Internal and external stereoisomers of squaraine rotaxane endoperoxide: Synthesis, chemical differences, and structural revision

Article abstract of DOI:10.1021/jo402564k

Photooxygenation of permanently interlocked squaraine rotaxanes with anthracene-containing macrocycles produces the corresponding squaraine rotaxane endoperoxides (SREPs) quantitatively. SREPs are stored at low temperature, and upon warming, they undergo

Full text of DOI:10.1021/jo402564k

Article
pubs.acs.org/joc
Internal and External Stereoisomers of Squaraine Rotaxane
Endoperoxide: Synthesis, Chemical Differences, and Structural
Revision
Carleton G. Collins,† Joshua M. Lee,† Allen G. Oliver,† Olaf Wiest,†,‡ and Bradley D. Smith*,†
†Department of Chemistry and Biochemistry, 236 Nieuwland Science Hall, University of Notre Dame, Notre Dame, Indiana 46556,
United States
‡Lab of Computational Chemistry and Drug Design, Laboratory of Chemical Genomics, Peking University Shenzhen Graduate
School, Shenzhen 518055, P.R. China
S
* Supporting Information
ABSTRACT: Photooxygenation of permanently interlocked squaraine
rotaxanes with anthracene-containing macrocycles produces the corresponding
squaraine rotaxane endoperoxides (SREPs) quantitatively. SREPs are stored at
low temperature, and upon warming, they undergo clean cycloreversion,
releasing singlet oxygen and emitting light. The structural elucidation in 2010
assigned the structure as the SREP-int stereoisomer, with the endoperoxide
unit directed inside the macrocycle cavity. New experimental and computa-
tional evidence reported here proves that the initial, kinetic photooxygenation
product is the less stable SREP-ext stereoisomer with the endoperoxide unit
directed outside the macrocycle. The photophysical properties and subsequent
reactivity of mechanically strained SREP-ext depend on the size of the end
groups of the encapsulated squaraine dye. If the end groups are sufficiently
large to prevent dissociation of the interlocked components, the strained
SREP-ext stereoisomer undergoes clean thermal cycloreversion. However, smaller squaraine end groups allow transient
dissociation, resulting in a pseudorotaxane dissociation/association process that produces SREP-int as the thermodynamic
stereoisomer that does not cyclorevert. The large difference in endoperoxide reactivity for the two SREP stereoisomers illustrates
the power of the mechanical bond to induce cross-component steric strain and selective enhancement of a specific reaction
pathway. The new insight enabled synthetic development of triptycene-containing squaraine rotaxanes with high fluorescence
quantum yields and large Stokes shifts.
cycloreversion.5−7 In this paper, experimental and computa-
tional results are presented that necessitate a revision of our
initial assignment of SREP stereochemistry and greatly expand
INTRODUCTION
■
In 2010, we reported that red light irradiation of an anthracene-
containing squaraine rotaxane (SR) in aerated organic solution
our understanding of the structural factors that control SREP
led to quantitative formation of a squaraine rotaxane
endoperoxide (SREP) (Scheme 1).1 SREPs can be stored
reactivity.
The limited structural data in 2010 led us to conclude that
the endoperoxide unit common to all SREPs was directed
indefinitely at temperatures below −20 °C, but upon warming
to body temperature they undergo a chemiluminescent
endoperoxide cycloreversion reaction that regenerates the
parent SR. Furthermore, the color of the emitted light can be
inside the surrounding macrocycle (i.e., the SREP-int stereo-
isomer in Scheme 1a).8 But new evidence reported here
demonstrates that the initial kinetic product formed by oxygen
controlled over the range of green to near-infrared by simply
changing the structure of the encapsulated squaraine dye.2
Mechanistic studies have shown that the thermal cycloreversion
process releases molecular oxygen, primarily in its singlet
cycloaddition to SR is the less stable SREP-ext stereoisomer.
The photophysical properties and subsequent reactivity of the
mechanically strained SREP-ext depend on the size of the end
groups attached to the encapsulated squaraine dye. If the end
excited state, and the light emission appears to be mediated by
energy transfer to the encapsulated squaraine dye.2 The unique
groups are large enough to prevent dissociation of the
interlocked components, the strained SREP-ext stereoisomer
undergoes the thermal chemiluminescent cycloreversion
reaction in Scheme 1b. However, if the squaraine end groups
combination of near-infrared fluorescence and chemilumines-
cence makes SREPs especially attractive as storable dyes for
optical imaging applications in living subjects.3,4 An ongoing
research goal is to enhance the chemiluminescence intensity,
and we have investigated different strategies to increase
rotaxane mechanical bond strain and accelerate the rate of
Received: November 19, 2013
Published: January 15, 2014
© 2014 American Chemical Society
1120
dx.doi.org/10.1021/jo402564k | J. Org. Chem. 2014, 79, 1120−1130

The Journal of Organic Chemistry
Article
the production of 1EP. The structure of squaraine 2 is
endowed with several important features that facilitate the
cyclic synthetic sequence in Scheme 2. The two internal, H-
Scheme 1
Scheme 2. Recyclable Templated Synthesis of
Monoendoperoxide Macrocycle 1EP
are small enough to allow transient dissociation, the result is a
pseudorotaxane dissociation/association process that produces
SREP-int as a thermodynamically more stable stereoisomer
that does not undergo cycloreversion and does not emit light
(Scheme 1c). The insight provided by this discovery has
permitted us to develop new synthetic chemistry that adds
benzyne to the exterior surface of the surrounding anthracene-
containing macrocycle in SR to produce new squaraine
rotaxane architectures with triptycene-containing macrocycles
and enhanced fluorescence emission properties.
bonding hydroxyl groups on the squaraine core improve dye
stability and also reduce dye affinity for the macrocyclic
cavity.11 In addition, the N-ethyl-N-nonylacetylene groups at
each end of squaraine 2 are large enough to effectively block
dissociation of the monoendoperoxide pseudorotaxane product,
3EP-ext (structure elucidation of 3EP-ext is described below),
that is formed quantitatively by irradiating 3 in aerated
chloroform solution.12 Previous studies have shown that
squaraine pseudorotaxane association constant is greatly
reduced in polar aprotic organic solvents such as acetone.11
Thus, forming 3EP-ext in chloroform and then dissolving it in
acetone at 5 °C leads to slow but quantitative dethreading over
a 12 h period and subsequent isolation of both components
(squaraine 2 and macrocycle 1EP) in good yield using column
chromatography. As expected for a 9,10-dialkylanthracene-9,10-
endoperoxide, the empty macrocycle 1EP does not undergo
cycloreversion at room temperature but instead slowly
decomposes through rearrangement pathways upon standing
for a long time (t1/2 ∼ 10 days at 22 °C).13 Variable-
RESULTS AND DISCUSSION
■
Isolation and Structure of Macrocycle Endoperoxide
1EP. The surrounding macrocycle in a generic SREP is the
tetralactam monoendoperoxide EP. To assess the effect of
mechanical bond strain on EP reactivity, the free mono-
endoperoxide macrocycle 1EP was prepared and its properties
were characterized. We have previously reported that exposure
of the parent tetralactam macrocycle 1 to singlet oxygen rapidly
produces the corresponding bis(anthracene-9,10-endoperox-
ide) adduct with no measurable accumulation of monoendoper-
oxide 1EP.9 Thus, in order to isolate 1EP, a stepwise
templation process was developed that first produced a SREP
by photooxygenation and then removed the internal squaraine
dye.10 After some experimentation, it was determined that
squaraine dye 2 serves as an excellent recyclable template for
1
temperature H NMR studies of pure 1EP show no peak
splitting down to −80 °C.
A sample of pure macrocycle 1EP was crystallized from a
mixed organic solvent system that included THF. The X-ray
crystal structure in Figure 1 shows the endoperoxide group
directed into the macrocycle cavity, which also contains a THF
1121
dx.doi.org/10.1021/jo402564k | J. Org. Chem. 2014, 79, 1120−1130

The Journal of Organic Chemistry
Article
Figure 1. Side and top views of the X-ray structure of 1EP·THF with
the endoperoxide oxygen atoms colored green for clarity.
molecule that is held by bifurcated hydrogen bonds with two
amide NH residues. The structure is consistent with the
molecular modeling results described below, indicating that the
internal endoperoxide conformation of 1EP (1EP-int) is more
stable than the alternative external endoperoxide conformation
(1EP-ext). The same internal endoperoxide stereochemistry
was found in the X-ray crystal structure of a closely related
bis(anthracene-9,10-endoperoxide) adduct.9
Discovery of SREP-int and SREP-ext Stereoisomers.
