the Art; Kluwer Academic
Publishers: Dordrecht, 2004, p 101ꢀ107.
(67) Schmidt, B. Eur. J. Org. Chem. 2004, 1865ꢀ1880.
(68) Otsuka, S.; Tani, K., Isomerization of olefin and the related reactions. In Transition Metals for Organic
Synthesis (2nd Edition), WileyꢀVCH: Weinheim, 2004; Vol. 1, pp 199ꢀ209.
(69) Seayad, A.; Ahmed, M.; Klein, H.; Jackstell, R.; Gross, T.; Beller, M. Science 2002, 297, 1676ꢀ1678.
(70) Scarso, A.; Colladon, M.; Sgarbossa, P.; Santo, C.; Michelin, R. A.; Strukul, G. Organometallics 2010,
29, 1487ꢀ1497.
(71) Casey, C. P.; Cyr, C. R. J. Am. Chem. Soc. 1973, 95, 2240ꢀ2248.
(72) KroghꢀJespersen, K.; Czerw, M.; Zhu, K.; Singh, B.; Kanzelberger, M.; Darji, N.; Achord, P. D.;
Renkema, K. B.; Goldman, A. S. J. Am. Chem. Soc. 2002, 124, 10797ꢀ10809.
(73) Romero, P. E.; Whited, M. T.; Grubbs, R. H. Organometallics 2008, 27, 3422ꢀ3429.
(74) Koch, W.; Holthausen, M. C. A Chemist’s Guide to Density Functional Theory; Wiley: New York, 2001.
(75) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys Rev. Lett. 1996, 77, 3865ꢀ3868.
(76) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299ꢀ310.
(77) Roy, L. E.; Hay, P. J.; Martin, R. L. J. Chem. Theory Comput. 2008, 4, 1029–1031.
(78) (a) Ditchfield, R.; Hehre, W. J.; Pople, J. A. J. Chem. Phys. 1971, 54, 724ꢀ728. (b) Hariharan, P. C.;
Pople, J. A. Molecular Physics 1974, 27, 209ꢀ214. (c) Raghavachari, K.; Binkley, J. S.; Seeger, R.;
Pople, J. A. J. Chem. Phys. 1980, 72, 650ꢀ654. (d) McLean, A. D.; Chandler, G. S. J. Chem. Phys.
1980, 72, 5639ꢀ5648.
(79) Frisch, Æ.; Frisch, M. J.; Clemente, F. R.; G. W. Trucks Gaussian 09 User's Reference, Gaussian,
Incorporated, Wallingford, CT, 2009; 147.
(80) McQuarrie, D. A. Statistical Thermodynamics; Harper and Row: New York, 1973.
(81) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.;
Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian,
30
ACS Paragon Plus Environment