Dynamics of capsuleplex formed between octaacid and organic guest molecules - Photophysical techniques reveal the opening and closing of capsuleplex

Article abstract of DOI:10.1139/V10-104

This manuscript is concerned with the opening and closing of a capsuleplex made up of organic guest molecules and two cavitand molecules known as octaacid (OA). The capsuleplex is loosely held together in water through weak interactions. We have investiga

Full text of DOI:10.1139/V10-104

203
Dynamics of capsuleplex formed between
octaacid and organic guest molecules —
Photophysical techniques reveal the opening and
closing of capsuleplex
Nithyanandhan Jayaraj, Steffen Jockusch, Lakshmi S. Kaanumalle, N.J. Turro, and
V. Ramamurthy
Abstract: This manuscript is concerned with the opening and closing of a capsuleplex made up of organic guest molecules
and two cavitand molecules known as octaacid (OA). The capsuleplex is loosely held together in water through weak in-
teractions. We have investigated the opening–closing of the OA capsuleplex by monitoring the quenching of excited sin-
glet and (or) triplet states of nine different guest molecules encapsulated within a capsule by molecular oxygen that is
dissolved in water surrounding the capsule. The rate constants for oxygen quenching of the excited guest molecules were
estimated by monitoring the intensity of fluorescence/phosphorescence or lifetimes of excited states of guest molecules in
the presence of different concentrations of oxygen in water. Guest molecules were chosen such that one could probe the
opening–closing process during a time range of 0.05 to 922 ms. We believe that the oxygen quenching constant reflects
the accessibility of oxygen to the guest enclosed within the OA capsuleplex, and this in turn depends on the capsule open-
ing–closing rate constant. Based on the quenching studies, we conclude that guests whose lifetimes are shorter than 5 ms
are inaccessible to oxygen. Results presented in this report suggest that the extent and time scale for opening depends on
the guest.
Key words: capsuleplex, octaacid, energy transfer, quenching by oxygen, capsule opening and closing.
´
´
´
´
´
´
Resume : Dans ce travail on a etudie l’ouverture et la fermeture d’un capsuleplexe qui est forme de molecules organiques
ˆ
´
hotes et de deux molecules de cavitands et qui est connu comme l’octa-acide (OA). Le capsuleplexe faiblement retenu en-
´
´
semble dans l’eau par le biais de faibles interactions. On a etudie le cycle d’ouverture–fermeture du capsuleplexe OA en
´
´
´
´
´
´
´
suivant la desactivation des etats singulet excite et triplet des neuf molecules invitees differentes encapsulees par une cap-
`
´
´
´
´
sule d’oxygene moleculaire qui est dissous dans l’eau entourant la capsule. On a evalue les constantes de vitesse de desac-
´
´
´
`
´
tivation des molecules invitees excitees par l’oxygene en suivant l’intensite de fluorescence/phosphorescence ou les temps
´
´
´
´
´
`
´
de vie des etats excites des molecules invitees en presence de concentrations variables d’oxygene dans l’eau. Les molecu-
´ ´
´
´
les invitees ont ete choisies d’une fac¸on telle qu’on puisse etudier d’ouverture–fermeture durant un intervalle de temps al-
`
´
`
´
`
lant de 0,05 a 922 ms. On croit que la constante de desactivation par l’oxygene est un reflet de l’accessibilite de l’oxygene
`
´
´
´
`
a la molecule invitee incluse dans le capsuleplexe d’OA qui dependrait a son tour sur la constante de vitesse d’ouverture–
´
´
´
´
fermeture de la capsule. Sur la base des etudes de desactivation, on conclut que les invites dont les temps de vie sont infe-
`
`
`
´
´
`
´
´
rieurs a 5 ms ne sont pas accessibles a l’oxygene. Les resultats rapportes ici suggerent que le degre et l’echelle de temps
´
´
pour l’ouverture dependent de l’invite.
´
´
´
`
Mots-cles : capsuleplexe, octa-acide, transfert d’energie, desactivation par l’oxygene, ouverture et fermeture de la capsule.
´
[Traduit par la Redaction]
cules that are loosely held by weak interactions in water.
Examples of 1:2 and 2:2 (guest:host ratio) capsuleplexes of
octaacid are known, and this article deals with both types of
capsuleplexes.2
We have recently established that an organic guest mole-
cule, confined within a capsule made up of two molecules of
OA, undergoes motion corresponding to rotations along its
Introduction
This manuscript is concerned with the dynamics of a cap-
suleplex made up of an organic guest and two host mole-
cules. The host octaacid (OA, 1, Scheme 1), having a
cuplike shape, belongs to a class of hosts known as cavi-
tands.1 The capsuleplex we are concerned with in this report
is made up of two OA cavitand molecules and guest mole-
Received 26 April 2010. Accepted 17 June 2010. Published on the NRC Research Press Web site at canjchem.nrc.ca on 21 January 2011.
This article is part of a Special Issue dedicated to Professor J. C. Scaiano.
N. Jayaraj, L.S. Kaanumalle, and V. Ramamurthy.1 Department of Chemistry, University of Miami, Coral Gables, FL 33124, USA.
