Traceless directing groups in radical cascades: From oligoalkynes to fused helicenes without tethered initiators

Article abstract of DOI:10.1021/ja510563d

We report the first example of a traceless directing group in a radical cascade. The chemo- and regioselectivity of the initial attack in skipped oligoalkynes is controlled by propargyl OR moiety. Radical translocations lead to the boomerang return of the radical center to the site of initial attack where it assists the elimination of the directing functionality via β-scission in the last step of the cascade. The Bu3Sn moiety continues further via facile reactions with electrophiles as well as Stille and Suzuki cross-coupling reactions. This selective radical transformation opens a new approach for the controlled transformation of skipped oligoalkynes into polycyclic ribbons of tunable dimensions. (Chemical Equation Presented).

Full text of DOI:10.1021/ja510563d

Subscriber access provided by UNIV OF MD - ALBIN KUHN LIB
Article
Traceless Directing Groups in Radical Cascades: From
Oligoalkynes to Fused Helicenes without Tethered Initiators
Igor V. Alabugin
J. Am. Chem. Soc., Just Accepted Manuscript • Publication Date (Web): 24 Dec 2014
Downloaded from http://pubs.acs.org on December 25, 2014
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Journal of the American Chemical Society
This document is confidential and is proprietary to the American Chemical Society and its authors. Do not
copy or disclose without written permission. If you have received this item in error, notify the sender and
delete all copies.
Traceless Directing Groups in Radical Cascades: From
Oligoalkynes to Fused Helicenes without Tethered Initiators
Journal: Journal of the American Chemical Society
Manuscript ID: ja-2014-10563d.R1
Manuscript Type: Article
Date Submitted by the Author: 22-Dec-2014
Complete List of Authors: Pati, Kamalkishore; Floridastate University, Chemistry and biochemistry
Gomes, Gabriel; Florida State University, Chemistry and Biochemistry
Harris, Trevor; Florida State University, Chemistry and Biochemistry
Hughes, Audrey; Florida State University, Chemistry and Biochemistry
Phan, Hoa; Florida State University, Department of Chemistry and
Biochemistry
Banerjee, Tanmay; Florida State University, Department of Chemistry and
Biochemistry
Hanson, Kenneth; Florida State University, Department of Chemistry and
Biochemistry
Alabugin, Igor; Florida State University, Chemistry and Biochemistry
ACS Paragon Plus Environment

Page 1 of 45
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Traceless Directing Groups in Radical Cascades:
From Oligoalkynes to Fused Helicenes without
Tethered Initiators
Kamalkishore Pati, Gabriel dos Passos Gomes, Trevor Harris, Audrey Hughes, Hoa Phan,
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
Tanmay Banerjee, Kenneth Hanson and Igor V. Alabugin
Department of Chemistry and Biochemistry, Florida State University,
Tallahassee, Florida 32306-4390
Abstract: We report the first example of a traceless directing group in a radical cascade. The
chemo- and regioselectivity of the initial attack in skipped oligoalkynes is controlled by
propargyl OR moiety. Radical translocations lead to the boomerang return of radical center to
the site of initial attack where it assists to the elimination of the directing functionality via -
scission in the last step of the cascade. The Bu3Sn moiety allows further via facile reactions
with electrophiles as well as Stille and Suzuki cross-coupling reactions. This selective radical
transformation opens a new approach for the controlled transformation of skipped oligoalkynes
into polycyclic ribbons of tunable dimensions.
Introduction:
Chemists have a love/hate relationship with protecting and directing groups.1 Although
such groups are very often needed to achieve the desired chemo- and regioselectivity of the key
chemical transformation in the presence of multiple targets, but like lingering guests, they
remain long after they have overstayed their welcome. However, use of such groups is often
unavoidable for achieving sufficient selectivity in cascade transformations of multifunctional
reactants.
The problem of selectivity comes to the fore in cascades aimed at the preparation of
polyaromatic ribbons from conjugated2 and skipped3 oligoalkynes. These processes correspond
to controlled “polymerization” of alkyne moieties positioned between the two rows of aromatic
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 45
1
2
3
4
5
6
7
8
9
rings via a selective sequence of “all-exo” cyclizations.4 Although the regioselectivity of
cyclizations is well controlled by the exo-preference for alkyne cyclizations,5 the key remaining
challenge in the design of these cascades involves achieving control over chemoselectivity of
initial radical attack. The secondary challenge lies in avoiding the formation of pentagonal
units at initiation and termination points of the oligoalkyne cascade. We discuss these
challenges below .
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1. Connection between benzannelated oligoalkynes and graphene ribbons
For the “polymerization cascade” to proceed to completion by utilizing each of the triple
bonds, the initial radical attack should proceed exclusively at the central alkyne of the
oligoalkyne precursor (Scheme 1).5a Such chemoselectivity is difficult to achieve because of the
close similarity in the electronic and steric properties of the multiple alkyne units. Initially, we
addressed this challenge via covalent attachment of a tethered initiator (“the weak link”) which
is directed at the correct alkyne target by geometric restraints imposed by the intramolecular
ACS Paragon Plus Environment

