Selective Photo-Oxygenation of Light Alkanes Using Iodine Oxides and Chloride

Article abstract of DOI:10.1002/cctc.201901175

Partial oxidation of light alkanes to generate alkyl esters has been achieved under photochemical conditions using mixtures of iodine oxides and chloride salts in trifluoroacetic acid (HTFA). The reactions are catalytic in chloride and are successful using compact fluorescent light, but higher yields are obtained using a mercury lamp. In this photo-initiated oxyesterification process, the robust alkyl ester products are resistant to over-oxidation, and under optimized conditions yields for alkyl ester production of ～50 % based on methane, ～60 % based on ethane (with a total functionalized yield of EtX (X=TFA or Cl) of 80 %) and ～30 % based on propane have been demonstrated. The reaction also proceeds in aqueous HTFA and dichloroacetic acid with lower yields. Mechanistic studies indicate that the process likely operates by a chlorine hydrogen atom abstraction pathway wherein alkyl radicals are generated, trapped by iodine, and converted to alkyl trifluoroacetates in situ.

Full text of DOI:10.1002/cctc.201901175

Full Papers
DOI: 10.1002/cctc.201901175
1
2
3
4
5
6
7
8
9
Selective Photo-Oxygenation of Light Alkanes Using Iodine
Oxides and Chloride
Nichole S. Liebov,[a] Jonathan M. Goldberg,[b] Nicholas C. Boaz,[b, c] Nathan Coutard,[a]
Steven E. Kalman,[a, d] Thompson Zhuang,[b] John T. Groves,*[b] and T. Brent Gunnoe*[a]
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
Partial oxidation of light alkanes to generate alkyl esters has
been achieved under photochemical conditions using mixtures
of iodine oxides and chloride salts in trifluoroacetic acid (HTFA).
The reactions are catalytic in chloride and are successful using
compact fluorescent light, but higher yields are obtained using
a mercury lamp. In this photo-initiated oxyesterification process,
the robust alkyl ester products are resistant to over-oxidation,
and under optimized conditions yields for alkyl ester production
of ~50% based on methane, ~60% based on ethane (with a
total functionalized yield of EtX (X=TFA or Cl) of 80%) and
~30% based on propane have been demonstrated. The
reaction also proceeds in aqueous HTFA and dichloroacetic acid
with lower yields. Mechanistic studies indicate that the process
likely operates by a chlorine hydrogen atom abstraction path-
way wherein alkyl radicals are generated, trapped by iodine,
and converted to alkyl trifluoroacetates in situ.
Introduction
goal and major challenge is to develop processes that
selectively produce mono-functionalized products.[6]
Natural gas is an abundant fuel resource and chemical
feedstock.[1] Traditionally, light alkanes are converted to higher
value chemicals by processes that require high temperature
and pressure. For example, methane can be functionalized by
initial steam reforming to generate synthesis gas (H2/CO) and
subsequently converted to methanol or long-chain hydro-
carbons using Fischer-Tropsch chemistry.[2] However, this proc-
ess for the conversion of methane to liquid fuels and chemicals
is capital- and energy-intensive.[3] The ethane and propane
components of natural gas are cracked at high temperatures (>
Several strategies for metal-catalyzed partial oxidation of
alkanes have been pursued. Here, we divide these strategies
into two categories, those that form catalytic intermediates
with metal-carbon bonds and those that generate free alkyl
radical intermediates without metal-carbon bond formation.
Arguably, the former strategy has been dominated by soluble
molecular catalysts wherein the CÀ H bond can be cleaved by a
number of reactions (e.g., oxidative addition, σ-bond meta-
thesis, electrophilic substitution, etc.).[3c,d,6b,7] Thus, a variety of
homogeneous catalysts have been examined for light alkane
functionalization, with emphasis on electrophilic metals. Often,
electrophilic metal catalysts are believed to function via initial
coordination of the alkane followed by proton removal to
generate a M–alkyl intermediate, which is then susceptible to a
reductive nucleophilic functionalization step (often after oxida-
tion of the metal-alkyl intermediate). For example, the Catalytica
system, one of the most effective and well-known electrophilic
catalysts, converts methane to methyl bisulfate with high yield
°
900 C) to generate ethylene and propylene, respectively, which
serve as precursors for a variety of higher value chemicals.[4] The
demand for efficient use of energy and the need for on-site
conversion of natural gas to liquids have provided motivation
to develop new methods for direct and selective low-temper-
[5]
°
ature (<200 C) light alkane oxy-functionalization. But, a key
and selectivity, >70% and >90%, respectively, using
a
[a] Dr. N. S. Liebov, Dr. N. Coutard, Dr. S. E. Kalman, Prof. T. B. Gunnoe
Department of Chemistry
platinum catalyst (bpym)PtCl2 (bpym=2,2’-bipyrimidinyl) in
oleum.[8] Germane here, it has been proposed that the electron-
withdrawing nature of the bisulfate group protects methyl
bisulfate from over-oxidation by the electrophilic catalyst.[9] This
strategy has been extended to catalysts based on Pd, Au, Hg, Tl
and Sb,[10] and stoichiometric alkane CÀ H activation with main
group elements has been demonstrated recently.[11] Sulfonation
using SO3 has also been demonstrated to generate methane-
sulfonic acid from methane.[12] However, electrophilic metal
catalysts have limitations. For example, they are often inhibited
by even weak bases and, thus, often operate effectively only in
strongly acidic media (often only in superacids).[13]
University of Virginia
Charlottesville VA 22904 (USA)
E-mail: tbg7h@virginia.edu
[b] Dr. J. M. Goldberg, Dr. N. C. Boaz, T. Zhuang, Prof. J. T. Groves
Department of Chemistry
Princeton University
Princeton NJ 08544 (USA)
E-mail: jtgroves@princeton.edu
[c] Dr. N. C. Boaz
Department of Chemistry
North Central College
Naperville IL 60540 (USA)
[d] Dr. S. E. Kalman
Chemistry Program
School of Natural Sciences and Mathematics
Stockton University
In contrast to the “organometallic” approach for alkane
functionalization, which features catalytic intermediates with
metal-carbon bonds, many heterogeneous catalysts for light
alkane oxidation are thought to operate by homolytic CÀ H
Galloway NJ 08205 (USA)
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/cctc.201901175
ChemCatChem 2019, 11, 5045–5054
5045
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
cleavage to generate free alkyl radicals.[14] Examples include
iron- and copper-based zeolites and transition metal oxides.[15]
These catalysts typically require high temperatures, often
based on ethane. These yields provide a substantial increase in
efficiency compared to the previously reported thermal reac-
tions.