Since our initial report in 2010,1 we have confirmed repeatedly
that photooxygenation of permanently interlocked versions of
the SR structure with very large squaraine end groups produces
the corresponding SREP quantitatively. Moreover, the
subsequent thermal cycloreversion cleanly releases molecular
oxygen (mostly in the singlet excited state) and emits light.7
Studies of various homologues show that this reversible oxygen
capture and release cycle occurs equally well with rotaxane
structures that encapsulate symmetric or unsymmetric squar-
aine dyes1−6
An important point with most of the SR structures in this
study is the relatively small size of at least one of the two
squaraine end groups (with exceptions discussed later). This
enabled pseudorotaxane formation by mixing millimolar
concentrations of the appropriate squaraine dye with macro-
cycle 1 in chloroform solution. In all cases, the yield of
pseudorotaxane was essentially quantitative, which was
expected since previous studies have shown that the association
constant is around 2 × 105 M−1.14 As indicated in the section
above, conversion of a SR to the corresponding SREP-ext
constricts the surrounding macrocycle. As a result, the steric
barrier for pseudorotaxane association/dissociation becomes
more sensitive to the size of the squaraine end groups. In
particular, larger squaraine end groups slow the rates of
association/dissociation. A manifestation of this squaraine end
group size effect became apparent when we compared the
reactivity of two closely related unsymmetric squaraine
pseudorotaxanes, 4 and 5. The structure of 4 (Figure 2) has
a comparatively large cyclohexamethyleneimine group at one
end of the encapsulated squaraine. As shown by the changes in
1H NMR spectra, irradiation of an aerated solution of 4 (Figure
2a) with red light for about 1 h at 0 °C produced the
corresponding endoperoxide 4EP-ext (Figure 2b, the structural
elucidation is described below), which subsequently underwent
clean cycloreversion at 38 °C (t1/2 = 5.2 h) to regenerate 4
(Figure 2c). Thus, the relatively large size of the cyclo-
1
Figure 2. Chemical structures with atom assignments and H NMR
spectra in CDCl3 and 22 °C showing: (a) starting sample of 4, (b)
4EP-ext formed by quantitative photooxygenation of 4, (c)
regenerated 4 due to cycloreversion of 4EP-ext. The red dotted
lines are provided to guide the eye.
hexamethyleneimine end group inhibits pseudorotaxane
dissociation sufficiently that 4EP-ext behaves like a perma-
nently interlocked structure. The intercomponent mechanical
bond strain selectively accelerates the endoperoxide cyclo-
reversion reaction (Scheme 1b) so that it is preferred over the
alternative endoperoxide rearrangement pathways that are
typically observed with 9,10-dialkyl anthracene-9,10-endoper-
oxides.13
Shown in Figure 3 are chemical structures and NMR spectra
for the analogous photooxygenation of pseudorotaxane 5
whose structure differs from 4 by only having a slightly smaller
piperidine group at one end of the encapsulated squaraine. As
1
indicated by the changes in H NMR spectra, irradiation of 5
(Figure 3a) with red light at 0 °C cleanly produced 5EP-ext
(Figure 3b, the structural elucidation is described below). But
in contrast to the behavior of 4EP-ext, analogue 5EP-ext
subsequently underwent a spontaneous isomerization process
over 3 h at 0 °C to form 5EP-int as a thermodynamically more
stable isomer (Figure 3c). Upon further standing at 38 °C,
5EP-int did not cyclorevert back to the starting pseudorotaxane
5 but instead decomposed very slowly over several weeks (see
the Supporting Information, section B). Thus, the smaller
piperidine end group permitted a low barrier dissociation/
association process to occur that converted the kinetic 5EP-ext
stereoisomer to the thermodynamic 5EP-int stereoisomer. The
stepwise reaction sequence in Figure 3 is a specific example of
the generalized process shown in Scheme 1c and is supported
by the following set of independent evidence.
1122
dx.doi.org/10.1021/jo402564k | J. Org. Chem. 2014, 79, 1120−1130

The Journal of Organic Chemistry
Article
Scheme 3. Squaraine Dyes That Were Tested for Inclusion
Inside Empty Macrocycle Endoperoxide 1EP To Form
SREP-int
squaraine end groups is highlighted by the different outcomes
obtained with squaraines 11 and 13, where a switch from N-
methyl to N-ethyl in the squaraine end groups is enough to
completely block dye inclusion within the cavity of macrocycle
1EP.15,16
1
It is worth noting that a broad comparison of H NMR
spectra for all of the different SREP compounds in this study
revealed that the chemical shift of proton C in the bridging
isophthalamide unit of the surrounding macrocycle is highly
diagnostic of the SREP stereochemistry. Compared to the
precursor SR, the signal for proton C is downfield in SREP-int
and upfield in SREP-ext (Table 1).
1
Figure 3. Chemical structures with atom assignments and H NMR
spectra of the same sample in CDCl3 and 22 °C showing: (a) starting
squaraine rotaxane 5, (b) the initial kinetic product, 5EP-ext, formed
by quantitative photooxygenation of 5, (c) the thermodynamic
product 5EP-int formed after allowing the sample of 5EP-ext to sit
for 3 h at 0 °C. The red dotted lines are provided to guide the eye.
Blocking Stereoisomerization by Converting a Pseu-
dorotaxane into a Rotaxane. The pseudorotaxane dissoci-
ation/association process in Scheme 1 predicts that SREP
stereoisomerization can be blocked by attaching large stopper
Independent Formation of SREP-int. Strong evidence
that the H NMR spectrum in Figure 3c corresponds to 5EP-
1
int was gained by demonstrating that the identical NMR
spectrum is produced when a sample of empty macrocycle
endoperoxide 1EP in CDCl3 is mixed with 1 molar equiv of the
unsymmetric squaraine dye 8. If the thermodynamically favored
isomer in Figure 3c is 5EP-int, then it is logical to conclude that
the precursor kinetically favored isomer in Figure 3b is 5EP-ext.
Table 1. Chemical Shift of Surrounding Macrocycle Proton
a
C.
encapsulated squaraine
dye
SREP-ext
δ (ppm)
SR
δ (ppm)
SREP-int
δ (ppm)
1
b
Furthermore, the H NMR spectra for 5EP-ext and 4EP-ext
6
9.38
9.37
9.37
9.23
9.37
9.38
9.37
9.39
9.39
9.54 (0.16)
9.54 (0.17)
9.53 (0.16)
9.40 (0.17)
9.57 (0.20)
(Figure 2b) are highly homologous.
7
9.24 (−0.14)
Additional 1H NMR experiments combined separate samples
of empty macrocycle 1EP with the different symmetric and
unsymmetric squaraine dyes, 6-11, shown in Scheme 3. These
squaraines have at least one relatively small squaraine end
group, and in each case, the corresponding pseudorotaxane
(SREP-int) was formed in quantitative yield and did not
undergo a cycloreversion reaction at 38 °C. In contrast,
squaraines 12 and 13, with slightly larger end groups, did not
form a pseudorotaxane, even when mixed with 10 molar equiv
of 1EP. The extraordinary sensitivity to the size of the
8
9.24 (−0.13)
b
9
10
11
12
13
19
9.21 (−0.16)
9.26 (−0.12)
9.28 (−0.09)
9.31 (−0.08)
9.26 (−0.13)
9.59 (0.21)
b
b
9.62 (0.23)
a
All signals were obtained at 22 °C and referenced to CHCl3 at 7.27
b
ppm. Not measured.
1123
dx.doi.org/10.1021/jo402564k | J. Org. Chem. 2014, 79, 1120−1130

The Journal of Organic Chemistry
Article
groups to the ends of the enapsulated squaraine dye. This
hypothesis was confirmed by comparing separate samples of
structurally related pseudorotaxane monoendoperoxide 14EP
and permanently interlocked rotaxane monoendoperoxide
15EP. The squaraine end group in pseudorotaxane 14 includes
an N-methyl group that is small enough to allow transient
dissociation of the pseudorotaxane after conversion to the
monoendoperoxide 14EP-ext.16 As expected, red light
irradiation of an aerated CDCl3 solution of pseudorotaxane
14 generated 14EP-ext in quantitative yield (Figure 4). This
cycloreverted to the parent rotaxane 15 (Figure 5) with a half-
life of 5 h at 38 °C.17
1
Figure 5. Chemical structures with atom assignments and H NMR
spectra in CDCl3 and 22 °C showing: (a) starting sample of rotaxane
15, (b) 15EP-ext formed by quantitative photooxygenation of 15, (c)
regenerated 15 due to cycloreversion of 15EP-ext over 11 h at 38 °C.
The red dotted lines are provided to guide the eye.
Benzyne Cycloaddition to External Face of Anthra-
cene Squaraine Rotaxane. The concept that dienophiles can
add to the external face of the anthracene units in the SR
structure was tested by conducting a cycloaddition reaction
using the highly reactive dienophile benzyne.18,19 Squaraine
rotaxane 16 was selected because the squaraine end groups
were unlikely to undergo side reactions with the excess benzyne
that was generated by decomposing benzenediazonium-2-
carboxylate. As summarized in Scheme 4, the benzyne
cycloaddition reaction produced two major products, single
addition adduct 17 in 57% isolated yield and the double
addition adduct 18 in 4% isolated yield. Not only do these
synthetic results strongly support the notion that dienophiles
(such as singlet oxygen) can add to the exterior surface of the
SR structure, the cycloaddition products 17 and 18 are
promising new examples of triptycene-containing SR architec-
tures with potentially useful spectral properties.20 Compared to
precursor 16, the triptycene versions 17 and 18 exhibit 40%
higher fluorescence quantum yields (Table 2). In addition, the
Stokes shifts for 17 and 18 are >40 nm, which is unusually large
for squaraines and near-infrared dyes in general.21,22
1
Figure 4. Chemical structures with atom assignments and H NMR
spectra of the same sample in CDCl3 and 22 °C showing: (a) starting
squaraine rotaxane 14, (b) the initial kinetic product, 14EP-ext,
formed by quantitative photooxygenation of 14, (c) the thermody-
namic product 14EP-int formed after allowing the sample of 14EP-ext
to sit for 3 h at 22 °C. The red dotted lines are provided to guide the
eye.
kinetic product subsequently isomerized over 3 h at 22 °C to
become the thermodynamically favored 14EP-int which did
not undergo any measurable endoperoxide cycloreversion. In
contrast, red light irradiation of the permanently interlocked
rotaxane 15 (prepared by clicking large stopper groups to the
ends of pseudorotaxane 14) produced the expected mono-
endoperoxide 15EP-ext which did not isomerize but instead
Molecular Modeling. A series of molecular modeling
studies compared the conformations and energetics of key
structures in Scheme 1. A central question is the relative
stability of the two different SREP stereoisomers. Our original
study included a modest set of calculations suggesting that the
1124
dx.doi.org/10.1021/jo402564k | J. Org. Chem. 2014, 79, 1120−1130

The Journal of Organic Chemistry
Article
a
Scheme 4. Cycloaddition of Benzyne and Squaraine
Rotaxane 16
Scheme 5. Model SREP System Used for DFT Calculations
a
The lowest energy conformation of SREP-int is 4.0 kcal/mol more
stable than the lowest energy conformation of SREP-ext.
Table 2. Spectral Data
16
657
697
5.11
0.48
17
655
696
5.10
18
652
700
4.89
λabs (nm)
λem (nm)
log ε (M−1 cm−1
Φf
Molecular modeling of the empty endoperoxide macrocycle
1EP determined that the lowest energy 1EP-int conformation
is 5.5 kcal/mol more stable than the alternative 1EP-ext
conformation and that conversion of 1EP-ext to 1EP-int occurs
spontaneously at room temperature (see section I in the
Supporting Information). This suggests dissociation of kinetic
isomer SREP-ext, followed by rapid switching of the free
macrocycle conformation 1EP-ext to low energy 1EP-int, and
subsequent association to form thermodynamic isomer SREP-
int as one possible mechanism for the SREP stereoisomerism
sequence in Scheme 1c, although other variations of this theme
are conceivable. For example, the SREP-ext dissociation step
could be concerted with the macrocycle conformational switch
to 1EP-int.
We then evaluated how strain in the anthracene-9,10-
endoperoxide section of macrocycle 1EP was affected by the
presence of the encapsulated squaraine dye, particularly in
SREP-ext. Scheme 6 shows overlays of the analogous
anthracene-9,10-endoperoxide moieties from ext and int
conformers of 1EP and the corresponding SREP stereoisomers.
These results show that the anthracene-9,10-endoperoxide
sections from 1EP-ext and SREP-ext (top) are both noticeably
flatter than the overlapped sections from 1EP-int and SREP-int
(bottom), reflecting the increased strain and deviation from
ideal bond angles for the ext stereoisomers. The relative strain
was analyzed by comparing the dihedral angle (φ) formed by
the planes of the two aryl rings in the anthracene-9,10-
endoperoxide moieties. In the case of the two int stereoisomers,
the φ values of 124.1° for SREP-int and 120.2° for 1EP are
quite similar and correlate with a relatively unstrained
a
a
)
b
b
0.62
0.67
a
b
Taken from ref 5. Quantum yield measurements ( 5%) were
determined in CHCl3 using 4,4-[bis(N,N- dimethylamino)phenyl]-
squaraine dye as a reference (Φf = 0.70 in CHCl3)23
SREP-int stereoisomer was more stable than SREP-ext.1 A
more thorough series of calculations, employing density
functional theory (DFT), confirms this conclusion. Specifically,
extensive DFT calculations using the model SREP system
shown in Scheme 5 indicate that the lowest energy
conformation of SREP-int is 4.0 kcal/mol more stable than
the lowest energy conformation of SREP-ext. In other words,
the kinetic product of SR photooxygenation, SREP-ext, is the
less stable SREP stereoisomer, and isomerization via a
pseudorotaxane dissociation/association equilibrium strongly
favors SREP-int as the thermodynamic product. The SREP
structure in our 2010 publication exhibited NMR peak splitting
at −50 °C which, at the time, was attributed (incorrectly) to a
macrocycle rocking motion by a putative SREP-int structure.1
We have subsequently found that some SREP compounds do
not exhibit any peak splitting down to −80 °C; therefore, the
dynamic NMR behavior exhibited by the original SREP system
is not a general phenomenon and cannot be used as a
diagnostic indicator of SREP stereoisomeric structure. For a
more detailed discussion of the dynamic NMR behavior of
SREP-ext, see section D in the Supporting Information.
1125
dx.doi.org/10.1021/jo402564k | J. Org. Chem. 2014, 79, 1120−1130

The Journal of Organic Chemistry
Article
Another notable finding of this study is that SREPs with
relatively small squaraine end groups undergo spontaneous
stereoisomerization from SREP-ext to SREP-int. The isomer-
ization mechanism is a pseudorotaxane dissociation/association
process that enables the initial kinetic photooxidation product,
SREP-ext, to alleviate mechanical bond strain by converting to
a thermodynamically more stable pseudorotaxane stereoisomer,
SREP-int, that does not cyclorevert (Scheme 1c). The
literature on interlocked molecules includes structural examples
with functional groups forced to adopt high-energy con-
formations.29 The SREP system extends this concept to
functional group reactivity. The large difference in endoper-
oxide reactivity for the two SREP stereoisomers illustrates the
power of a mechanical bond to induce cross-component steric
strain and selective enhancement of a specific reaction
pathway.30
Scheme 6. Overlapped Pictures of the anthracene-9,10-
endoperoxide Sections Excised from the Calculated Low
Energy Conformations of Empty Macrocycle 1EP
a
Conformer and Corresponding SREP Stereoisomer
EXPERIMENTAL SECTION
■
Compounds 1,14 3,11 6,20 9,31 13,14 and 165 have been previously
reported, and spectra of the samples used in this study are provided in
the Supporting Information. The synthesis and characterization of
anilines 20, 21, and 22 are detailed below. Anilines 23 and 24 are
commercially available and were used as received. The synthesis and
characterization of anilines 25,32 26,14 and 2711 have been published
elsewhere and are not included here. Structures of some
aforementioned compounds are provided in Chart 1.
a
In each case, the value (φ) corresponds to the dihedral angle formed
by planes of the two aryl rings.
anthracene-9,10-endoperoxide system.13,24 With the two ext
stereoisomers, the φ value of 136.3° for 1EP reflects moderate
strain of the anthracene-9,10-endoperoxide section, and the φ
value of 152.4° for SREP-ext indicates significantly higher
strain induced by the encapsulated squaraine dye.25 These
results are consistent with the finding that the endoperoxide
carbon−oxygen bonds are elongated by 0.2 Å in SREP-ext
compared to SREP-int. These computational results help
rationalize the experimental data. The SREP-ext stereoisomer
undergoes a relatively facile cycloreversion reaction because the
anthracene-9,10-endoperoxide is relatively strained due, in part,
to cross-component steric interactions with the encapsulated
squaraine dye. In contrast, the anthracene-9,10-endoperoxide in
SREP-int stereoisomer is relatively unstrained and cyclo-
reversion is not the lowest activation energy reaction pathway.
Chart 1. Synthetic Precursors
CONCLUSIONS
■
General Procedure for the Synthesis of Unsymmetrical
Squaraines 7, 8, and 12. Squaraine precursor 3-(4-(dibenzylamino)-
phenyl)-4-hydroxycyclobut-3-ene-1,2-dione (0.30 mmol, prepared
using literature methods13) was dissolved in anhydrous 2-propanol
(30 mL) under an atmosphere of argon. The reaction flask was
charged with a solution of the appropriate aniline (0.30 mmol) in
anhydrous 2-propanol (15 mL). Drying agent, tri-n-butyl orthoformate
(1.5 mL), and anhydrous benzene (60 mL) were added, and the
reaction was refluxed for 16 h with a Dean−Stark apparatus.
Concentration under reduced pressure afforded crude material that
was purified by silica gel column chromatography to give the desired
squaraine derivative as a blue crystalline solid in the yields presented
below.
Photooxygenation of permanently interlocked squaraine
rotaxanes (SR) with anthracene-containing macrocycles and
very large squaraine stopper groups produces the correspond-
ing monoendoperoxide SREP-ext in quantitative yield (Scheme
1). The photochemical process involves cycloaddition of
photosensitized singlet-state molecular oxygen to the sterically
accessible exterior surface of one of the macrocycle’s
anthracene units.26,27 Although mechanically strained, samples
of rotaxane SREP-ext can be stored indefinitely at low
temperature, and upon warming to body temperature they
undergo a clean thermal cycloreversion that releases singlet
oxygen and emits light.28 Structural elucidation of SREP-ext as
the kinetic product of SR photooxygenation is a revision of
earlier publications that incorrectly assigned the stereoisomer as
SREP-int.1−6 The realization that dienophiles can be added to
the exterior surface of anthracene-containing squaraine
rotaxanes to form the SREP-ext stereoisomer led us to
synthesize two new squaraine rotaxane structures with
triptycene-containing macrocycles (17 and 18) and potentially
useful fluorescence emission properties.
Data for Squaraine 7. A 20:80 to 30:70 (v/v) ethyl acetate/
chloroform eluent was used to obtain pure 7 in 18% yield (27 mg,
1
0.054 mmol): mp 180−185 °C dec; H NMR (600 MHz, CDCl3) δ
8.40 (d, J = 8.8 Hz, 2H), 8.33 (d, J = 8.8 Hz, 2H), 7.35−7.38 (m, 4H),
7.29−7.32 (m, 2H), 7.20−7.21 (m, 4H), 6.85 (d, J = 9.1 Hz, 2H), 6.63
(d, J = 9.1 Hz, 2H), 4.77 (s, 4H), 3.50−3.53 (m, 4H), 2.08−2.10 (m,
4H); 13C NMR (125 MHz, CDCl3) δ 190.2, 187.2, 154.3, 153.2,
136.2, 134.0, 132.9, 129.0, 127.6, 126.4, 120.8, 119.8, 113.3, 112.9,
54.0, 48.3, 25.2; HRMS (ESI-TOF) calcd for C34H31N2O2 [M + H]
499.2380, found 499.2373.
1126
dx.doi.org/10.1021/jo402564k | J. Org. Chem. 2014, 79, 1120−1130

The Journal of Organic Chemistry
Article
Data for Squaraine 8. A 10:90 to 20:80 (v/v) ethyl acetate/
chloroform eluent was used to obtain pure 8 in 14% yield (22 mg,
0.042 mmol): mp 128−133 °C dec; H NMR (600 MHz, CDCl3) δ
8.39 (d, J = 9.1 Hz, 2H), 8.36 (d, J = 9.1 Hz, 2H), 7.35−7.39 (m, 4H),
7.29−7.33 (m, 2H), 7.20−7.22 (m, 4H), 6.89 (d, J = 9.7 Hz, 2H), 6.88
(d, J = 9.4 Hz, 2H), 4.79 (s, 4H), 3.59−3.61 (m, 4H), 1.69−1.77 (m,
6H); 13C NMR (125 MHz, CDCl3) δ 190.4, 188.6, 183.4, 155.3,
154.8, 136.3, 134.1, 133.3, 129.3, 127.9, 126.7, 121.1, 120.2, 113.5,
113.2, 54.2, 48.5, 26.0, 24.5; HRMS (ESI-TOF) calcd for C35H33N2O2
[M + H] 513.2537, found 513.2543.
7.33 (m, 4H), 7.08 (d, J = 9.2 Hz, 2H), 6.86 (d, J = 9.2 Hz, 2H), 6.72
(dd, J = 3.2 Hz, J = 7.0 Hz, 4H), 6.53 (dd, J = 3.0 Hz, J = 6.8 Hz, 4H),
6.39 (d, J = 9.4 Hz, 2H), 5.93 (d, J = 9.4 Hz, 2H), 5.39 (dd, J = 5.6 Hz,
J = 14.8 Hz, 4H), 5.04 (dd, J = 2.5 Hz, J = 14.6 Hz, 4H), 4.88 (s, 4H),
3.56 (t, J = 5.6 Hz, 4H), 1.82−1.88 (m, 4H), 1.68−1.71 (m, 4H), 1.52
(s, 18H); 13C NMR (125 MHz, CDCl3) δ 184.1, 181.2, 178.8, 167.3,
154.3, 153.9, 153.0, 136.5, 133.5, 133.3, 133.2, 130.7, 130.6, 129.4,
129.2, 128.8, 128.3, 126.6, 126.2, 125.8, 124.4, 123.9, 122.7, 118.8,
116.5, 112.3, 111.8, 60.6, 55.2, 50.5, 35.6, 31.6, 27.6, 26.7; HRMS
(ESI-TOF) calcd for C92H86N6O6Na [M + Na] 1393.6501, found
1393.6526.
1
Data for Squaraine 12. A 10:90 (v/v) ethyl acetate/chloroform
eluent was used to obtain pure 12 in (31% yield, 49 mg, 0.093 mmol):
mp 172−176 °C dec; 1H NMR (500 MHz, CDCl3) δ 8.40 (d, J = 9.4
Hz, 2H), 8.35 (d, J = 9.2 Hz, 2H), 7.35−7.39 (m, 4H), 7.29−7.33 (m,
2H), 7.20−7.23 (m, 4H), 6.87 (d, J = 9.4 Hz, 2H), 6.79 (d, J = 9.3 Hz,
2H), 4.79 (s, 4H), 3.65 (t, J = 6.0 Hz, 4H), 1.82−1.88 (m, 4H), 1.58−
1.62 (m, 4H); 13C NMR (125 MHz, CDCl3) δ 190.3, 187.3, 183.5,
155.0, 154.5, 136.4, 134.2, 133.1, 129.2, 127.9, 126.7, 121.1, 119.9,
113.1, 112.6, 54.2, 50.6, 27.3, 26.7; HRMS (ESI-TOF) calculated for
C36H35N2O2 [M + H] 527.2693; found 527.2686.
1
Data for pseudorotaxane 5: H NMR (600 MHz, CDCl3) δ 9.37
(t, J = 1.4 Hz, 2H), 8.52 (d, J = 1.5 Hz, 4H), 8.22 (dd, J = 2.8 Hz, J =
5.1 Hz, 4H), 7.74 (dd, J = 3.2 Hz, J = 6.7 Hz, 4H), 7.72 (dd, J = 3.2
Hz, J = 6.7 Hz, 4H), 7.47−7.50 (m, 4H), 7.39−7.42 (m, 2H), 7.30−
7.33 (m, 4H), 7.09 (d, J = 9.1 Hz, 2H), 6.85 (d, J = 9.1 Hz, 2H), 6.73
(dd, J = 2.9 Hz, J = 7.0 Hz, 4H), 6.53 (dd, J = 3.2 Hz, J = 7.0 Hz, 4H),
6.40 (d, J = 9.1 Hz, 2H), 6.01 (d, J = 9.4 Hz, 2H), 5.39 (dd, J = 5.6 Hz,
J = 14.7 Hz, 4H), 5.07 (dd, J = 2.9 Hz, J = 14.9 Hz, 4H), 4.88 (s, 4H),
3.52 (t, J = 5.6 Hz, 4H), 1.83−1.88 (m, 2H), 1.73−1.78 (m, 4H), 1.52
(s, 18H); 13C NMR (125 MHz, CDCl3) δ184.1, 181.1, 179.5, 154.3,
154.1, 153.1, 136.4, 133.4, 133.2, 130.7, 130.6, 129.5, 129.2, 128.8,
128.4, 126.6, 126.2, 126.0, 124.4, 123.9, 122.7, 118.8, 116.6, 112.6,
112.3, 55.2, 48.6, 38.1, 35.6, 31.6, 26.6, 24.7; HRMS (ESI-TOF) calcd
for C91H84N6O6Na [M + Na] 1379.6345, found 1379.6325.
Data for squaraine 9: 1H NMR (500 MHz, CDCl3) δ 8.62 (d, J =
9.0 Hz, 4H), 7.10 (d, J = 9.0 Hz, 4H), 4.00 (s, 6H); 13C NMR was not
acquired due to the very poor solubility and stability of this squaraine;
HRMS (ESI-TOF) calcd for C18H15O4 [M + H] 295.0965, found
295.0992.
Data for pseudorotaxane 14: 1H NMR (500 MHz, CDCl3) δ 9.39
(t, J = 1.4 Hz, 2H), 8.55 (d, J = 1.2 Hz, 4H), 8.27 (t, J = 4.0 Hz, 4H),
7.77 (dd, J = 3.4 Hz, J = 6.8 Hz, 8H), 7.04 (d, J = 9.2 Hz, 4H), 6.68
(dd, J = 3.2 Hz, J = 7.0 Hz, 8H), 6.10 (d, J = 9.4 Hz, 4H), 5.26 (d, J =
4.2 Hz, 8H), 3.47 (t, J = 7.6 Hz, 4H), 3.14 (s, 6H), 2.19 (td, J = 2.6
Hz, J = 7.0 Hz, 4H), 1.94 (t, J = 2.6 Hz, 2H), 1.71 (quint, J = 7.2 Hz,
4H), 1.55 (s, 18H), 1.32−1.44 (m, 24H); 13C NMR (150 MHz,
CDCl3) δ 184.3, 179.7, 167.3, 154.0, 153.1, 133.5, 133.0, 130.7, 129.2,
128.8, 126.0, 124.2, 122.8, 117.3, 111.6, 84.9, 68.4, 53.0, 39.2, 38.2,
35.6, 31.7, 29.8, 29.7, 29.3, 28.9, 28.6, 27.8, 27.3, 18.6; HRMS (ESI-
TOF) calcd for C96H104N6O6Na [M + Na] 1459.7910, found
1459.7932.
General Procedure for the Synthesis of Symmetrical
Squaraines 10, 11, and 19. Squaric acid (0.30 mmol) was
suspended in anhydrous n-butanol (30 mL), and the reaction flask
was charged with a solution of the appropriate aniline (0.60 mmol) in
anhydrous 2-propanol (15 mL). Anhydrous benzene (60 mL) was
added, and the reaction was refluxed for 16 h with a Dean−Stark
apparatus under an atmosphere of argon. Concentration under
reduced pressure afforded crude material that was purified by column
chromatography to give the desired squaraine derivatives as blue
crystalline solids.
Data for Squaraine 10. A 30:70 (v/v) ethyl acetate/chloroform
eluent was used to obtain pure 10 in 13% yield (15.6 mg, 0.0389
mmol): mp 210−216 °C dec; 1H NMR (600 MHz, CDCl3) δ 8.38 (d,
J = 9.1 Hz, 4H), 6.90 (d, J = 9.1 Hz, 4H), 3.57−3.60 (m, 8H), 1.69−
1.75 (m, 12H); 13C NMR (125 MHz, CDCl3) δ 188.7, 183.7, 155.0,
133.6, 120.4, 113.5, 48.5, 25.9, 24.6; HRMS (ESI-TOF) calcd for
C26H29N2O2 [M + H] 401.2224, found 401.2236.
General Procedures for the Preparation of Squaraine
Pseudorotaxane Endoperoxides. Method A: Monoendoperoxide
macrocycle 1EP was mixed with separate samples of squaraines 6−11
in CDCl3 at 22 °C, quantitatively generating SREP-int in <5 min as
1
confirmed by H NMR. All endoperoxide pseudorotaxane complexes
were isolated as green amorphous solids. Method B: An aerated
solution of squaraine pseudorotaxane 3, 4, 5, or 14 (1−3 mM in
CDCl3) was cooled to 0 °C and irradiated with >540 nm filtered light
(150 W xenon lamp) for 1.0−1.5 h to give the corresponding SREP-
Data for Squaraine 11. A 20:70 (v/v) ethyl acetate/chloroform to
a 10:90 (v/v) methanol/chloroform eluent was used to obtain pure 11
1
in 3.5% yield (18 mg, 0.039 mmol): mp 210−213 °C dec; H NMR
(500 MHz, CDCl3) δ 8.41 (d, J = 9.4 Hz, 4H), 6.79 (d, J = 9.4 Hz,
4H), 4.17 (d, J = 2.4 Hz, 4H), 3.75−3.79 (m, 8H), 3.23 (s, 6H), 2.44
(t, J = 2.4 Hz, 2H);13C NMR (125 MHz, CDCl3) δ 189.5, 183.6,
154.6, 133.5, 120.3, 112.7, 79.3, 75.2, 67.5, 58.8, 52.5, 40.1; HRMS
(ESI-TOF) calcd for C28H29N2O4 [M + H] 457.2122, found 457.2147.
Data for Squaraine 19. A 10:90 to 20:80 (v/v) ethyl acetate/
chloroform eluent was used to obtain pure 19 in 33% yield (86.8 mg,
1
ext in quantitative yield as judged by H NMR. The sample must be
stored ≤0 °C to avoid either, cycloreversion back to the precursor
squaraine pseudorotaxane in the case of 3EP-ext or 4EP-ext, or
stereoisomerization to the more stable SREP-int in the case of 5EP-
ext or 14EP-ext. All endoperoxide pseudorotaxane complexes were
isolated as green amorphous solids.
1
1
Data for pseudorotaxane endoperoxide 3EP-ext: H NMR (500
0.146 mmol): mp 178−186 °C dec; H NMR (500 MHz, CDCl3) δ
MHz, CDCl3) δ 10.02 (s, 2H), 9.34 (s, 2H), 8.59 (s, 2H), 8.47 (s,
2H), 8.23 (d, J = 6.2 Hz, 2H), 8.11 (d, J = 9.0 Hz, 2H), 7.61 (d, J = 8.8
Hz, 2H), 7.35 (d, J = 7.6 Hz, 2H), 7.16 (d, J = 4.6 Hz, 2H), 7.13 (dd, J
= 6.6 Hz, J = 8.4 Hz, 2H), 6.96 (d, J = 7.4 Hz, 2H), 6.92 (t, J = 7.4 Hz,
2H), 6.79 (d, J = 9.4 Hz, 2H), 6.63 (dd, J = 6.6 Hz, J = 8.8 Hz, 2H),
6.22 (t, J = 7.4 Hz, 2H), 5.62 (d, J = 7.2 Hz, 2H), 5.59 (dd, J = 6.4 Hz,
J = 15.4 Hz, 2H), 5.54 (s, 2H), 5.18 (d, J = 15.0 Hz, 2H), 4.45 (dd, J =
6.4 Hz, J = 13.8 Hz, 2H), 4.03 (dd, J = 1.2 Hz, J = 13.4 Hz, 2H), 3.43−
3.53 (m, 4H), 3.33−3.43 (m, 4H), 2.20 (td, J = 2.6 Hz, J = 7.0 Hz,
4H), 1.94 (t, J = 2.8 Hz, 2H), 1.66−1.75 (m, 4H), 1.54 (s, 18H),
1.28−1.45 (m, 24H). 13C NMR (150 MHz, CDCl3, 0 °C) δ 182.4,
167.9, 167.5, 163.1, 155.8, 153.0, 135.9, 134.5, 134.0, 133.3, 133.2,
131.2, 131.1, 130.8, 130.4, 129.7, 129.3, 128.4, 126.9, 125.2, 123.8,
123.6, 122.6, 122.2, 121.3, 106.8, 100.2, 85.0, 81.6, 68.4, 51.2, 46.2,
39.3, 35.6, 32.1, 31.6, 29.9, 29.8, 29.7, 29.6, 28.9, 28.6, 27.2, 22.9, 18.6,
14.4; HRMS (ESI-TOF) calcd for C98H108N6O10 [M + ] 1528.8121,
found 1528.8112.
8.38 (d, J = 9.2 Hz, 4H), 6.75 (d, J = 9.3 Hz, 4H), 3.48 (t, J = 7.6 Hz,
4H), 3.16 (s, 6H), 2.19 (td, J = 2.7 Hz, J = 7.1 Hz, 4H), 3.48 (t, J = 2.7
Hz, 2H), 1.63−1.69 (m, 4H), 1.53 (pent, J = 7.0 Hz, 4H), 1.37−1.44
(m, 4H), 1.28−1.37 (m, 16H); 13C NMR (125 MHz, CDCl3) 188.3,
183.7, 154.4, 133.5, 119.9, 112.4, 84.9, 68.3, 53.1, 39.1, 29.6, 29.5, 29.2,
28.9, 28.6, 27.5, 27.2, 18.6; HRMS (ESI-TOF) calcd for C40H53N2O2
[M + H] 593.4102, found 593.4082.
General Procedure for the Preparation of Squaraine
Pseudorotaxanes. A squaraine dye (6−11) and macrocycle 1 (1
molar equiv) were dissolved in chloroform (1−5 mL) at room
temperature, generating a pseudorotaxane complex quantitatively in
under 5 min at 0.5−3.5 mM concentrations. All pseudorotaxane
complexes were isolated as green amorphous solids.
1
Data for pseudorotaxane 4: H NMR (500 MHz, CDCl3) δ 9.37
(t, J = 1.2 Hz, 2H), 8.52 (d, J = 1.9 Hz, 4H), 8.24 (dd, J = 2.8 Hz, J =
5.4 Hz, 4H), 7.74 (dd, J = 3.2 Hz, J = 6.8 Hz, 4H), 7.72 (dd, J = 3.4
Hz, J = 7.0 Hz, 4H), 7.47−7.51 (m, 4H), 7.38−7.42 (m, 2H), 7.30−
1127
dx.doi.org/10.1021/jo402564k | J. Org. Chem. 2014, 79, 1120−1130

The Journal of Organic Chemistry
Article
1
Data for pseudorotaxane endoperoxide 4EP-ext: H NMR (600
MHz, CDCl3, 0 °C) δ 9.26 (s, 2H), 8.55−8.58 (m, 4H), 8.43 (s, 2H),
7.86 (dd, J = 3.2 Hz, J = 6.8 Hz, 2H), 7.76 (dd, J = 3.2 Hz, J = 7.0 Hz,
2H), 7.45 (t, J = 7.6 Hz, 4H), 7.38 (t, J = 7.3 Hz, 2H), 7.34 (d, J = 9.1
Hz, 2H), 7.28−7.32 (m, 6H), 7.17 (dd, J = 3.5 Hz, J = 5.0 Hz, 2H),
7.03 (dd, J = 3.8 Hz, J = 4.7 Hz, 2H), 6.98 (d, J = 9.1 Hz, 2H), 6.93
(dd, J = 2.9 Hz, J = 7.0 Hz, 2H), 6.60 (d, J = 9.1 Hz, 2H), 6.48 (dd, J =
2.9 Hz, J = 7.0 Hz, 2H), 6.39 (dd, J = 2.9 Hz, J = 5.6 Hz, 2H), 6.33
(dd, J = 2.9 Hz, J = 5.1 Hz, 2H), 5.82 (d, J = 9.4 Hz, 2H), 5.74 (dd, J =
6.8 Hz, J = 15.0 Hz, 2H), 5.01 (d, J = 15.0 Hz, 2H), 4.93 (s, 4H), 4.48
(dd, J = 6.7 Hz, J = 13.8 Hz, 2H), 4.03 (d, J = 13.8 Hz, 2H), 3.63−3.71
(m, 2H), 3.35−3.44 (m, 2H), 1.70−1.92 (m, 8H), 1.50 (s, 18H); 13C
NMR (150 MHz, CDCl3, 0 °C) δ 184.2, 180.1, 176.7, 168.0, 167.8,
153.9, 153.8, 152.8, 136.3, 135.9, 135.3, 134.2, 134.1, 133.8, 133.1,
130.8, 130.6, 130.3, 130.1, 130.0, 129.8, 129.4, 128.9, 128.2, 126.4,
126.3, 124.8, 124.7, 123.7, 122.8, 122.2, 121.6, 120.4, 115.7, 112.4,
111.8, 81.2, 55.2, 50.2, 38.7, 37.0, 35.4, 31.5, 27.6, 26.7; HRMS (ESI-
TOF) calcd for C92H87N6O8 [M + H] 1403.6580, found 1403.6587.
(11.4 mg, 26.3 μmol) in chloroform (5 mL). N,N-Diisopropylethyl-
amine (2 drops) and tris[(1-benzyl-1H-1,2,3-triazol-4-yl)methyl]-
amine copper(I)bromide (2.5 mg, 3.7 μmol) were added, and the
reaction was stirred at rt for 16 h. The reaction was washed with
saturated aqueous EDTA solution (5 mL), and the organic layer was
collected and evaporated under reduced pressure. The crude product
that was purified using silica gel column chromatography, using a
20:80 to 50:50 (v/v) ethyl acetate/chloroform eluent gradient to
obtain pure 15 (84% yield, 13.0 mg, 5.64 μmol) as a green amorphous
solid: 1H NMR (600 MHz, CDCl3) δ 9.39 (t, J = 1.7 Hz, 2H), 8.55 (d,
J = 0.9 Hz, 4H), 8.27 (t, J = 4.4 Hz, 4H), 7.77 (dd, J = 3.5 Hz, J = 7.1
Hz, 8H), 7.22−7.27 (m,16H), 7.16−7.22 (m, 16H), 7.09 (d, J = 8.8
Hz, 4H), 7.03 (d, J = 9.4 Hz, 4H), 6.74 (d, J = 8.8 Hz, 4H), 6.67 (dd, J
= 3.2 Hz, J = 7.0 Hz, 8H), 6.09 (d, J = 9.4 Hz, 4H), 5.25 (d, J = 3.8 Hz,
8H), 4.38 (t, J = 7.3 Hz, 4H), 3.94 (t, J = 5.8 Hz, 4H), 3.46 (t, J = 7.9
Hz, 4H), 3.13 (s, 6H), 2.70 (t, J = 7.9 Hz, 4H), 2.08 (pent, J = 7.4 Hz,
4H), 1.77 (pent, J = 5.8 Hz, 4H), 1.65−1.73 (m, 4H), 1.54 (s, 18H),
1.37−1.44 (m, 20H); 13C NMR (150 MHz, CDCl3) δ 184.3, 179.6,
167.3, 156.8, 154.0, 153.1, 148.6, 147.2, 139.3, 133.5, 133.0, 132.4,
131.3, 130.6, 129.2, 128.8, 127.6, 126.1, 126.0, 124.2, 122.8, 120.7,
117.3, 113.3, 111.6, 66.9, 64.5, 53.0, 50.0, 39.2, 38.2, 35.6, 31.7, 29.9,
29.8, 29.7, 29.6, 29.5, 29.4, 27.8, 27.5, 27.2, 26.5, 25.9; HRMS (ESI-
TOF) calcd for C154H159N12O8 [M2+] 1152.1196, found 1152.1183.
1
Data for pseudorotaxane endoperoxide 5EP-ext: H NMR (600
MHz, CDCl3) δ 9.24 (s, 2H), 8.62 (d, J = 6.4 Hz, 2H), 8.59 (t, J = 1.6
Hz, 2H), 8.45 (t, J = 1.8 Hz, 2H), 7.87 (dd, J = 3.2 Hz, J = 6.7 Hz,
2H), 7.80 (dd, J = 3.2 Hz, J = 6.8 Hz, 2H), 7.43−7.46 (m, 4H), 7.41
(d, J = 9.1 Hz, 2H), 7.37 (t, J = 7.8 Hz, 2H), 7.30−7.32 (m, 4H), 7.21
(d, J = 7.3 Hz, 2H), 7.18 (dd, J = 3.2 Hz, J = 5.6 Hz, 2H), 7.16 (dd, J =
2.3 Hz, J = 6.2 Hz, 2H), 6.99−7.03 (m, 4H), 6.94 (dd, J = 2.9 Hz, J =
7.0 Hz, 2H), 6.63 (d, J = 9.4 Hz, 2H), 6.50 (dd, J = 2.9 Hz, J = 6.8 Hz,
2H), 6.39−6.41 (m, 4H), 5.87 (d, J = 9.4 Hz, 2H), 5.77 (dd, J = 7.1
Hz, J = 15.0 Hz, 2H), 5.04 (dd, J = 1.2 Hz, J = 14.6 Hz, 2H), 4.92 (s,
4H), 4.47 (dd, J = 6.5 Hz, J = 13.5 Hz, 2H), 4.04 (dd, J = 2.0 Hz, J =
13.5 Hz, 2H), 3.49 (t, J = 6.5 Hz, 4H), 1.82−1.86 (m, 2H), 1.74−1.78
(m. 4H), 1.51 (s, 18H).
1
Squaraine rotaxane endoperoxide 15EP-ext: H NMR (600
MHz, CDCl3) δ 9.25 (s, 2H), 8.66 (t, J = 4.1 Hz, 2H), 8.61 (t, J = 1.7
Hz, 2H), 8.47 (t, J = 1.7 Hz, 2H), 7.87 (dd, J = 3.2 Hz, J = 7.0 Hz,
4H), 7.22−7.28 (m,18H), 7.16−7.20 (m, 18H), 7.12 (dd, J = 3.2 Hz, J
= 5.6 Hz, 4H), 7.09 (d, J = 8.8 Hz, 4H), 6.76 (dd, J = 2.9 Hz, J = 7.0
Hz, 4H), 6.73 (d, J = 9.1 Hz, 4H), 6.43 (dd, J = 2.9 Hz, J = 5.6 Hz,
4H), 6.13 (d, J = 9.1 Hz, 4H), 5.42 (d, J = 4.1 Hz, 4H), 4.37 (t, J = 7.0
Hz, 4H), 4.25 (d, J = 4.1 Hz, 4H), 3.93 (t, J = 5.9 Hz, 4H), 3.45 (t, J =
7.9 Hz, 4H), 3.13 (s, 6H), 2.68 (t, J = 7.6 Hz, 4H), 2.07 (pent, J = 7.6
Hz, 4H), 1.77 (pent, J = 5.9 Hz, 4H), 1.62−1.71 (m, 8H), 1.52 (s,
18H), 1.38−1.42 (m, 8H), 1.32−1.37 (m, 12H); 13C NMR (150
MHz, CDCl3, 0 °C) δ 184.5, 178.4, 168.1, 167.8, 156.7, 153.7, 152.8,
148.6, 147.0, 139.2, 135.7, 134.0, 133.9, 133.2, 132.2, 131.2, 130.8,
130.6, 130.2, 129.9, 129.0, 127.6, 126.0, 125.5, 124.4, 122.8, 121.9,
120.7, 117.8, 113.1, 111.6, 81.2, 66.8, 64.3, 53.0, 50.0, 39.2, 38.7, 37.1,
35.5, 31.5, 29.9, 29.7, 29.6, 29.5, 29.4, 27.8, 27.5, 27.2, 26.3, 25.8, 22.9;
HRMS (ESI-TOF) calcd for C154H159N12O10 [M + H] 2336.2297,
found 2336.2312.
Monoendoperoxide Macrocycle 1EP. Endoperoxide pseudor-
otaxane 3EP (19 mg, 12 μmol) was dissolved in acetone (20 mL) and
allowed to dethread over 12 h at 5 °C. The resulting blue solution was
evaporated under reduced pressure and purified by silica gel column
chromatography. A 10:90 (v/v) ethyl acetate/chloroform eluent was
used to obtain 1EP (91% yield, 9.7 mg, 11 μmol) as a white solid. On
occasion, unoxidized macrocycle 1 contaminates the sample, which
can be removed via selective templation of 1 with squaraine 13 (1EP
does not complex 13). The resulting pseudorotaxane can be separated
from 1EP by silica gel column chromatography using a 20:20:60 (v/v/
v) ethyl acetate/hexanes/chloroform eluent to provide purified 1EP as
an amorphous white solid: 1H NMR (500 MHz, CDCl3) δ 8.41 (t, J =
2.2 Hz, 2H), 8.39 (dd, J = 3.4 Hz, J = 7.1 Hz, 4H), 8.34 (t, J = 1.8 Hz,
2H), 7.59 (dd, J = 3.2 Hz, J = 7.0 Hz, 4H), 7.37 (dd, J = 3.4 Hz, J = 5.6
Hz, 4H), 7.28 (t, J = 1.4 Hz, 2H), 7.20 (dd, J = 3.2 Hz, J = 5.6 Hz,
4H), 6.40 (t, J = 5.6 Hz, 2H), 6.36 (t, J = 4.0 Hz, 2H), 5.68 (d, J = 4.4
Hz, 4H), 4.81 (d, J = 5.8 Hz, 4H), 1.45 (s, 18H); 13C NMR (150
MHz, CDCl3) δ 167.2, 166.8, 153.7, 138.5, 133.8, 132.8, 130.6, 130.4,
130.0, 129.6, 128.3, 127.0, 124.9, 121.6, 119.1, 81.3, 38.6, 37.5, 31.4,
29.9; HRMS (ESI-TOF) calcd for C56H52N4O6Na [M + Na]
899.3761, found 899.3779.
1
Data for endoperoxide pseudorotaxane 5EP-int: H NMR (600
MHz, CDCl3) δ 9.53 (s, 2H), 8.74 (dd, J = 3.5 Hz, J = 6.5 Hz, 2H),
8.59 (t, J = 1.7 Hz, 2H), 8.45 (t, J = 1.7 Hz, 2H), 8.10 (dd, J = 3.8 Hz, J
= 8.2 Hz, 2H), 7.95 (dd, J = 3.8 Hz, J = 8.3 Hz, 2H), 7.75 (dd, J = 5.3
Hz, J = 9.4 Hz, 2H), 7.65 (d, J = 11.0 Hz, 2H), 7.41 (t, J = 7.7 Hz,
4H), 7.32−7.37 (m, 2H), 7.16−7.22 (m, 6H), 7.06 (dd, J = 3.2 Hz, J =
5.6 Hz, 2H), 6.92 (d, J = 7.3 Hz, 2H), 6.85 (dd, J = 2.9 Hz, J = 5.6 Hz,
2H), 6.79 (dd, J = 3.2 Hz, J = 5.6 Hz, 2H), 6.56 (dd, J = 2.9 Hz, J = 5.6
Hz, 2H), 6.38 (d, J = 9.7 Hz, 2H), 5.78 (dd, J = 7.0 Hz, J = 14.9 Hz,
2H), 5.67 (d, J = 9.1 Hz, 2H), 5.32 (d, J = 14.6 Hz, 2H), 5.10 (dd, J =
8.5 Hz, J = 15.2 Hz, 2H), 4.70 (s, 4H), 4.15 (d, J = 15.2 Hz, 2H), 3.31
(t, J = 5.6 Hz, 4H), 1.70−1.77 (m, 4H), 1.63−1.66 (m, 2H), 1.47 (s,
18H); 13C NMR (150 MHz, CDCl3) δ 186.1, 167.7, 166.4, 154.3,
154.1, 152.9, 139.6, 138.4, 136.3, 133.5, 133.3, 133.0, 132.9, 131.0,
130.8, 129.9, 129.5, 129.3, 129.2, 128.1, 127.3, 126.9, 126.8, 126.7,
126.3, 125.9, 125.0, 124.3, 122.2, 121.6, 121.3, 120.7, 112.5, 112.4,
80.6, 54.3, 48.1, 38.3, 36.9, 35.5, 31.6, 31.4, 26.3; HRMS (ESI-TOF)
calcd for C91H84N6O8K [M + K] 1427.5988, found 1428.5989.
Data for endoperoxide pseudorotaxane 14EP-ext: 1H NMR (600
MHz, CDCl3) δ 9.26 (t, J = 1.4 Hz, 2H), 8.67 (t, J = 4.4 Hz, 2H), 8.62
(t, J = 1.6 Hz, 2H), 8.48 (t, J = 1.5 Hz, 2H), 7.88 (dd, J = 3.2 Hz, J =
6.7 Hz, 4H), 7.22 (t, J = 4.6 Hz, 2H), 7.13 (dd, J = 3.2 Hz, J = 5.4 Hz,
4H), 6.77 (dd, J = 3.1 Hz, J = 6.9 Hz, 4H), 6.43 (dd, J = 2.9 Hz, J = 5.6
Hz, 4H), 6.14 (d, J = 9.4 Hz, 4H), 5.43 (d, J = 4.1 Hz, 4H), 4.28 (d, J
= 4.1 Hz, 4H), 3.47 (t, J = 7.7 Hz, 4H), 3.15 (s, 6H), 2.18 (td, J = 2.6
Hz, J = 7.3 Hz, 4H), 1.95 (t, J = 2.6 Hz, 2H), 1.68−1.73 (m, 4H),
1.62−1.68 (m, 4H), 1.53 (s, 18H), 1.30−1.38 (m, 20H).
Data for endoperoxide pseudorotaxane 14EP-int: 1H NMR (500
MHz, CDCl3) δ 9.62 (t, J = 1.6 Hz, 2H), 8.80 (t, J = 5.0 Hz, 2H), 8.64
(t, J = 1.8 Hz, 2H), 8.48 (t, J = 1.7 Hz, 2H), 8.05 (dd, J = 3.4 Hz, J =
7.0 Hz, 4H), 7.82 (t, J = 5.8 Hz, 2H), 7.46 (d, J = 8.8 Hz, 4H), 6.94
(dd, J = 3.2 Hz, J = 5.6 Hz, 4H), 6.89 (dd, J = 3.2 Hz, J = 7.0 Hz, 4H),
6.71 (dd, J = 3.1 Hz, J = 5.6 Hz, 4H), 5.92 (d, J = 8.4 Hz, 4H), 5.56 (d,
J = 4.4 Hz, 4H), 4.62 (br. s, 4H), 3.31 (t, J = 7.6 Hz, 4H), 2.99 (s, 6H),
2.19 (td, J = 2.6 Hz, J = 7.2 Hz, 4H), 1.94 (t, J = 2.6 Hz, 2H), 1.47−
1.56 (m, 8H), 1.52 (s, 18H), 1.30−1.38 (m, 20H); HRMS (ESI-TOF)
calcd for C96H105N6O8 [M + H] 1470.8021, found 1470.8037.
Squaraine Rotaxane 15. Pseudorotaxane 14 (9.6 mg, 6.7 μmol)
was combined with 1-(4-azidobutoxy)-4-(triphenylmethyl)benzene
Procedure for the Preparation of Triptycene Squaraine
Rotaxanes 17 and 18. Anthranilic acid (515 mg, 3.76 mmol) and
trichloroacetic acid (10 mg, 0.061 mmol) were dissolved in anhydrous
THF (20 mL) under a dry atmosphere of argon. The reaction was
cooled to −5 °C, and isopentyl nitrite (0.82 mL, 0.72 g, 6.1 mmol)
was added over 10 min. The mixture was stirred for 45 min at 0 °C
and then allowed to warm to room temperature for 1 h (a blast shield
was used during this sequence). The benzenediazonium-2-carboxylate
1128
dx.doi.org/10.1021/jo402564k | J. Org. Chem. 2014, 79, 1120−1130

The Journal of Organic Chemistry
Article
precipitate was collected via filtration using minimal suction and
washed with cold THF, making sure not to let the filtrate go dry
(Safety warning: this diazonium salt is potentially shock sensitive and
may violently detonate when dry). The wet filtrate was immediately
transferred to a solution of rotaxane 16 (23.3 mg, 15.9 μmol) in
dichloroethane (25 mL), and the mixture was heated at 40 °C for 15
min. After the reaction mixture was cooled to room temperature, the
solution was evaporated under reduced pressure and purified via silica
gel column chromatography using a 10:20:70 (v/v/v) ethyl acetate/
hexanes/chloroform eluent to provide a mixture of cycloadducts 17
and 18. Rotaxanes 17 (57% yield, 14 mg, 9.0 μmol) and 18 (4% yield,
1.1 mg, 0.68 μmol) were separated using multiple preparative TLC
purifications with a 10:10:80 (v/v/v) ethyl acetate/hexanes/chloro-
form solvent system. They were isolated as green amorphous solids.
3.58 mmol), and potassium carbonate (4.10 g, 28.9 mmol) in
acetonitrile was refluxed for 18 h. The resulting mixture was cooled to
room temperature and filtered over Celite. The filtrate was collected,
dried over MgSO4, and evaporated under reduced pressure to yield
crude material that was purified by silica gel column chromatography.
A 50:50 (v/v) chloroform/hexanes eluent was used to obtain pure 22
(38% yield, 253 mg, 0.982 mmol) as a clear, viscous oil: 1H NMR (600
MHz, CDCl3) δ 7.29−7.33 (m, 2H), 7.75−7.79 (m, 3H), 3.38 (t, J =
7.7 Hz, 2H), 3.00 (s, 3H), 2.27 (td, J = 2.6 Hz, J = 7.0 Hz, 2H), 2.02
(t, J = 2.6 Hz, 2H), 1.66 (pent, J = 7.6 Hz, 2H), 1.61 (pent, J = 7.6 Hz,
2H), 1.46−1.51 (m, 2H), 1.37−1.42 (m, 8H); 13C NMR (150 MHz,
CDCl3) δ 149.4, 129.2, 115.9, 112.2, 84.8, 68.3, 52.9, 38.4, 29.6, 29.2,
28.8, 28.6, 27.3, 26.8, 18.5; HRMS (ESI-TOF) calcd for C18H28N [M
+ H] 258.2216, found 258.2205.
Computational Methodology. Density functional theory (DFT)
calculations were performed using Gaussian 09 (see the Supporting
Information for the complete reference). All structures were optimized
without constraints at the M06/6-31G* level of theory. Thermal
corrections were calculated at the same level of theory. All calculations
used a polarizable continuum model (PCM) with parameters for
chloroform to account for solvation effects. This level of theory
generally allowed calculations of our SREP model to optimize within 2
days, a practically acceptable duration. Single-point energies of the
optimized structures were calculated at the M06/6-311+G** level of
theory with the PCM model for chloroform and added to thermal
corrections to obtain free energy values. The crystal structure of
monoendoperoxide macrocycle 1EP and a previously published crystal
structure of rotaxane 16 were modified to provide the starting
structures for the DFT optimizations.5 Optimized structures for the
lowest energy conformations and background calculations are
provided in the Supporting Information.
1
Data for squaraine rotaxane 17: H NMR (600 MHz, CDCl3) δ
9.52 (t, J = 1.5 Hz, 2H), 8.62 (t, J = 3.5 Hz, 2H), 8.57 (t, J = 1.8 Hz,
2H), 8.51 (t, J = 1.8 Hz, 2H), 7.72 (dd, J = 3.2 Hz, J = 6.7 Hz, 4H),
7.45 (t, J = 3.8 Hz, 2H), 7.40−7.44 (m, 8H), 7.34−7.37 (m, 4H),
7.27−7.29 (m, 8H), 7.11 (d, J = 9.4 Hz, 4H), 6.95 (dd, J = 3.5 Hz, J =
5.9 Hz, 2H), 6.87 (dd, J = 3.2 Hz, J = 5.6 Hz, 4H), 6.70 (dd, J = 3.2
Hz, J = 5.8 Hz, 2H), 6.58 (dd, J = 2.9 Hz, J = 7.0 Hz, 4H), 6.30 (d, J =
9.1 Hz, 4H), 5.88 (dd, J = 2.9 Hz, J = 5.9 Hz, 4H), 5.34 (d, J = 3.8 Hz,
4H), 4.83 (s, 8H), 4.74 (d, J = 3.7 Hz, 4H), 1.50 (s, 18H); 13C NMR
(150 MHz, CDCl3) δ 183.6, 180.8, 168.9, 167.8, 154.5, 152.6, 149.5,
143.0, 136.2, 134.9, 133.9, 133.6, 130.8, 130.2, 129.8, 129.5, 128.7,
128.5, 128.0, 126.8, 126.7, 125.9, 124.5, 124.2, 123.1, 121.9, 119.7,
118.6, 114.1, 112.9, 55.2, 51.7, 35.5, 31.6; HRMS (ESI-TOF) calcd for
C106H93N6O6 [M + H] 1545.7151, found 1545.7121; λmax abs 655 nm;
λ
em 696 nm.
max
1
Data for squaraine rotaxane 18: H NMR (600 MHz, CDCl3) δ
10.1 (t, J = 1.7 Hz, 2H), 8.55 (d, J = 1.4 Hz, 4H), 7.68 (t, J = 4.1 Hz,
4H), 7.35−7.39 (m, 8H), 7.29−7.34 (m, 4H), 7.18 (d, J = 9.1 Hz,
4H), 6.83 (dd, J = 3.2 Hz, J = 5.6 Hz, 8H), 6.74 (dd, J = 2.9 Hz, J = 5.3
Hz, 4H), 6.54 (dd, J = 2.9 Hz, J = 5.6 Hz, 4H), 6.18 (d, J = 9.4 Hz,
4H), 5.95 (dd, J = 2.9 Hz, J = 5.9 Hz, 8H), 4.79 (s, 8H), 4.72 (d, J =
4.1 Hz, 8H), 1.48 (s, 18H); HRMS (ESI-TOF) calcd for
C112H96N6O6Na [M + Na] 1643.7284, found 1643.7287; λmax abs
652 nm; λmax em 700 nm.
Determination of SREP-int Stereoisomer Stability. The rate of
1
decomposition of SREP-int was monitored with H NMR spectros-
copy. A sealed sample of 5EP-int in CDCl3 was stored in the dark at
38 °C, and decomposition was monitored over 30 days. The sample
was doped with anisole, which was used as an internal standard.
Anisole is a stable, high boiling compound (bp = 154 °C), and
therefore, we assumed no internal standard was lost through
evaporation or decomposition. Decomposition kinetics were deter-
mined by integrating the bridging isophthalamide proton C of the
surrounding macrocycle relative to the anisole methyl signal. The 5EP-
int half-life was ∼350 h.
N-Phenylazepane (20). Iodobenzene (1.64 g, 8.04 mmol) and
hexamethyleneimine (0.875 g, 8.83 mmol) were combined in a dry
round-bottom flask containing dimethylethanolamine (13 mL), CuI
(210 mg, 1.10 mmol), and K3PO4 (4.49 g, 21.2 mmol). The mixture
was heated at 100 °C for 20 h under an atmosphere of argon, at which
point the reaction was quenched with water (150 mL) and the product
extracted with ether (3 × 50 mL). The ether was dried over MgSO4
and evaporated under reduced pressure to yield a crude material that
was purified by silica gel column chromatography. A 30:70 (v/v)
chloroform/hexanes eluent was used to obtain pure 20 (3.4% yield,
ASSOCIATED CONTENT
■
S
* Supporting Information
Spectral data, kinetic studies, X-ray structure details, and
expanded computational results. This material is available free
of charge via the Internet at http://pubs.acs.org.
1
48.8 mg, 0.278 mmol) as a clear viscous oil: H NMR (500 MHz,
CDCl3) δ 7.22−7.27 (m, 2H), 6.71−6.75 (m, 2H), 6.67 (tt, J = 1.0 Hz,
J = 7.2 Hz, 1H), 3.47−3.51 (m, 4H), 1.79−1.85 (m, 4H), 1.58 (quint,
J = 2.6 Hz, 4H); 13C NMR (125 MHz, CDCl3) δ 149.1, 129.4, 115.3,
111.3, 49.2, 28.0, 27.4.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: smith.115@nd.edu.
Notes
N-Methyl-N-(2-(prop-2-yn-1-yloxy)ethyl)aniline (21). A NaOH
(aq) solution was prepared by dissolving NaOH (18 g) in H2O (30
mL). Reactant 2-(methylphenylamino)ethanol (3.0 mL, 31 mmol) was
dissolved in toluene (40 mL) and slowly added over the basic aqueous
layer. Phase-transfer catalyst tetrabutylammonium bisulfate (90 mg,
0.12 mmol) and propargyl bromide (18 mL, 200 mmol) were added to
the toluene layer, and the reaction was gently stirred for 18 h. The
organic layer was isolated, dried over MgSO4, and evaporated under
reduced pressure, affording 21 as a light brown oil (73% yield, 4.27 g,
22.7 mmol) that was used without further purification: 1H NMR (500
MHz, CDCl3) δ 7.30−7.34 (m, 2H), 6.78−6.83 (m, 3H), 4.22 (d, J =
2.4 Hz, 2H), 3.77 (t, J = 6.1 Hz, 2H), 3.63 (d, J = 6.1 Hz, 2H), 3.06 (s,
3H), 2.50 (t, J = 2.4 Hz, 1H); 13C NMR (125 MHz, CDCl3) δ 149.2,
129.3, 116.5, 112.3, 79.8, 74.7, 67.5, 58.5, 52.4, 39.0; HRMS (ESI-
TOF) calcd for C12H16NO [M + H] 190.1226, found 190.1215.
N-Methyl-N-(undec-10-yn-1-yl)aniline (22). A mixture of 11-
bromo-1-undecyne (567 mg, 2.61 mmol), N-methylaniline (384 mg,
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the University of Notre Dame and
the NSF. We thank Dr. Per-Ola Norrby and the anonymous
peer reviewers of this article for their valuable comments.
REFERENCES
■
(1) Baumes, J. M.; Gassensmith, J. J.; Giblin, J.; Lee, J. J.; White, A.
G.; Culligan, W. J.; Leevy, W. M.; Kuno, M.; Smith, B. D. Nature
Chem. 2010, 2, 1025.
(2) Collins, C. G.; Baumes, J. M.; Smith, B. D. Chem. Commun. 2011,
47, 12352.
1129
dx.doi.org/10.1021/jo402564k | J. Org. Chem. 2014, 79, 1120−1130

The Journal of Organic Chemistry
Article
(3) Lee, J. J.; Goncalves, A.; Smith, B.; Palumbo, R.; White, A. G.;
Smith, B. D. Aust. J. Chem. 2011, 64, 604.
(4) Lee, J. J.; White, A. G.; Rice, D. R.; Smith, B. D. Chem. Commun.
(25) In response to a reviewer request, we calculated how the energy
of free 9,10-dimethylantharacene-9,10-endoperoxide changes with
dihedral angle. The results of the calculations are shown in Figure
S24 (Supporting Information) and indicate a quadratic dependency.
(26) For a recent mechanistic study of singlet oxygen cycloaddition
to anthracene, see: Klaper, M.; Linker, T. Angew. Chem., Int. Ed. 2013,
52, 11896.
(27) For examples of steric control of singlet oxygen cycloaddition in
synthetic organic chemistry, see: (a) Adam, W.; Bosio, S. G.; Turro, N.
J. J. Am. Chem. Soc. 2002, 124, 8814. (b) Adam, W.; Guthlein, M.;
Peters, E. M.; Peters, K.; Wirth, T. J. Am. Chem. Soc. 1998, 120, 4091.
(28) Zehm, D.; Fudickar, W.; Linker, T. Angew. Chem., Int. Ed. 2007,
46, 7689.
2013, 49, 3016.
(5) Baumes, J. M.; Murgu, I.; Oliver, A.; Smith, B. D. Org. Lett. 2010,
12, 4980.
(6) Baumes, J. M.; Murgu, I.; Connell, R. D.; Culligan, W. J.; Oliver,
A. G.; Smith, B. D. Supramol. Chem. 2012, 24, 14.
(7) For further discussion of the mechanism of self-illumination and
the hypothesis that the rate of SREP cycloreversion controls the
chemiluminescence emission intensity, see: Smith, B. D. Proc. SPIE
2011, 791013.
(8) For a systematic summary of rotaxane isomerism, see: Yerin, A.;
Wilks, E. S.; Moss, G. P.; Harada, A. Pure Appl. Chem. 2008, 80, 2041.
(9) Gassensmith, J. J.; Baumes, J. M.; Eberhard, J.; Smith, B. D. Chem.
Commun. 2009, 2517.
(10) Typically, the role of a molecular template in a templated
synthesis is to promote a specific reaction pathway, but in this case, the
purpose of the encapsulated squaraine template is to prevent an
undesired cycloaddition reaction of singlet oxygen with the second
anthracene unit in macrocycle 1.
(29) Leigh, D. A.; Lusby, P. J.; Slawin, A. M.; Walker, D. B. Angew.
Chem., Int. Ed. 2005, 44, 4557.
(30) For discussions of strain energy in enzymology, see: (a) Bugg,
T. D. H. Wiley Encyclopedia of Chemical Biology; Begley, T. P., Ed.;
Wiley: Hoboken, 2009; Vol. 1, p 653. (b) Hayashi, H.; Mizuguchi, H.;
Miyahara, I.; Islam, M. M.; Ikushiro, H.; Nakajima, Y.; Hirotsu, K.;
Kagamiyama, H. Biochim. Biophys. Acta 2003, 1647, 103. For examples
of small molecule strain and reactivity changes upon binding, see:
(c) Kraka, E.; Tuttle, T.; Cremer, D. Chem.Eur. J. 2007, 13, 9256.
(11) Collins, C. G.; Peck, E. M.; Kramer, P. J.; Smith, B. D. Chem. Sci.
2013, 4, 2557.
(12) If the encapsulated squaraine template has smaller end groups,
the SREP is susceptible to transient dethreading in chloroform
solution and undesired cycloaddition of another molecule of singlet
oxygen to the free macrocycle 1EP.
(d) Bahr, N.; Guller, R.; Reymond, J.-L.; Lerner, R. A. J. Am. Chem.
̈
Soc. 1996, 118, 3550.
(31) Farnum, D. G.; Webster, B.; Wolf, A. D. Tetrahedron Lett. 1968,
5003.
(32) Fang, Y.; Zheng, Y. Q.; Wang, Z. Y. Eur. J. Org. Chem. 2012,
1495.
(13) Aubry, J. M.; Pierlot, C.; Rigaudy, J.; Schmidt, R. Acc. Chem. Res.
2003, 36, 668.
(14) Gassensmith, J. J.; Arunkumar, E.; Barr, L.; Baumes, J. M.;
DiVittorio, K. M.; Johnson, J. R.; Noll, B. C.; Smith, B. D. J. Am. Chem.
Soc. 2007, 129, 15054.
(15) While squaraines 12 and 13 cannot enter the cavity of
macrocycle monoendoperoxide 1EP, they readily associate inside the
more spacious macrocycle 1.
(16) High sensitivity of pseudorotaxane threading/unthreading to the
size of the squaraine end groups has been observed in a related system
with a different macrocycle. X-ray crystal structures suggest that a
squaraine N-methyl group is constrained to remain within the
squaraine aromatic conjugation plane, whereas the terminal methyl
of an N-ethyl group protrudes out of the plane and creates significant
steric bulk Lai, C. C.; Liu, Y. H.; Wang, Y.; Peng, S. M.; Chiu, S. H.
Org. Lett. 2007, 9, 4523. Hsueh, S. Y.; Lai, C. C.; Liu, Y. H.; Peng, S.
M.; Chiu, S. H. Angew. Chem., Int. Ed. 2007, 46, 2013.
(17) It is worth noting that the same rate of cycloreversion was
obtained when rotaxane 15-ext was dissolved in acetone. In other
words, there was no evidence that the more polar solvent allowed the
oxidized macrocycle to alleviate mechanical bond strain by trans-
locating away from the central squaraine station to a sterically less
demanding location over one of the rotaxane’s two long alkyl chains.
Additional experiments using even more polar solvents such as DMSO
or DMF were not possible due to nucleophilic attack of the exposed
squaraine by the solvent.
(18) Klanderman, B. H.; Criswell, T. R. J. Org. Chem. 1969, 34, 3426.
(19) Masci, B.; Pasquale, S.; Thuery, P. Org. Lett. 2008, 10, 4835.
(20) Xue, M.; Su, Y. S.; Chen, C. F. Chem.Eur. J. 2010, 16, 8537.
(21) Luo, S. L.; Zhang, E. L.; Su, Y. P.; Cheng, T. M.; Shi, C. M.
Biomaterials 2011, 32, 7127.
(22) Hilderbrand, S. A.; Weissleder, R. Curr. Opin. Chem. Biol. 2010,
14, 71.
(23) Arun, K. T.; Ramaiah, D. J. Phys. Chem. A 2005, 109, 5571.
(24) For discussions of conventional intramolecular steric effects on
addition and release of oxygen to aromatic compounds, see:
(a) Fudickar, W.; Linker, T. Chem.Eur. J. 2006, 12, 9276.
(b) Wasserman, H. H.; Wiberg, K. B.; Larsen, D. L.; Parr, J. J. Org.
Chem. 2005, 70, 105. (c) Schmidt, R.; Brauer, H.-D. J. Photochem.
1986, 34, 1. (d) Lapouyade, R.; Desvergne, J. P.; Bouas-Laurent, H.
Bull. Soc. Chim. Fr. 1975, 9, 2137. (e) Castillo, A.; Greer, A. Struct.
Chem. 2009, 20, 399.
1130
dx.doi.org/10.1021/jo402564k | J. Org. Chem. 2014, 79, 1120−1130