Page 3 of 45
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
trajectory.5b Although this approach solves the problem of chemoselectivity of activation, the
solution is not ideal because atoms of the tether remain in the molecule after the cascade.
Furthermore, when conjugated oligoalkynes are used as reactants, ring formation from
the first two alkyne moieties of benzannelated oligoalkynes, has to proceed via a 5-exo-dig path
(Scheme 2). As a consequence, the “polyacetylene ribbon” formed from the oligoalkyne always
contained a pentagonal unit. We have shown that the presence of this pentagon can be avoided
when the first reaction between the two alkynes are engineered to follow a 6-exo path, when the
starting oligoalkynes are changed from conjugated to “skipped” by adding one extra carbon.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2: Evolution of molecular design for the conversion of oligoalkynes into polyaromatic
ribbons
In this work, we disclose a new strategy for chemoselective reactions of oligoalkynes
that does not rely on the weak link for radical initiation. Instead, it achieves selective
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 45
1
2
3
4
5
6
7
8
9
intermolecular attack by using a propargylic methoxy group at the skipped alkynes for directing
tin radical at the correct position where it initiates the full cascade. The use of O-directing
functionalities in Sn-radical attack at alkynes has been documented6 with many interesting
examples in the recent work of Hale7 and Organ.8
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 order to avoid incorporation of the directing group in the final product, the sequence
of bond forming and bond breaking steps in the cascade has to be coordinated to eliminate the
directing group at the last step of the cascade, with concomitant aromatization of the polycyclic
ribbon. Our hypothesis was that the removal of the OMe would benefit from translocation of the
radical center if one can design a process where the radical, as a boomerang, returns to the
position of the initial intermolecular radical attack to the β-carbon relative to the departing
group.
Scheme 3. Translocation of radical coupled with loss of X
Under this condition, the loss of directing group occurs via fast β-scission as the last part
of a one-pot transformation and, thus, such group can be considered a traceless directing group.9
ACS Paragon Plus Environment

Page 5 of 45
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Not only does this new strategy provide a concise synthetic approach to the defect-free
preparation of the top part of the polyaromatic ribbon, but it also retains the Bu3Sn-substituent
as a convenient synthetic handle for further synthetic modifications of the polycyclic
framework.
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:
The starting 1,(2-alkynylphenyl)propargyl methyl ethers 1 are readily prepared from
commercially available 2-bromobenzaldehydes via a combination of Sonogashira cross-
coupling and nucleophilic addition of acetylide anions to the aldehyde (Table 1). The lithiated
alkoxide formed at the latter step can be converted either into a propargylic alcohol via
protonation or into propargyl methyl ethers via in situ addition of iodomethane. The propargyl
methyl ethers 1 were synthesized with a variety of substituents on rings A, B, and C and in good
yields (55-90%).
Table 1: 1-(2-alkynylphenyl)propargyl methyl ethers
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 45
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
Table 2: Optimization of reaction conditions
Entry
Reagent/initiatora
Condition
Yieldb
ACS Paragon Plus Environment