1
2
3
4
5
6
7
8
9
°
>250 C, and because the rate of reaction is often directly
dependent on the strength of CÀ H bonds, mono-functionalized
products are often more reactive than starting alkanes. Thus, Results and Discussion
although catalytic oxyhalogenation offers a route for light
alkane functionalization by converting alkanes to alkyl
halides,[16] the alkyl halide has weaker CÀ H bonds relative to the
starting alkane,[17] and over-oxidation is significant at high
alkane conversions.[17c,18]
Our groups have previously demonstrated thermal function-
alization via oxyesterification (OxE) of light alkanes in trifluoro-
acetic acid using chloride and iodine oxide salts.[19] Using iodate
as the oxidant, methane, ethane and propane are converted to
the corresponding trifluoroacetate esters in high substrate-
based yields, ~25% for methane, ~30% for ethane, and ~20%
for propane using iodate as the oxidant. The alkyl esters can
undergo hydrolysis to generate alcohol and regenerate HTFA.
Reactions proceed over a wide range of temperatures (100–
From our initial reports of the iodate/chloride OxE process
operated under thermal conditions, the highest yield, based on
methane, of MeX (X=TFA, Cl) using KCl/NH4IO3 was 24%.[19a] An
initial screening of the photochemical reactivity of the OxE
process found that high yields relative to alkane (16% MeX
relative to methane with a 35:1 ratio of MeTFA:MeCl) could be
obtained using a mercury lamp (Scheme 1A). Control reactions
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
°
235 C) and pressures (35–1000 psi). Although this thermal OxE
process has some similarities to oxychlorination with an oxidant
and a chloride source, it exhibits unique selectivity for the
mono-functionalized ester product. Recently, combined exper-
imental and computational studies found evidence for a
reaction pathway in which free radical intermediates mediate
homolytic alkane CÀ H bond-breaking.[20] DFT calculations
indicated that IO2· and Cl· radicals formed in situ under OxE
conditions (i.e., chloride and iodate or periodate in HTFA) can
abstract a hydrogen atom from methane.[20a] Thus, the reactions
are catalytic in chloride, which is a differentiating factor
compared to traditional catalytic alkane oxychlorination. In our
previous studies, the formation of key reaction intermediates
was proposed to depend on the acidity of HTFA. Thus, reactions
in acetic acid were less successful. Iodine, which forms in situ,
was proposed to serve as an efficient alkyl radical trap to
generate alkyl iodides that are converted to R-TFA products in
HTFA solvent. Importantly, the electron-withdrawing TFA ester
functionality was demonstrated to protect the alkyl products
from over-oxidation.[20a,21] Thus, we have proposed that this OxE
process is similar to the strategy using electrophilic metals to
produce methyl bisulfate in oleum (see above) but is successful
in a less acidic solvent (i.e., HTFA vs. oleum).
Scheme 1. (A) Photochemical partial oxidation of methane using a mercury
lamp. Conditions: CH4 (100 psi), KCl (0.67 mmol), NH4IO3 (7.7 mmol), HTFA
(8 mL), 24 h photolysis using the Hg lamp. The percent yield of MeTFA is
based on the amount of added methane. The standard deviation is based on
a minimum of three experiments. (B) Oxidation of methane using the OxE
system under ambient light. Conditions: CH4 (100 psi), KCl (0.67 mmol),
NH4IO3 (7.7 mmol), HTFA (8 mL), 24 h exposure to laboratory light in a fume
°
hood or 24 h heating at 50 C in an oil bath.
°
using ambient light or heating at 50 C gave minimal methane
functionalization (Scheme 1B). Similar to the previously re-
ported thermal reaction,[19] the production of ~4 mmol of
MeTFA reveals a likely catalytic role for chloride since only
0.67 mmol of KCl are used.
In the absence of chloride, trace MeTFA is produced
(Figure 1A). Thus, chloride is essential to the methane function-
alization. Increasing the starting chloride amount increases the
MeX yield (Figure 1A). MeTFA yields of ~25%, based on
methane, with 94% or 91% selectivity for MeTFA, were
observed with 2.01 or 3.35 mmol of starting KCl, respectively
(selectivity refers to the ratio of MeTFA to MeCl). With higher
chloride starting amounts (�6.7 mmol), the selectivity shifts
toward MeCl, and a small amount of undesired over-oxidation
to form dichloromethane is observed (Figure 1A). Also, with
6.7 mmol of KCl, the overall yield of MeX (X=TFA or Cl) drops
to 20.7(7)%, with selectivity for MeTFA decreasing to 70%. We
speculate (see below) that a larger KCl amount allows chlorine
to compete with iodine to trap the putative methyl radical,
which forms MeCl. Unlike MeI, MeCl is not readily converted to
While there is a significant body of work regarding the
functionalization of light alkanes under thermal conditions,
photo-mediated partial oxidation of light alkanes has limited
precedent.[22] We speculated that a photo-initiated oxyesterifica-
tion (photo-OxE) process might offer advantages over the
thermal process. For example, oxidation of HTFA is an issue for
the OxE process with iodate and chloride, which limits alkyl
ester yield,[19a] and we considered that using lower temperatures
with a photo-initiated process might enhance yields of the alkyl
ester product by minimizing oxidation of the solvent. Thus, we
pursued photolysis as a method to generate reactive intermedi-
ates under milder reaction conditions. Herein, we report a
photo-OxE process for light alkanes that achieves high alkyl
ester and halide yields of ~50% based on methane and ~80%
ChemCatChem 2019, 11, 5045–5054
www.chemcatchem.org
5046
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
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
Figure 1. (A) Effect of starting KCl amount on MeX yield. Conditions: CH4 (100 psi), KCl (0.67–6.7 mmol), NH4IO3 (7.7 mmol), HTFA (8 mL), 24 h photolysis using
the Hg lamp. Error bars denote standard deviations based on at least three experiments. *With 6.7 mmol, 1.73(2)% yield of dichloromethane is also observed.
(B) Effect of starting NH4IO3 amount on MeX yield. Conditions: CH4 (100 psi), KCl (2.01 mmol), NH4IO3 (1.0–7.7 mmol), HTFA (8 mL), 24 h photolysis using the Hg
lamp. Error bars denote standard deviations based on at least two experiments. (C) Photochemical partial oxidation of methane by KCl/NH4IO3 in different
solvent combinations. Dilute HTFA is 3:1 mol:mol HTFA:H2O. MeX is the corresponding acetate ester for each acid. Conditions: CH4 (100 psi), KCl (2.01 mmol),
NH4IO3 (7.7 mmol), solvent (8 mL), 24 h photolysis using the Hg lamp. Error bars denote standard deviations based on at least three experiments. (D) Effect of
methane pressure (with longer reaction times) on photochemical OxE of methane. Conditions: CH4 (25, 50, or 100 psi), KCl (2.01 mmol), NH4IO3 (7.7 mmol),
HTFA (8 mL), 48 h photolysis using the Hg lamp. Error bars denote standard deviations based on at least three experiments.