S. Jockusch and N.J. Turro. Department of Chemistry, Columbia University, New York, NY 10027, USA.
1Corresponding author (e-mail: murthy1@miami.edu).
Can. J. Chem. 89: 203–213 (2011)
doi:10.1139/V10-104
Published by NRC Research Press

204
Can. J. Chem. Vol. 89, 2011
Scheme 1. (i) Structure and dimensions of the octaacid (OA) and its complexes; (a–h) represent the various protons in OA. The dimensions
indicated for OA refer to atom-to-atom in (ii) and include van der Walls radius in (iii). Cartoon representations of the 1:2 and 2:2 octaacid
complexes are shown in (iv).
Scheme 2. Three types of motions exerted by guests within octaacid (OA) capsuleplexes are illustrated in (a–c). The equilibrium between
capsuleplex opening and closing is illustrated in (d).
short and long axes.3 Examination of a number of guest
molecules by H NMR and EPR experiments suggested that
cally? If the latter, what is the time scale in which such a
process takes place?’’
1
In this study, we have probed the opening–closing of a
capsuleplex by monitoring the quenching of excited singlet
and (or) triplet states of guest molecules (Scheme 3) present
inside a capsule by molecular oxygen that is present outside
the capsule. The rate of quenching was estimated by moni-
toring the intensity of fluorescence/phosphorescence or life-
times of excited states of guest molecules in the presence of
different concentrations of oxygen in water. Guest molecules
were chosen such that one could probe the opening–closing
process during a wide time range. Structures and molecular
dimensions of guests employed whose excited-state lifetimes
cover 0.05 to 922 ms are listed in Scheme 3. Based on the
results presented in the following, we conclude that guests
whose lifetimes are shorter than 5 ms are not accessible to
the extent of rotation depends on the structure of the guest
molecule. In addition, when two guest molecules form a
capsuleplex with OA, they can undergo sliding motion.
These three types of motion that we have established as oc-
curring within the OA capsuleplex are illustrated in a car-
toon fashion in Schemes 2a–2c. In this manuscript, we are
concerned with yet another motion, namely partial opening
and closing of the capsuleplex (Scheme 2d). We define
opening to mean that the two cavitand molecules that make
up the capsule come apart enough for a quencher such as
oxygen to access the guest present inside. Since the capsule-
plex is made up of two weakly interacting host molecules, a
legitimate question comes to mind: ‘‘Does the capsuleplex
remain closed all the time or does it open and close periodi-
Published by NRC Research Press

Jayaraj et al.
205
Scheme 3. Structures and dimensions of the guests used in this study. The dimensions indicated for various guests refer to atom-to-atom and
do not include the van der Waals radius.
oxygen. However, guests with lifetimes longer than 15 ms
are accessible to oxygen. Based on the measured quenching
rate constants, we conclude that the extent and time scale
for opening depends on the guest.
0.5 equiv, there are changes in the host region. Addition of
more than 0.5 equiv of the guest did not result in any further
changes in the spectra, suggesting that 8 forms a 1:2 com-
plex with OA. Also, addition of more than 0.5 equiv of 8 to
the OA solution resulted in turbidity, suggesting that 8 is in-
soluble in water in the absence of OA. The appearance of
sharp signals owing to guest 8 and the absence of a shift in
the guest peaks with varying ratios of host-to-guest sug-
gested that in the NMR time scale, 8 remains within the
OA capsule. Capsule formation was further confirmed by
Results and discussion
Capsuleplexes composed of guests 2–10 and OA were
prepared in water at pH ~9.0 (borate buffer).4,5 Details are
provided in the Experimental section of the Supplementary
data. 1H NMR spectral data confirmed the inclusion of
guests within the OA capsule and provided information
about the stoichiometry of the complex. NOESY experi-
ments helped us infer the orientation of guest molecule(s)
1
measuring the diffusion constant of the capsuleplex by H
NMR DOSY experiments. For example, free OA has a dif-
fusion constant of 1.88 ꢀ 10–6 cm2 s–1, whereas the 8@OA2
complex has a diffusion constant of 1.38 ꢀ 10–6 cm2 s–1.
Similar diffusion constants were observed for complexes of
2–7, 9, and 10 (see Supplementary data). Guest-to-host ra-
tios for all capsuleplexes discussed in this study are included
in Table 1.
