A methodology for the photocatalyzed radical trifluoromethylation of indoles: A combined experimental and computational study

Article abstract of DOI:10.1016/j.jfluchem.2018.08.005

A methodology for the direct introduction of the trifluoromethyl group on to indole scaffolds is presented. The procedure involves the use of sodium trifluoromethylsulfinate (Langlois reagent) as the source of the trifluoromethyl radical, and is performed photochemically with 2-tert-butylanthraquinone as a photocatalyst. The reaction has also been probed computationally. Reaction kinetics and molecular orbital analyses from our quantum chemical computations successfully predict and rationalize the formation of the experimentally observed product and, in the case of 1-methylbenzimidazole, even reproduce the same qualitative trends in regioisomer preference.

Full text of DOI:10.1016/j.jfluchem.2018.08.005

Accepted Manuscript
Title: A methodology for the photocatalyzed radical
trifluoromethylation of indoles: a combined experimental and
computational study
Authors: Shelli A. Miller, Bas van Beek, Trevor A. Hamlin, F.
Matthias Bickelhaupt, Nicholas E. Leadbeater
PII:
S0022-1139(18)30283-5
DOI:
Reference:
https://doi.org/10.1016/j.jfluchem.2018.08.005
FLUOR 9209
To appear in:
FLUOR
Received date:
Revised date:
Accepted date:
11-7-2018
9-8-2018
9-8-2018
Please cite this article as: Miller SA, van Beek B, Hamlin TA, Bickelhaupt FM,
Leadbeater NE, A methodology for the photocatalyzed radical trifluoromethylation
of indoles: a combined experimental and computational study, Journal of Fluorine
Chemistry (2018), https://doi.org/10.1016/j.jfluchem.2018.08.005
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

A methodology for the photocatalyzed radical trifluoromethylation of
indoles: a combined experimental and computational study
Shelli A. Miller,a Bas van Beek,b Trevor A. Hamlin,*b F. Matthias Bickelhaupt,b,c and Nicholas
E. Leadbeater*a
a Department of Chemistry, University of Connecticut, 55 North Eagleville Road, Storrs,
Connecticut 06269-3060, United States
b Department of Theoretical Chemistry and Amsterdam Center for Multiscale Modeling
(ACMM), Vrije Universiteit Amsterdam, De Boelelaan 1083, NL-1081 HV Amsterdam, The
Netherlands
c Institute for Molecules and Materials (IMM), Radboud University, Heyendaalseweg 135, NL-
6525 AJ Nijmegen, The Netherlands
Graphical Abstract
HIGHLIGHTS
 A methodology for the direct introduction of the trifluoromethyl group on to indole
scaffolds is presented.
 Sodium trifluoromethylsulfinate (Langlois reagent) is used as the source of the
trifluoromethyl radical
 Reactions are performed photochemically with 2-tbutylanthraquinone as a
photocatalyst.

 Quantum chemical computations successfully predict and rationalize the formation
of the experimentally observed products.
ABSTRACT
A methodology for the direct introduction of the trifluoromethyl group on to indole scaffolds
is presented. The procedure involves the use of sodium trifluoromethylsulfinate (Langlois
reagent) as the source of the trifluoromethyl radical, and is performed photochemically with
2-tert-butylanthraquinone as a photocatalyst. The reaction has also been probed
computationally. Reaction kinetics and molecular orbital analyses from our quantum
chemical computations successfully predict and rationalize the formation of the
experimentally observed product and, in the case of 1-methylbenzimidazole, even reproduce
the same qualitative trends in regioisomer preference.
KEYWORDS
Photocatalysis, Trifluoromethylation, Langlois reagent, Indoles, Density functional
calculations

1. INTRODUCTION
The incorporation of fluorine atoms or fluorine-containing groups into organic molecules is
a current topic of considerable interest both in academic and industrial settings.1,2 As a
result, synthetic strategies which achieve fluorination are desirable based on the properties
conferred to the fluorinated product.3 For example, carbon-fluorine bonds are not only
stronger than analogous C-H bonds, but are also more polarized, making them more resistant
against oxidative metabolism.4 While monofluoro- and difluoro-moieties are certainly of
interest, the trifluoromethyl functionality (-CF3) has received significant attention, likely
because of its ability to serve as a bioisostere for several functional groups.5 This bioisosteric
property can be used to adjust the steric and electronic properties of a compound, or to
prevent metabolic degradation.6 A range of synthetic strategies are available for performing
trifluoromethylation reactions,7 examples being the use of the Umemoto,8 Togni,9 Langlois,10
and Rupert-Prakash11 reagents (Figure 1).
Figure 1. Structures of common trifluoromethylating reagents
One approach to the direct trifluoromethylation of substrates involves the use of
trifluoromethyl radical sources.12 Other methods of trifluoromethylating, such as cross-
coupling reactions, are at a disadvantage to direct radical methods because they require pre-
functionalized substrates, while direct radical methods do not. In the realm of radical
trifluoromethylation reactions, the Langlois reagent has proven particularly popular, being