MeTFA,[19a] and, hence, the formation of MeCl likely removes the
chlorine catalyst from the reaction.
In previously published work, the solvent scope of the
thermal OxE process was examined, and it was found that HTFA
was the most effective solvent. We also sought to test the
influence of the reaction solvent on methane functionalization
for the photochemical OxE reaction. While the highest yields of
functionalized products are obtained with HTFA, MeX yields
(X=TFA or dichloroacetate) of 16(2)% and 17(1)% are obtained
after 24 h using dilute HTFA (3:1 mol:mol HTFA:H2O) and the
weaker acid dichloroacetic acid (pKa 1.35 versus 0.52 for HTFA)
respectively, as reaction solvents (Figure 1D). The tolerance of
water is notable, as many light alkane functionalization
processes in acidic media exhibit poor activity in non-super-
acidic media. Acetic acid was among those tested as a potential
solvent for thermal methane functionalization using the iodate/
chloride OxE process. Although MeOAc was produced from the
reaction of methane with KCl/NH4IO3 in HOAc, under thermal
conditions it was demonstrated that MeOAc originates from
oxidation of HOAc rather than methane functionalization.
However, only trace MeOAc, ~0.1% based on methane, is
observed for the photo-OxE reaction with methane, indicating
The effect of varying the amount of NH4IO3 was also studied
(Figure 1B). Decreasing the amount of NH4IO3 from 7.7 mmol
resulted in slightly higher yields of MeX relative to methane
when using 2 or 3 mmol of the oxidant. However, using 1 mmol
of NH4IO3, the yield of MeX was significantly lower.
Longer reaction times and lower methane pressures were
tested with the optimal starting KCl amount of 2.01 mmol. With
100 psi methane (~25 mmol of methane), 39(4)% yield of
MeTFA was observed after 48 h of photolysis (Figure 1C), higher
À
than optimized yields for IO3 /ClÀ under thermal conditions
°
(180 C, 1 h, 24% yield MeX). Also, the product concentration
obtained using photolysis is much greater (~1.2 M vs ~0.23 M)
than the highest yield for the thermal OxE process obtained
with iodate. With 50 psi of methane, a 48(4)% yield of MeTFA
was obtained with 95% selectivity. With 25 psi of methane,
lower average yields are obtained with higher variability
between experiments.
ChemCatChem 2019, 11, 5045–5054
www.chemcatchem.org
5047
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
that solvent oxidation is minimal under these milder conditions.
This lends evidence that the photolysis conditions reduce the
extent of solvent oxidation (compared to thermal reactions),
which might be responsible for the higher yields of MeOAc
under thermal conditions. Further, the inability of acetic acid to
facilitate methane functionalization could indicate that it is too
weakly acidic to generate the necessary active species, which is
consistent with our previous studies. Previous UV/Vis spectro-
scopy of reaction mixtures indicated that an acidic solvent is
required to generate Cl2 and interhalogen species (i.e., ICl and/
or ICl3) that are expected to be precursors of active intermedi-
ates in the functionalization process. HTFA, the effective solvent
in the thermal KCl/NH4IO3 system, generates these species
in situ, but HOAc does not.
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
Figure 3. Photochemical functionalization of ethane with KCl/NH4IO3. Con-
ditions: CH3CH3 (25, 50 or 100 psi), KCl (2.01 mmol), NH4IO3 (7.7 mmol), HTFA
(8 mL), 24 h photolysis using the Hg lamp. Error bars denote standard
deviations based on three experiments.
To test the ability of visible light to functionalize methane,
irradiation using a compact fluorescent light (CFL) was exam-
ined. As shown in Figure 2, irradiation with a 105 W CFL results
es S5 and S6). There is coupling between a doublet at 1.72 (J=
5.8 Hz) and a quartet at 6.39 ppm (J=5.8 Hz), which is similar to
the reported resonances for 1,1-bis(trifluoroacetoxy)ethane.
Assuming the doublet corresponds to a methyl peak, there is a
1.6(8)% yield of this product relative to ethane with 25 psi, with
only trace formation observed under 50 or 100 psi of ethane.
These observations suggest that at high conversions of alkane,
over-oxidation occurs but only to a very minimal degree.
Similarly, in addition to the major product, 2-trifluoroacetox-
ypropane, a variety of products, including di-functionalized
propyl products, are generated in the photochemical oxidation
of propane (Figure 4).
Periodate was also examined as an oxidant for the photo-
chemical process (Figure 5). Although similar MeTFA yields were
Figure 2. Comparison of irradiation by a mercury lamp and a 105 W compact
fluorescent light (CFL) for photochemical partial oxidation of methane by
KCl/NH4IO3. Conditions: CH4 (100 psi), KCl (2.01 mmol), NH4IO3 (7.7 mmol),
solvent (8 mL), 24 h photolysis. Error bars denote standard deviations based
on at least three experiments.
in methane functionalization; however, yields are lower than
those obtained with the mercury lamp. After 24 h of irradiation,
~10% yield of MeTFA is obtained using the 105 W CFL, while
~25% yield is obtained with the mercury lamp. After 72 h,
105 W CFL irradiation generates MeTFA in ~25% yield, but
reaction times longer than 72 h do not result in improved yields
of MeX.
Photo-OxE reactions of ethane and propane were also
examined using the KCl/NH4IO3 mixture in HTFA with the
mercury lamp. For ethane, the selectivity for EtTFA and EtCl is
high with minimal formation of the di-functionalized products
1-iodo-2-trifluoroacetoxyethane,
1-chloro-2-trifluoroacetoxy-
ethane, 1,2-bis(trifluoroacetoxy)ethane and 1,2-dichloroethane
(Figure 3). Using 25 psi of ethane, an optimized EtX (X=TFA or
Cl) yield of 82(5)% was demonstrated. There is also evidence for
Figure 4. Photochemical functionalization of propane (25–100 psi) with KCl
(2.01 mmol) and NH4IO3 (7.7 mmol) in HTFA with 24 h of photolysis using the
Hg lamp. Error bars denote standard deviations based on three experiments.
1
a 1,1-difunctionalized product at low ethane pressures by H
1
NMR spectroscopy and from the H,1H-COSY spectrum (Figur-
ChemCatChem 2019, 11, 5045–5054
www.chemcatchem.org
5048
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
1
2
3
4
5
6
7
8
Scheme 2. EtTFA decay in the presence of KCl, NH4IO3 and HTFA after 24 h
photolysis using the Hg lamp.
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
tially broad applicability of the ester protection strategy beyond
the initial thermal conditions. The rates of methane functional-
ization and MeTFA decay cannot be compared directly as the
concentration of methane in HTFA solution is unknown.
However, to approximate the relative rate of methane function-
alization versus MeTFA decay, the slopes of product vs time
plots were compared. As the plot for methane functionalization
is linear through 24 h, the fit of the points through the first 24 h
was used to obtain the slope for functionalization (Figure S11).
