Radical Reactions in Cavitands Unveil the Effects of Affinity on Dynamic Supramolecular Systems

Article abstract of DOI:10.1021/jacs.9b11595

Radical reduction of alkyl halides and aerobic oxidation of alkyl aromatics are reported using water-soluble container compounds (1 and 2). The reductions involve α,ω-dihalides (4-8 and 10) with radical initiators in cavitand hosts with varied binding affinities. Product distributions lead to general guidelines for the use of dynamic supramolecular systems with fast reactions. The binding of guest substrates in the hosts must show high affinities (K_a> 10³M^-1) to ensure that the reactions take place under confinement in the containers.

Full text of DOI:10.1021/jacs.9b11595

Subscriber access provided by Macquarie University
Article
Radical Reactions in Cavitands Unveil the Effects
of Affinity on Dynamic Supramolecular Systems
Manuel Petroselli, Venkatachalam Angamuthu, Faiz-Ur Rahman, Yang Yu, and Julius Rebek, Jr.
J. Am. Chem. Soc., Just Accepted Manuscript • Publication Date (Web): 08 Jan 2020
Downloaded from pubs.acs.org on January 8, 2020
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 10
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
Radical Reactions in Cavitands Unveil the Effects of Affinity
on Dynamic Supramolecular Systems
Manuel Petroselli†, Venkatachalam Angamuthu†, Faiz-Ur Rahman†, Yang Yu†* and Julius Rebek, Jr.†,‡*
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
†Center for Supramolecular Chemistry & Catalysis and Department of Chemistry, College of Science, Shanghai University,
99 Shang-Da Road, Shanghai 200444, P. R. China.
‡The Skaggs Institute for Chemical Biology and Department of Chemistry, The Scripps Research Institute, 10550 North
Torrey Pines Road, La Jolla, California 92037, USA.
KEYWORDS. free radical, water-soluble cavitand, binding constant, dynamic equilibrium.
Deep cavitands such as 1 (Figure 1) are water-soluble
ABSTRACT: Radical reduction of alkyl halides and aerobic
container hosts that allow the study of guest molecules under
oxidation of alkyl aromatics are reported using water-soluble
confinement where reactivity may be affected by the limited
container compounds (1 and 2). The reductions involve -
dihalides (4-8 and 10) with radical initiators in cavitand hosts
space [1]. They are conformationally dynamic systems that
interconvert between a receptive vase shape, stabilized by
with varied binding affinities. Product distributions lead to
binding of suitable guests, and a flattened form, called velcrand
general guidelines for the use of dynamic supramolecular
or kite form. In the absence of suitable guests, the kite form, as
systems with fast reactions. Binding of guest substrates in the
its more stable dimeric form, called velcraplex, is not detectable
hosts must show high affinities (Ka > 103 M-1) to ensure that the
by 1H NMR spectroscopy. Characteristic methine triplets can be
reactions take place under confinement in the containers (11 and
12).
observed in the 1H NMR spectrum at 5.6 ppm and around 4 ppm
for “vase” cavitand and velcrand, respectively (Figure 1 -
bottom). Host 2 is less flexible and appears as the vase in its
resting state [2]
.
INTRODUCTION
Figure 1. Cartoons and structures of host 1 and 2 (Top). Modeled
host 1 conformations: vase (A), kite (B) or velcrand and the dimeric
velcraplex (C). Hydrogen atoms and “feet” are removed for clarity.
[3-5]
Reactions such as cyclizations
functionalizations
or selective mono-
[6-8]
confined in 1 have been reported and
show some parallels to the action of enzymes in biological
systems [9, 10]. The affinity of a guest to any host is described by
the binding constant (KA) [1] that reflects the kinetics of the in-
out equilibrium (guest uptake and release) of the complex
(Figure 2). High values of KA often correspond to slow release
with high affinity to the host while lower values can involve fast
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 10
the present research. Cavitand 1 and 2 were taken as model host:
release and lower affinity. The binding constant (KA) reflects
the ratio of these rates; the concentration of the free guest is also
a factor. Low and high concentrations are observed for high and
low KA values, respectively. Other factors can affect the KA
value, including solvent, the guest’s solubility, temperature and
the intramolecular forces involved in the host-guest recognition.
For example, highly water-soluble guests as cyclic alkyl amines
are relatively happy in solution and show lower KA values than
the less water-soluble cyclic alkyl halides. Indeed, free guest is
observed in the 1H NMR spectrum when cyclohexyl amine and
1 are mixed in water in 1:1 ratio, while no free guest is observed
1
2
3
4
5
6
7
8
The absence of hydrogen-bond donors in either host prevents
formation of capsules and both containers maintain an opening to
the solution. As model radical reactions in water, we chose radical
reduction of alkyl halides with Ph3SnH and aerobic oxidation of
alkyl aromatic compounds.
RESULTS AND DISCUSSION
Radical Reduction of Alkyl Halides in 1. Radical reduction of
alkyl halides, using reducing agents such as triphenyltin hydride
(Ph3SnH) or the “greener” trialkylsilanes (R3SiH), is a classical
chain reaction [22]. Historically used for the synthesis of alkanes [23,
24] or as a tool for generating building blocks [25, 26]; recently it has
even been applied in DNA-encoded library (DEL) settings [27]. The
alkyl halide quickly reacts with the tin radical in solution, leading
to tin halide and a carbon-centered radical.
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
when cyclohexyl halides are used under the same conditions [11]
.
External
Stimuli
I1
P1
G1
G1
External
Stimuli
I2
P2
G1
KA
In
I2@1
P2@1
G1@1
1
H
Ph3Sn
Ph3Sn
Figure 2. General reactivity in bulk solution (Top) General
reactivity in Host 1 (Bottom). G: Guest, I: Intermediate, P: Product.
H
In
Ph3Sn
H
Ph3Sn
X
R
H
R-
R-
Ph3Sn
X
Ph3Sn
Where the products are formed is a reasonable question for
reactions involving dynamic containers, as it cannot be assumed
that the reaction takes place in the host. A general guest G1 in
bulk solution, exposed to external stimuli (i.e. radical initiators
or light), leads to the formation of product P1 through
intermediate I1. The same guest G1 in presence of 1 may be
bound in a host-guest complex G1@1. The confined space in 1
forces G1 to assume unusual conformations in the cavity, and
can promote alternatives after exposure to external stimuli, with
consequent formation of “alternative” product P2 (Figure 2).
The necessary and unique conditions needed to observe this
Scheme 1. Generation of tin radical in presence of radical initiator
(Top). General mechanism for radical reduction of alkyl halides (RX
where X = Br or I) using Ph3SnH as reducing agent (Bottom).
The radical is rapidly quenched by the remaining tin hydride (k
≈ 106 M-1 s-1 in benzene) producing the relevant alkane and
regenerating the tin radical for the next cycle (Scheme 1).
Theoretical TS calculations were performed on bromo
cyclohexane and trimethylsilyl radical (taken as model alkyl
halide and reducing agent, respectively) in order to investigate
the orientation of the attack.
behavior are a host-guest complex that is kinetically stable on
the NMR chemical shift timescale
excludes the possibility of generating “side” product P1 from
“free” G1 in solution.
[12,
13]. This condition
We used cavitands in reactions where radicals are involved. They
are generally hard to control and modulate; the fast kinetics (often
k > 103 M-1 s-1) and the possible side reactions make these reactions
neither “user friendly” nor easy to manage. Even so, radical
processes currently enjoy popularity and efforts have been made to
investigate their chemistry in supramolecular contexts such as
capsules [14, 15], cages [16, 17] and organometallic systems [18]. The use
of external chemical agents is ineffective in these systems, due to
the protective action of the capsule (or cage) that maintains a
mechanical barrier between guest and reaction medium. This aspect
strongly limits the use of these systems to mostly photo-activated
processes. For example, homolytic cleavage of C-C bonds in
carbonyl compounds (Norrish reactions) or C-O bonds in esters
(Fries reactions) with subsequent radical-radical coupling
processes are well known [19, 20]. Also, electron transfers through
the organic wall after photo-irradiation are reported between
encapsulated guest and external compounds [21]. To our knowledge,
no radical reactions involving radical initiators, have been reported
in dynamic hosts with open ends. This unexplored area prompted
Figure 3. (Left) Alkyl halides used in this investigation (3-10);
(Middle) Calculated TS for radical dehalogenation of bromo
cyclohexane with trimethylsilyl radical at DFT/6-31G(d,p) level of
theory; and (Right) cartoon of the 3@1. Hydrogens on calculated
structure are omitted for better clarity.
A linear TS structure with attack in the cyclohexyl plane is the
most stable structure, which would be compatible with the
expected orientation of the guest in 1 (Figure 3). Triphenyltin
hydride was chosen as reducing agent for its faster hydrogen
donation compared to silanes (106 vs 10≤ 5 M-1 s-1 in benzene).
Radical reduction of 3 using Ph3SnH (1 eq.) and a catalytic
ACS Paragon Plus Environment

Page 3 of 10
Journal of the American Chemical Society
Figure 4. Partial 1H NMR (600 MHz, D2O, 298 K) spectra of: Initial
amount of AAPH as radical initiator was performed in D2O but
1
2
3
4
5
6
7
8
host-guest complex 6@1 (A), after addition of Ph3SnH (2 eq.),
AAPH (cat.) after 12 h at 40 °C (B), authentic 1-iodododecane in 1
(C) and authentic dodecane in 1 (D). Partially-reduced and
fully-reduced product are marked in the reaction mixture with red
and black stars, respectively (top). Cartoon of 6@1 involved in the
radical reduction (bottom).
no products were detected after 12 h at 40°C (See SI14). The
same reaction using tris(trimethylsilyl)silane ((TMS)3SiH) as
reducing agent in toluene led to the formation of the reduction
product (cyclohexane) in 97% yield [28]. The iodine atom of 3 as
a guest in 1 may not be exposed enough to the reaction medium
to engage in the radical reduction cycle. Indeed, the NMR signal
of the α-hydrogen in bound 3 shows a ∆ value of -3.84 ppm,
placing it, on average, deep within the cavity [11]. Moreover, the
size of Ph3SnH with three phenyl groups prevents its binding in
1 (See SI11) and makes its approach to bound 3 even less likely.
Guest 4, having an additional iodo atom in position 4, appears
to be more exposed to the reaction medium and the ∆ value for
the same hydrogen is now -2.83 ppm (averaged value) (See
SI35). Unfortunately, guest 4 shows the same inertia as 3 and
no reduction product was observed (See SI15). Linear dialkyl
halides were therefore selected as substrates to increase
accessibility in their complexes: one of the terminal halogen
atoms is always exposed to the reaction medium and their
conformations in 1 are dynamic [7]. Commercially available
guest 5 was tested but no reactivity was observed in the complex
under the usual conditions (See SI16), yet it easily reacts with
Ph3SnH in bulk solution [29]. The ∆ value for the α-hydrogens
Alkyl iodides are more reactive and 6, in which bromo atoms
were replaced with iodo atoms, was tested. Reduced products
were finally detected by 1H NMR (See Figure 4). This result is
not attributed to the intrinsic reactivity of guests 3 or 4 but to
the enhanced reactivity of 6 in 1. In bulk solution 3 and 4 are
more reactive than 6, due to the formation of a secondary carbon
centered radical rather than a primary one. Binding of guest 6
features the same behavior reported earlier for guest 5 (See
SI37) in which iodo atoms quickly exchange their positions in
the cavitand (See Figure 4 - Equilibrium I). The high reactivity
of iodo atom towards radical reduction allows the formation of
“Complex A” that corresponds to the mono-reduced product
observed in the 1H NMR spectrum (See Figure 4B – red stars).
The remaining iodo atom is at the bottom of the cavity, where
it is protected and inaccessible to radical initiators.
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
of bound 5 was -1.93 ppm (averaged value), confirming the
But “Complex A” is dynamic, in equilibrium with “Complex
B” where the iodo atom is now exposed to the reaction medium
and theoretically accessible to the tin radical (See Figure 4 –
Equilibrium II), leading to the formation of dodecane (See
Figure 4B – black stars). The ∆ value for α-hydrogens of
bound iodododecane is -2.63 ppm (averaged value) (See SI39)
relatively close to the value of 2.83 ppm, observed for bound 4
which was buried and unreactive. Thus the iodo atom of
“Complex B” (Figure 4) would not be accessible to the radical
initiator. Accordingly, the alkane detected (dodecane) could not
be generated inside the host. The intermediate involved was
confirmed to be a radical since the reduction of 6 in 1 under
oxygen atmosphere led to the detection of the corresponding
mono-alcohol as the major product (See SI28 and SI29).
[7]
better exposure of the carbons bearing the halogen atoms
.
The inertia could be caused by the lower reactivity of bound 5
towards tin radical and not by its accessibility: radical
dehalogenation using reducing agents as Ph3SnH is well-known
to be affected by the nature of the halogen atom.
The effect of the lipophilic tail on the selectivity for the fully-
reduced product (alkane) using the shorter guests 7 and 8 was
also explored. These guests should be “less exposed”: placing
the iodo atom much closer to the bottom of the cavity in
“Complex B” decreases the reactivity as earlier observed for
guests 3 and 4. Not much difference in conversion (≈ 75-67%)
and selectivity in alkane (≈ 30-35%) were observed between
guest 6 and guest 7 (See SI30 and SI31) but differences were
obtained with the shortest guest 8 (See Fig. 5). Selectivity of
13% in fully-reduced product (heptane) was reached with 8,
confirming the better (but not complete) protection of the iodo
atom in the intermediate (See SI32). The ∆ value for α-
hydrogens of bound iodoheptane was -3 ppm (averaged value)
(See SI45) that is higher respect to the value found for bound 4
which was completely stable under the same conditions.
C12
I
I
Ph3Sn
I
Ph3Sn
No Reaction
I
I
I
Int. A
II
I
C12
C1
III
Ph3Sn
C1
C12
C12
Int. B
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 10
Surprisingly, radical reduction of 10 using Ph3SnH showed a
different result (See Fig. 6).
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 5. Partial 1H NMR (600 MHz, D2O, 298 K) spectra of: Initial
host-guest complex 8@1 (A), after addition of Ph3SnH (2 eq.),
AAPH (cat.) after 12 h at 40 °C (B), authentic 1-iodoheptane in 1 (C)
and authentic heptane in 1 (D). Partially-reduced and fully-reduced
products are marked in the reaction mixture with red and black
stars, respectively.
Figure 6. Partial 1H NMR (600 MHz, D2O, 298 K) spectra of: Initial
host-guest complex 10@1 (A), after addition of Ph3SnH (2 eq.),
AAPH (cat.) after 12 h at 40 °C (B), authentic 2-bromononane in 1 (C)
and authentic nonane in 1 (D). Partially-reduced and fully-reduced
products are marked in the reaction mixture with red and black
stars, respectively.
The product alkane (heptane) should not be attributed to the
reactivity of “Complex B” in the host but it should come from
another process. The dynamic equilibrium of 8 in 1 makes an
unambiguous interpretation of the result difficult. Therefore,
guests with less dynamic intermediates in 1 were tested in order
to further confirm this hypothesis.
Partially-reduced product was observed as expected (See Figure
6B – red stars) but the fully-reduced product (nonane) was also
present in the 1H NMR spectra (Figure 6B – black stars).
Br
C9
Br
I
Ph3Sn
Ph3Sn
No Reaction
Br
Br
Br
9a@1
II
Br
C1
C9
III
Ph3Sn
C1
C9
C9
9b@1
Scheme 2. Cartoons of host-guest complexes involved in the
radical reduction of 10.
Secondary alkyl halides such as guest 10, where the partially
reduced product (9) is “fixed” and protected in 1 (Scheme 2),
should show higher selectivity for mono-functionalization.
Guest 10 in 1 assumes the same motion observed previously for
guests 5-8 (See SI49). Better reactivity than guest 5 was
guaranteed by the formation of a secondary carbon center
radical. The branched ends still give kinetically stable host-
guest complexes (on the NMR timescale) in 1 but the signals
are broadened due to the faster in-out exchange rates (See SI48).
In what follows, the different ends in 9 are called “bromo end”
and “methyl end” in order to describe the relative stability of
the conformations. The ∆ values of -3.72 and -3.44 ppm were
calculated for the methyl group and hydrogen of the Br-CH
group of the bromo-end, respectively, confirming their position
at the bottom of the cavity (See SI47). Guest 9a@1 should
therefore be protected by the host’s walls and no reduction
should be observed as previously reported for guests 3 and 4.
Figure 7. Partial 1H NMR (600 MHz, D2O, 298 K) spectra of: Initial
host-guest complex 9@1 (A), after addition of Ph3SnH (2 eq.),
AAPH (cat.) after 12 h at 40 °C (B) and authentic nonane in 1 (C).
Reduced product (nonane) is marked in the reaction mixture with
red stars.
As mentioned above, intermediate 9a@1 should not be reactive
due to the protected orientation of the bromo atom inside the
cavity – indicated by 1H COSY NMR and ∆ values (See SI46
and SI47). Radical reduction on guest 9 with Ph3SnH was
performed in order to confirm this surmise (Fig. 7). The host-
guest complex between 9 and 1 is kinetically stable (on the
NMR timescale) but not protected. A conversion of 29% (See
SI33) to the alkane (nonane) was observed (Figure 7B). The
kinetic stability of the host-guest complex on the NMR
timescale was a fundamental and only requirement with
dynamic supramolecular systems such as host 1. Instead, the
formation of kinetically stable host-guest complex on the NMR
ACS Paragon Plus Environment

Page 5 of 10
Journal of the American Chemical Society
time scale seems to be a necessary but not sufficient condition host-guest complexes at NMR time scale are detected when
1
2
when fast processes such as radical reactions are involved in
dynamic supramolecular systems. To repeat, kinetically stable
guest 9 and guest 10 are encapsulated in 1, but fully-reduced
product (nonane) is still formed.
3
4
5
6
7
8
Br
Br
C9
C9
Ph3SnH
106
Ph3Sn
10>6
1
Host
Br Br
D2O
5
Br
Br
Br
Br
9
Ph3Sn
Ph3SnI
10
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10
1
@
9
1
@
kA
Ph3Sn
Br
1
Host
Br Br
Host-Guest
Complexes
+
+
5
D2O
5
5
Scheme 3. Cartoons of the host-guest complexes and the mechanism of the radical reduction of 10.
If the radical reduction of 10 takes place inside 1, only
reactivity. Guest 6 was taken as a model alkyl iodide due to its
similarities with 7 and 8 while guest 10 shows broad peaks,
making the NMR titration hard to follow. This problem,
compounded by the low solubility of the guests in water
prevents the accurate measure of in-out rate constants by, say,
EXSY experiments. The broad peaks in 10@1 point to a faster
in-out exchange or a lower KA value than guests 6 or 9. Guest 6
shows a KA value of 0.7  103 M-1 (Table 1, Entry 3) while guest
9 is 0.9  103 M-1 (Table 1, Entry 4), confirming the hypothesis.
These host-guest complexes must show a KA of at least 1.2 
103 M-1 to guarantee that the reaction takes place in the host.
compound 9 should be detected as final product. If the reaction
takes place outside the host 1, the relevant alkane (nonane) must
also be detected (See Scheme 3). No free guest was observed in
1
the H NMR spectrum after the encapsulation of 10 (or any
other guest reported here) in 1, leading us to propose that the
“free-guest” should come from the in-out exchange, regulated
by the binding constant (KA). If this hypothesis was correct, a
relation between binding constants of each involved guest and
the observed reactivity towards tin radical should be found. For
a relatively low KA value (usually fast in-out exchange), the tin
radical can capture the guest during the short time it will be in
solution, whereas higher KA values (slower in-out exchange)
result in lower concentrations of the guest in solution, making
the free radical trapping less favored. A binding constant of 5 
103 M-1 was calculated for guest 4 after NMR titration (See
Table 1, Entry 1) while a value of 1.2  103 M-1 was obtained
for guest 5 (See Table 1, Entry 2).
Aerobic Oxidation of Alkyl Aromatic Compounds in 1. The
role of the KA for radical processes was further tested on aerobic
oxidation of alkyl aromatic compounds (11 and 12) in 1, and
show this behavior is independent of the nature of the radical
process. The relative low C-H BDE for substituted alkyl
aromatic compounds (from 82 to 87 kcal/mol) makes these
compounds easily oxidized though a radical mechanism under
aerobic conditions. Auto-oxidation of cumene, for example, is
reported under air or oxygen atmosphere at high temperature (>
80°C). Milder conditions can be effective using radical
initiators such as AIBN ((2,2’-azobis(2-methylpropionitrile))
Table 1. Binding constants (KA) for representative guests in host
1.
Entry
Guest
KA (M-1)
[30]
[31-33]
and with specific catalysts as N-hydroxy compounds
or metal-complexes [34]
5.0  103
1
4
.
1.2  103
0.7  103
0.9  103
2
3
4
5
6
9
Scheme 4. General pathway for A) auto-oxidation of alkyl aromatic
compound (R-H) ; B) and oxidation with a radical initiator (In.)
under aerobic conditions.
Both 4 and 5 are unreactive in the presence of tin radical and no
reduction products are detected after the reaction. A KA value
of 1.2  103 M-1 appears high enough to prevent the reaction
outside the host. Reactive guests as 6 or 9 should have a KA
value lower than 1.2  103 M-1 in order to explain the observed
Oxygen abstracts hydrogen from an alkyl aromatic compound,
giving a carbon centered radical that, under an aerobic
atmosphere, is quenched at diffusion-controlled rates to form a
peroxyl radical. The peroxyl radical may undergo termination
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 10
processes and consequent formation of neutral products as Azobis(2-amidinopropane) dihydrochloride (AAPH) is shown
1
2
3
4
5
6
alcohols, ketones, aldehydes or carboxylic acids, depending on
the substrate [35] (Scheme 4A). The addition of radical initiators
does not alter the mechanism of the oxidation but acts on the
activation step’s kinetics. The kinetic constant KH2 is very often
higher than the kinetic constant KH1 (Scheme 4B). Oxidation
mechanism of 11 and 12 as guests used the radical initiator 2,2’-
(Scheme 5). AAPH was selected for its high water-solubility,
its thermal activation at relatively low temperatures (≈36°C)
and very low affinity and reactivity towards host 1: no oxidative
products were observed after treatment of 1 with 2 eq. of AAPH
after 5h at 55°C (See SI9 and SI10).
7
8
9
R2
R2
R2
NH
R1
R2
H2N
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
Host
Disprop.
D2O
R1
R1
R1
R2
R1
NH
H2N
109
O2
11
:
R1:
R2: Me
R2
OO
R2
O
H
R2
OO
H
12
:
R1:
Disprop.
R2: H
R1
R1
R1
Scheme 5. Mechanism of aerobic oxidation of alkyl aromatic compounds (11 and 12) in 1.
In the absence of 1, quantitative conversion to the
corresponding diol was reached with 11 and AAPH (1.1 eq) in
D2O at 40°C after 12h. This result reflects the high reactivity of
11 and alkyl aromatic compounds in general towards AAPH in
bulk solution (See SI12). A KA value for 11 in 1 of 12.1  103
M-1 was determined (See SI), predicting no reaction outside the
host. Calculations indicated the most stable trajectory of the
attack is perpendicular to the phenyl plane (Figures 8A and 8B)
but the orientation of 11 (or 12) in host 1 makes the approach
inaccessible to the radical initiator (Figure 8C).
the reaction takes place outside the host, but this is prevented
by the KA value of 12.1  103 M-1. In the experiment, guest 11
was completely stable in 1 after 5 h at 50°C in the presence of
AAPH (1 eq.). All the characteristic signals of bound 11 remain
after the reaction and no new signals are observed (Figures 9B
and SI50), confirming the hypothesis concerning the role of the
binding constant (KA). Similar behavior was observed for guest
12 under the same conditions (See SI51 and SI53). In order to
demonstrate the importance of KA (rather than on the host)
radical reduction of 10 using a new metallo cavitand 2 [2] are
described below.
Figure 8. (Right) Theoretical TS-anti (A) and TS-syn (B) conformers
for HAT reaction between “AAPH radical” and cumene at B3LYP/6-
31G(d,p) level of theory. (Left) Cartoon of 11@1 (C). Hydrogens
not involved in the HAT are missed for better clarity.
Figure 9. 1H NMR (600 MHz, D2O, 298 K) spectra of 11 in 1 in 1:1
ratio in D2O (A) and after addition of AAPH (1 eq.) after 5 h
at 55 °C (B). Characteristic signals of guest 11 are marked with red
stars (See SI50 for detail).
Experimentally, the ∆ observed for the isopropyl group at the
open end of 1 (Figure 9 – Red Isopropyl group) is 0.2 ppm (See
SI52), placing it close to the open end of the host but not outside
the cavity. Oxidation product(s) are therefore predicted only if
ACS Paragon Plus Environment

Page 7 of 10
Journal of the American Chemical Society
Radical Reduction of 10 in 2. Guests bound in 2 experience a The decrease of diastereotopic signals, due to the loss of the
[2]
1
2
3
4
5
6
7
8
narrower space and movements compared to 1 are limited
.
stereogenic center (Br-*CHMe-) further confirms the reaction
takes place inside the cavity of the host (See SI62). Quantitative
conversion and selectivity higher than 95% for mono-reduced
For example, a KA value of 1.5  105 M-1 was obtained by
isothermal titration calorimetry (ITC) using n-BuOH as a guest
in 2 [2]. This value is two orders of magnitude higher than KA
values of any guest in 1, and a similar KA value was determined
for bound 10 in 2 by direct competition experiments with n-
1
product were detected. This was confirmed by the H NMR
spectrum of authentic mono-reduced product 9 bound in 2
(Figure 10). Accordingly, the high KA makes the protection of
intermediate 9 complete. The in-out exchange is now so slow
as to be irrelevant.
1
BuOH (See SI61). Two complexes are detected by H NMR
spectrum when 9 is bound in 2 (Figure 10C). The complexes
correspond to the two possible orientations of 9 in 2, but the
limited space of the cavity makes their interconversion
extremely slow (Scheme 6). If the reduction of 10 takes place
in the cavity, only complex 9a@2 should be detected, while
both complexes must be observed if the reduction takes place
outside the host. Reduction of 10 in 2 gave bound 9 in a single
orientation and no nonane, (the product detected when the
reaction involved 1) was observed (Figure 10).
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
CONCLUSION
Earlier reports [3-7] suggest that host-guest complexes need only
to be kinetically stable on the NMR chemical shift timescale to
guarantee the reaction takes place in the host. Radical
reductions of alkyl halides investigated in host 1 indicate this
condition is necessary but not sufficient when reactions with
fast kinetics such as radical reactions (often k > 103 M-1 s-1) are
involved. Instead, a KA value larger than 1.2  103 M-1 for the
complex guarantees the reaction occurs inside the container.
Lower KA values, calculated for alkyl α,ω-diiodides, gave
reduced products (alkanes), arising from the reaction outside
the host 1. These guests are weakly bound, despite showing
kinetically stable complexes on the NMR time scale. The role
of the binding constant appears independent of the radical
processes. Alkyl aromatic compounds (11 and 12) with high KA
values (KA ≈ 1.2  104 M-1), are stable in 1 in presence of
AAPH. No oxidation products are observed, despite showing
high reactivity under same conditions in bulk solution. Radical
reduction of secondary alkyl dibromide 10 in rigidified host 2
was also investigated. It showed high reactivity yet partially
reduced product 9 was completely protected in the host cavity,
due to its high KA value (≈ 1.5  105 M-1). We hope these results
act as guidelines for the use of dynamic hosts in the field of
radical chemistry.
Br
Me
Me
Br
2
+
Me
Me
D2O
Me
5
Me
9b
Br
9
2
@
9a
2
@
(major)
(minor)
Br
Me
Me
Slow
Me
Me
Br
9a
2
@
9b
2
@
Scheme 6. Cartoons and orientations of 9@2 (Top) and the
equilibrium between host-guest complexes (Bottom).
ASSOCIATED CONTENT
Supporting Information. Synthesis and experimental procedures,
1H and 13C NMR spectra for synthetized guests, stability and
control experiments, ∆ calculation for all host-guest complex and
binding constant determination are reported in the supporting
information.
AUTHOR INFORMATION
Corresponding Author
*
To whom correspondence may be addressed. E-mail:
jrebek@scripps.edu or yangyu2017@shu.edu.cn
ACKNOWLEDGMENT
This work was supported by the National Natural Science
Foundation of China (Grant 21801164), the US National Science
Foundation (CHE 1801153) and by Shanghai University (N.13-
G210-19-230), Shanghai, China. Y.Y. thanks the Program for
Professor of Special Appointment (Donfang Scholarship) of the
Shanghai Education Committee.
Figure 10. A) Upfield regions of the 1H NMR spectra (600 MHz, 305
K) of 10 (10 µL, 50 mM acetone-d6 then removed) in a solution of
2 (1 mM) in 0.5 ml of D2O. B) After 12 h at 40 °C in presence of
Ph3SiH/AAPH under nitrogen atmosphere. C) Authentic 2-
bromononane (9) in 2. D) Authentic alkane (nonane) in 2. Signals
of 9a@2 and 9b@2 are marked in red and black stars, respectively.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 10
20. Ramamurthy, V., Photochemistry within a Water-Soluble Organic
REFERENCES
Capsule. Acc Chem Res, 2015, 48, 2904-2917.
1
2
3
4
5
6
7
8
21. Raj, A. M.; Porel, M.; Mukherjee, P.; Ma, X.; Choudhury, R.;
Galoppini, E.; Sen, P.; Ramamurthy, V., Ultrafast Electron
Transfer from Upper Excited State of Encapsulated Azulenes to
Acceptors across an Organic Molecular Wall. J Phys Chem C,
2017, 121 (37), 20205-20216.
22. Chatgilialoglu, C.; Studer, A., Encyclopedia of Radicals in
Chemistry, Biology and Materials, 2012, John Wiley & Sons, Ltd.
23. Cole, S. J.; Kirwan, N.; Roberts, B. P.; Willis, C. R., Radical Chain
Reduction of Alkyl Halides, Dialkyl Sulphides and O-Alkyl S-
Methyl Dithiocarbonates to Alkanes by Trialkylsilanes. J Chem
Soc Perkin Trans, 1991, 1, 103-112.
24. Lesage, M.; Martinho Simoes, J. A.; Griller, D., Triphenylsilane:
A Useful Radical-Based Reducing Agent. J Org Chem, 1990, 55,
5413-5414.
25. Renaud, P.; Dénès, F.; Beaufils, F., Preparation of Five-Membered
Rings via the Translocation-Cyclization of Vinyl Radicals. Synlett,
2008, (16), 2389-2399.
26. Zhang, P.; Le, C. C.; MacMillan, D. W., Silyl Radical Activation
of Alkyl Halides in Metallaphotoredox Catalysis: A Unique
Pathway for Cross-Electrophile Coupling. J Am Chem Soc, 2016,
138 (26), 8084-7.
27. Qin, T.; Malins, L. R.; Edwards, J. T.; Merchant, R. R.; Novak, A.
J.; Zhong, J. Z.; Mills, R. B.; Yan, M.; Yuan, C.; Eastgate, M. D.;
Baran, P. S., Nickel-Catalyzed Barton Decarboxylation and Giese
Reactions: A Practical Take on Classic Transforms. Angew Chem
Int Ed Engl, 2017, 56 (1), 260-265.
28. Bellestri, M.; Chatgilialoglu, C., Tris(trimethylsilyl)silane as a
Radical-Based Reducing Agent in Synthesis. J Org Chem, 1991,
56, 678-683.
29. Ingold, K. U.; Bowry, V. W., Why are organotin hydride
reductions of organic halides so frequently retarded? Kinetic
studies, analyses, and a few remedies. J Org Chem, 2015, 80 (3),
1321-31.
1. Murray, J.; Kim, K.; Ogoshi, T.; Yao, W.; Gibb, B. C., The aqueous
supramolecular chemistry of cucurbit[n]urils, pillar[n]arenes and
deep-cavity cavitands. Chem Soc Rev, 2017, 46 (9), 2479-2496.
2. Rahman, F. U.; Li, Y. S.; Petsalakis, I. D.; Theodorakopoulos, G.;
Rebek, J., Jr.; Yu, Y., Recognition with metallo cavitands. Proc
Natl Acad Sci U S A, 2019, 116 (36), 17648-17653.
3. Shi, Q.; Masseroni, D.; Rebek, J., Jr., Macrocyclization of Folded
Diamines in Cavitands. J Am Chem Soc, 2016, 138 (34), 10846-8.
4. Wu, N. W.; Petsalakis, I. D.; Theodorakopoulos, G.; Yu, Y.;
Rebek, J., Jr., Cavitands as Containers for alpha,omega-Dienes and
Chaperones for Olefin Metathesis. Angew Chem Int Ed Engl, 2018,
57 (46), 15091-15095.
5. Wu, N. W.; Rebek, J., Jr., Cavitands as Chaperones for
Monofunctional and Ring-Forming Reactions in Water. J Am
Chem Soc, 2016, 138 (24), 7512-7515.
6. Masseroni, D.; Mosca, S.; Mower, M. P.; Blackmond, D. G.;
Rebek, J., Jr., Cavitands as Reaction Vessels and Blocking Groups
for Selective Reactions in Water. Angew Chem Int Ed Engl, 2016,
55 (29), 8290-8293.
7. Angamuthu, V.; Petroselli, M.; Rahman, F. U.; Yu, Y.; Rebek, J.,
Binding orientation and reactivity of alkyl alpha,omega-
dibromides in water-soluble cavitands. Org Biomol Chem, 2019,
17 (21), 5279-5282.
8. Angamuthu, V.; Rahman, F.-U.; Petroselli, M.; Li, Y.; Yu, Y.;
Rebek, J., Mono epoxidation of α,ω-dienes using NBS in a water-
soluble cavitand. Org Chem Front, 2019,6, 3220-3223.
9. Hooley, R. J.; Rebek, J., Jr., Chemistry and catalysis in functional
cavitands. Chem Biol, 2009, 16 (3), 255-264.
10. Yu, Y.; Rebek, J., Jr., Reactions of Folded Molecules in Water. Acc
Chem Res, 2018, 51 (12), 3031-3040.
11. Feng, H.-N.; Petroselli, M.; Zhang, X.-H.; Rebek, J. J.; Yu, Y.,
Cavitands: capture of cycloalkyl derivatives and 2-
methylisoborneol (2-MIB) in water. Supramol Chem, 2019, 31 (3),
108-113.
12. Biros, S. M.; Ullrich, E. C.; Trembleau, L.; Rebek Jr, J., Kinetically
Stable Complexes in Water: The Role of Hydration and
Hydrophobicity. J Am Chem Soc, 2004, 126, 2870-2876.
13. Avram, L.; Wishard, A. D.; Gibb, B. C.; Bar-Shir, A., Quantifying
Guest Exchange in Supramolecular Systems. Angew Chem Int Ed
Engl, 2017, 56 (48), 15314-15318.
14. Jagadesan, P.; Samanta, S. R.; Choudhury, R.; Ramamurthy, V.,
Container Chemistry: Manipulating excited state behavior of
organic guests within cavitands that form capsules in water. J Phys
Org Chem, 2017, 30 (9), e3728.
15. Ayhan, M. M.; Casano, G.; Karoui, H.; Rockenbauer, A.; Monnier,
V.; Hardy, M.; Tordo, P.; Bardelang, D.; Ouari, O., EPR Studies of
the Binding Properties, Guest Dynamics, and Inner-Space
Dimensions of a Water-Soluble Resorcinarene Capsule. Chem Eur
J, 2015, 21 (46), 16404-10.
16. Ouari, O.; Bardelang, D., Nitroxide Radicals with Cucurbit
[n]urils and Other Cavitands. Israel J Chem, 2018, 58 (3-4), 343-
356.
17. Galan, A.; Ballester, P., Stabilization of reactive species by
supramolecular encapsulation. Chem Soc Rev, 2016, 45 (6), 1720-
37.
18. Olivo, G.; Farinelli, G.; Barbieri, A.; Lanzalunga, O.; Di Stefano,
S.; Costas, M., Supramolecular Recognition Allows Remote, Site-
Selective C-H Oxidation of Methylenic Sites in Linear Amines.
Angew Chem Int Ed Engl, 2017, 56 (51), 16347-16351.
19. Kaanumalle, L. S.; Gibb, C. L.; Gibb, B. C.; Ramamurthy, V.,
Photo-Fries reaction in water made selective with a capsule. Org
Biomol Chem, 2007, 5 (2), 236-8.
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
30. Lissi, E. A.; Faure, M.; Montoya, N.; Videla, L. A., Reactivity of
Indole Derivatives towards Oxygenated Radicals. Free Rad Res
Comms, 1991, 15 (4), 211-222.
31. Melone, L.; Petroselli, M.; Pastori, N.; Punta, C., Functionalization
of Cyclodextrins with N-Hydroxyphthalimide Moiety: A New
Class of Supramolecular Pro-Oxidant Organocatalysts. Molecules,
2015, 20 (9), 15881-92; Dobras G., Sitko M., Petroselli M., Caruso
M., Cametti M., Punta C., Orlinska B., Chem Cat Chem, 2019,
DOI: 10.1002/cctc.201901737.
32. Petroselli, M.; Franchi, P.; Lucarini, M.; Punta, C.; Melone, L.,
Aerobic oxidation of alkylaromatics using
a lipophilic N-
hydroxyphthalimide: overcoming the industrial limit of catalyst
solubility. Chem Sus Chem, 2014, 7 (9), 2695-703.
33. Petroselli, M.; Melone, L.; Cametti, M.; Punta, C., Lipophilic N-
Hydroxyphthalimide Catalysts for the Aerobic Oxidation of
Cumene: Towards Solvent-Free Conditions and Back. Chem Eur
J, 2017, 23 (44), 10616-10625.
34. Matsui, S.; Fujita, T., New cumene-oxidation systems O2 activator
effects and radical stabilize effects. Catal Today, 2001, 71, 145-
152.
35. Ingold, K. U., Peroxyl Radicals. Acc Chem Res, 1969, 2 (1), 1-9.
ACS Paragon Plus Environment

Page 9 of 10
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
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 10
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
Encapsulation in host-guest complexes reveals effect of affinity on radical reactivity in synthetic receptors.
85x47mm (300 x 300 DPI)
ACS Paragon Plus Environment