1
within the OA capsule. The H NMR experimental details
for one guest molecule, 4,4’-dimethoxythiobenzophenone
(8), as a representative of guests 2–10, are described in the
following. A similar approach yielded information about the
inclusion of other guest molecules and are listed in
The experiments we performed to gain information con-
cerning the capsule opening–closing process rely on the
quenching of excited states (S1 or T1) of guest molecules by
molecular oxygen.8 Guest molecules 2–8, having very high
rate constants of intersystem crossing (ISC) from S1 to T1,
show phosphorescence at room temperature in solution (see
Table 1 for the quenching of the excited state).8,9 On the
other hand, probes 9 and 10, having low rate constants of
ISC from S1 to T1, fluoresce at room temperature. In the
first set of compounds (2–8), the T1 state (phosphorescence),
and in the latter two molecules (9 and 10), the S1 state (flu-
orescence), are quenched by molecular oxygen via energy
transfer in organic solvents with rate constants >5 ꢀ 108
(mol/L)–1 s–1 (Table 1).10,11 Since pyrene (9) did not phos-
phoresce at room temperature, quenching in the case of the
pyrene triplet was monitored by its T–T absorption. As indi-
cated in Table 1, the pyrene triplet was quenched by oxygen
in a cyclohexane solution with a rate constant of 1.6 ꢀ 109
(mol/L)–1 s–1.10 Since the excited states of all the guest
1
Scheme 3. The H NMR spectra of free OA in water, 8 in
CDCl3, and 8@OA2 capsuleplex (represented as guest@host;
stoichiometry indicated as a subscript) in water are provided
in Fig. 1. In this case, the signals owing to OCH3 in com-
pound 8 appear at d 3.9 ppm in CDCl3, whereas within OA
in D2O, it is upfield shifted to d –0.81 ppm. Such an upfield
shift owing to diamagnetic shielding by the aromatic ring
that forms the OA capsule has been established earlier.6,7
Similar upfield shifts for various protons in compounds 2–7,
9, and 10 confirmed the inclusion of these probe molecules
within the OA capsule (the spectra are provided as Supple-
mentary data). As seen in Fig. 1, in addition to the shift in
1
the H signals of the guest, the signals owing to the OA
molecule were also affected. Interpretation of these changes
has been discussed in prior publications and will not be re-
1
peated here. H NMR titration experiments established the
stoichiometry of the 8@OA2 complex to be 1:2. Titration
spectra in the case of OA with 8 are provided in Fig. 2. As
can be seen in the spectra, upon addition of the guest up to
Published by NRC Research Press

206
Can. J. Chem. Vol. 89, 2011
1
Fig. 1. H NMR spectra of (i) 4,4’-dimethoxythiobenzophenone (8) in CDCl3, (ii) capsuleplex, octaacid (OA) :8 (2:1), [OA] = 1 mmol/L in
10 mmol/L tetraborate D2O buffer, and (iii) OA (1 mmol/L in 10 mmol/L tetraborate D2O buffer). Signals marked with * represent the
residual proton resonances from the corresponding solvents.
molecules we chose were quenched with very high rate con-
stants in solution,12 we believed that oxygen should be able
to quench the excited states of guest molecules 2–10 en-
trapped within the OA, provided that the excited guest mol-
ecules were accessible to oxygen during their lifetime.
Alternatively, the quenching could occur through the wall
of the capsuleplex. We believed that information on the rate
constants of oxygen quenching of the excited states of vari-
ous guests would be able to distinguish between the two
mechanistic possibilities. If the quenching occurs through
donor–acceptor contact, the accessibility of oxygen to the
excited guest molecule would depend on the opening–
closing of the capsuleplex. The cavitand that forms the cap-
sule has two ports, one small and one large. Oxygen with a
as a representative of guests 2–6. Experimental data for
guests 2 and 4–6 are provided in the Supplementary data.
Room-temperature emission spectra of 3@OA2 in water (bo-
rate buffer) consisting of fluorescence, delayed fluorescence
(480–520 nm), and phosphorescence (530–650 nm) are
shown in Fig. 3a. As illustrated in the figure, the phosphor-
escence intensity is significantly reduced in the presence of
oxygen compared with that in nitrogen-saturated buffer sol-
ution. The displayed spectra clearly show that emission in-
tensities are dependent on the concentration of oxygen in
water. The phosphorescence (530–650 nm) was significantly
quenched by oxygen, whereas the fluorescence (direct and
delayed fluorescence at 480–520 nm) shows only minor
oxygen quenching. Consistent with the steady-state emission
data, the excited triplet state decay profiles displayed in
Fig. 3b also exhibit oxygen-concentration dependence.
Rate constants for quenching of the benzil triplet by oxy-
gen were determined by monitoring the decay profiles of
benzil triplet states at varying concentrations of oxygen.
Triplet-state decay profiles were monitored by two inde-
pendent methods, phosphorescence (Fig. 3b) and triplet–
triplet absorption measurements. The rate constants for
quenching were obtained from the slope of the plots of the
pseudo-first-order decay rate constants of benzil triplet states
vs the oxygen concentration (Fig. 4). Quenching rate con-
stants obtained by the two methods (included in the figure)
are identical within experimental error.13,14 These data sug-
gest that oxygen is able to quench the benzil triplet in spite
of it being enclosed within the OA capsule. However, the
rate constant for quenching of benzil in a capsuleplex is
˚
molecular width of ~3 A (including the van der Waals’ ra-
dius) will not be able to access the entrapped guest through
the smaller port of the OA capsuleplex, which has a diame-
˚
ter of ~2 A (including the van der Waals’ radius, see
Scheme 1). Therefore, we believe that oxygen and the en-
trapped guest could meet only through the larger port that is
present at the middle of the capsuleplex. The larger port
would become available only if the capsuleplex opens at
least partially. We need to address several questions before
drawing conclusions regarding the opening–closing process
of the capsuleplex from the oxygen quenching data: (i) Is
the probe molecule entrapped within a capsule accessible to
oxygen? (ii) Is the quenching static or dynamic? (iii) What
are the rates of quenching of the capsule-trapped guests? (iv)
What is the rate-determining step in the quenching process?