All of the data points for MeTFA decay were used to obtain a
linear fit of those results. The selectivity for methane function-
alization vs. MeTFA decay is 13, similar to the ratio observed in
the thermal process, 13.5. Although the rate of EtTFA oxidation
was not quantified, EtTFA was also found to be stable after 24 h
of photolysis (Scheme 2).
We speculate that chlorine radical is an important inter-
mediate in the photo-OxE process. N-chlorosuccinimide (NCS)
was used to compare the photo-OxE process to a traditional
radical-based functionalization method. With benzoyl peroxide
as a radical initiator, the photolysis of methane with NCS
generates MeCl and dichloromethane. However, upon addition
of iodine to the system, MeTFA is produced (Scheme 3), similar
to observations reported for the thermal process.[19] MeI has
also been demonstrated to convert rapidly to MeTFA under the
photochemical conditions (Scheme 4). Thus, this result is
consistent with chlorine-mediated formation of methyl radical,
iodine trapping to form MeI, and conversion of MeI in HTFA to
Figure 5. Comparison of photochemical functionalization of methane with
iodate versus periodate. Conditions: CH4 (100 psi), KCl (2.01 mmol), NH4IO3 or
KIO4 (7.7 mmol), HTFA, (8 mL), 24 or 48 h photolysis using the Hg lamp. Error
bars denote standard deviations based on three trials for each condition
except for the 48 h trial with periodate, which is based on six trials.
obtained using either iodate or periodate, the standard
deviations associated with MeTFA were greater when using
periodate.
In previous studies we found that the thermal OxE
chemistry likely operates by a radical process involving H-atom
abstraction from alkanes, resulting in the selective formation of
alkyl trifluoroacetates in HTFA. The alkyl trifluoroacetates were
demonstrated experimentally to be stable under the OxE
reaction conditions, minimizing over-oxidation and conse-
quently enabling high yields and selectivities. Thus, we sought
to determine and compare the rates of methane functionaliza-
tion and MeTFA decay under photolysis.
MeTFA was found to be stable under photochemical
conditions in the presence of iodate and chloride. Figure 6
shows a comparison of methane functionalization and MeTFA
decomposition under the conditions optimized for methane
functionalization. The stability of MeTFA indicates the poten-
Figure 6. Methane functionalization vs. MeTFA decay in the presence of KCl, NH4IO3 and HTFA under photochemical reaction conditions as a function of time.
ChemCatChem 2019, 11, 5045–5054
www.chemcatchem.org
5049
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
1
2
3
4
5
6
Scheme 3. Addition of iodine to chlorination by N-chlorosuccinimide and benzoyl peroxide results in formation of MeTFA under photolysis using the Hg lamp.
Standard deviations are based on at least three experiments.
7
8
9
cyclohexane studies might not directly translate to methane
due to its higher CÀ H BDE (~105 kcalmolÀ 1).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
We screened cyclohexane functionalization under stoichio-
metric conditions with equimolar substrate and NH4IO3 oxidant.
°
Under thermal conditions (100 C, 1 h), functionalization of
cyclohexane with 1 equiv. of NH4IO3 and 0.25 equiv. of KCl,
Scheme 4. MeI conversion to MeTFA under photolysis using the Hg lamp.
The standard deviation is based on at least three experiments.
relative to cyclohexane, in HTFA solvent results in the formation
of five products as detected by GC-MS. The observed mono-
functionalized products were cyclohexyl-TFA and cyclohexyl-
chloride, which likely result from a hydrogen atom abstraction
pathway. We hypothesize that the three di-functionalized
products result from initial desaturation of cyclohexane to give
cyclohexene, and that addition of in situ-generated Cl2, ICl or
ICl3 across the double bond in cyclohexene yields a dihalocyclo-
hexane intermediate; the detection of trans-1,2-dichlorocyclo-
Scheme 5. Methane functionalization by chlorine radical wherein the methyl
radical formed following H-atom abstraction is trapped by iodine to
generate MeI, which is converted to MeTFA in HTFA solvent.
hexane as
a stable product in one of these reactions
corroborates this hypothesis. Solvolysis of the dihalo intermedi-
ate by HTFA yields both 1,2-bis(trifluoroacetoxy)cyclohexane
and 2-chloro-1-trifluoroacetoxy-cyclohexane, as detected by
GC-MS and 1H NMR spectroscopy (Scheme 6).
Other light sources were also examined (Table 1). The
105 W CFL gave similar results to the 26 W CFL, with ~25%
yield of functionalized products based on cyclohexane (Table S1
in the Supporting Information provides results under additional
conditions). Use of a Rayonet UV Photochemical Reactor (UV-
reactor) gave further improved yields, 45(2)%, similar to the
best yields obtained with methane under optimized conditions.
The reaction time using the UV-reactor was limited to 18 h due
to the build-up of pressure under reaction conditions.
MeTFA (Scheme 5). While it is possible that functionalization
occurs through a carbocationic reaction pathway, we believe it
is unlikely under our reaction conditions.12 These data demon-
strate that iodine is likely an efficient radical trap in the
photochemical process, similar to the thermal process. Over-
oxidation to form dichloromethane is also suppressed when
iodine is added. For example, without added iodine, dichloro-
methane accounts for 25% of functionalized products, but
<9% of products are due to dichloromethane when
0.067 mmol of I2 is added. The addition of iodine to the
photochemical iodate/chloride reaction mixture was examined
and did not result in improved yields (see Supporting
Information, Figure S14).
Since we observed a significant improvement in product
conversion under photo-OxE conditions compared to thermal
reactivity, we were interested in probing possible mechanistic
differences between the thermal and photo-initiated reactions.
We previously examined kinetic isotope effects (KIE) for thermal
OxE using adamantane-h16/d16 [kH/kD =1.52(3)] as a model
system for methane. However, at room temperature, adaman-
tane is poorly soluble in HTFA, rendering interpretation of KIE
experimental results ambiguous. Instead, we examined
cyclohexane-h12/d12 under the new photochemical conditions
and also subjected these substrates to thermal OxE conditions
to provide a direct comparison. We used cyclohexane as a
model substrate for our KIE studies since we can more easily
control stoichiometric reactions and the CÀ H bond dissociation
energy (BDE) for cyclohexane (99.5 kcalmolÀ 1) is quite similar to
the light alkanes propane (100.9 kcalmolÀ 1 for À CH3 and
98.1 kcalmolÀ 1 for À CH2À ) and ethane (100.5 kcalmolÀ 1). A
caveat here is that the mechanistic conclusions from our
Table 1. Cyclohexane functionalization under OxE conditions (thermal and
photochemical).[a]
°
100 C
1 h
26 W CFL
24 h
105 W CFL
24 h
UV-reactor
18 h
R1 =TFA
R1 =Cl
2.7(8)
1.6(1)
trace
trace
trace
13.3(3)
4.5(1)
1.1(1)
1.6(1)
*6.4(3)
11.7(6)
4.7(3)
1.3(1)
2.2(3)
*5.6(3)
30(2)
ND
4.7(2)
ND
*10.1(5)
R1 =R2 =TFA
R1 =R2 =Cl
R1 =Cl;
R2 =TFA
Total Product
Yield
4.3(8)
26.9(5)
25.5(8)
45(2)
[a] Product yields (%) are shown. Products were quantified using GC-MS vs.
an internal standard (tetradecane). *Denotes detection and quantification
by H NMR spectroscopy. ND=not detected.