a relatively inexpensive, bench stable solid. It has garnered much attention since Baran and
co-workers reported its application in an efficient methodology for the trifluoromethylation
of heterocycles in 2011 (Figure 2).13 Since then, a number of peroxide-free methods have
been proposed. These generally rely on the use of either transition metals, inert atmosphere,
harsh conditions, or combination thereof.14-17
In hopes of developing an efficient and mild approach to radical trifluoromethylation,
some have been turning to the use of photocatalysis. The popularity of photocatalysis as a
tool in preparative organic chemistry has been growing exponentially in recent years.18
These methods have opened avenues to chemical transformations that are otherwise
unachievable. Photochemical approaches to radical trifluoromethylation using the Langlois
reagent have been reported, but they tend to require rigorously anhydrous and anaerobic
conditions, as well as the use of transition metals (Figure 3).18-21 Nicewicz found that the
Langlois reagent could be used, in conjunction with an organic acridinium based
photocatalyst, to facilitate the hydrotrifluoromethylation of alkenes, resulting in 25-74%
yields.21 This method, using blue LED light (λ = 450 nm), requires stoichiometric amounts of
methyl thiosalicylate or thiophenol, which act as hydrogen atom donors. Similarly, Lefebvre
later reported a hydrotrifluoromethylation method which explored the reactivity of
electron-deficient substrates.20 They employ a benzophenone-based photocatalyst in
conjunction with UV light (λ = 355 nm) and use hexafluoroisopropanol (HFIP) as a super-
stoichiometric additive, lending some credence to previous observations that HFIP can act
as a proton donor rather than a hydrogen atom donor. They also demonstrate that similar
yields can be obtained when employing an iridium-based photocatalyst and blue light. Itoh
and co-workers have reported the use of an anthraquinone derivative as an organic

photocatalyst for the trifluoromethylation of arenes and heteroarenes, however, this method
still requires an inert atmosphere, and has a limited substrate scope.19 Interestingly, when
there is more than one reactive site, a degree of regioselectivity is observed in electron-rich
arenes.
Figure 2. Examples of the use of the Langlois reagent in trifluoromethylation reactions

Figure 3. Examples of the use of the Langlois reagent in photo-mediated
trifluoromethylation reactions
Unrelated to organofluorine chemistry, the Itoh group have recently reported a
methodology for the visible-light mediated cross-dehydrogenative C-H amination of indoles
with phthalimides, employing 2-tert-butylanthraquinone as a photocatalyst (Figure 4).23
Having already developed a number of non-radical approaches to trifluoromethylation,24 the
reports by Itoh piqued our interest and we sought to develop a methodology for the photo-
mediated radical trifluoromethylation of indoles using Langlois reagent and an
anthraquinone photocatalyst. Incorporation of fluorine into indole scaffolds is very
attractive due to the prevalence of indoles in natural products and pharmaceuticals. Photo-
mediated trifluoromethylation of indoles has been the subject of previous studies, although
not using Langlois reagent (Figure 5).25 We wanted to expand on these previous reports,
offering an additional approach for the synthesis of these valuable products. Our metal-free
method is convenient in that rigorously anhydrous and / or anaerobic conditions are not
required. While product yields are moderate, they are not unlike many of the other methods

reported in the literature. To augment our synthetic chemistry, we employed computational
chemistry to help us understand why yields for 2- or 3-substituted trifluoromethylated
indoles can be quite low.
Figure 4. The visible-light mediated cross-dehydrogenative C-H amination of indoles
Figure 5. Key previous methods for photo-mediated trifluoromethylation of indoles, and our
approach
2. RESULTS AND DISCUSSION
As a substrate for reaction optimization, we selected 1-methyl-2-phenylindole (1a).
Employing DMF as our solvent (0.1 M in substrate), 1 eq of Langlois reagent, 0.6 eq of
potassium carbonate as a base, 10 mol% 2-tert-butylanthraquinone (AQN) as a
photocatalyst, and a blue LED light source, we obtained a 10 % yield by NMR of the desired