Page 7 of 45
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
Bu3SnH/AIBN
Et3SiH/AIBN
Et3SiH/AIBN
Bu3SnH/AIBN
Ph3SnH/AIBN
Bu3SnH/ABCN
65%
--c
Benzene, 16 h, 80°C
Benzene, 16 h, 80°C
Toluene, 14 h, 110°C
Toluene, 14 h, 110°C
Toluene, 14 h, 110°C
Toluene, 14 h, 100°C
--d
86%
40%
60%
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
7
8
9
Bu3SnH/DTBPB
Et3B/Bu3SnH
Et3B/Bu3SnH
40%
-- e
Toluene, 14 h, 100°C
THF, 12 h, rt
Benzene, 14 h, 70°C
-- f
a 1.3 equiv of reagent and 0.4 equiv initiator, b Isolated yields from silica chromatography,
Unreacted starting material (60%, 30%, 65%, and 75% respectively) and complex mixture of
products
c-f
Table 2 shows screening of various radical reagents and initiators for the model
transformation of 1a to 2a. The combination of Bu3SnH and AIBN in refluxing toluene was the
most efficient. The failure of silicon reagents indicates that a suitable organotin radical
precursor was essential for the success of this cascade. The reaction conditions were further
optimized by changing molar ratios of Bu3SnH / AIBN and varying flow rate using a syringe
pump (see Table 3). We found that maintaining steady concentration of AIBN is preferred over
the addition of initiator at once in the beginning of reaction. The addition of 0.04M
Bu3SnH/0.01M AIBN mixture at the flow rate of 1 mL/hr to the 0.03M solution of substrate in
toluene (110 °C, 14 h) provided 2a in excellent (86%) yield. Use of lower amounts of
Bu3SnH/AIBN led to incomplete conversions.
Table 3: Further Optimization of Bu3Sn-Addition to Alkyne 1a
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 45
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
Entry
AIBN (eq)
Bu3SnH (eq)
Rate
Conversion
Yieldb
__________________________________________________________________________
1
2
3
1
1
1
1
all at oncec
60%
70%
80%
40%
60%
70%
2
all at once
1.5
AIBN at once,
Bu3SnH 2 mL/ h
4
5
0.5
0.5
1.5
1.5
Bu3SnH + AIBN,
2 mL/ hd
100%
100%
76%
86%
Bu3SnH + AIBN,
1 mL/ h
6
7
0.4
0.2
1.3
1.5
Bu3SnH + AIBN,
1 mL/ h
100%
60%
86%
50%
Bu3SnH + AIBN,
1 mL/ h
______________________________________________ ________________________
b
aEquivalents are based on proportion to starting material, Isolated yields after column chromatography,
d
cTotal concentration 0.03M, Condition A: AIBN and Bu3SnH were dissolved in 2 mL toluene together.
Condition B: AIBN and Bu3SnH were dissolved separately and added simultaneously. Condition A and B
showed no change in yield (all reactions performed under argon atmosphere)
The scope of this reaction was verified by the successful cascade cyclization of
propargyl methyl ethers shown in Table 4. The range of substituents includes donor and
acceptor groups in the core benzene ring as well as at the both alkyne termini. As the result,
selective introduction of substituents of varying electronic nature at specific positions of rings
A, B, and C (Table 4) of the polycyclic product is possible. Although reaction conditions were
not optimized for each of the substrates 2b-2r, the isolated yields remained consistently high for
ACS Paragon Plus Environment

Page 9 of 45
Journal of the American Chemical Society
1
2
3
4
5
6
both the donor and the acceptor substituents (70-88%), indicating generality and robustness of
this cascade transformation.
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
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 45
1
2
3
4
Table 4: Radical Cascade of Propargyl Methyl Ethers
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 Isolated yields after column chromatography
ACS Paragon Plus Environment