1
ChemCatChem 2019, 11, 5045–5054
www.chemcatchem.org
5050
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
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
Scheme 6. Proposed synthetic routes to products from cyclohexane functionalization. Products in dotted boxes were detected and quantified by GC-MS and/
or 1H NMR spectroscopy for X=Cl only. It is proposed that for X=I, the compounds readily undergo solvolysis by HTFA or exchange with ClÀ so the alkyl
iodides are not detected.
Since a significant amount of difunctionalized product was
observed under both thermal and photochemical pathways,
multiple reactive pathways could be occurring simultaneously.
However, it is still possible to make a reasonable estimate of the
isotope effect since we are only probing the initial hydrogen
atom abstraction of cyclohexane (Scheme 6). To compare the
two pathways, we examined the ratio of the combined
protiated products vs. the combined deuterated products (kH/
kD, Table 2). The thermal iodate/chloride system gave a kH/kD of
We previously showed that upon combining iodate and
chloride in HTFA, Cl2 and ICl/ICl3 could be detected by UV-vis
spectroscopy (see Supporting Information Figure S39. To probe
if chlorine atoms were intermediates in our photochemical
iodate/chloride system, we also examined the kH/kD values for
reactions of cyclohexane-h12/d12 with ICl, ICl3, and periodate/
chloride, and compared those to the iodate/chloride system
and the reported Cl2 reaction (see Supporting Information for
stoichiometric reaction yields). By UV-vis spectroscopy, it has
been demonstrated that combining periodate and chloride in
HTFA also gives Cl2 and ICl/ICl3 (see Supporting Information,
Figure S38), albeit at much lower concentrations than analo-
gous conditions with iodate. Examination of the thermal kH/kD
values for reaction of ICl3 {kH/kD =2.3(2)} and periodate/chloride
{kH/kD =1.94(3)} with cyclohexane gave similar isotope effects to
iodate/chloride {kH/kD =2.1(2)}. However, the kH/kD for the
thermal reaction of ICl with cyclohexane gave a more significant
kH/kD of 4.3(1). These results indicate that the thermal reaction is
not consistent with solely a chlorine atom abstraction pathway.
Since certain thermal kH/kD values (entries 1–3, Table 2) were
in line with each other, we hypothesize that these species arise
from the same mechanistic intermediate. Therefore, we propose
a scenario where iodate disproportionates into I(III) and I(VII)
species that are both able to facilitate hydrogen atom
abstraction (Scheme 7). As we previously reported, the calcu-
Table 2. OxE kinetic isotope effect experiments with cyclohexane under
thermal and photochemical conditions.[a]
[b]
Entry
Thermal/
Condition
kH/kD
Photochemical
1
2
3
4
5
6
7
8
KCl/NH4IO3
KCl/KIO4
ICl3
2.1(2); 1.9(1)][c]
1.94(3); [2.2(1)][c]
2.3(2)
4.3(1)
1.06(3)
1.08(3)
1.15(1)
1.21(4)
Thermal
ICl
KCl/NH4IO3
KCl/KIO4
ICl3
Photochemical
ICl
*
lated intermediate IO2 may be an in situ generated active
[a] Conditions: NH4IO3 (0.25 mmol), KCl (0.14 mmol); KIO4 (0.25 mmol), KCl
(0.14 mmol); ICl3 (0.25 mmol); ICl (0.28 mmol), [b] kH/kD ratios reflect total
protiated vs. total deuterated products, [c] KIE reaction performed at
species. Alternatively, iodate might be reduced with subsequent
oxidation of chloride to generate Cl2 and an I(III) species. In
either scenario, given the kH/kD of 4.3(1) for ICl with
cyclohexane, we do not expect a significant iodine(I) compo-
nent. In contrast, the kH/kD values for photochemical
cyclohexane functionalization by iodate/chloride, periodate/
chloride, ICl and ICl3 are nearly identical to the reported Cl2
system (kH/kD �1.1–1.2), indicating the possibility that the photo-
OxE process operates exclusively through hydrogen atom abstrac-
tion by a chlorine atom. Since ICl, ICl3 and Cl2 are all observed
upon combining iodate and chloride in HTFA, any of these
species could serve as chlorine atom sources.
°
130 C, 1 h. Product ratios were determined by GC-MS. Standard deviations
are based on at least three experiments.
2.1(2), which differs from the kH/kD of the photochemical
reaction {kH/kD =1.06(3)}, suggesting two different reaction
mechanisms. Previous work by Li and Pirasteh found a kH/kD of
1.14(1) for the reaction of atomic chlorine with cyclohexane-h12/
d12, in line with our photo-OxE cyclohexane functionalization.
While the thermal iodate/chloride yields are low for
cyclohexane functionalization, the kH/kD values did not signifi-
cantly change upon increasing the temperature from 100 to
°
130 C indicating that the reactive species is not readily affected
by a change in temperature (Table 2).
ChemCatChem 2019, 11, 5045–5054
www.chemcatchem.org
5051
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
1
2
3
4
5
6
Scheme 7. Proposed species resulting from thermal or photochemical iodate/chloride mixtures in HTFA. For each condition, any of the species in dotted
boxes may facilitate hydrogen atom abstraction from an alkane.
7
8
9
acid, dichloroacetic acid, benzene-d6, nitromethane, benzoyl
peroxide, methyl trifluoroacetate, cyclohexane, and cyclohexane-d12
were purchased commercially and used as received. N-Chlorosucci-
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
Conclusions
nimide was purchased commercially and recrystallized from glacial
acetic acid prior to use. Methane, ethane, propane and argon were
purchased from GTS-Welco and used as received. The mercury
lamp (Hanovia, medium pressure, UV lamp, 450 W) was purchased
from Ace Glass and used as received. The 105 W compact
fluorescent lightbulb (Overdrive CFL, 420 W equivalent, 5000 K full
spectrum) was purchased from Amazon and used as received. The
26 W CFL (2700 K) was purchased from Grainger and used as
received. The UV reactor is a Rayonet Photochemical Reactor (RPR-
100) containing six 8 watt RPR2537 A low pressure mercury lamps.