trifluoromethylated product after 24 h (Table 1, entry 1). The reaction was performed in a
sealed Pyrex test tube using approximately 50 mg of 4 Å molecular sieves to sequester water.
The λ-max of AQN is 255 nm, and keeping all parameters constant with the exception of
switching the light source to 255 nm LEDs and using a quartz reaction tube increased the
yield to 32% (entry 2). Changing the photocatalyst to 9-mesityl-10-methyl acridinium
tetrafluoroborate and using blue light, a 10% yield was obtained, suggesting that AQN was
superior (entry 3). Doubling both the photocatalyst loading and also the quantity of Langlois
reagent increased the yield to 47% (entry 4). Changing the solvent to acetonitrile not only
increased the product yield to 50%, but also dramatically decreased the necessary reaction
time to 3 h (entry 5). Cesium carbonate and ammonium carbonate were screened as
alternative bases and both found to be on par with potassium carbonate (entries 6 and 7).
Thus, we decided to proceed using ammonium carbonate due to its increased solubility in
acetonitrile, thereby allowing for the more effective stirring of the reaction mixture.
Performing the reaction either open to the atmosphere or under a balloon of air slightly
increased product conversion, but at the expense of increasing reaction time (entries 8 and
9). An atmosphere of nitrogen doubled the reaction time with no effect on yield (entry 10).
Under an oxygen atmosphere, the reaction was greatly accelerated at the expense of the
desired product (entry 11), a number of side products being formed. While these were not
identified, they are believed to be polymeric in nature and if the reaction is left for longer,
decomposition occurs. To avoid loss of solvent due to evaporation and the formation of these
undesired side products, we decided to proceed using a balloon of air. Knowing that HFIP
can play a profound effect on reaction rate and outcome,26 we performed two trials with the
addition of 0.5 eq and 0.25 eq of HFIP to the reaction mixture. Using 0.5 eq HFIP increased

the yield to 57% after 48 h (entry 12), but when reduced to 0.25 eq HFIP we obtained a 61%
yield after just 24 h (entry 13). These conditions were taken on to the substrate screen. Three
negative control experiments were also run to confirm that all components of the reaction
mixture were necessary. The reaction did not occur in the absence of light, photocatalyst, or
base (entries 14-16). In addition, performing the reaction under an atmosphere of nitrogen
along with the addition of HFIP did not have a discernable effect on the outcome of the
reaction (entry 17).
Table 1. Optimization of reaction conditionsa
Entry
Solvent
DMF
Catalyst (mol%)
AQN (10)
Base
CF3SO2Na (eq)
Additive (eq)
Light
blue
255
Atm
Time (h)
20
Yield (%)
1
2
K2CO3
K2CO3
1
1
-
-
Closed
Closed
10
32
20
DMF
AQN (10)
20
3
DMF
9-Ms-10-MeAcr (10)
K2CO3
1
-
blue
Closed
10
4
5
6
7
8
DMF
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
AQN (20)
AQN (20)
AQN (20)
AQN (20)
AQN (20)
AQN (20)
AQN (20)
AQN (20)
AQN (20)
AQN (20)
-
K2CO3
K2CO3
Cs2CO3
2
2
2
2
2
3
3
3
2
2
2
2
2
3
-
-
-
-
-
-
-
-
255
255
255
255
255
255
255
255
255
255
255
255
-
Closed
Closed
Closed
Closed
Open
Lab air
N2
O2
Lab air
Lab air
Lab air
Lab air
Lab air
N2
20
3
3
3
9
47
50
46
50
57
55
57
37
57
(NH4)2CO3
(NH4)2CO3
(NH4)2CO3
(NH4)2CO3
(NH4)2CO3
(NH4)2CO3
(NH4)2CO3
(NH4)2CO3
-
9
8
10
11
12
13
14
15
16
17
48
2.5
48
24
24
24
24
48
HFIP (0.5)
HFIP (0.25)
HFIP (0.25)
HFIP (0.25)
HFIP (0.25)
HFIP (0.5)
61
no product
no product
no product
54
AQN (20)
AQN (20)
AQN (20)
(NH4)2CO3
(NH4)2CO3
255
a Reactions were performed on the 0.3 mmol scale at a concentration of 0.1 M, using 0.6 eq base. They were monitored by TLC and worked up
after the disappearance of the starting material. b Yield by quantitative 1H NMR.
We began our substrate scope using 2-substituted indoles as reagents (Table 2). We
determined product yields initially using quantitative NMR spectroscopy on the crude
product mixture and then isolating the products using preparative thin-layer

chromatography. Substrates that were not N-methyl-substituted were found to have similar
reactivity to their N-methyl-substituted analogues (Table 2, entries 1-3). The substituent at
the 2-position was varied. Placing a methyl group at the 2-position led to lower conversions
(entries 4 and 5). One other nitrogen protecting group, MOM, was found to be comparable to
the N-methyl protected analogue (entry 6). Unfortunately, when trying to expand the scope
to non-nitrogen containing systems, we found benzofurans and the more biologically active
uracil were unreactive (entries 7 and 8).
Table 2. Substrate screening of 2-substituted indoles
Entry
1
Product
Yieldb (%)
46 (61)
2
3
4
5
47 (67)
30 (61)
16 (33)
30 (48)
6
7
8
38 (46)
N.R.
N.R.

a
Indole (0.3 mmol, 1 eq), sodium trifluoromethylsulfinate (0.6 mmol, 2 eq), 2-tert-butyl anthraquinone (0.06 mmol, 0.2 eq), ammonium carbonate
(0.18 mmol, 0.6 eq), HFIP (0.075 mmol, 0.25 eq), and acetonitrile (3 mL, 0.1 M) in a sealed quartz tube with a balloon of air and irradiated at 255
nm for 24 h. N.R. = no reaction. b Isolated yields; values in parentheses are yields obtained using quantitative 1H-NMR spectroscopy.
When using 3-substituted indoles as substrates, product conversions were lower
than in the case of the 2-substituted analogues (Table 3). The formation of multiple
unidentified side products and decomposition was observed, thus making for a challenging
isolation of the desired product. In the case of unsubstituted indoles, both 2- and 3-
trifluoromethylated products were formed, this not being unexpected (Table 4). In light of
the range of side products formed in some of these reactions, we decided to screen 1-
methylbenzimidazole as a substrate and we again obtained a range of trifluoromethylated
products, but in this case, we characterized and isolated each of them (Table 5).
Table 3. Substrate screening of 3-substituted indoles
Entry
1
Product
Yieldb (%)
3 (10)
2
3
(5)
25 (39)
a Indole (0.3 mmol, 1 eq), sodium trifluoromethylsulfinate (0.6 mmol, 2 eq), 2-tert-butyl anthraquinone (0.06 mmol, 0.2 eq), ammonium
carbonate (0.18 mmol, 0.6 eq), HFIP (0.075 mmol, 0.25 eq), and acetonitrile (3 mL, 0.1 M) in a sealed quartz tube with a balloon of air and
irradiated at 255 nm for 24 h. b Isolated yields; values in parentheses are yields obtained using quantitative 1H-NMR spectroscopy.

Table 4. Substrate screening of unsubstituted indoles
Entry
1
Product
Yieldb (%)
(12)
2
3
4
(26)
(4)
(5)
a Indole (0.3 mmol, 1 eq), sodium trifluoromethylsulfinate (0.6 mmol, 2 eq), 2-tert-butyl anthraquinone (0.06 mmol, 0.2 eq), ammonium carbonate
(0.18 mmol, 0.6 eq), HFIP (0.075 mmol, 0.25 eq), and acetonitrile (3 mL, 0.1 M) in a sealed quartz tube with a balloon of air and irradiated at 255
nm for 24 h. b Values in parentheses are yields obtained using quantitative 1H-NMR spectroscopy
Table 5. Product distribution obtained in the reaction of 1-methylbenzimidazole (3)
Entry
1
Product
Yieldb (%)
7
2
19
3
4
6 (1:1 ratio)
3
a 1-Methylbenzimidazole (0.3 mmol, 1 eq), sodium trifluoromethylsulfinate (0.6 mmol, 2 eq), 2-tert-butyl anthraquinone (0.06 mmol, 0.2 eq),
ammonium carbonate (0.18 mmol, 0.6 eq), HFIP (0.075 mmol, 0.25 eq), and acetonitrile (3 mL, 0.1 M) in a sealed quartz tube with a balloon of air
and irradiated at 255 nm for 24 h. b Isolated yields.
In order to probe the reaction in more detail, the regioselectivity of the aromatic
CF3 radical addition was evaluated using DFT calculations at the (U)BP86/TZ2P level27-

29 with the ADF program.30,31 Acetonitrile (ε = 37.5) was simulated in all optimizations using
the COSMO solvation model.32-35 Gibbs free energies (ΔG) account for zero-point and thermal
energy, changes in volume and pressure, and entropy effects at 298.15 K and 1 atm. The
regioselectivity of the trifluoromethylation of 1-methyl-2-phenyl indole (1a) and 1-
methylbenzimidazole (3) have been evaluated, as the former serves as a good model system
with a common core structure and the latter has all other identified minor products fully
characterized (Table 5).
The potential energy surface was calculated based on the generally accepted model
of the photochemical generation of the CF3 radical (4) and subsequent addition to the indole
(1a) or benzimidazole (3). This approach is in line with the method proposed by Houk and
co-workers, which assumes that the energy associated with the aromatic CF3 radical adduct
(2.IC) is an adequate approximation to the energy of the actual transition state (Figure 6).36
The low experimentally and computationally determined barriers reported for CF3 radical
additions motivate the use of this approach.37-40 Barriers are expected to be further lowered
due to the stabilizing influence of the aromatic rings.41
Figure 6. Reaction scheme for CF3 functionalization. HAr = 1-methy-2-phenylindole (1a) or
1-methylbenzimidazole (3)

The Gibbs free energies (∆GIC – intermediate complex, ∆GP – products) associated
with trifluoromethyl radical addition to compounds 1a and 3 are shown in Table 6. These
substrates are primarily trifluoromethylated at C3 and C4, respectively, and analysis of the
∆GIC values reveals that radical addition at those positions is preferred compared to the other
possible sites. The reaction energy (∆GP) does not correlate with the experimentally
observed major product, thus these reactions likely operate under kinetic control. The low
barrier for CF3 radical addition leads to the intermediate complex (IC), which is then rapidly
quenched via a highly exergonic reaction leading to the product (P). Interestingly, CF3 radical
addition to 1a at C3 has a very small ∆∆GIC value (ca 0.3 kcal mol-1), possibly explaining the
poor yields.
Table 6: Computed reaction energiesa for the stationary points of 1a and 3 [intermediate
complexes (IC) and products (P)]. Energetically preferred pathways are highlighted in bold
Position:
2
-
3
4
5
6
7
IC
P
–8.9
–8.6
–2.2
–5.9
–3.1
-
–60.4 –61.7 –63.8
–61.8
–13.7
–71.9
–55.9
–12.7
–66.9
IC –14.8
-
-
–16.6 –11.6
–70.8 –72.3
P
–65.3
a Gibbs free energies (kcal mol-1, 298.15 K, 1 atm) computed at the COSMO(MeCN)-(U)BP86/TZ2P level
Focusing on the analysis of compound 3, for which we have experimentally identified
and quantified all possible regioisomers, the same general trends in regioselectivity are
observed as initially shown for 1a. The ∆GIC energies correctly rationalize the same

qualitative trends in regioselectivity. Nevertheless, while the computations reproduce the
correct trend, a systematic overestimation of the regioselectivity is observed (Table 7), i.e.
the computed values of ∆∆GIC are consistently too large. This may be the result of neglecting
the TS associated with CF3 radical addition, with a ∆∆GTS of merely 0.5 kcal mol-1 being
sufficient to reproduce the observed ratio between positions 4 and 7.
Table 7: Experimental (Exp.) and computed (Comp.) product ratios. Computed ratios based
on ∆∆GIC.
Position:
2
4
5
6
7
Exp.
0.20
0.54
0.17a
0.09
Comp.
0.06
0.93
0.00
0.01
0.00
a Combination of regioisomer 5 and 6
Next, we rationalized the preference for the observed reaction sites based on a
frontier molecular orbital (FMO) analysis. The key FMO interaction is a two-center/three-
electron bond between the HOMO1a,3 and SOMOCF3, with calculated energy gaps of 1.7 and
1.1 eV for the IC of 1 and 3a, respectively (Figure 7). The other FMO interaction, between the
SOMOCF3 and LUMO1a,3, is weaker with larger energy gaps (1a – 4.0 eV; 3 – 4.0 eV). The small
HOMO1a,3–SOMOCF3 gap, combined with a favorable orbital overlap (1a – 0.29; 3 – 0.27),
emphasizes the established electrophilic nature of the CF3 radical and the primary role of the
SOMOCF3 as an electron-acceptor, rather than donor.42 Additionally, we found that the carbon
with the largest HOMO1a,3 coefficient is also the one that leads to the kinetically favorable IC
(Figure 8, 1a – C3, 3 – C4). This evidence highlights the key FMO interactions underpinning
the observed reactivity of the indoles in this radical addition reaction.

LUMOCF3
LUMO1a
LUMO3
HOMO1a
HOMO3
SOMOCF3
4
3
N
Ph
CF3
N
N
4
1a
3
De HOMO1a,3–SOMOCF3 (eV)
Orbital Overlap (S)
0.29
1a
3
1.7
1.1
0.27
De SOMOCF3LUMO1a,3 (eV)
Orbital Overlap (S)
0.17
1a
3
4.0
4.0
0.23
Figure 7. Indole/benzimidazole (1a and 3) and CF3 (4) frontier molecular orbital (FMO)
gaps and overlaps of the kinetically most favorable intermediate complexes (C3 and C4,
respectively). FMO overlap and energy gaps are based on restricted open-shell molecular
fragments
Figure 8: Qualitative representation of the HOMO1a,3 and LUMO1a,3 based on MO coefficients
calculated at the COSMO(MeCN)-(U)BP86/TZ2P level

CONCLUSION
We have developed a methodology for the direct introduction of the trifluoromethyl group
on to indole scaffolds. The procedure involves the use of sodium trifluoromethylsulfinate
(Langlois reagent) as the source of the trifluoromethyl radical and is performed
photochemically with 2-tert-butylanthraquinone as a photocatalyst. We have also probed
the reaction computationally. Our computed reaction kinetics successfully predict the
formation of the experimentally observed product and, in the case of 1-
methylbenzimidazole, even reproduce the same qualitative trends in regioisomer
preference. This regioselectivity was rationalized by means of an orbital analysis, which
highlighted the key HOMO1a,3|SOMOCF3 interaction due to its small energy gap and favorable
overlap. Additionally, we discovered that the carbon associated with the kinetically most
favorable site consistently has the largest contribution to the HOMO1a,3. This orbital analysis
serves as an affordable method to predict the experimental regioselectivity of this
trifluoromethyl radical addition reaction.
4. EXPERIMENTAL SECTION
4.1. Determination of product conversion using quantitative NMR spectroscopy
NMR Spectra (1H, 13C, 19F) were performed at 298 K on a Brüker Avance III nanobay 400 MHz
1
NMR spectrometer equipped with 5 mm BBFO probe. H-NMR Spectra obtained in CDCl3
were referenced to residual non-deuterated chloroform (7.26 ppm) in the deuterated
solvent. 13C-NMR Spectra obtained in CDCl3 were referenced to chloroform (77.2 ppm). 19F-
NMR spectra were referenced to hexafluorobenzene (–162.4 ppm). Product conversions
determined using quantitative NMR spectroscopy were obtained via the ERETIC method

(Electronic REference To access In-vivo Concentrations)43 in Brüker TopSpin 3.5 and
calculated using 1,2,4,5-tetramethylbenzene (83.44 mM) as an external standard, with a P1
value of 15.62 microseconds and D1 value of 5 s.
4.2. General procedure for the preparation of trifluoromethyl-substituted indoles: The
desired indole (0.3 mmol, 1 eq) was mixed with sodium trifluoromethylsulfinate (0.6 mmol,
2 eq), 2-tert-butylanthraquinone (0.06 mmol, 0.2 eq), ammonium carbonate (0.18 mmol, 0.6
eq), hexafluoroisopropyl alcohol (0.075 mmol, 0.25 eq), and acetonitrile (3 mL, 0.1 M) in a
sealed quartz tube with a balloon of air and irradiated at 255 nm for 24 h. The reaction was
quenched with water and extracted with dichloromethane or EtOAc (3 x 30 mL). The
combined organic layers were dried over sodium sulfate, filtered, and the solvent removed
in vacuo. The desired product was then isolated using preparative TLC with the appropriate
solvent system.
1-Methyl-2-phenyl-3-(trifluoromethyl)-1H-indole (2a)17 was purified using 95:5 Hex:EtOAc to
give a yellow solid (38 mg, 46%). 1H NMR (400 MHz, CDCl3) δ ppm 7.88-7.78 (dd, 1H), 7.55-
7.47 (m, 3H), 7.45-7.37 (m, 3H), 7.34 (m, 1H), 7.27 (td, J = 7.5, 7.0, 1.2 Hz, 1H), 3.54 (s, 3H).
13C NMR (100 MHz, CDCl3, 100 MHz) δ ppm 140.9 (q, J = 4.0 Hz), 136.5, 130.6 (q, J = 1.2 Hz),
130.3, 129.5, 128.5, 124.9 (q, J = 267.2 Hz), 124.6 (q, J = 1.8 Hz), 123.1, 121.7, 120.0 (q, J = 1.5
Hz), 110.1, 104.2 (q, J = 35.2 Hz), 30.9. 19F NMR (376 MHz, CDCl3) δ ppm -53.39.
5. ACKNOWLEDGEMENTS
This work was supported financially by the University of Connecticut. Additionally, we thank
the Netherlands Organization for Scientific Research (NWO) for financial support through

the Planetary and Exo-Planetary Science program (PEPSci) and the Dutch Astrochemistry
Network (DAN). We would also like to thank undergraduate students Joshua Paolillo and
Madeline Williams (University of Connecticut) for contributions to the experimental
components of this project.
6. SUPPORTING INFORMATION
1
13
19
Experimental and computational details, H, C, and F NMR spectra of the compounds
prepared.

REFERENCES
1.
2.
A. Harsanyi, G. Sandford, Green Chem. 17 (2015) 2081-2086.
(a) W. K. Hagmann, J. Med. Chem. 51 (2008) 4359-4369. (b) D. O’Hagan, Chem. Soc.
Rev. 37 (2008) 308-319.
3.
4.
(a) T. Liang, C. N. Neumann, T. Ritter, Angew. Chem. Int. Ed. 52 (2013) 8214-8264;
(b) H. Vorbrueggen, Helv. Chim. Acta 94 (2011) 947-965; (c) K. Müller, C. Faeh, F.
Diederich, Science 317 (2007) 1881-1886.
(a) R. R. Frey, C. K. Wada, R. B. Garland, M. L. Curtin, M. R. Michaelides, J. Li, L. J.
Pease, K. B. Glaser, P. A. Marcotte, J. J. Bouska, S. S. Murphy, S. K. Davidsen, Bioorg.
Med. Chem. Lett. 12 (2002) 3443-3447; (b) A. S. Madsen, C. A. Olsen, Med. Chem.
Commun. 7 (2016) 464-470.
5.
(a) N. A. Meanwell, J. Med. Chem. [Online early access]. DOI:
10.1021/acs.jmedchem.7b01788. Published Online: February 5, 2018.
https://pubs.acs.org/doi/abs/10.1021%2Facs.jmedchem.7b01788 (accessed July 6,
2018). (b) I. G. Molnár, C. Thiehoff, M. C. Holland, R. Gilmour, ACS Catal. 6 (2016)
7167-7173; (c) N. A. Meanwell, J. Med. Chem. 54 (2011) 54, 2529-2591.
M. Jagodzinska, F. Huguenot, G. Candiani, M. Zanda, ChemMedChem 4 (2009) 49-51.
S. Barata‐Vallejo, B. Lantaño, A. Postigo, Chem. Eur. J. 20 (2014) 16806-16829.
T. Umemoto, S. Ishihara, Tetrahedron Lett. 31 (1990) 3579-3582.
6.
7.
8.
9.
P. Eisenberger, S. Gischig, A. Togni, Chem. Eur. J. 12 (2006) 2579-2586.
B. R. Langlois, E. Laurent, N. Roidot, Tetrahedron Lett. 32 (1991) 7525-7528.
10.

11.
(a) R. Krishnamurti, D. R. Bellew, G. K. S. Prakash, J. Org. Chem. 56 (1991) 984-989.
(b) I. Ruppert, K. Schlich, W. Volbach, Tetrahedron Lett. 25 (1984) 2195-2198.
A. Studer, Angew. Chem. Int. Ed. 51 (2012) 8950-8958.
12.
13.
Y. Ji, T. Brueckl, R. D. Baxter, Y. Fujiwara, I. B. Seiple, S. Su, D. G. Blackmond, P. S.
Baran, Proc. Natl. Acad. Sci. U.S.A. 108 (2011) 14411-14415.
14.
A. Deb, S. Manna, A. Modak, T. Patra, S. Maity, D. Maiti, Angew. Chem. Int. Ed. 52
(2013) 9747-9750.
15.
16.
T. Patra, A. Deb, S. Manna, U. Sharma, D. Maiti, Eur. J. Org. Chem. (2013) 5247-5250.
Y.-D. Yang, K. Iwamoto, E. Tokunaga, N. Shibata, Chem. Commun. 49 (2013) 5510-
5512.
17.
18.
(a) B. Zhang, A. Studer, Org. Lett. 16 (2014) 1216-1219. (b) Y. Ye, K. P. S. Cheung, L.
He, G. C. Tsue, Org. Lett. 20 (2018) 1676-1679.
(a) C. K. Prier, D. A. Rankic, D. W. C. Macmillan, Chem. Rev. 113 (2013) 5322-5363;
(b) J. M. R. Narayanam, C. R. J. Stephenson, Chem. Soc. Rev. 40 (2011) 102-113; (c) N.
A. Romero D. A. Nicewicz, Chem. Rev. 116 (2016) 10075-10166; (d) M. H. Shaw, J.
Twilton, D. W. C. Macmillan, J. Org. Chem. 81 (2016) 6898-6926.
L. Cui, Y. Matusaki, N. Tada, T. Miura, B. Uno, A. Itoh, Adv. Synth. Catal. 355 (2013)
2203-2207.
19.
20.
(a) Q. Lefebvre, N. Hoffmann, M. Rueping, Chem. Commun. 52 (2016) 2493-2496. (b)
L. Zhao, P. Li, X. Xie, L. Wang, Org. Chem. Front. 5 (2018) 1689-1697. (c) H.-T. Qin, S.-
W. Wu, J.-L. Liu, F. Liu, Chem. Commun. 53 (2017) 1696-1699
D. J. Wilger, N. J. Gesmundo, D. A. Nicewicz, Chem. Sci. 4 (2013) 3160-3165.
21.