In this section, we present experimental data for benzil (3)
Published by NRC Research Press

Jayaraj et al.
207
1
Fig. 2. H NMR titration studies of the octaacid (OA) with 8 (500 MHz, D2O). [OA] = 1 mmol/L in 10 mmol/L tetraborate D2O buffer.
Signals marked with * represent the residual proton resonances from the solvent.
significantly lower (1.2 ꢀ 107 (mol/L)–1 s–1 vs 6.0 ꢀ 108
(mol/L)–1 s–1) than in an organic solvent (Table 1). Rate
constants for oxygen quenching of the triplets of guest mol-
ecules 2–6 and 9 captured with the OA capsule are provided
in Table 1. As noted, the numbers are more than an order of
magnitude lower in the capsule than in an organic solvent.
To interpret the data, we need to ascertain whether the
quenching is static or dynamic. The data provided in Fig. 3
suggest that the quenching is a dynamic process. Had it been
a fully static process, no variation in lifetime would have
occurred. Further support comes from the Stern–Volmer
plots with points taken from both steady-state emission and
lifetime data (Fig. 3c). The linear lines with almost identical
slopes are consistent with a dynamic process. This conclu-
sion is also consistent with the difference in rise time of sin-
glet oxygen phosphorescence and decay time of benzil
phosphorescence (Fig. 5c).
To probe the mechanism of the quenching process, phos-
phorescence emission from singlet oxygen was monitored
upon quenching the 3@OA2 triplet by ground-state oxygen.
As illustrated in Fig. 5a, quenching of benzil triplet resulted
in phosphorescence from singlet oxygen, suggesting that
quenching is the result of energy transfer. Furthermore, the
difference in emission intensities (Fig. 5a) and lifetimes
(Figs. 5b and 5c; 5 and 41 ms in H2O and D2O buffer, re-
spectively)15 of singlet oxygen observed between H2O and
D2O solutions suggested that singlet oxygen, generated by
the above energy-transfer process, spends most of its life-
Published by NRC Research Press

208
Can. J. Chem. Vol. 89, 2011
Table 1. Excited-state lifetime and oxygen-quenching data for capsuleplexes made up of octaacid (OA) and organic guests 2–10 listed in
Scheme 3.
Excited state
O2 quenching in
OA (%)
kq (O2) ((mol/L)–1 s–1)
kq (O2) ((mol/L)–1 s–1)
in solution
Compound
Pyrene (9)
H:G
2:1
2:1
2:2
2:1
2:2
2:2
2:1
2:1
Type
T1
T1
T1
T1
Lifetime in OA (ms)a
in OAa
922b
596e
187e
151e
65e
38c
85f
55f
63f
59f
48f
3f
5ꢀ105a
1.6ꢀ109d
8.2ꢀ108b,g
>109
4,4’-Dimethylbenzil (2)
Fenchthione (4)
Benzil (3)
Camphorthione (5)
Adamantanethione (6)
Xanthione (7)
4,4’-Dimethoxy
thiobenzophenone (8)
Pyrene (9)
8.7ꢀ107a 7.9ꢀ107b
1.6ꢀ106b
1.3ꢀ107a 1.2ꢀ107e
2.4ꢀ107e
2.8ꢀ107e
—
6.0ꢀ108b,g
T1
T1
>109
17e
>109
T1
T1
4.8e
0.62e
>109
0f
—
>109
2:1
2:1
S1
S1
0.34h
0.05h
0f
0f
—
—
2.5ꢀ1010d
2.3ꢀ1010d
Phenanthrene (10)
Note: H, host; G, guest; and kq, quenching rate constant.
aError limit ~10%.
bDetermined by triplet–triplet absorption.
cDetermined by comparing the triplet absorption decay traces in both nitrogen- and oxygen-saturated solutions of guest@OA capsuleplexes.
dIn cyclohexane solution.
eDetermined by phosphorescence.
fDetermined by comparing the relative luminescence intensities (at lmax) of the spectra recorded in both nitrogen- and oxygen-saturated solutions of
guest@OA capsuleplexes.
gIn acetonitrile solution.
hDetermined by fluorescence.
time in water. Since our earlier studies have shown that the
capsuleplex is nonpolar and has no water molecules inside
the capsule,16 we believe that singlet oxygen following en-
ergy transfer, if it occurs within the capsule, escapes to
water. Alternatively, the energy transfer might be occurring
when the capsule is partially open so that water molecules
surround the donor and acceptor pair. We favor the latter
model.
both triplet (7 and 8) and excited singlet (9 and 10) states,
we believe that the absence of quenching in the second set
does not depend on the spin state. Closer examination of Ta-
ble 1 suggests that the lifetime of the excited state is likely
to play a role in the quenching process. Although in solution
they are quenched at diffusion-controlled rate constants,
guest molecules whose lifetimes are shorter than 5 ms are
not quenched by oxygen when they are enclosed in the OA
capsule. Also, it is important to note that the five guest mol-
ecules whose lifetimes span between 17 and 922 ms are
Triplets of xanthione (7) and 4,4’-dimethoxythiobenzophe-
none (8) included within OA capsule were not quenched by
oxygen, although they were quenched at diffusion-limited
rates in organic solvents.17–19 Phosphorescence spectra and
triplet decay profiles in the presence of oxygen and nitrogen,
shown in Fig. 6 for xanthione, are distinctly different from
those shown in Fig. 3 for benzil. In Fig. 6, the phosphores-
cence intensities and triplet decays are almost identical for
nitrogen- and oxygen-saturated solutions. This behavior also
extended to S1 of pyrene (9) and phenanthrene (10). As il-
lustrated in Fig. 7a, the fluorescence spectra of pyrene@OA2
in nitrogen-saturated and oxygen-saturated buffer solutions
are identical. The S1 lifetime of pyrene was nearly identical
under nitrogen- and oxygen-saturated conditions (Fig. 7b).
This is distinctly different from that in cyclohexane solution
where pyrene S1 is quenched by oxygen at a diffusion-
limited rate. Thus, while the S1 states of pyrene and
phenanthrene are quenched at diffusion-controlled rates in
cyclohexane,10,11 they are unquenched when they are in-
cluded within the OA.
quenched with rate constants varying between
5
ꢀ
105 (mol/L)–1 s–1 and 8 ꢀ 107 (mol/L)–1 s–1. We need to ad-
dress three important questions: (i) What exactly do the
quenching rate constants imply? (ii) Why are some encapsu-
lated guests quenched but not the others? (iii) Why do the
quenching rate constants vary between the guests?
In Scheme 4, the sequence of steps involved in the
energy-transfer process are summarized with rate constants
listed above each step. The first step is the diffusion of the
excited capsuleplex and oxygen to form a collision complex
with a rate constant of kd. The second step represents the ac-
cess of oxygen to the excited guest present within the cap-
sule. This step requires partial opening of the capsuleplex
and the rate constant is represented as ko. The third step is
the true energy-transfer process between ground-state oxy-
gen and the excited guest that are close to each other. This
step is represented by the rate constant kE0 T. To differentiate
between the energy transfer occurring across the wall of the
capsuleplex and the one involving the collisional complex
between ground-state oxygen and the excited guest (present
within the host OA), the former process is represented by a
rate constant of kET.
Examination of Table 1 reveals that there are two sets of
molecules, one (triplet 2–6 and triplet 9) includes the mole-
cules that are quenched by oxygen when enclosed within the
OA capsule, and the other (7–10) represents the molecules
that are not quenched by oxygen when enclosed within the
OA capsule. Since the second set of molecules involves
The sequence of steps involved in the quenching of
guest@OA2 by oxygen is shown in Scheme 4. We visualize
Published by NRC Research Press

Jayaraj et al.
209
Fig. 3. (a) Steady-state emission spectra of 3@OA2 in solutions sa-
turated with various oxygen–nitrogen mixtures. OA, octaacid. (b)
Phosphorescence decay profile of 3@OA2 observed at 560 nm after
excitation (308 nm, 15 ns pulse width). (c) Stern–Volmer plots with
various concentrations of oxygen using steady-state phosphores-
cence intensity (Io/I) and triplet lifetimes (to/t).
Fig. 4. Determination of the quenching rate constant (kq) of benzil
(3) triplets by oxygen. (a) Pseudo-first-order decay constant of the
triplet–triplet absorption of 3 observed at 460 nm after excitation
with laser pulses (355 nm, 5 ns pulse width) of 3@OA2 solutions at
different oxygen concentrations. OA, octaacid. (b) Phosphorescence
decay constant of 3 observed at 560 nm after excitation with laser
pulses (308 nm, 15 ns pulse width) of 3@OA2 solutions at different
oxygen concentrations. The black lines show the linear fits.
three possible mechanisms for the energy-transfer process:
(i) Energy transfer occurs between the excited guest en-
trapped within the OA capsule and oxygen in water through
the walls of the capsule. This type of energy transfer has
been demonstrated to occur between benzil entrapped within
a covalently linked capsule (carceplex) and oxygen.20–24 The
upper limit of the rate constant for this process was esti-
mated to be 104 (mol/L)–1 s–1. Since the measured rate con-
stants in the current study are orders of magnitudes higher,
we believe that the energy transfer does not occur through
the walls of the capsule. (ii) Energy transfer occurs between
the donor guest and acceptor oxygen (both being present
within the capsule). Under such conditions, static quenching
is expected. Such a possibility has been ruled out by the re-
sults suggesting that the quenching is a dynamic rather than
a static process (see Figs. 3 and 5 and related discussion).
(iii) The third possibility, which we favor, is the one where
quenching is controlled by the capsule opening–closing
process. In this case, the quenching would be expected to
be dynamic. In solution, the rate constant for energy transfer
between benzil and oxygen was measured to be 6 ꢀ
108 (mol/L)–1 s–1, which is higher than the number measured
in the presence of the OA.13,14 Since the rate constants we
have measured for benzil@OA2 and five other guest@OA2
capsuleplexes (Table 1) are lower than those in solution (in
the absence of the OA), we believe that the measured rate
constants do not represent the energy-transfer process, but
they represent the accessibility of oxygen to the entrapped
guest, which in turn depends on the capsule opening–closing
process.
To fully understand the meaning of the quenching con-
stant, we need to know exactly what the measured kq repre-
sents. The steps involved in the quenching are the formation
of the encounter complex between molecular oxygen and the
excited guest – OA capsuleplex (kd), the decomplexation of
the encounter complex between oxygen and capsuleplex
without energy transfer (k–d), and the capsule opening that
facilitates the interaction between oxygen and the excited
encapsulated guest (ko). Upon close interaction between
ground-state oxygen and the excited guest, we assume that
Published by NRC Research Press

210
Can. J. Chem. Vol. 89, 2011
Fig. 5. Quenching of benzil (3) phosphorescence and generation of
singlet oxygen. (a) Steady-state phosphorescence of singlet oxygen
from excitation (310 nm) of air-saturated buffer solutions (D2O or
H2O) of 3@OA2. OA, octaacid. (b) (H2O) and (c) (D2O): Phos-
phorescence decay trace of 3 (blue trace observed at 560 nm) and
1O2 (red trace observed at 1270 nm) after excitation with a laser
pulse (308 nm, 15 ns pulse width). The 3@OA2 solutions in buffer
were saturated with N2/O2 gas mixture (1:1). The black lines show
the kinetic fits of benzil phosphorescence decay and growth + de-
Fig. 6. (a) Steady-state phosphorescence spectra of 7@OA2 in oxy-
gen and nitrogen, excitation wavelength (400 nm). OA, octaacid.
(b) Phosphorescence decay profile of 7@OA2 observed at 670 nm
after excitation (355 nm, 5 ns pulse width) of 7@OA2 solutions in
oxygen and nitrogen.
1
cay of O2 phosphorescence. Bz in the figure represents benzil.
would be between 0.1 and 0.5 ns.26,27 Thus, the value for
k–d would be between 2 ꢀ 109 and 1 ꢀ 1010 s–1. Under
such conditions, ko would be between 0.3 and 1.3 ꢀ kq. On
the basis of these assumptions, we believe that the measured
quenching constants listed in Table 1 indirectly reflect the
rate constants for capsule opening and that the values vary
over at least two orders of magnitude between the pyrene
triplet and the 4,4’-dimethylbenzil triplet.
kdko
½1ꢁ
½2ꢁ
½3ꢁ
kq ¼
kq ¼
ko ¼
kꢂd þ ko
kdko
kꢂd
kꢂdkq
kd
the energy transfer occurs instantaneously, and the slow step
is the accessibility of oxygen to the trapped excited guest
molecule. Under these conditions, the measured kq would
be given by eq. [1]. If we assume that k–d > ko, then kq and
ko would be given by eqs. [2] and [3], respectively. Since no
specific interactions are expected between oxygen and the
capsuleplex, we believe that k–d would be determined by
the lifetime of the encounter complex (capsuleplex and oxy-
gen) in water and kd would be the diffusion rate constant in
water, which is 7.4 ꢀ 109 (mol/L)–1 s–1.25 Based on litera-
ture reports, we assume that the lifetime of the encounter
complex between oxygen and the capsuleplex in water
Finally, we address the question of why the excited states
of xanthione (T1, 4.7 ms), 4,4’-dimethoxythiobenzophenone
(T1, 0.6 ms), pyrene (S1, 0.3 ms), and phenanthrene (S1,
0.05 ms) included within OA capsule are not quenched by
oxygen (Table 1). To understand this process, one needs to
recognize that guests with excited-state lifetimes longer than
17 ms are quenched by oxygen to varying percentages (see
Table 1). For example, the phosphorescence of adamantane-
thione (6) with a lifetime of 17 ms is quenched to 48% by
oxygen (based on relative phosphorescence intensities of
Published by NRC Research Press

Jayaraj et al.
211
Scheme 4. Schematic representation of energy transfer between the encapsulated guest and oxygen. The rate constants indicated in the
scheme are defined in the text.
Fig. 7. (a) Steady-state fluorescence spectra of 9@OA2 in oxygen
and nitrogen. OA, octaacid. (b) Fluorescence decay profile of
9@OA2 observed at 380 nm after excitation (330 nm) of 9@OA2
solutions in oxygen and nitrogen.
quenched only to 38% as monitored by its T–T absorption.
The most noteworthy feature is that molecules whose ex-
cited-state lifetimes are shorter than 5 ms remain unquenched
by oxygen. We believe that 5 ms may be the cutoff point in
terms of the accessibility of oxygen to the excited guest
molecule. This suggests that the trapped guest excited state
must live for longer than 5 ms for oxygen to access it. One
could interpret this to mean that the capsule takes longer
than 5 ms to open (at least in the case of xanthione).
As indicated by the quenching percentage (Table 1), the
time it takes for the capsule to open may vary with the
guest. The excited-state lifetime alone does not determine
the quenching percentage. As seen in Table 1, there is no
direct correlation between the lifetime and quenching per-
centage (compare the triplets of 2, 6, and 9). The observed
kq range from 5 ꢀ 105 to 9 ꢀ 107 (mol/L)–1 s–1, depending
on the guest molecule (Table 1). This variation in kq values
indicates that the structure of the guest plays an important
role in the stability of the host–guest complex and the ki-
netics of the accessibility of oxygen to the entrapped guest.