Fisher-Porter reactors were purchased from Andrews Glass and
used with custom-built Swagelok reactor tops. 1H NMR spectra
were recorded on a Varian Inova 500, Bruker Avance 500, Avance
DRX 600 or an Avance III 600 spectrometer. NMR spectra were
obtained using neat reaction mixtures with C6D6 inserted in a
sealed capillary tube as an internal lock reference. Chemical shifts
are reported relative to the added internal standard (δ 4.18 for
nitromethane, 2.04 for glacial acetic acid, or 7.26 for chloroform).
GC-MS analysis was performed using an Agilent Technologies
7890 A gas chromatograph equipped with a fused silica column
(crossbond 35% diphenyl-65% dimethyl polysiloxane; 30 m
×0.32 mm; 0.5 μm thickness) and electron impact mass analyzer.
The photo-OxE process for light alkanes with iodate-chloride
mixtures is selective for partial oxidation products using
chloride as catalyst. The yields for methane and ethane (~50%
and 80%, respectively) are notable for a process that likely
involves homolytic hydrogen atom abstraction. The photo-
chemical process exhibits tolerance to water and weaker acid
solvents, but yields are highest when using neat HTFA as the
solvent. Generally, the new photo-reaction yields are higher
than those obtained using the same reaction under thermal
conditions (i.e., optimal yields of 24% and ~50% for thermal
and photochemical reactions, respectively), which could be due
to the reduction of reactive intermediates under the photo-
chemical conditions. The higher yields could also be the result
of a reduction of solvent oxidation under the milder photo-
chemical conditions.
Investigation into the mechanism indicates a chlorine atom
abstraction radical pathway for the photochemical process,
which differs from the thermal pathway where multiple active
species may be present. When iodine is added to the methane
chlorination reaction with N-chlorosuccinimide and benzoyl
peroxide, MeTFA is generated. In the absence of iodine, MeTFA
is not produced. This demonstrates iodine’s efficiency as a
radical trap, as even with a large excess of a chlorine radical
source, MeI can be generated in situ, similar to what was
observed thermally. The addition of chloride is necessary, and
slight increases in chloride concentration has a beneficial effect
on product yield. However, the use of high concentrations of
KCl result in lower yields of functionalized product, potentially
due to the ability of chlorine radical to trap active intermediates
and remove the catalytic amount of chlorine from the cycle.
The stability and protecting effect of the product ester group
that was found under thermal conditions is also transferable to
the photochemical system.
General procedure for photochemical oxidation of light alkanes.
A Fisher-Porter reactor was charged with a stir bar, potassium
chloride, ammonium iodate and HTFA (8 mL). The reactor was then
sealed and pressurized with alkane. The reactor was then placed on
a stir plate 13 cm away from the lamp and stirred. The Hg lamp was
in a closed chamber for safety purposes. Following the reaction, the
lamp was turned off, and the reactor was vented. Internal standard
was then added to the reaction mixture, and the reaction mixture
was stirred. An aliquot was removed for centrifugation, and the
1
supernatant was used for H NMR analysis. Similar procedures were
used for the other light sources.
MeTFA decay under photochemical conditions. A Fisher-Porter
reactor was charged with a stir bar, potassium chloride (2.01 mmol),
ammonium iodate (7.7 mmol), MeTFA (5.37 mmol) and HTFA
(8 mL). The reactor was sealed and pressurized with argon. The
reactor was then placed on a stir plate in front of the Hg lamp and
stirred in the chamber, which was closed before turning the lamp
on. Following the reaction, the lamp was turned off, and the reactor
was vented. Internal standard was added to the reaction mixture,
and the mixture was stirred. An aliquot was removed for
Experimental Section
1
Caution: Many of the reagents and conditions described herein are
potentially hazardous. Appropriate safety procedures should be
consulted prior to handling concentrated acids, strong oxidants,
and mixtures of hydrocarbon substrates and oxygen, especially
under pressure.
centrifugation, and the supernatant was used for H NMR analysis.
Chlorination using N-chlorosuccinimide and benzoyl peroxide. A
Fisher-Porter reactor was charged with a stir bar, recrystallized N-
chlorosuccinimide (3.5 mmol), benzoyl peroxide (0.035 mmol), and
HTFA (8 mL) in the glovebox. If used, iodine (0.067 mmol) was also
added to the reactor. HTFA was heated to reflux for 1 h under N2
before being added to the reactor. The reactor was then sealed and
pressurized with 100 psi of methane. The reactor was placed on a
stir plate in front of the lamp and set to stir in the Hg lamp
General Comments and Materials. All reactions were carried out
under ambient atmosphere unless indicated otherwise. Potassium
chloride, ammonium iodate, potassium periodate, iodine, iodine
trichloride, iodine monochloride, trifluoroacetic acid, glacial acetic
ChemCatChem 2019, 11, 5045–5054
www.chemcatchem.org
5052
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
chamber, which was closed before turning the lamp on. Following
the reaction, the lamp was turned off and the reactor was vented.
Internal standard was added to the reaction mixture, and the
Keywords: methane · CÀ H functionalization · photochemistry ·
catalysis · oxidation
1
2
3
1
mixture was stirred. An aliquot was removed and used for H NMR
analysis. Centrifugation was not required as the mixture was
homogeneous.
4
5
6
7
8
9
[1] a) Q. Wang, X. Chen, A. N. Jha, H. Rogers, Renewable Sustainable Energy
Rev. 2014, 30, 1–28; b) BP Statistical Review of World Energy 2014,
British Petroleum, London, 2014.
[2] G. A. Olah, Á. Molnár, in Hydrocarbon Chemistry, 2nd ed., John Wiley &
Sons, 2003.
[3] a) Z. Zakaria, S. K. Kamarudin, Renewable Sustainable Energy Rev. 2016,
65, 250–261; b) J. A. Martens, A. Bogaerts, N. De Kimpe, P. A. Jacobs, G.
Marin, K. Rabaey, M. Saeys, S. Verhelst, ChemSusChem 2017, 10, 1039–
1055; c) N. J. Gunsalus, A. Koppaka, S. H. Park, S. M. Bischof, B. G.
Hashiguchi, R. A. Periana, Chem. Rev. 2017, 117, 8521–8573; d) R. H.
Crabtree, Chem. Rev. 1995, 95, 987–1007; e) P. Schwach, X. Pan, X. Bao,
Chem. Rev. 2017, 117.
[4] M. S. Matar, M. J. Mirbach, H. A. Tayim, in Catalysis in Petrochemical
Processes, Kluwer Academic Publishers, Boston, MA, 1988.
[5] X. Meng, X. Cui, N. P. Rajan, L. Yu, D. Deng, X. Bao, Chem. 2019, 5, 1–30.
[6] a) F. Schuth, Science 2019, 363, 1282–1283; b) A. Caballero, P. J. Pérez,
Chem. Soc. Rev. 2013, 42, 8809–8820.