Page 11 of 45
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
To examine the role of the methoxy substituent in the radical cascade, two alternate
substrates were prepared: a free propargyl alcohol 1s and a “deoxygenated” analogue with an
unsubstituted methylene bridge 1t (synthesis of these compounds is described in the SI part).
Even though the reaction of the free alcohol was noticeably slower than the reaction of
propargylic ether 1a, it still displayed comparable level of selectivity, providing an attractive
synthetic alternative to the reactions of OMe-substituted substrates. On the other hand, the
reaction of the “deoxygenated” substrate 1t is significantly less selective (Scheme 4). Yields
and reaction times for these experiments suggest that the presence of the oxygen substituent
plays an important role in the selectivity of this cascade reaction.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 4: Control experiments: effect of propargylic substitution on selectivity
Structure Determination:
1
Structures of the products were determined by the combination of H and 13C NMR
spectroscopy, which was complemented, in selected cases, by X-ray crystallography.10 The
1
characteristic H NMR features include disappearance of the reactants’ OCH3 and the
propargylic H peaks (at 3.5-3.7 and 5.3-5.9 ppm, respectively) to give rise to a 5.1-5.8 ppm
singlet for the remaining sp3 C-H in the products. In 13C NMR, disappearance of the OCH3 (56-
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 45
1
2
3
4
5
6
7
8
58 ppm), propargylic CH (71-73 ppm), and the four alkynyl carbons (84-92 ppm) along with the
appearance of a signal at 52-55 ppm for the sp3 carbon in the partially reduced five-membered
ring indicated completion of the cascade.
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 1: The ORTEP diagram for 2a
The naphthalene and phenylene moieties in the benzofluorene skeleton of 11-phenyl-
11H-benzo[a]fluorene 2a are approximately coplanar suggesting that efficient conjugation and
electronic communication should be possible between the different parts of the molecule. On
the other hand, the exocyclic aryl group is projected away from the polycyclic plane and
oriented in a way that minimizes steric repulsion with the aromatic hydrogens (Figure 1).
Figure 2: The ORTEP diagram for 2p-Sn
Chemo- and regioselectivity of Sn addition was confirmed by X-ray analysis of Sn-
precursor of 14-fluoro-11-phenyl-11H-benzo[a]fluorene 2p-Sn (Figure 2). As expected, the Sn
atom has approximately tetrahedral geometry with connections to one sp2 and three sp3 carbon
ACS Paragon Plus Environment

Page 13 of 45
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
atoms. Surprisingly, the average of Sn-Csp3 bond lengths (2.163 Å) is smaller than the Sn-Csp2
distance (2.171 Å).11 The anomalously long bond length in the latter case can be attributed to
the steric repulsion between the bulky Bu3Sn group and ortho-hydrogens of the polyclic moiety.
Furthermore, the steric clash of the butyl groups and the polycyclic structure leads to the
deviation of C-Sn-C angles (106.93o to 114.96o) from the ideal 109.5o tetrahedral valence angle
and the slight (~6o) deviation of Sn atom from the benzo[a]fluorene plane.
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
Computational Details:
The computational analysis of potential energy profiles involved in this multistep
transformation was performed in Gaussian 0912 with unrestricted M06-2X functional13 due to its
relatively accurate description of reaction and activation energies for a variety of chemical
processes including radical reactions.14 The LanL2DZ basis set was used for Sn. Chemcraft
1.715 and CYLView16 were used to render the molecules and orbitals. Frequency calculations
were performed to confirm each stationary point as either a minimum or a first-order saddle
point. NBO 3.0 program was used to analyze electronic properties of reactive intermediates.17
Results and Discussion:
Chemoselectivity in radical cascades:
Chemoselectivity is a key challenge in the design of cascade transformations of
multifunctional substrates. In a system, where functional groups are similar or almost identical
(i.e., oligoalkynes), such challenge is especially formidable. We had shown earlier that a new
application of dynamic covalent chemistry towards reversible radical reactions of Sn-radicals
and -systems can be used for the design of surprisingly selective transformation18 For example,
in reactions of aromatic enynes (Scheme 5) where an intermolecular radical attack can lead to
the formation of four different benzylic radicals (each capable of two cyclizations), we observed
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 45
1
2
3
4
5
6
7
8
9
the formation of a single product out of the 8 possible cyclic structures. The preferred product is
formed from the attack of the most reactive (vinyl) radical at the most reactive (alkene) π-bonds
via the most stereoelectronically favorable (exo) path. Assuming that the radicals can equilibrate
prior to the cyclization, this process can be characterized as a kinetic self-sorting of a pool of the
four equilibrating radicals via reaction with the lowest activation barrier. In accord with the
Curtin-Hammett principle, under the conditions where addition of Sn-radicals is reversible, it is
not the relative stability of radicals but the absolute energies of the respective cyclization
transition states that matter.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 5. Illustration of the proposed radical pool concept for the cyclization of conjugated en
ynes18a
The equilibrium between vinyl radicals is supported by the literature reports confirming
that addition of vinyl stannanes to double bonds is reversible.19 For example, Sn radicals are
known to isomerize alkenes fast and much faster than Si-based radicals. However, at this point,
the “radical pool” is a hypothetical model to fit the observed results.
Alternatively, a simpler source of selectivity can be found in a kinetic preference for the
Sn-radical attack at the propargylic position. Such attack would directly provide the correct
radical, positioned properly for initiating the cascade. In order to test for applicability of the two
models to the present system, we evaluated the relative reactivity of the two types of substituted
ACS Paragon Plus Environment

Page 15 of 45
Journal of the American Chemical Society
1
2
3
4
5
6
alkynes separately by carrying out an intermolecular competition experiment between propargyl
ether 3 and the diaryl alkyne 4 to determine the relative reactivity of these functionalities
towards Bu3SnH/AIBN system.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 6. Intermolecular competition in radical hydrostannation of alkynes - propargylic ether
3 vs. diphenyl acetylene (tolane) 4
Scheme 6 summarizes competition experiments for the 1:1 ratio mixture of propargylic
ether 3 and diphenyl acetylene 4 at the previously optimized cyclization conditions. Whereas
the propargylic ether was completely consumed, tolane 4 remained unreacted. The simplest
rationale for these experimental result is that the presence of the propargylic methoxy
substituent directs Sn addition by accelerating kinetic attack at the adjacent alkyne carbon.20
Guided by these results, we found that the calculated barrier for SnMe3 radical addition
at the propagylic position is remarkably low: ΔG≠ (110˚C) = 4.2 kcal/mol. The relatively low
energy of the TS can be rationalized by the several interactions between oxygen lone pairs, the
π system and the Sn radical as illustrated by the NBO energies in Scheme 7. Such Transition
State stabilization derives, at least partially, from unusual three-electron interactions between
the lone-pair of oxygen and the radical orbital at Sn to guide radical attack at the required
position.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 45
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 7: Selected electronic interactions (NBO analysis in kcal/mol) contributing to the
structure and stability of the transition state of the Me3Sn radical addition to the bis-alkyne 1a to
form the most stable radical A. Note that the radical center (nSn) is half occupied and, unlike a
lone pair, can serve as either a donor or an acceptor in 2nd order delocalizing interactions Top
left: (- and -spins summed) πC-C ⟶ σ*C-O; Top center: (α-spin) sp1.12 nO ⟶ nSn ; Top right:
(α-spin) n*Sn ⟶ π*C-C. This interaction is responsible for the formation of Sn-C bond. Bottom
left: (β-spin) sp3.67 nO ⟶ nSn; Bottom center: (β-spin) sp3.07 nO ⟶ nSn; Bottom right: (β-spin) πC-
C ⟶ nSn.
To better understand the thermodynamic landscape, we also compared the relative
stabilities of the four vinyl radicals that can be formed from the skipped diyne 1a and R3Sn
radical. Furthermore, we had also considered the fifth possibility – formation of a 2c,3e bond
between the radical and the oxygen’s lone pair but these species did not correspond to the
separate energy minimum and converged to the most stable vinyl radical A without barrier.
Both ΔE and ΔG (taking the energy of the separated enediyne and the SnR3 radical as the
reference point) were calculated for the four intermediates A, B, C, and D (using a truncated,
Me3Sn substituent). In this system, the radical intermediate leading to the experimentally
ACS Paragon Plus Environment

Page 17 of 45
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
observed pathway was found to be significantly more stable than the other vinyl radicals in
Scheme 8. This difference in stability originates from a combination of several factors. First, the
two more stable radicals (A and D) enjoy benzylic stabilization. However, efficiency of
benzylic resonance can be attenuated by steric effects. Furthermore, other electronic effects can
also offer significant stabilization and complicate the overall situation. For example, the only
non-benzylic radical (B) is, somewhat surprisingly, more stable than radical C. This inversion
of stability may be at least partially derived from the captodative stabilization of the radical
center by a vicinal acceptor (C-O) and a vicinal donor (C-Sn) bonds flanking radical 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
Scheme 8. Relative free energies of the four vinyl radicals formed by radical attack at the triple
bonds of the skipped enediyne 1a at 110oC.
Figure 3 illustrates the calculated geometries for the four vinyl radicals A - D. In all four
radicals, the vicinal C-Sn bond is aligned with the radical orbital in order to maximize
stabilizing hyperconjugative interaction with the radical. Despite a literature suggestion that this
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 45
1
2
3
4
5
6
7
8
9
interaction is unimportant,21 NBO analysis find this interaction to the strongest among the
interactions of Sn-C with the vicinal orbitals. Although this interaction is common for all four
species and can be neglected in the discussion of their relative stability, it is likely to impose
significant effect on the subsequent reactivity for each of the radicals.
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 3: The calculated geometries of the four vinyl radicals A, B, C and D at UM06-
2X/LanL2DZ level of theory.
Although the most stable radical A is the only one where the strong hyperconjugative
donor (C-Sn) is positioned vicinally to a strong hyperconjugative acceptor (*C-O), the
respective C-Sn ⟶ *C-O interactions is relatively small (3.65 kcal) – considerably smaller than
interaction of the C-Sn bond with the radical center (26 kcal). The high acceptor ability of
radical center in comparison to that of the C-O bond can be explained by the combination of
ACS Paragon Plus Environment

Page 19 of 45
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
more favorable stereoelectronic arrangement with the lower energy of the acceptor radical
orbital. In particular, while the gap between the C-Sn bond is 0.43 a.u., the gap between C-Sn
and σ*C-O is 0.73 a.u., as shown at the Scheme 9. The efficiency of overlap can be evaluated via
the comparison of Fock matrix elements between the donor and acceptor orbitals. Interaction
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
( )
2
energies can be evaluated via second order perturbative approach as ∆퐸푖푗
=
푞푖|퐹 |2
푖푗
. In this case, the 퐹 term for C-Sn bond and radical center is 0.143 a.u.
⁄
푖푗
(휀( ) − 휀푖( )
)
푁퐿
퐿
푗
whereas the same term for C-Sn ⟶ *C-O is ~0.046 a.u. Interestingly, the NBO analysis also
finds a remote through-space nO ⟶ σ*Sn-C interaction (2.5 kcal) between propargylic oxygen
and Me3Sn-moiety. The other vinyl radicals showed slightly different combinations of similar
interactions, suggesting that the overall energies result from a complex interplay between
numerous electronic and steric effects (Scheme 9).
Scheme 9. Selected electronic interactions (NBO Energies in kcal/mol) contributing to the
structure and stability of three vinyl radicals A, C and D. Top left: σSn-C ⟶ radical, σSn-C
⟶
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 20 of 45
1
2
3
4
5
6
7
8
9
σ*C-O and nO ⟶ σ*Sn-C interactions for radical A. Top right: Rationale for the greater acceptor
ability of radical center in comparison to σ*C-O orbital based on contrasting energy gaps
between these two acceptor orbitals and the donor σSn-C orbital. Differences in the overlap (not
shown) play an additional role. Bottom left: σSn-C ⟶ radical and nO ⟶ σ*Sn-C interactions for
radical C. Bottom right: σSn-C ⟶ radical and nO ⟶ σ*Sn-C (not shown for clarity, 0.7 kcal/mol)
interactions for radical D.
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
Barriers for the cyclizations of vinyl radicals
We also calculated the barriers for potentially important cyclizations for the three stable
radicals: the most stable radical that can do a 6-exo-dig ring closure and the less stable radicals
that can do the potentially faster 5-exo-dig cyclizations. All values are for ΔG in kcal/mol at
110˚C (the experimental temperature).
The 13.8 kcal/mol barrier for 6-exo-dig cyclization of the most stable radical A suggests
that this reaction will provide the lowest energy path for the escape from the pool, when the
next most stable radical D has the barrier for its fastest cyclization that substantially exceeds 6.0
kcal (=13.8-7.8, Scheme 8). The cyclization is highly exothermic, exergonic and effectively
irreversible.
ACS Paragon Plus Environment