22.
(a) J. F. Berrien, M. Ourévitch, G. Morgant, N. E. Ghermani, B. Crousse, D. Bonnet-
Delpon, J. Fluorine Chem. 128 (2007) 839-843; (b) B. Satpathi, S. S. V. Ramasastry,
Angew. Chem. Int. Ed. 55 (2016) 1894-1898.
23.
24.
T. Yamaguchi, E. Yamaguchi, A. Itoh, Org. Lett. 19 (2017) 1282-1285.
(a) T. A. Hamlin, G. M. L. Lazarus, C. B. Kelly, N. E. Leadbeater, Org. Process Res. Dev.
18 (2014) 1253-1258; (b) T. A. Hamlin, C. B. Kelly, R. M. Cywar, N. E. Leadbeater, J.
Org. Chem. 79 (2014) 1145-1155. (c) T. A. Hamlin, C. B. Kelly, N. E. Leadbeater, Eur.
J. Org. Chem. (2013) 3658-3661; (d) C. B. Kelly, M. A. Mercadante, T. A. Hamlin, M. H.
Fletcher, N. E. Leadbeater, J. Org. Chem. 77 (2012) 8131-8141; (e) D. M. Rudzinkski,
C. B. Kelly, N. E. Leadbeater, Chem. Commun. 48 (2012) 9610-9612.
For key examples, see: (a) J. W. Beatty, J. J. Douglas, K. P. Cole, C. R. J. Stephenson,
Nat. Commun. 6 (2015) 7919; (b) D. A. Nagib, D. W. C. MacMillan, Nature 480 (2016)
224-228; (c) J. Jacquet, S. Blanchard, E. Derat, D. E.-L. Murr, L. Fensterbank, Chem.
Sci. 7 (2016) 2030-2036.
25.
26.
(a) B. Satpathi, S. S. V. Ramasastry, Angew. Chem. Int. Ed. 55 (2016) 1894-1898; (b)
B. Elsler, A. Wiebe, D. Schollmeyer, K. M. Dyballa, R. Franke, S. R. Waldvogel, Chem.
Eur. J. 21 (2015) 12321-12325; (c) Y.-L. Liu, T.-D. Shi, F. Zhou, X.-L. Zhao, X. Wang, J.
Zhou, Org. Lett. 13 (2011) 3826-3829.
27.
28.
29.
30.
A. D. Becke, Phys. Rev. A 38 (1988) 3098-3100.
J. P. F. Perdew, Phys. Rev. B 33 (1986) 8822-8824.
E. van Lenthe, E. J. Baerends, J. Comput. Chem. 24 (2003) 1142-1156.
G. te Velde, F. M. Bickelhaupt, E. J. Baerends, C. Fonseca Guerra, S. J. A. van Gisbergen,
J. G. Snijders, T. Ziegler, J. Comput. Chem. 22 (2001) 931-967.

31.
C. Fonseca Guerra, J. G. Snijders, G. te Velde, E. J. Baerends, Theor. Chem. Acc. 99
(1998) 3910-403.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
C. C. Pye, T. Ziegler, Theor. Chem. Acc. 101 (1999) 396-408.
A. Klamt, G. Schüürmann, J. Chem. Soc. Perkin Trans. 2 (1993) 799-805.
A. Klamt, J. Phys. Chem. 99 (1995) 2224-2235.
A. Klamt, V. Jonas, J. Chem. Phys. 105 (1996) 9972-9981.
Y. Itoh, K. N. Houk, K. Mikami, J. Org. Chem. 71 (2006) 8918-8925.
J. M. Sangster, J. C. J. Thynne, J. Phys. Chem. 73 (1969) 2746-2750.
J. M. Tedder, J. C. Walton, Adv. Phys. Org. Chem. 16 (1978) 51-86.
A. Bottoni, J. Chem. Soc. Perkin Trans. 2 (1996) 2041-2047.
J. Korchowiec, T. Uchimaru, J. Phys. Chem. A 102 (1998) 2439-2442.
A. P. Stefani, M. Szwarc, A. Kuboyama, S. Matsuzaki, H. Takagi, H. Arano, T. Zytowski,
H. Fischer, J. Korchowiec, T. Uchimaru, J. Am. Chem. Soc. 102 (1997) 2439-2442.
A. P. Stefani, M. Szwarc, J. Am. Chem. Soc. 84 (1962) 3661-3666.
S. Akoka, L. Barantin, M. Trierweiler, Anal. Chem. 71 (1999) 2554-2557.
42.
43.