[7] a) T. B. Gunnoe, in Physical Inorganic Chemistry: Reactions, Processes, and
Applications (Ed.: A. Bakac), John Wiley & Sons, Hoboken, New Jersey,
2010, pp. 495–549; b) B. G. Hashiguchi, S. M. Bischof, M. M. Konnick,
R. A. Periana, Acc. Chem. Res. 2012, 45, 885–898; c) E. G. Chepaikin, Russ.
Chem. Rev. 2011, 80, 363–396.
[8] R. A. Periana, D. J. Taube, S. Gamble, H. Taube, T. Satoh, H. Fujii, Science
1998, 280, 560–564.
Functionalization of cyclohexane. An 8 mL microwave vial was
charged with 98 mg (0.51 mmol) NH4IO3 and 10 mg (0.13 mmol)
KCl. HTFA (2 mL) and cyclohexane (54 μL, 0.5 mmol) were added to
the vial, generating a golden yellow mixture. The vial was sealed
with a crimp cap containing a septum. Thermal: the vial was placed
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
°
in an oil bath preheated to 100 C and heated for 1 h with vigorous
stirring. The vial was then allowed to cool to room temperature.
The reaction mixture appeared as a golden yellow solution with
abundant white solid. Photochemical: the vial was placed in front
of the corresponding light source (~3 cm) and allowed to stir at
room temperature for 24 h unless otherwise specified. After 24 h,
the reaction mixture appeared as a magenta solution with white
solid. Workup: the reaction mixture was removed from the vial and
filtered through a PTFE filter into a vial containing tetradecane
(0.24 mmol). The reaction vial was rinsed with dichloromethane
(5 mL) and filtered. The solutions were combined and washed two
times with 10 mL H2O. The organic layer was isolated and dried
over MgSO4. An aliquot was taken from the dichloromethane
solution and analyzed by GC-MS to quantify cyclohexyl-TFA,
cyclohexyl-chloride, 1,2-bis(trifluoroacetoxy)cyclohexane, and 1,2-
dichlorocyclohexane vs. tetradecane as the internal standard. The
solvent was removed from the dichloromethane solution and the
residue was dissolved in CDCl3 for analysis and quantification of 2-
chloro-1-trifluoroacetoxy-cyclohexane by 1H NMR spectroscopy.
Reactions were run in triplicate for each set of conditions.
[9] M. Ahlquist, R. J. Nielsen, R. A. Periana, W. A. Goddard, III., J. Am. Chem.
Soc. 2009, 131, 17110–17715.
[10] a) C. J. Jones, D. Taube, V. R. Ziatdinov, R. A. Periana, R. J. Nielsen, J.
Oxgaard, W. A. Goddard, Angew. Chem. Int. Ed. 2004, 116, 4726–4729;
b) R. A. Periana, D. J. Taube, E. R. Evitt, D. G. Löffler, P. R. Wentrcek, G.
Voss, T. Masuda, Science 1993, 259, 340–343; c) B. G. Hashiguchi, M. M.
Konnick, S. M. Bischof, S. J. Gustafson, D. Devarajan, N. Gunsalus, D. H.
Ess, R. A. Periana, Science 2014, 343, 1232–1237; d) L. C. Kao, A. C.
Hutson, A. Sen, J. Am. Chem. Soc. 1991, 113, 700–701; e) A. Koppaka,
S. H. Park, B. G. Hashiguchi, N. Gunsalus, C. R. King, M. M. Konnick, D. H.
Ess, R. A. Periana, Angew. Chem. Int. Ed. 2019, 58, 2241–2245; f) Z. An, X.
Pan, X. Y. Liu, X. Han, X. Bao, J. Am. Chem. Soc. 2006, 128, 16028–16029;
g) N. Basickes, T. E. Hogan, A. Sen, J. Am. Chem. Soc. 1996, 118, 13111–
13112; h) A. Sen, M. A. Benvenuto, M. R. Lin, A. C. Hutson, N. Basickes, J.
Am. Chem. Soc. 1994, 116, 998–1003; i) M. Muehlhofer, T. Strassner,
W. A. Herrmann, Angew. Chem. Int. Ed. 2002, 41, 1745–1747; Angew.
Chem. 2002, 114, 1817–1819.
Kinetic isotope effect studies. An 8 mL microwave vial was charged
with 52 mg (0.27 mmol) NH4IO3 and 10 mg (0.14 mmol) KCl, 57 mg
KIO4 (0.25 mmol) and 10 mg KCl (0.14 mmol), 58 mg ICl3
(0.25 mmol), or 45 mg ICl (0.28 mmol). HTFA (2 mL), cyclohexane-
h
12 (27 μL, 0.25 mmol), and cyclohexane-d12 (27 μL, 0.25 mmol) were
added to the vial. The vial was sealed with a crimp cap containing a
septum. The reaction mixture was subjected either to thermal or
photochemical reaction conditions as described above. The reac-
tion mixture was removed from the vial and filtered through a PTFE
filter. The reaction vial was rinsed with dichloromethane (5 mL) and
filtered. The solutions were combined and washed two times with
10 mL H2O. The organic layer was isolated and dried over MgSO4.
An aliquot was taken from the dichloromethane solution and
analyzed by GC-MS. Reactions were run in triplicate for each set of
conditions.
[11] a) J. S. J. Hargreaves, G. J. Hutchings, R. W. Joyner, S. H. Taylor, Appl.
Catal. A 2002, 227, 191–200; b) J. Li, S. Zhou, J. Zhang, M. Schlangen, D.
Usharani, S. Shaik, H. Schwarz, J. Am. Chem. Soc. 2016, 138, 11368–
11377.
[12] a) V. A. Roytman, D. A. Singleton, Science 2019, 364; b) C. Diaz-Urrutia, T.
Ott, Science 2019, 363, 1326-+.
[13] a) B. L. Conley, W. J. Tenn, K. J. H. Young, S. K. Ganesh, S. K. Meier, V. R.
Ziatdinov, O. Mironov, J. Oxgaard, J. Gonzales, W. A. Goddard, R. A.
Periana, J. Mol. Catal. A: Chem 2006, 251, 8–23; b) T. Zimmermann, M.
Soorholtz, M. Bilke, F. Schuth, J. Am. Chem. Soc. 2016, 138, 12395–
12400.
Acknowledgements
[14] R. Horn, R. Schlögl, Catal. Lett. 2015, 145, 23–39.
[15] a) J. R. Webb, T. Bolaño, T. B. Gunnoe, ChemSusChem 2011, 4, 37–49;
b) B. E. R. Snyder, M. L. Bols, R. A. Schoonheydt, B. F. Sels, E. I. Solomon,
Chem. Rev. 2018, 118, 2718–2768; c) P. Tomkins, M. Ranocchiari, J. A.
van Bokhoven, Acc. Chem. Res. 2017, 50, 418–425; d) M. Dusselier, M. E.