Page 21 of 45
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4: Geometries of radical A and its transition state for the 6-exo-dig cyclization
Indeed, although the 11.8 kcal energy barrier for the 5-exo-dig cyclization of radical D is
lower than the 6-exo barrier for radical A if the respective radicals are taken as the reference
points, the absolute energy of the 6-exo-dig barrier is lower. This 5-exo barrier is higher than a
typical 5-exo-dig barrier22 due to a stereoelectronic penalty that the reacting radical has to pay
for rotating out of conjugation with the central benzene ring in order to attack the triple bond in
the transition state, as seen in Figure 5.23
Figure 5: Geometries of the radical D and its transition state for the 5-exo-dig cyclization
Such penalty is unique for this radical because, unlike radicals A and C, radical D is
conjugated with an aromatic system that is annealed to the forming ring. Due to this structural
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 22 of 45
1
2
3
4
5
6
7
8
9
constraint, the central aromatic system is incapable of the same facile rotation that terminal aryl
groups in radical A and C can do without penalty. Since such rotation can result in a ~10 kcal
penalty,22b this loss of conjugation of radical center with the vicinal phenyl group should
account for this relatively high barrier.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Because the above 5-exo-dig cyclization is unusually slow, we also analyzed 5-exo-dig
cyclization of radical B which does not have to pay the same penalty in the cyclization TS.
However, the barrier is only slightly lower (11.3 kcal), presumably due to steric interaction in
the TS.
Figure 6: Geometries of the radical B and its transition state for the 5-exo-dig cyclization
Overall, the calculated energies for the competing radical reactions of the key
components of the radical pool can also rationalize the observed selectivity of the radical
cascade as long as the barriers for radical equilibration are sufficiently low. Note that the
present situation is different from that in the radical pool presented in Scheme 5 because the
desired 6-exo-dig cyclization will have to compete with two 5-exo cyclizations. As we
discussed above, 5-exo-dig barriers are generally lower than the 6-exo-dig alternatives.12,13
Consequently, in order to achieve selective reaction in this scenario one has either to stabilize
ACS Paragon Plus Environment

Page 23 of 45
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
the reactant that has to undergo the usually slower 6-exo-path to the extent where its cyclization
has the lowest absolute energy out of the all possible barriers or destabilize the 5-exo TS for the
less stable radicals. The above discussed stereoelectronic penalty for the unusually slow 5-exo-
dig cyclization of radical D leads to ~5.7 kcal/mol preference for the 6-exo-dig cyclization of
radical 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
58
59
60
Scheme 10. Comparison of relative energies of the four vinyl radicals and barriers for their exo-
dig cyclizations
At the present time, we do not have the sufficiently accurate information about the
barriers for C-Sn bond fragmentation in the vinyl radicals A-D and the barriers for the
interconversion in these radicals via metallotropic shifts. In order for the radical pool concept to
operate, such barriers have to be lower than the absolute barrier for the 6-exo-dig cyclization of
radical A. Based on the kinetic preference for the formation of radical A (Scheme 8) and the
Occam’s razor, we rationalize the observed selectivities as a direct facilitating effect of the OR
group on the Bu3Sn attack at the propargylic position followed by trapping of the kinetically
formed vinyl radical by a fast and irreversible 6-exo cyclization.
Full radical cascade and return of the radical at the -carbon:
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 24 of 45
1
2
3
4
5
6
7
8
9
The selective 6-exo-dig cyclization discussed in the previous section initiates the
experimentally observed cascade. Scheme 11 shows a proposed mechanism involving an initial
formation of a radical intermediate A formed via the attack of the Bu3Sn radical at the central
alkyne. The subsequent 6-exo-dig cyclization forms the second vinyl radical intermediate F.
Attack of the neighboring phenyl π-system results in a delocalized radical G, leading to
rearomatization via a 1,5-hydrogen shift.24 Finally, fragmentation of intermediate H allows for
the aromatization of the top ring into a naphthalene moiety. This explanation supports the
journey of the radical in this system: it begins with the Bu3Sn radical attacking the alkyne
position vicinal to the directing group starting a boomerang-like radical journey. The vinyl
radical is formed, leading to the 5-exo-dig cyclization: this is the furthest position of the radical
from where it started. The formation of the 5-membered ring makes the radical closer to its
origins but not enough: another resonance structure can be written. At this point the boomerang
is back to its initial position, vicinally to the directing and now leaving group. This position is
especially interesting because it will end forming a new π bond and making the directing OR
group leave the system. Most importantly, the last step removes the directing group from the
product, making this directing approach traceless. The other approach is that G’ can undergo
the β-scission of the OR first and then do the 1,5-shift to rearomatize the system. Both
possibilities are presented in Scheme 11 and Scheme 12.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 25 of 45
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 11. Proposed mechanism of cascade cyclization: note the continuous translocation of
the radical center through the molecule and conversion of -radicals (A and F) to delocalized -
radicals G and H where the radical is positioned next to the departing directing group. Color
coding for the last two steps corresponds to the color coding in the potential energy surface in
Scheme 12.
The proposed reaction mechanism is supported by the calculated free energy profile for
the full cascade transformation (Scheme 12). As discussed above, the initial 6-exo-dig
cyclization is moderately fast but highly exergonic and effectively irreversible. Subsequent
attack at the aromatic ring proceeds via a slightly higher (15.2 kcal) barrier. This process is
mildly exergonic and moves system even down the potential energy path. As a consequence,
even the relatively high barrier for the subsequent 1,5-hydrogen shift (18.6 kcal) is still lower in
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 26 of 45
1
2
3
4
5
6
7
8
9
the absolute energy than the two previous barriers, so the cascade can move forward. Because
the 1,5-shift leads to aromatization, it is highly exergonic and takes the system ~ 62 kcal lower
than the original vinyl radical A. The key final -scission of the directing OR group has a
barrier of only 10 kcal for R=Me (a slightly higher 15.4 kcal barrier was calculated for R=H,
where reaction was found to proceed slower experimentally). This process is exergonic by ~5
kcal/mol. Furthermore, the released OR radicals can be quickly intercepted via other
thermodynamically favorable processes (e.g., H-abstraction from Bu3Sn-H, propagating the
cascade).
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
We had also calculated the alternative pathway where fragmentation precedes 1,5-shift.
Overall, the highest barrier on this path is only marginally higher. Considering the challenges in
accurate computing of reaction barriers for large Sn-containing species, this path may serve as
an alternative to the mechanism outlined above. Interestingly, the barriers for the same
reactions are lower when they terminate the cascade in comparison to the analogous barriers
when these reactions serve as penultimate step (18.6 vs. 7.3 kcal for the 1,5-shift, 10.0 vs. 20.6
kcal for the fragmentation). This is a direct evidence for thermodynamic contribution to reaction
barrier as outlined by the Marcus theory.24 When the two reactions terminate the cascade, they
are assisted by significant aromatic stabilization (formation of one aromatic ring for
fragmentation and two aromatic rings for 1,5-shift).
ACS Paragon Plus Environment

Page 27 of 45
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 12. The calculated energy profile for the cascade radical transformation of diyne 1a.
All energies are given relative to radical A except for the smaller numbers in italics that
correspond to the activation barrier heights relative to the preceding energy minima.
An interesting stereoelectronic feature is observed in the computed fragmentation TS
where steric repulsion between the bulky Me3Sn moiety and the OMe group helps in pushing
the OMe group away from planarity and aligning breaking C-O bond with the aromatic -
system. This effect, which can be classified as steric assistance, should be increased further in
the real system which has an even more bulky Bu3Sn moiety instead of the Me3Sn group used in
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 28 of 45
1
2
3
4
5
the computational analysis. If these expectations are correct, the experimental fragmentation
barrier should be lowered even further.
6
7
8
9
Possible involvement of cationic intermediates:
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
An altnernative explanation to the directing effect of the propagylic OR group on the
regioselective addition of Sn-radicals to alkynes was made by Organ and coworkers who, on the
basis of a careful experimental study, suggested the involvement of Bu3Sn cations in the
selectivity-determining step.7a This applicability of this scenario to our system is further
discussed and analyzed computationally in the Supporting Information. In short, our results
suggest that cationic pathway is unlikely to be involved in our radical cascade.
Functionalization of stannyl benzofluorenes:
Two practical conclusions can be obtained from these results. First, this cascade
provides a practical synthetic approach to substituted benzo[a]fluorenes, the class of compounds
with known anticancer activity.25 Furthermore, the Sn-containing products can be introduced in
reactions with a variety of electrophiles. For example, they can be readily protodestannylated to
yield the respective hydrocarbon and introduced in Stille cross-coupling. Alternatively, they can
be converted into an electrophilic benzofluorenone building blocks via reaction with iodine
and/or reacted with carbon nucleophiles, i.e., boronic acids under Suzuki conditions (Scheme
13).26
ACS Paragon Plus Environment