For example, for benzil, a kq of 1.3 ꢀ 107 (mol/L)–1 s–1 was
observed, whereas for longer dimethylbenzil (11.8 vs
˚
13.7 A; see Scheme 3), a kq approximately six times larger
(8 ꢀ 107 (mol/L)–1 s–1) was observed (see Scheme 5 for car-
toon representations of the capsuleplexes). This suggests that
for the capsule to open just enough for oxygen to reach the
excited guest, it takes about six times longer for benzil than
for dimethylbenzil. Clearly, the length of the molecule that
determines the compactness of the capsuleplex is important
in the opening–closing process.
The most interesting observation was noticed with pyrene.
In this case, a much smaller rate constant of 5 ꢀ 105
(mol/L)–1 s–1 was observed. This small kq could be ex-
plained by the different nature of the pyrene triplet state
(p–p*) compared with the benzil derivatives with an n–p*
nature. Triplet states of n–p* are localized at the C=O
oxygen- and nitrogen-saturated solutions). On the other
hand, 4,4’-dimethylbenzil with a lifetime of 596 ms is
quenched to 85% of its initial phosphorescence intensity. In-
creased quenching cannot be solely due to the longer life-
time of the 4,4’-dimethylbenzil triplet. For example, the
pyrene triplet with a much longer lifetime (922 ms) is
Published by NRC Research Press

212
Scheme 5. Structures of the guest@host complexes based on H NMR data.
Can. J. Chem. Vol. 89, 2011
1
chromophore, which is located at the capsule entrance for
benzil derivatives, and therefore, probably more accessible
to oxygen. Pyrene triplet states (p–p*) are delocalized over
the entire molecule, which probably makes them less acces-
sible to oxygen quenching at the capsule entrance. Thus, the
measured quenching constant reflects the accessibility of
oxygen to the excited guest for energy transfer to take place.
Therefore, the extent of the required opening for each guest
could vary, and this in turn may depend on the structure of
the guest. We caution that the measured kqs listed in Table 1
do not represent the same extent of opening in all cases.
However, the lower than diffusion-limited value and higher
than the energy-transfer rate constant (kET, Scheme 4) re-
ported for a carcerand suggest that the measured kqs provide
information concerning the opening of the capsuleplex.
From the quenching studies, we conclude that the capsule-
plex is dynamic and it periodically opens and closes in the
ms time scale.
times (17 vs 151 ms). From these studies, it is clear that cap-
suleplexes of organic guests and OA are dynamic, and one
of the motions they exhibit is the opening and closing of
the capsule (Scheme 2d). The time scale for this process de-
pends on the structure and hydrophobicity of the guest. The
extent to which the capsuleplex has to open for quenching to
occur may vary with the orientation of the guest within the
capsule.
Conclusion
In this study, we show that excited states of guests en-
closed as capsuleplexes could be quenched by oxygen via
an energy-transfer process. The rate constant measured for
the quenching reflects the accessibility of oxygen to the
guest enclosed within the OA capsuleplex, and this in turn
depends on the capsule opening–closing. The rate constant
of this process most likely depends on the structure of the
guest.
The difference in rate constants between the three thiones
4, 5, and 6 is also interesting. In spite of having similar triplet-
state lifetimes in solution, the triplet-state lifetimes and
quenching constants in OA capsuleplexes differ by an order
of magnitude between fenchthione and adamantanethione.
Examination of the orientation of the thiones within the cap-
Supplementary data
Supplementary data for this article are available on the
journal Web site (canjchem.nrc.ca).
1
sule accounts for the difference.9 Orientations based on H
Acknowledgements
NMR data of the three thiones within the capsule are shown
in Scheme 5. NMR experiments showed that the thiocar-
bonyl chromophore of fenchthione is deeply buried at the
bottom of the OA cap, which shields the excited state from
oxygen quenching and, consequently, a lower kq is observed
(1.6 ꢀ 106 (mol/L)–1 s–1). Clearly in this case, a small open-
ing as in the case of benzil would not suffice for oxygen to
reach the excited chromophore. On the other hand, in the
case of adamantanethione, where the thione chromophore is
in the middle of the capsule, it could be accessed (similar to
benzil) with a slight opening of the capsule. The observed
quenching constant for adamantanethione is comparable to
that of benzil, although they have distinctly different life-
V.R. and N.J.T. are grateful to the National Science Foun-
dation (NSF), USA, for generous financial support (CHE-
0848017 and CHE-0717518, respectively). The authors
thank Professor C. Bohne for insightful discussions.
References
(1) Gibb, C. L. D.; Gibb, B. C. J. Am. Chem. Soc. 2004, 126
(37), 11408. doi:10.1021/ja0475611. PMID:15366865.
(2) Jayaraj, N.; Zhao, Y.; Parthasarathy, A.; Porel, M.; Liu, R. S.
H.; Ramamurthy, V. Langmuir 2009, 25 (18), 10575. doi:10.
1021/la901367k. PMID:19496576.
(3) Kulasekharan, R.; Jayaraj, N.; Porel, M.; Choudhury, R.;
Published by NRC Research Press

Jayaraj et al.
Sundaresan, A. K.; Parthasarathy, A.; Ottaviani, M. F.; Jock-
usch, S.; Turro, N. J.; Ramamurthy, V. Langmuir 2010, 26
(10), 6943. doi:10.1021/la904196g. PMID:20055365.
(4) Parthasarathy, A.; Kaanumalle, L. S.; Ramamurthy, V. Org.
Lett. 2007, 9 (24), 5059. doi:10.1021/ol702322u. PMID:
17956116.
(5) Sundaresan, A. K.; Ramamurthy, V. Org. Lett. 2007, 9 (18),
3575. doi:10.1021/ol0713990. PMID:17665921.
(6) Kaanumalle, L. S.; Gibb, C. L. D.; Gibb, B. C.;
Ramamurthy, V. J. Am. Chem. Soc. 2005, 127 (11), 3674.
doi:10.1021/ja0425381. PMID:15771483.
213
(15) Hurst, J. R.; Schuster, G. B. J. Am. Chem. Soc. 1983, 105
(18), 5756. doi:10.1021/ja00356a009.
(16) Porel, M.; Jayaraj, N.; Kaanumalle, L. S.; Maddipatla, M. V.
S. N.; Parthasarathy, A.; Ramamurthy, V. Langmuir 2009,
25, 3473. doi:10.1021/la804194w. PMID:19708142.
(17) Ramesh, V.; Ramnath, N.; Ramamurthy, V. J. Photochem.
1982, 18 (3), 293. doi:10.1016/0047-2670(82)85014-4.
(18) Rajee, R.; Ramamurthy, V. J. Photochem. 1979, 11 (2), 135.
doi:10.1016/0047-2670(79)80046-5.
(19) Rajee, R.; Ramamurthy, V. Tetrahedron Lett. 1978, 19 (51),
5127. doi:10.1016/S0040-4039(01)85829-9.
(7) Natarajan, A.; Kaanumalle, L. S.; Jockusch, S.; Gibb, C. L.
D.; Gibb, B. C.; Turro, N. J.; Ramamurthy, V. J. Am. Chem.
Soc. 2007, 129 (14), 4132. doi:10.1021/ja070086x. PMID:
17362015.
(8) Turro, N. J.; Ramamurthy, V.; Scaiano, J. C. Modern Mole-
cular Photochemistry of Organic Molecules; University
Science Books: Sausalito, CA, 2010.
(20) Parola, J. A.; Pina, F.; Ferreira, E.; Maestri, M.; Balzani, V.
J. Am. Chem. Soc. 1996, 118 (46), 11610. doi:10.1021/
ja962450q.
(21) Pina, F.; Parola, A. J.; Ferreira, E.; Maestri, M.; Armaroli,
N.; Ballardini, R.; Balzani, V. J. Phys. Chem. 1995, 99 (34),
12701. doi:10.1021/j100034a003.
´
(22) Farran, A.; Deshayes, K. D. J. Phys. Chem. 1996, 100 (9),
(9) Jayaraj, N.; Maddipatla, M. V. S. N.; Prabhakar, R.;
Jockusch, S.; Turro, N. J.; Ramamurthy, V. J. Phys. Chem.
B 2010, 114 (45), 14320. doi:10.1021/jp911698s. PMID:
20356085.
(10) Patterson, L. K.; Porter, G.; Topp, M. R. Chem. Phys. Lett.
1970, 7 (6), 612. doi:10.1016/0009-2614(70)87019-1.
(11) Saltiel, J.; Atwater, B. W. Adv. Photochem. 1988, 14, 1.
doi:10.1002/9780470133446.ch1.
3305. doi:10.1021/jp9531467.
(23) Place, I.; Farran, A.; Deshayes, K.; Piotrowiak, P. J. Am.
Chem. Soc. 1998, 120 (48), 12626. doi:10.1021/ja982538z.
(24) Romanova, Z. S.; Deshayes, K.; Piotrowiak, P. J. Am. Chem.
Soc. 2001, 123 (44), 11029. doi:10.1021/ja011325y. PMID:
11686708.
(25) Turro, N. J.; Ramamurthy, V.; Scaiano, J. C. Modern Mole-
cular Photochemistry of Organic Molecules; University
Science Books: Sausalito, CA, 2010; pp 469–478.
(26) Noyes, R. M. J. Am. Chem. Soc. 1955, 77 (8), 2042. doi:10.
1021/ja01613a003.
(12) Montalti, M.; Credi, A.; Prodi, L.; Teresa Gandolfi, M.
Handbook of Photochemistry; 3rd ed.; Taylor & Francis:
Boca Raton, 2006.
(13) Fang, T.-S.; Brown, R. E.; Kwan, L. C.; Singer, L. A. J.
Phys. Chem. 1978, 82 (23), 2489. doi:10.1021/j100512a008.
(14) Parker, C. A.; Joyce, T. A. Chem. Commun. 1968, 1421.
(27) Czapski, G.; Ogdan, J.; Ottolenghi, M. Chem. Phys. Lett.
1969, 3 (6), 383. doi:10.1016/0009-2614(69)80143-0.
Published by NRC Research Press