Davis, Chem. Rev. 2018, 118, 5265–5329.
[16] a) J. A. Labinger, J. Mol. Catal. A 2004, 220, 27–35; b) G. Zichittella, V.
Paunovic, A. P. Amrute, J. Perez-Ramirez, ACS Catal. 2017, 7, 1805–1817;
c) V. Paunović, G. Zichittella, P. Hemberger, A. Bodi, J. Pérez-Ramírez,
ACS Catal. 2019, 9, 1710–1725; d) S. G. Podkolzin, E. E. Stangland, M. E.
Jones, E. Peringer, J. A. Lercher, J. Am. Chem. Soc. 2007, 129, 2569–2576.
[17] a) E. Tschuikow-Roux, S. Paddison, Int. J. Chem. Kinet. 1987, 19, 15–24;
b) A. Fattahi, R. E. McCarthy, M. R. Ahmad, S. R. Kass, J. Am. Chem. Soc.
2003, 125, 11746–11750; c) A. Breed, M. F. Doherty, S. Gadewar, P.
Grosso, I. M. Lorkovic, E. W. McFarland, M. Weiss, Catal. Today 2005, 106,
301–304.
This research was sponsored by the U.S. Department of Energy,
Office of Energy Efficiency and Renewable Energy, Advanced
Manufacturing Office, under contract DE-AC05-00OR22725 with
UT-Battelle, LLC. Preliminary research by N.C.B. and T.Z. was
supported by the U.S. National Science Foundation award CHE-
1464578 to J.T.G.
Conflict of Interest
The authors declare no conflict of interest.
[18] a) M. C. Sze, H. Riegel, H. D. Schindler, The Lummus Company,
US4207268A, 1980; b) J. L. Dunn Jr., B. Posey Jr., Dow Chemical
ChemCatChem 2019, 11, 5045–5054
www.chemcatchem.org
5053
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Company, US2866830A, 1958; c) J. F. Hartwig, J. Am. Chem. Soc. 2016,
138, 2–24.
[19] a) G. C. Fortman, N. C. Boaz, D. Munz, M. M. Konnick, R. A. Periana, J. T.
Groves, T. B. Gunnoe, J. Am. Chem. Soc. 2014, 136, 8393–8401; b) S. E.
Kalman, D. Munz, G. C. Fortman, N. C. Boaz, J. T. Groves, T. B. Gunnoe,
Dalton Trans. 2015, 44, 5294–5298.
[26] a) G. M. Kramer, Tetrahedron 1986, 42, 1071–1077; b) E. S. Rudakov, L. K.
Volkova, Russ. Chem. Bull. 2008, 57, 1611–1626; c) D. R. Adams, P. D.
Bailey, I. D. Collier, S. A. H. Leah, C. Ridyard, Chem. Commun. 1996, 333–
334.
1
2
3
[27] CRC Handbook of Chemistry and Physics, 99 ed. (Ed.: J. R. Rumble), CRC
4
Press/Taylor & Francis, Boca Raton, FL, 2018.
[20] a) R. Fu, R. J. Nielsen, N. S. Liebov, W. A. Goddard III, T. B. Gunnoe, J. T.
Groves, J. Phys. Chem. C 2019, 123, 15674–15684; b) N. A. Schwartz, N. C.
Boaz, S. E. Kalman, T. Zhuang, J. M. Goldberg, R. Fu, R. J. Nielsen, W. A.
Goddard III, J. T. Groves, T. B. Gunnoe, ACS Catal. 2018, 8, 3138–3149.
[21] a) C. Walling, M. F. Mayahi, J. Am. Chem. Soc. 1959, 81, 1485–1489;
b) J. M. Tedder, Angew. Chem. Int. Ed. 1982, 21, 401–410; Angew. Chem.
1982, 94, 433–442.
[22] a) H. Kunkely, A. Vogler, Z. Naturforsch. 2013, 68b, 891–894; b) K.
Ohkubo, K. Hirose, Angew. Chem. Int. Ed. 2017, 56, 1–5; Angew. Chem.
2017, 129, 1–1; c) A. Hu, J.-J. Guo, H. Pan, Z. Zuo, Science 2018, 361,
668–672; d) M. Zhao, W. Lu, Org. Lett. 2017, 19, 4560–4563; e) M. Zhao,
W. Lu, Org. Lett. 2018, 20, 5264–5267; f) J. Xie, R. Jin, A. Li, Y. Bi, Q. Ruan,
Y. Deng, Y. Zhang, S. Yao, G. Sankar, D. Ma, J. Tang, Nature Catal. 2018,
1, 889–896.
[28] D. D. Tanner, J. E. Rowe, A. Potter, J. Org. Chem. 1986, 51, 457–460.
[29] H. Böhme, R. Schmitz, Chem. Ber. 1955, 88, 357–361.
5
6
7
8
[30] F. Cech, M. Linskeseder, E. Zbiral, Monatsh. Chem. 1976, 107, 1429–1436.
[31] a) Z. J. Li, A. Pirasteh, Int. J. Chem. Kinet. 2006, 38, 386–398; b) S. M.
Aschmann, R. Atkinson, Int. J. Chem. Kinet. 1995, 27, 613–622.
[32] G. Schmitz, Phys. Chem. Chem. Phys. 1999, 1, 1909–1914.
[33] a) Z. Li, A. Pirasteh, Int. J. Chem. Kinet. 2006, 38, 386–398; b) S. M.
Aschmann, R. Atkinson, J. Chem. Kinet. 1995, 27, 613–622.
[34] a) M. I. Christie, R. S. Roy, B. A. Thrush, Trans. Faraday Soc. 1959, 55,
1149–1152; b) W. W. Hess, E. S. Huyser, J. Kleinber, J. Org. Chem. 1964,
29, 1106–1109; c) C. C. Kelly, W. H. S. Yu, M. H. J. Wijnen, Can. J. Chem.
1970, 48, 603–606; d) dC. C. Kelly, M. H. J. Wijnen, J. Phys. Chem. 1969,
73, 2447–2448.
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
[23] CRC Handbook of Chemistry and Physics, 97th ed. (Ed.: W. M. Haynes),
Taylor & Francis Group, Boca Raton, FL, 2017, pp. 88–97.
[24] R. A. Periana, O. Mirinov, D. J. Taube, S. Gamble, Chem. Commun. 2002,
2376–2377.
Manuscript received: July 2, 2019
Revised manuscript received: August 7, 2019
Version of record online: September 25, 2019
[25] a) A. L. Ternay Jr., D. Deavenport, G. Bledsoe, J. Org. Chem. 1974, 39,
3268–3271; b) T. Hogan, A. Sen, J. Am. Chem. Soc. 1997, 119, 2642–2646.
ChemCatChem 2019, 11, 5045–5054
www.chemcatchem.org
5054
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim