Pillar[5]arene as a Co-factor in templating rotaxane formation

Article abstract of DOI:10.1021/ja407229h

After the manner in which coenzymes often participate in the binding of substrates in the active sites of enzymes, pillar[5]arene, a macrocycle containing five hydroquinone rings linked through their para positions by methylene bridges, modifies the binding properties of cucurbit[6]uril, such that the latter templates azide-alkyne cycloadditions that do not occur in the presence of only the cucurbit[6]uril, a macrocycle composed of six glycoluril residues doubly linked through their nitrogen atoms to each other by methylene groups. Here, we describe how a combination of pillar[5]arene and cucurbit[6]uril interacts cooperatively with bipyridinium dications substituted on their nitrogen atoms with 2-azidoethyl- to 5-azidopentyl moieties to afford, as a result of orthogonal templation, two [4]rotaxanes and one [5]rotaxane in >90% yields inside 2 h at 55 C in acetonitrile. Since the hydroxyl groups on pillar[5]arene and the carbonyl groups on cucurbit[6]uril form hydrogen bonds readily, these two macrocycles work together in a cooperative fashion to the extent that the four conformational isomers of pillar[5]arene can be trapped on the dumbbell components of the [4]rotaxanes. In the case of the [5]rotaxane, it is possible to isolate a compound containing two pillar[5]arene rings with local C₅ symmetries. In addition to fixing the stereochemistries of the pillar[5]arene rings, the regiochemistries associated with the 1,3-dipolar cycloadditions have been extended in their constitutional scope. Under mild conditions, orthogonal recognition motifs have been shown to lead to templation with positive cooperativity that is fast and all but quantitative, as well as being green and efficient.

Full text of DOI:10.1021/ja407229h

Subscriber access provided by DUESSELDORF LIBRARIES
Article
Pillar[5]arene as a Co-Factor in Templating Rotaxane Formation
Chenfeng Ke, Nathan L. Strutt, Hao Li, Xisen Hou, Karel J. Hartlieb, Paul R.
McGonigal, Zhidong Ma, Julien Iehl, Charlotte L. Stern, Chuyang Cheng, Zhixue Zhu,
Nicolaas A. Vermeulen, Thomas J. Meade, Youssry Y Botros, and J. Fraser Stoddart
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/ja407229h • Publication Date (Web): 23 Sep 2013
Downloaded from http://pubs.acs.org on September 26, 2013
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 47
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Pillar[5]arene as a Co-Factor in Templating Rotaxane Formation
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
†
†
†
†
†
†
Chenfeng Ke, Nathan L. Strutt, Hao Li, Xisen Hou, Karel J. Hartlieb, Paul R. McGonigal,
†
†
†
†
†
†
Zhidong Ma, Julien Iehl, Charlotte L. Stern, Chuyang Cheng, Zhixue Zhu, Nicolaas A. Vermeulen,
†
‡, ‖
,
†
Thomas J. Meade, Youssry Y. Botros, J. Fraser Stoddart*
† Department of Chemistry, Northwestern University, 2145 Sheridan Road, Evanston, IL 60208, United States
‡
Intel Labs, Building RNBꢀ6ꢀ61, 2200 Mission College Boulevard., Santa Clara, California 95054ꢀ1549,
United States
‖
National Center for Nano Technology Research, King Abdulaziz City for Science and Technology, P.O. Box
6
086, Riyadh 11442, Kingdom of Saudi Arabia
REVISED VERSION
*
Correspondence Address
Professor J Fraser Stoddart
Center for the Chemistry of Integrated Systems
Department of Chemistry
Northwestern University
2145 Sheridan Road, Evanston, IL 60208 (USA)
Fax: (+1)ꢀ847ꢀ491ꢀ1009
Email: stoddart@northwestern.edu
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 47
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ABSTRACT: After the manner in which coꢀenzymes often participate in the binding of
substrates in the active sites of enzymes, pillar[5]arene − a macrocycle containing five
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
hydroquinone rings linked through their para positions by methylene bridges − modifies the
binding properties of cucurbit[6]uril, such that the latter templates azideꢀalkyne cycloadditions
that do not occur in the presence of only the cucurbit[6]uril − a macrocycle comprised of six
glycoluril residues doubly linked through their nitrogen atoms to each other by methylene groups.
Here, we describe how a combination of pillar[5]arene and cucurbit[6]uril interacts cooperatively
with bipyridinium dications substituted on their nitrogen atoms with 2ꢀazidoethylꢀ to 5ꢀ
azidopentyl moieties to afford, as a result of orthogonal templation, two [4]rotaxanes and one
o
[
5]rotaxane in > 90% yields inside two hours at 55 C in acetonitrile. Since the hydroxyl groups
on pillar[5]arene and the carbonyl groups on cucurbit[6]uril form hydrogen bonds readily, these
two macrocycles work together in a cooperative fashion to the extent that the four
conformational isomers of pillar[5]arene can be trapped on the dumbbell components of the
[
4]rotaxanes. In the case of the [5]rotaxane, it is possible to isolate a compound containing two
pillar[5]arene rings with local C symmetries. In addition to fixing the stereochemistries of the
5
pillar[5]arene rings, the regiochemistries associated with the 1,3ꢀdipolar cycloadditions have
been extended in their constitutional scope. Under mild conditions, orthogonal recognition
motifs have been shown to lead to templation with positive cooperativity that is fast and all but
quantitative, as well as being green and efficient.
1. INTRODUCTION
2
ACS Paragon Plus Environment

Page 3 of 47
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Complex biological molecules and assemblies are, more often than not, constructed under the
guidance of nature’s catalysts − the enzymes – which exploit noncovalent bonding interactions to
recognize and bring together substrates with high specificities, lowering the transition state
energies for the chemical reactions they catalyze. In certain cases, nonꢀprotein molecules −
known as cofactors − are necessary to trigger the activity of the otherwise latent enzymic
machinery. Although the catalytic power of enzymes is rooted in a number of different
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
phenomena operating in concert, preorganization which orients the enzyme’s active sites and
substrates in space relative to each other is undoubtedly of paramount importance. Inspired by
nature, chemists have developed elegant supramolecular systems which exploit noncovalent
2
3
4
bonding interactions, e.g., hydrogen bonding, hydrophobic forces and van der Waals
5
interactions to control the relative positioning of substrates in receptors. These systems, which
act as simplified models of enzymes, can (i) accelerate reactions by increasing the effective
molarity and/or preorganizing reactants, (ii) direct substrates along reaction paths which they
would not otherwise follow, and (iii) discriminate between compounds bearing similar functional
1
groups as a consequence of selective binding. Many successful designs utilize macrocyclic
6
7
8
9
10
scaffolds such as cyclodextrins, crown ethers, metallocycles, calixarenes and cucurbiturils
as receptors to encapsulate substrates in their cavities, thus increasing the effective
concentrations, and so accelerating the reaction.
Among these macrocylic compounds, cucurbit[6]uril (CB) was first reported by Mock et
1
1
12
al. to act as a catalyst in the alkyneꢀazide 1,3ꢀdipolar cycloaddition (AAC) − the Cu(I)ꢀ
1
3
catalyzed variant of which has been popularized as the quintessential “click” reaction − giving
rise to 1,4ꢀdisubstituted triazoles regioselectively with a considerable acceleration of the reaction
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 47
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
5
rate, often ca. 10 ꢀfold compared to the uncatalyzed reaction. Since its discovery by Mock, the
CBꢀcatalyzed 1,3ꢀdipolar cycloaddition (CBꢀAAC) has found application in the construction of
1
4
15
polymers, mechanically interlocked molecules (MIMs) and pH responsive controlledꢀrelease
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
16
systems.
Despite the fact that both the CBꢀAAC and the Cu(I)ꢀcatalyzed cycloaddition (CuAAC)
exhibit favorable kinetics and regiospecificities, the former has found comparatively few
applications compared to the now ubiquitous CuAAC. Perhaps the fact that CB is not as freely
1
7
available as copper salts and is less convenient to handle because of its poor solubility in
1
8
19
water and other common laboratory solvents, add up to explanations of a sort. The greatest
drawback of the CBꢀAAC, however, is its substrate scope − to date, all reports are restricted to
describing reactions between propargylammonium and azidoethyl-ammonium derivatives. This
limited scope can be rationalized by considering the cyclization mechanism and noncovalent
bonding interactions which underpin the reaction. CBꢀAAC proceeds by means of the initial
formation of a heteroꢀternary complex, which renders the cyclization a pseudoꢀunimolecular
process as a result of bringing the triple bond in the propargylammonium ion and the azide
function in the azidoethylammonium ion into close proximity as well as aligning them so that
1
1a, 20
they are poised to undergo trizaole ring formation.
The entropic cost of bringing the CB and
2
1
AAC precursors together is compensated for by the favorable binding enthalpy and the release
of high energy water molecules from the cavity of the CB. Chargeꢀdipole and hydrogen bonding
interactions between the secondary dialkylammonium ions and the carbonyl groups around the
+
rim of the CB are maximized when the NH centers are close to the planes of carbonyl oxygens,
2
a requirement which dictates the geometry of the ternary complex. Consequently, altering the
4
ACS Paragon Plus Environment

Page 5 of 47
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
+
distance between the NH2 centers and the alkyne/azide groups in the substrates disturbs the
nearꢀperfect alignment in the ternary complex, a geometry which is necessary in order to to
2
2
lower the transition state energy of the AAC.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
23, 24
Recently, we developed a cooperative capture strategy
for the assembly of MIMs
which exploits the acceleration of the CBꢀAAC (Scheme 1a) in the presence either βꢀ
cyclodextrin (βꢀCD) or γꢀcyclodextrin (γꢀCD), resulting in rapid and quantitative formation of
[
4]rotaxanes and higher oligorotaxanes. Hydrogen bonding between the rims of two CB rings
2
5
and a CD torus gives rise to positive cooperativity in the multiꢀcomponent assembly, boosting
the rate of the CDꢀCBꢀAAC (Scheme 1a) on account of the ensemble’s greatly enhanced binding
affinity for functionalized secondary dialkylammonium guests. Based on these initial findings,
we envisaged that the observed positive cooperativity need not be restricted to CDs – other
macrocycles, which are capable of multivalent (also bifurcated) hydrogen bonding, and are
complementary in shape or size to CB, might also aid and abet the efficient synthesis of
oligorotaxanes.
Here, we report that the alkyneꢀazide 1,3ꢀdipolar cycloaddition (AAC) templated by
cucurbit[6]uril (CB) − call it (Scheme 1b) the PꢀCBꢀAAC reaction − is also promoted by
2
6
pillararenes − a family of macrocycles composed of five to ten hydroquinone rings linked
through their paraꢀpositions by methylene bridges − as well as discovering that this combination
of macrocycles is more accommodating to cationic substrates of different lengths, e.g., from 2ꢀ to
5
ꢀazidopentylbipyridinium units. It transpires that pillar[5]arene (P) alleviates the strict
conformational preference of the PꢀCBꢀAAC reaction by stabilizing the geometry in which the
2
+
27,
positively charged bipyridinium (BIPY ) units move away from the portals of the CB, acting
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 47
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
28
as a ‘molecular gasket’. In the same manner as a cofactor may be necessary to stabilize the
catalytically active geometry of an enzyme, P modifies the binding properties of CB such that it
will template the cycloaddition of substrates which are unreactive in the presence of CB alone.
When the substrate length is increased, then two P gaskets bridge the longer gaps between CB
rings, allowing both Pꢀcontaining [4]ꢀ and [5]rotaxanes to be obtained in high yields. The four
possible conformational isomers of the P rings that result from steric hindrance of the oxygenꢀ
throughꢀtheꢀannulus rotation of the hydroquinone rings are trapped as mixtures on the dumbbell
components of these rotaxanes.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2. RESULTS AND DISCUSSION
24, 25
23
In order to harness the positive cooperativity
observed in our preliminary investigations on
oligoꢀ and polyrotaxanes incorporating CB and CDs, it is useful to have a good understanding of
the distribution of charge on the rims of these two rings. Examination of the calculated
electrostatic potential map of CB reveals that the carbonyl oxygen atoms which adorn the
periphery (Figure 1a) support high electron densities. The hydroxyl groups encircling the rims of
βꢀCD (Figure 1b) are complementary, bearing as they do weak positive charges capable of acting
as hydrogen bond donors. Although the rims of CB and βꢀCD are not matched in size, we
2
3
discovered in our previous investigations that βꢀCD still forms sufficiently strong hydrogen
bonding networks with two CB rings to favor [4]rotaxane formation. Pillar[5]arene (P) presents
an array of five phenolic hydroxyl groups on each face; analysis of their electrostatic potential
map (Figure 1c) confirms that the hydroxyl groups on these macrocycles are more polarized than
the hydroxyl groups on βꢀCD (Figure 1b), suggesting that they should take part in stronger
2
9
hydrogen bonding interactions with CB. In addition, P is the most easily accessible and
6
ACS Paragon Plus Environment

Page 7 of 47
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
thermodynamically stable pillararene homologue. While it does not commute perfectly with the
sixꢀfold symmetry of CB, P is highly complementary in terms of rim size. For these reasons, it
3
0
appeared to be a promising candidate to enhance the CBꢀAAC. Accordingly, P was selected for
further investigation of the cooperative capture process during the synthesis of oligorotaxanes.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Synthesis of hetero[n]rotaxanes. On account of the different solubility profiles for CB and P,
we were unable to find a solvent combination to dissolve both rings simultaneously: although CB
can be induced into an aqueous solution in the presence of cations, P is insoluble in water. The
problem was overcome by preforming the organicꢀsoluble alkyne⊂CB precursor by an
31
“
inclusionꢀfollowedꢀbyꢀprecipitation” process. A 1:1 mixture of Nꢀ(3,5ꢀdimethoxybenzyl)ꢀ
propargylammonium chloride (1ꢁCl) with CB in aqueous solution was treated with NH PF , to
4
6
3
2
afford the ternary complex [1 ⊂CB]ꢁ[PF ] which was isolated by filtration. Alternatively, the
2
6 2
use of KPF as the counterion exchange reagent furnished the 1:1 complex [1⊂CB]ꢁ[PF ]
6
6
3
3
because of the weaker CBꢀpotassium compared to CBꢀammonium interaction. See SI for
details. These organic soluble complexes can then be employed as a convenient source of CB
3
4
and the alkyne precursors.
When contemplating the use of P in the cooperative capture strategy, we posited that
2+
26a,35
cationic BIPY derivatives CVꢁ2PF (Scheme 2), which are known
to have a strong binding
6
affinity for P in polar organic solvents (e.g., Me CO, MeCN, Me SO) would be more appropriate
2
2
3
6
2+
guests than previously employed azidoalkylammonium salts. As far as we are aware, BIPY
37
salts have not been used in the CBꢀAAC process. A test reaction of [1⊂CB]ꢁ[PF ] with
6
2CVꢁ2PF in MeCN afforded the [3]rotaxane 2C3Rꢁ4PF in an isolated yield of 30% and
6
6
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 47
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2+
confirms the fact that BIPY salts are viable substrates. See the SI for full details. The reaction
was sluggish, however, and failed to reach completion, even after 2 days at 55 °C.
An initial trial of the PꢀCBꢀAAC (Scheme 1b) was conducted by mixing the rodꢀlike
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
precursor 2CVꢁ2PF (1.0 equiv) and P (1.2 equiv) in MeCN before addition of the CBꢀalkyne
6
complex [1 ⊂CB]ꢁ[PF ] (2.2 equiv), which was followed by a change in color of the reaction
2
6 2
mixture from pale yellow to orange. Pleasingly, P brought about a considerable acceleration in
the reaction rate, and was incorporated into the product, 2C4R·4PF , confirming our hypothesis
6
that it can take part in the cooperative capture of rotaxanes. In three parallel experiments, carried
out at different temperatures and monitored by TLC, the reaction reached completion within 2
o
o
o
min (55 C), 40 min (20 C) and 2 h (−10 C), respectively. The [4]rotaxane 2C4Rꢁ4PF was
6
isolated (Table 1, entry 1, [1 ⊂CB]) by column chromatography in 60−70% yield. Although
2
satisfactory, the yield was somewhat lower than the cooperative capture mediated by CDs, which
2
3
affords [4]rotaxanes in up to 97%. The lowerꢀthanꢀexpected yield may be attributed to
competitive binding of the excess of the propargylammonium derivative present in the reaction
3
8
mixture. By simply substituting [1 ⊂CB]ꢁ[PF ] with the complex [1⊂CB]ꢁ[PF ], which has the
2
6 2
6
desired 1:1 CBꢀalkyne stoichiometry, the efficiency of the reaction was improved: all starting
1
materials were consumed, and H NMR spectroscopic analysis of the crude reaction mixture
indicated quantitative conversion to 2C4R·4PF . The isolated yield (Table 1, entry 1, [1⊂CB])
6
after column chromatography was 96%, indicating that the PꢀCBꢀAAC is as efficient as the CDꢀ
CBꢀAAC. It is worth noting that, in CD CN at room temperature, the stopper precursor
3
[
1⊂CB]ꢁ[PF ] disproportionates (Figure 2) overnight to afford the 2:1 complex [1 ⊂CB]ꢁ[PF ]
6
2
6 2
8
ACS Paragon Plus Environment

Page 9 of 47
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
and CB. This process is driven by the precipitation of CB from solution, pushing the equilibrium
towards [1 ⊂CB]ꢁ[PF ] .
2
6 2
Having established that the PꢀCBꢀAAC reaction tolerates an azidoethylbipyridinium
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2+
substrate, and that the reaction is accelerated by P, we sought to explore BIPY guests of
different lengths. In the absence of P, no triazole ring products were formed after mixing stopper
precursor [1⊂CB]ꢁ[PF ] (or [1 ⊂CB]ꢁ[PF ] ) with longer rod precursors 3CVꢁ2PF − 5CVꢁ2PF
6
2
6 2
6
6
1
4, 15, 16, 37
(Scheme 2). This observation is consistent with all previous reports
that the distance
between the alkyne/azide and the cation is critical for the supramolecular preorganization and,
consequently, templated by CB. In the presence of P, however, the reaction of [1⊂CB]ꢁ[PF ] with
6
3
CVꢁ2PF in MeCN afforded (Table 1, entry 5, [1⊂CB]) the hetero[4]rotaxane 3C4Rꢁ4PF
6
6
o
(
Scheme 2) in 95% isolated yield within 2 h at 55 C. P apparently enables the CBꢀAAC, which
3
9
does not take place in the presence of only CB or a CBꢀCD mixture, by stabilizing an otherwise
energetically unfavorable binding geometry of [1⊂CB]ꢁ[PF ] and 3CVꢁ2PF . This geometry
6
6
allows the azide group at the end of the outstretched propylene chain, to occupy the center of the
CB cavity alongside the alkyne and thus undergo cyclization. Tentatively, we suggest that the
2
+
2+
ring component of an initial 3CV ⊂P is relatively free to move along the vector of the BIPY
without paying a large energy penalty – that is, any increase in free energy as a result of
translation is compensated for by the ringꢀtoꢀring hydrogen bonding network with CB upon
quaternary complex formation. After installation of one triazole ring, the P moves to the other
2
+
end of the BIPY to stabilize the second PꢀCBꢀAAC reaction, installing the second stopper and
forging the mechanical bond of the [4]rotaxane.
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 47
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Upon lengthening the linear spacers yet further to butylene chains 4CVꢁ2PF , no
6
hetero[4]rotaxane was identified in the reaction containing stopper precusors, [1⊂CB]ꢁ[PF ] and
6
P. Instead, mother nature chooses (Scheme 2) a much more organized output − namely, a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
hetero[5]rotaxane 4C5Rꢁ4PF − which contains two P gaskets threaded onto the dumbbell. The
6
[
5]rotaxane 4C5Rꢁ4PF was isolated (Table 1, entry 7, [1⊂CB]) in 90% yield after reaction of
6
o
4CVꢁ2PF , [1 ⊂CB]ꢁ[PF ] and P in 1:2.5:2.5 ratio at 20 C for 1 h in MeCN solution. Decreasing
6
2
6
2+
the molar ratio of 4CVꢁ2PF and P (4CV :P = 1:1, 1:1.5, and 1:1.8) does not alter this outcome:
6
no [3]ꢀ or [4]rotaxane was observed and 4C5Rꢁ4PF was the sole interlocked product. This
6
3
,
5b 40
observation implies that the two adjacent P rings
on the [5]rotaxane 4C5Rꢁ4PF also
6
communicate with each other through the interꢀring hydrogen bonding, leading to increased
stability when compared to analogs with fewer rings and provides the lowest energy
cycloaddition pathway.
The pentylene derivative 5CVꢁ2PF gave rise to hetero[5]rotaxane 5C5Rꢁ4PF , although
6
6
the longer oligomethylene chain seems to impede the reaction, which requires 18 h to reach
completion, resulting (Table 1, entry 9, [1⊂CB]) in a substantially reduced isolated yield of 30%.
The decreased reaction rate, and the observation of the corresponding halfꢀreacted dumbbell
5HDꢁ3PF and monoꢀPꢀ[4]rotaxane analog (5C4Rꢁ4PF , see SI for details) as minor products,
6
6
suggest that the positive cooperativity has been curtailed, presumably because the four rings can
no longer bridge the length of the fully extended dumbbell effectively. These results are
consistent with a stepwise mechanism for the formation (Figure 3) of the [5]rotaxane 5C5R·4PF6:
after the first PꢀCBꢀAAC to introduce a stopper at one end of the dumbbell (i) the P ring may
2
+
shuttle along the BIPY subunit of the dumbbell to mediate the second PꢀCBꢀAAC and capture
1
0
ACS Paragon Plus Environment

Page 11 of 47
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
the [4]rotaxane 5C4Rꢁ4PF , (ii) the second PꢀCBꢀAAC can occur via an intermediate with two P
6
gaskets encircling the dumbbell, giving rise to the [5]rotaxane 5C5R·4PF with a hydrogen bond
6
network that spans all four rings, or (iii) if no further reaction occurs, the pseudorotaxane
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
intermediate dissociates during purification to yield the half dumbbell 5HDꢁ3PF6.
In keeping with these observations, the limit of this transformation was reached with the
hexylene derivative 6CVꢁ2PF which only affords trace amounts of the [5]rotaxane 6C5Rꢁ4PF ,
6
6
o
after stirring at 55 C for more than a week.
Solvent effects. Initially, we probed the PꢀCBꢀAAC in MeCN as it appeared to be the most
convenient solvent in which to dissolve the CBꢀcontaining precursors [1⊂CB]ꢁ[PF ] and
6
2
+
[
1 ⊂CB]ꢁ[PF ] . During the course of our investigations, we observed that the V ⊂P, which is
2
6 2
soluble in MeCN at the outset, gradually precipitates out of solution over a period of time (30
min – 1 h). This process appears to have little impact on reactions which proceed to completion
within this time period, as evidenced by the excellent isolated yields; we suspect, however, that
the lower yields, obtained over longer experiment times, might be attributed in part to this
process which effectively shuts down the reaction. Measuring the thermodynamic parameters for
binding (binding constants K , enthalpy
ꢀH and entropy
ꢀS) of the P and (2−5CV)ꢁ2PF6
a
complexes in MeCN, however, is not a trivial matter because of solubility issues. Switching to
2
+
other polar solvents, e.g., MeNO or Me CO, avoids the V ⊂P solubility issue, but raises new
2
2
problems − when the rotaxanation is performed in MeNO or Me CO it requires (Table 1, entry
2
2
2−3) much longer reaction times (more than 18 h) with significantly lower yields (< 50%)
because of the poor solubility of the stopper precursors [1 ⊂CB]ꢁ[PF ] and [1⊂CB]ꢁ[PF ].
2
6 2
6
1
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 47
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2+
P is known to form both 1:1 and 1:2 complexes with BIPY derivatives in Me SO
2
3
5b
2+
solutions: indeed, a homogenous solution of all reactants, including V , P and [1 ⊂CB]/
2
[
1⊂CB] can be achieved. Despite this practical breakthrough, no mechanically interlocked
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
product was obtained (Table 1, entry 4, 6, 8 and 10) in the reaction. It seems that solventꢀsolute
2
9
interactions dominate in Me SO because of its strong hydrogen bond acceptor capacity
2
(
βsulfoxide = 8.9 c.f. βnitrile = 4.7) which inhibits assembly of the productive multicomponent
complexes. This result implies that the interꢀring hydrogen bonding interactions between CB, P
and the substrates is the dominating factor in the PꢀCBꢀAAC.
Conformational analysis. In principle, P can adopt (Figure 4) four enantiomeric pairs of
diastereoisomeric conformations in solution on account of the oxygenꢀthroughꢀtheꢀannulus
4
1
rotation of the hydroquinone rings in P. These conformations undergo rapid inversion and
1
interconversion in solution on the H NMR timescale at room temperature. The rate of this
o
2+
rotation can be slowed down at low temperature (below –60 C) in the presence of a BIPY unit
42
as a guest. Following the formation of 2C4Rꢁ4PF , the change in conformation of the P ring is
6
2
+
impeded since its cavity is occupied by the BIPY unit in the dumbbell, such that there is no
space available for any of the hydroquinone rings to rotate through the annulus of P. For this
42
reason, the hetero[4]rotaxane 2C4Rꢁ4PF can exist as four conformational isomers − A, B, C
6
and D with their correspondent enantiomers (Figure 4) , , and . Conformational isomer A
has C symmetry, while the other three conformational isomers (B, C and D) all have C2
5
symmetry. Proton resonances (aromatic protons m and hydroxyl protons n, Figure 4) of P in
conformational isomer A should exhibit two individual singlet peaks while, in the other three
1
2
ACS Paragon Plus Environment

Page 13 of 47
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
conformational isomers (B, C and D) five singlet peaks for m and five singlet peaks for n should
1
be exhibited, respectively, in the H NMR spectrum.
The hetero[4]rotaxane 2C4Rꢁ4PF was purified by column chromatography (SiO ) and
6
2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
its formation was confirmed by the appearance of the resonance at 6.7 ppm for the triazole ring
1
proton H in the H NMR spectrum (Figure 4). Three sets of α proton resonances (δ 9.3−9.6 ppm
e
2+
in Figure 4) from the BIPY unit in the dumbbell and three β proton resonances (δ 5.3−6.0 ppm)
reveal the presence of three isomers associated with different P ring conformations. The β proton
resonances are moved upfield compared with those for the rod precursor 2CVꢁ2PF , while the α
6
proton resonances are shifted significantly downfield, indicating that the β protons are shielded
by P while the α protons are deshielded. This observation is consistent with the fact that the P
2
+
o
o
ring encircles the BIPY unit. On heating the hetero[4]rotaxane to 60 C in CD CN and 100 C
3
in (CD ) SO, we noted that the multiple α and β proton resonances do not undergo coalescence
3
2
1
on the H NMR timescale. In the HSQC spectrum (see the SI), a total of 22 different OH and
aromatic proton resonances (δ 6.1−8.3 ppm) can be observed (Figure 4) and assigned to the P
ring. These observations suggest that the P ring of the hetero[4]rotaxane 2C4Rꢁ4PF exists in
6
43
44
three different conformations, which are unable to undergo inversion or interconversion upon
o
heating to 100 C.
Conformational isomer A (in red, Figure 4) of the P ring in 2C4Rꢁ4PF , which gives rise
6
1
to the simplest spectrum on account of its C symmetry, was identified first of all in the H NMR
5
spectrum, with the aid of twoꢀdimensional NMR techniques (see SI for the COSY, NOESY,
HSQC and HMBC NMR spectra). Identification of the other two conformational isomers out of
the total three is not trivial: we begin with distinguishing proton m from n by HSQC and HMBC
1
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 47
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
spectra, followed by linking the allꢀneighboring protons sequentially in the same P ring by 2D
NOESY experiments. 2D NOESY Spectra of 2C4Rꢁ4PF have been obtained with different
6
mixing times (200 ms, 300 ms, and 400 ms, respectively) at 298 K. A full buildꢀup NOE signal
between protons m and n on the P rings requires a longer mixing time (≥ 400 ms). Based on an
analysis of the NOE correlation peaks (see the SI), two conformational isomers − namely, B (in
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
blue) and C (in green) − can be identified in the H NMR spectrum (Figure 4): conformational
isomer D is not observed. The populations of the conformational isomers in 2C4Rꢁ4PF6,
1
calculated from integration of their particular H NMR signals, are (Table 2) in the sequence of
B > A > C. In order to acquire a better understanding of the conformational composition,
o
o
2
C4Rꢁ4PF was synthesized at −10 C and 55 C, respectively. The kinetics of the rotaxane
6
formation are significantly different at these temperatures, − i.e., the reaction goes to completion
o
o
within 2 min at 55 C but requires 2 h to do so at −10 C. Once again, only three conformational
1
isomers (A, B and C) are observed in the H NMR spectra. When the temperature of the reaction
is low, only the population of conformational isomer A increases gradually, an observation
indicates that
is the thermodynamically most favorable product. Conformational
isomer A, which supports the least number of intraꢀring hydrogen bonds (OHꢀOH hydrogen
bonding) in the P ring, is able to form the strongest hydrogen bonding network with CBs on the
dumbbell of 2C4Rꢁ4PF . By contrast, the OH groups in conformational isomer D which has the
6
4
3
greatest propensity to from intraꢀring hydrogen bonds are less able to form hydrogen bonds
o
with CB. Even when the reaction is performed at −10 C, the major product is
, an
observation which suggests that the energy difference between conformational isomers A and B
is really quite small.
1
4
ACS Paragon Plus Environment

Page 15 of 47
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
1
Four conformational isomers (A, B, C and D) can be observed in the H NMR spectrum
of the hetero[4]rotaxane 3C4Rꢁ4PF . By following a similar procedure to that adopted for
6
2
C4Rꢁ4PF , all the conformational isomers could be identified and their relative populations
6
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
were found to be A: 30 %, B: 40 %, C: 23 %, and D: 7 %, respectively. This observation
suggests that all four conformational isomers of P can form stable complexes with 3CVꢁ2PF in
6
solution and hence be captured by the rod precursor [1⊂CB]ꢁ[PF ]. The relative populations of
6
the conformational isomers of P in 3C4Rꢁ4PF follow a similar trend to those of P in 2C4Rꢁ4PF .
6
6
2
+
The α and β protons of the BIPY unit in 3C4Rꢁ4PF are less shielded and deshielded by P,
6
respectively, than in 2C4Rꢁ4PF . This comparison implies that the P ring can shuttle along the
6
2
+
BIPY unit in 3C4Rꢁ4PF because of its extended oligomethylene chain length. In addition,
6
even although the P ring adopts the same conformations in 3C4Rꢁ4PF , the chemical shifts of
6
resonances for protons signals m and n in the conformational isomers A, B and C are very
different compared with those in the corresponding conformational isomers in 2C4Rꢁ4PF6.
In theory, the hetero[5]rotaxane 4C5Rꢁ4PF bearing two P rings on its dumbbell can exist
6
1
(
see the SI) as 20 different stereoisomers. The proton signals in the H NMR spectrum of
4C5Rꢁ4PF are not well separated, rendering the full analysis of the each P conformation
6
extremely challenging. Fortunately, the most abundant conformation present in 4C5Rꢁ4PF can
6
be separated by recycling reverse phase HPLC. There are two hydroxyl protons signals at 9.0
ppm as well as two aromatic protons (7.35 and 6.73 ppm) which can be assigned (Figure 5) to P.
This assignment indicates that in this fraction, the two P rings adopt conformation A/ with C5
symmetry, obligating one set of protons (n and m) to point towards the CB ring and another set
(n' and m') to point away. There are still, however two enantiomers (AA and
) and one other
1
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 47
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
stereoisomer (A ). It is worth noting that the resonance (δ = 4.67 ppm) for the α proton in the
2
+
BIPY unit is shifted upfield significantly whereas the resonance for the β proton is shifted
downfield. In addition, the resonances for protons g and h of the oligomethylene chain are
shielded and their resonances appear at high field compared to those for the rod precursor. This
phenomenon suggests that two P rings are threaded along the dumbbell symmetrically, each of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2+
them occupying half of the BIPY unit and the oligomethylene chain. The formation of a highly
symmetrical structure in the case of 4C5R is under thermodynamic control. The P ring
conformational preference in the hetero[5]rotaxane 4C5R is a consequence of the relative
strength of the hydrogen bonding interactions − whereas both the CBꢀP and CBꢀP interꢀring
A
B
interactions are of similar energy, optimal hydrogen bonding between P rings only occurs when
both adopt their C conformations.
5
Molecular Modeling. Slow vapor diffusion of iPr O into a MeNO solution of rotaxane 2C4R
2
2
afforded single crystals suitable for Xꢀray diffraction. Alas, the crystals were weakly diffracting
and the data could be refined to a resolution of 1.30 Å (see the SI): the situation is, no doubt,
complicated by the presence of isomers giving rise to disorder in the solid state. From the
partially resolved structure, it is possible to distinguish the three ring components threaded on the
dumbbell, with a P ring positioned between two CB rings. Since no detailed structural
information could be garnered from this data, we appealed to molecular mechanics simulations
in order to gain a better understanding of the noncovalent bonding interactions that direct
rotaxane formation and influence the choice and close up distribution of isomers. Simulations
were performed starting from all four P isomers (A–D in Figure 4) to determine the lowest
energy structure for the rotaxane 2C4R: see the SI. The simulated structure (Figure 6a) of 2C4R
1
6
ACS Paragon Plus Environment

Page 17 of 47
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
incorporating a P ring locked in conformation A closely matches the structure suggested by the
4
6
Xꢀray crystallographic analysis. Owing to its C symmetric conformation, all phenolic
5
hydroxyl groups in this conformation of P are able to participate in hydrogen bonding
interactions with the neighboring carbonyl groups of the CBs, thus maximizing the stabilizing
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
interꢀring interactions. The capacity of P ring in its C symmetric conformation to participate in
5
interꢀring hydrogen bonding is diminished in the case of the C symmetric conformations B–D as
2
a result of internal hydrogen bonding between some of the phenolic hydroxyl groups. The P ring
of the rotaxane 2C4R in conformation D was found to have the maximized intraꢀring hydrogen
bonding, in keeping with the fact that it has the fewest interꢀring hydrogen bonding interactions
with CB rings. This difference in the interꢀring hydrogen bonding networks may explain why
conformational isomer D was not observed for 2C4R in practice.
We have discovered that lengthening the rod precursor by two methylene units slows
down rotaxane formation from 2 min (2C4R) to 2 h (3C4R). It is evident from the simulated
structure of 3C4R that the contiguous hydrogen bonding network is disrupted, with fewer interꢀ
ring hydrogen bonds. The CB rings, which occupy the ammonium binding sites, are held much
further apart, preventing the P ring from sealing the gap between them completely. During
formation of 3C4R, the ensemble benefits much less from the positive cooperativity, generated
by interꢀring hydrogen bonding, leading to a reduced reaction rate. Additionally, the energy
difference between isomers of 3C4R is not as significant as it is in the case of 2C4R, in
accordance with the fact that all four isomers have been identified experimentally on the
dumbbell of 3C4R.
1
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 47
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
[
5]Rotaxanes 4C5R and 5C5R follow a similar trend to those of the [4]rotaxanes 2C4R
and 3C4R. The spacing of rings (Figure 6c) in 4C5R is optimal, leading to (i) a high degree of
positive cooperativity, (ii) a high yield, (iii) a short reaction time, and (iv) a noticeable preference
for the isomer (4C5R , 4C5RA and 4C5R ) in which the P rings adopt C conformation. The
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
AA
5
longer oligomethylene chains in 5C5R (Figure 6d) result in longer interꢀring hydrogen bonding
distances and, hence, weaker interactions. A consequence of changing the strength of the interꢀ
ring interactions is reflected in a change in the specificity of the reaction favoring the [5]rotaxane,
i.e., 4C5R is preferred over the [5]rotaxanes regardless of the reaction stoichiometry, but the
[
4]rotaxane analog (5C4R) of the [5]rotaxane 5C5R was identified as a side product.
Conclusions
2
3
The cooperative capture strategy for templating the formation of heterorotaxanes containing
cyclodextrins has been extended to the use of pillar[5]arene rings as promoters. Compared to the
interactions between cyclodextrins and cucurbiturils, the superior hydrogen bonding interactions
between the faces of cucurbit[6]uril and pillar[5]arene, provided the constitutions of the
dumbbells are of appropriate lengths, can (i) accelerate 1,3ꢀdipolar cycloadditions between
azides and alkynes, affording two threeꢀring rotaxanes and one fourꢀring rotaxane in high yields,
as well as (ii) support cycloadditions with a broader range of substrates, varying from 2ꢀ
azidoethylꢀ to 5ꢀazidopentylpyridinium ions. The siting of the pillar[5]arene rings around
bipyridinium units stabilizes otherwise energetically unfavorable coꢀconformations of the alkyne
and azide moieties within the cucurbit[6]uril cavity, and allows pillar[5]arene ring(s) to promote
cycloadditions at each end of the nascent dumbbells sequentially. All four conformations (A−D)
1
of pillar[5]arene have been observed experimentally and characterized fully by H NMR
1
8
ACS Paragon Plus Environment

Page 19 of 47
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
4+
spectroscopy in the case of 3C4R . The populations of the trapped and fixed conformational
4
2
isomers of the pillar[5]arenes in the hetero[4]/[5]rotaxanes appear to be at least partially under
thermodynamic control, while the relative yields of one (2C4R ) of the isomers is observed to
A
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
increase as the reaction temperature decreases. Molecular mechanical simulations also suggest
that the relative energies among the conformational isomers in these rotaxanes are similar. The
cooperative hydrogen bonding interactions between cucurbit[6]uril and pillar[5]arene rings give
rise to a high level of preorganization during rotaxane formation, which, on one hand, greatly
enhances cucurbit[6]uril’s activity towards templating the 1,3ꢀdipolar cycloaddition and, on the
other hand, controls the conformational distribution of the pillar[5]arene rings. These results add
yet another cooperative approach to the syntheses of mechanically interlocked molecules with
high efficiencies and good constitutional integrities. Furthermore, a new strategy for altering
template activity employing noncovalent bonding interactions has been demonstrated. These
findings expand the use of cucurbiturilꢀcatalyzed 1,3ꢀdipolar cycloadditions for the development
of highly ordered polyrotaxanes and add a fresh approach to the practice of supramolecular
catalysis.
Experiment Section
General Methods. All reagents were purchased from commercial suppliers (Aldrich and Fisher)
2
6h
47
48
and used as received. Pillar[5]arene
(P), cucurbit[6]uril (CB), and the rod precursors
(
2−6CVꢁ2PF ) were synthesized as reported in the literature. Thin layer chromatography (TLC)
6
was performed on silica gel 60 F254 (E. Merck). Column chromatography was carried out on
silica gel 60F (Merck 9385, 0.040–0.063 mm). Caution: Significant amounts of rotaxanes
2
C4R−5C5Rꢁ4PF are retained on silica columns and so it is important to use a minimum
6
amount of silica gel. UVꢀVis Spectra were measured on a Shimadzu 3600 UVꢀVisꢀNIR
1
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 20 of 47
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
spectrometer with a temperature control system, employing cuvettes with a pathway of either 2
or 10 mm. Nuclear magnetic resonance (NMR) spectra were recorded on Bruker Avance 500 or
1
6
00 spectrometers with working frequencies of 500 or 600 MHz for H and 125 or 150 MHz for
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
13
C nuclei, respectively. Chemical shifts are reported in ppm relative to the signals corresponding
to the residual nonꢀdeuterated solvents (CDCl : δ = 7.26 ppm, CD CN: δ = 1.94 ppm, D O: δ =
3
3
2
4
.62 ppm and (CD ) SO: δ = 2.50 ppm). Highꢀresolution mass spectra were measured on a
3 2
Finnigan LCQ iontrap mass spectrometer (HRꢀESI).
[1⊂CB]·[PF ]: 1ꢁCl (600 mg, 2.5 mmol) and CB (2.5 g, 2.5 mmol) in H O (70 mL) were stirred
6
2
o
at 90 C for 1 h, after which time the reaction mixture was cooled down to room temperature and
the insoluble residue removed by filtration. When KPF (5.0 g, 27.1 mmol) was added to the
6
aqueous solution, a white precipitate was generated. The solid was collected by filtration, washed
with excess of H O and dried under reduced pressure to afford [1⊂CB]ꢁ[PF ] (3.2 g, yield 95%).
2
6
1
H NMR (500 MHz, CD CN) δ = 7.34 (bs, 2H), 6.87 (d, J = 2.3 Hz, 2H), 6.56 (t, J = 2.2 Hz, 1H),
3
5.77 (d, J = 15.1 Hz, 12H), 5.37 (s, 12H), 4.28 (s, 2H), 4.16 (d, J = 15.2 Hz, 12H), 3.85 (s, 6H),
3.44 (s, 2H), 2.16 (s, 1H).
[1 ⊂CB]·[PF ] : 1ꢁCl (200 mg, 0.83 mmol) and CB (800 mg, 0.80 mmol) in H O (50 mL) were
2
6 2
2
o
stirred at 60 C overnight and the insoluble residue was removed by filtration. When NH PF
4
6
(1.62 g, 10 mmol) was added to the aqueous solution, a white precipitate was generated. The
solid was collected by filtration, washed with excess of H O and dried under reduced pressure to
2
1
afford [1 ⊂CB]ꢁ[PF ] (1.34 g, yield 95%). H NMR (500 MHz, CD CN) δ = 6.80 (bs, 4H), 6.58
2
6 2
3
(t, J = 2.3 Hz, 2H), 5.76 (d, J = 15.2 Hz, 12H), 5.38 (s, 12H), 4.32 (s, 4H), 4.17 (d, J = 15.2 Hz,
13
1
2H), 3.84 (s, 12H), 3.70 (s, 4H). C NMR (126 MHz, CD CN) δ = 160.9, 155.3, 155.1, 132.3,
3
2
0
ACS Paragon Plus Environment

Page 21 of 47
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
2+
1
07.5, 100.7, 69.4, 68.2, 54.9, 50.8, 50.3, 49.6, 49.5, 35.4. HRꢀESIꢀMS calcd for [M – 2PF6]
+
m/z = 704.2647, found m/z = 704.2643; calcd for [M – PF ] m/z = 1553.4941, found m/z =
6
1553.4897.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
C4R·4PF : (Approach 1) [1⊂CB]ꢁ[PF ] (172 mg, 127.6 µmol), 2CV·2PF (25 mg, 42.6 µmol)
6
6
6
and P (35 mg, 57.4 µmol) were dissolved in MeCN (30 mL) and stirred at room temperature. The
color of the reaction mixture turns orange immediately. The reaction was monitored by TLC
(SiO : Me CO with 2% NH PF (m/v), I stain, R = 0.3) and was found to be complete inside 1
2 2 4 6 2 f
h. The solvent was then removed under reduced pressure and the residue was purified by column
chromatography (SiO : Me CO with 0.2% NH PF and then Me CO with 2% NH PF (m/v)).
2
2
4
6
2
4
6
The fraction with R = 0.3 (TLC / SiO : Me CO with 2% NH PF (m/v)) was collected and the
f
2
2
4
6
solvent was removed under the reduced pressure. H O (20 mL) was added to the residue to
2
remove excess of NH PF and the product ‘crashes out’ as an orange precipitate. The solid was
4
6
collected by filtration, washed with an excess of H O to remove NH PF and dried under high
2
4
6
vacuum to afford the hetero[4]rotaxane 2C4R·4PF as an orange powder (145 mg, yield 95%).
6
(Approach 2) 2CV·2PF (10 mg, 17.1 µmol) and P (16 mg, 26.2 µmol) were dissolved in MeCN
6
o
o
o
(20 mL) and stirred at an appropriate temperature (−10 C, 25 C and 55 C, respectively) until
complex formation reaches equilibrium. Then, [1 ⊂CB][PF ] (65 mg, 38.3 µmol) was added in
2
6 2
one portion to the reaction mixture, whereupon the color of the mixture changed to orange. The
reaction was monitored by TLC (SiO : Me CO with 2% NH PF (m/v), I stain, R = 0.3). After
2
2
4
6
2
f
all the 2CV·2PF had been consumed, the solvent was removed under reduced pressure and the
6
residue was purified by column chromatography (SiO : Me CO with 0.2% NH PF and then
2
2
4
6
Me CO with 2% NH PF (m/v)). The fraction with R = 0.3 (TLC / SiO : Me CO with 2%
2
4
6
f
2
2
2
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 22 of 47
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
NH PF (m/v)) was collected and the solvent was removed under the reduced pressure. H O (20
4
6
2
mL) was added to the residue to remove excess of NH PF , whereupon the product ‘crashes out’
4
6
as an orange precipitate. The solid was collected by filtration, washed with an excess of H O to
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
remove NH PF and dried under high vacuum to afford the hetero[4]rotaxane 2C4R·4PF as an
4
6
6
orange powder (yield 60−70%).
1
In the H NMR spectrum of hetero[4]rotaxane 2C4R·4PF , three − namely 2C4R ꢁ4PF ,
6
A
6
2C4R ꢁ4PF and 2C4R ꢁ4PF − of the four possible conformational isomers 2C4R ꢁ4PF ,
B
6
C
6
A
6
1
2
C4R ꢁ4PF , 2C4R ꢁ4PF and 2C4R ꢁ4PF , are observed. No 2C4R ꢁ4PF was detected by H
B 6 C 6 D 6 D 6
NMR spectroscopy. When the separation of the three conformational isomers was attempted
using recycling reverse phrase HPLC, the isomers invariably eluted together as a mixture. HRꢀ
4+
ESIꢀMS of the threeꢀcomponent mixture of 2C4R·4PF : calcd for [M – 4PF ] m/z = 827.7894,
6
6
3
+
found m/z = 827.7890, [M – 3PF6] m/z = 1152.0406, found m/z = 1152.0403, [M – 3PF –
6
3
+
HPF6] m/z =1103.3835, found m/z = 1103.3829. By comparing the relative integrated
1
1
1
1
intensities for each of the proton resonances in the H NMR spectrum with 2D Hꢀ H COSY, Hꢀ
1
3
1
1
C HSQC and Hꢀ H NOESY spectra (see the Supporting Information), we were able to assign
1
every proton signal in the H NMR spectrum to one of the three conformational isomers. The
1
1
relative conformations (A, B and C) of the P rings were assigned by Hꢀ H NOESY spectra and
the relative abundances of the three isomers in the mixture obtained from integration are listed in
Table 2. We have extracted the proton signal assignments for each conformational isomer (A, B
and C) of 2C4R·4PF and listed them below. Note that the proton counts listed below are relative
6
ones for each conformational isomer.
2
2
ACS Paragon Plus Environment

Page 23 of 47
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
1
Conformational isomer 2C4R ꢁ4PF : H NMR (600 MHz, CD CN) δ: 9.35 (d, J = 6.6 Hz, 4H),
A
6
3
8
5
5
.07 (bs, 4H), 7.36 (s, 10H), 6.96 (m, 4H), 6.84 (s, 10H), 6.64 (s, 2H), 6.57 (t, J = 2.3 Hz, 2H),
.93 (dd, J = 15.5, 12H), 5.86 (d, J = 6.7 Hz, 4H), 5.77 (d, J = 15.4 Hz, 12H), 5.40 (s, 24H),
.21ꢀ5.10 (m, 4H), 4.54 (t, J = 6.0, 4H), 4.41 (t, J = 5.9 Hz, 4H), 4.32 (d, J = 15.5 Hz, 12H),
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4.24ꢀ4.20 (m, 4H), 4.17 (d, J = 15.4, 12H), 3.88 (s, 12H), 3.55 (s, 10H). Conformational isomer
1
2
C4R ꢁ4PF : H NMR (600 MHz, CD CN) δ: 9.57 (d, J = 6.5 Hz, 4H), 8.07 (bs, 4H), 7.82 (s,
B
6
3
2H), 7.81 (s, 2H), 7.65 (s, 2H), 7.30 (s, 2H), 7.26 (s, 2H), 6.96 (m, 4H), 6.91 (s, 2H), 6.71 (d, 2H),
6.70 (s, 2H), 6.64 (s, 2H), 6.57 (t, J = 2.3 Hz, 2H), 6.26 (s, 2H), 6.15 (s, 2H), 5.93 (dd, J = 15.5,
2H), 5.77 (d, J = 15.4 Hz, 12H), 5.52 (d, J = 6.7 Hz, 4H), 5.40 (s, 24H), 5.21ꢀ5.10 (m, 4H), 4.54
t, J = 6.0, 4H), 4.41 (t, J = 5.9 Hz, 4H), 4.32 (d, J = 15.5 Hz, 12H), 4.24ꢀ4.20 (m, 4H), 4.17 (d, J
1
(
= 15.4, 12H), 3.88 (s, 12H), 3.78 (d, J = 13.8 Hz, 2H), 3.69 (s, 2H), 3.57 (m, 4H), 2.94 (d, J =
1
1
6
7
3.9 Hz, 2H). Conformational isomer 2C4R ꢁ4PF : H NMR (600 MHz, CD CN) δ: 9.58 (d, J =
C
6
3
.6 Hz, 4H), 8.18 (s, 2H), 8.07 (bs, 4H), 7.47 (s, 2H), 7.18 (s, 2H), 7.15 (s, 2H), 7.14 (s, 3H),
.11 (s, 3H), 7.04 (s, 3H), 6.96 (m, 4H), 6.64 (s, 2H), 6.57 (t, J = 2.3 Hz, 2H), 6.49 (s, 2H), 6.42
(s, 2H), 6.41 (s, 2H), 5.93 (dd, J = 15.5, 12H), 5.77 (d, J = 15.4 Hz, 12H), 5.40 (s, 24H), 5.35 (d,
J = 6.1 Hz, 4H), 5.21ꢀ5.10 (m, 4H), 4.54 (t, J = 6.0, 4H), 4.41 (t, J = 5.9 Hz, 4H), 4.32 (d, J =
15.5 Hz, 12H), 4.24ꢀ4.20 (m, 4H), 4.17 (d, J = 15.4, 12H), 3.94 (d, J = 13.7 Hz, 2H), 3.88 (s,
12H), 3.62 (m, 4H), 3.40 (s, 2H), 3.05 (d, J = 14.1 Hz, 2H).
3C4R·4PF : (Approach 1) [1⊂CB]ꢁ[PF ] (172 mg, 127.6 µmol), 3CV·2PF (25 mg, 40.7 µmol)
6
6
6
o
and P (35 mg, 57.4 µmol) were dissolved in MeCN (30 mL) and stirred at 55 C temperature.
The color of the reaction mixture turns orange immediately. The reaction was monitored by TLC
(
SiO : Me CO with 2.5 % NH PF (m/v), I stain, R = 0.3) and was found to be complete inside
2 2 4 6 2 f
2
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 24 of 47
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
h. The solvent was removed under reduced pressure and the residue was purified by column
chromatography (SiO : Me CO with 0.2% NH PF and then Me CO with 3% NH PF (m/v)).
2
2
4
6
2
4
6
The fraction with R = 0.3 (TLC / SiO : Me CO with 2.5% NH PF (m/v)) was collected and the
f
2
2
4
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
solvent was removed under the reduced pressure. H O (20 mL) was added to the residue to
2
remove excess of NH PF , whereupon the product ‘crashes out’ as an orange precipitate. The
4
6
solid was collected by filtration, washed with an excess of H O to remove NH PF and dried
2
4
6
under high vacuum to afford the hetero[4]rotaxane 3C4R·4PF as an orange powder (145 mg,
6
yield 95%). (Approach 2) 3CV·2PF (12 mg, 18.7 µmol) and P (18 mg, 29.5 µmol) were
6
o
dissolved in MeCN (20 mL) and stirred at 55 C until complex formation reached equilibrium.
Then, [1 ⊂CB][PF ] (65 mg, 38.3 µmol) was added in one portion to the reaction mixture,
2
6 2
whereupon the color of the mixture changes to orange. The reaction was monitored by TLC
SiO : Me CO with 2.5% NH PF (m/v), I stain, R = 0.3). After all the 2CV·2PF had been
(
2
2
4
6
2
f
6
consumed, the solvent was removed under reduced pressure and the residue was purified by
column chromatography (SiO : Me CO with 0.2% NH PF and then Me CO with 3% NH PF
6
2
2
4
6
2
4
(m/v)). The fraction with R = 0.3 (TLC / SiO : Me CO with 2.5% NH PF (m/v)) was collected
f 2 2 4 6
and the solvent was removed under the reduced pressure. H O (20 mL) was added to the residue
2
to remove excess of NH PF , whereupon the product ‘crashes out’ as an orange precipitate. The
4
6
solid was collected by filtration, washed with an excess of H O to remove NH PF and dried
2
4
6
under high vacuum to afford the hetero[4]rotaxane 3C4R·4PF as an orange powder (yield 59%).
6
In the case of the hetero[4]rotaxane 3C4R·4PF , all four possible conformational isomers,
6
1
namely 3C4R ꢁ4PF , 3C4R ꢁ4PF 3C4R ꢁ4PF and 3C4R ꢁ4PF have been observed by H
A
6
B
6
C
6
D
6
NMR spectroscopy. When the separation of the four conformational isomers was attempted
2
4
ACS Paragon Plus Environment

Page 25 of 47
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
using recycling reverse phrase HPLC, the isomers invariably eluted together as a mixture. HRꢀ
4+
ESIꢀMS of the fourꢀcomponent mixture of 3C4R·4PF : calcd for [M – 4PF ] m/z = 834.7969,
6
6
3
+
found m/z = 834.7972, [M – 3PF6] m/z = 1161.3841, found m/z = 1161.3833, [M – 3PF –
6
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3+
HPF6] m/z =1112.7267, found m/z = 1112.7259. By comparing the relative integrated
1
1
1
1
intensities for each of the proton resonances in the H NMR spectrum with 2D Hꢀ H COSY, Hꢀ
1
3
1
1
C HSQC and Hꢀ H NOESY spectra (see the Supporting Information), we were able to assign
1
every proton signal in the H NMR spectrum to one of the four conformational isomers. The
1
relative conformations (A, B, C and D) of the P rings were assigned from an analysis of the Hꢀ
1
H NOESY spectra and the relative abundances of the four isomers in the mixture are listed in
Table 2. We have extracted the proton signal assignments for each conformational isomer (A, B,
C and D) of 3C4R·4PF and they are listed below. Note that the proton counts listed below are
6
the relative ones for each conformational isomer.
1
Conformational isomer 3C4R ꢁ4PF : H NMR (600 MHz, CD CN) δ: 8.63 (d, J = 6.5 Hz, 4H, α),
A
6
3
8.16 (bs, 4H, NH ), 7.38 (s, 10H, OH), 6.93 (m, 4H, b), 6.81 (s, 10H, ArH), 6.56 (m, 2H, a), 6.50
2
(
s, 2H, e), 5.95−5.71 (m, 24H, x + x ), 5.89 (d, J = 6.5 Hz, 4H, β), 5.33 (s, 24H, z), 4.80 (m, 4H,
a
b
h), 4.44−4.39 (m, 8H, c and d), 4.24 (d, J = 15.2 Hz, 12H, y ), 4.11 (d, J = 15.2 Hz, 12H, y ),
b
a
1
3
.85 (s, 12H, OMe), 3.97−0.85 (m, 28H, f + g+ k). Conformational isomer 3C4R ꢁ4PF : H
B
6
NMR (600 MHz, CD CN) δ: 8.62 (d, J = 6.5 Hz, 4H, α), 8.16 (bs, 4H, NH ), 7.96 (s, 2H, OH),
3
2
7
.82 (s, 2H, OH), 7.51 (s, 2H, OH), 7.39 (s, 2H, OH), 7.21 (s, 2H, ArH), 7.18 (s, 2H, ArH), 7.16
s, 2H, OH), 6.93 (m, 4H, b), 6.68 (s, 2H, ArH), 6.56 (m, 2H, a), 6.50 (s, 2H, e), 6.39 (s, 2H,
ArH), 6.22 (s, 2H, ArH), 5.95−5.71 (m, 24H, x + x ), 5.71 (d, J = 6.5 Hz, 4H, β), 5.33 (s, 24H,
(
a
b
z), 4.80 (m, 4H, h), 4.44−4.39 (m, 8H, c and d), 4.24 (d, J = 15.2 Hz, 12H, y ), 4.11 (d, J = 15.2
b
2
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 26 of 47
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Hz, 12H, y ), 3.85 (s, 12H, OMe), 3.97−0.85 (m, 28H, f + g+ k). Conformational isomer
a
1
3C4R ꢁ4PF : H NMR (600 MHz, CD CN) δ: 8.60 (d, J = 6.5 Hz, 4H, α), 8.16 (bs, 4H, NH ),
C
6
3
2
8
.05 (s, 2H, OH), 7.99 (s, 2H, OH), 7.68 (s, 2H, OH), 7.23 (s, 2H, OH), 7.23 (s, 2H, OH), 7.10 (s,
H, ArH), 6.98 (s, 2H, ArH), 6.93 (m, 4H, b), 6.56 (m, 2H, a), 6.53 (s, 2H, ArH), 6.51 (s, 2H,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
ArH), 6.51 (s, 2H, ArH), 6.50 (s, 2H, e), 5.95−5.71 (m, 24H, x + x ), 5.87 (d, J = 6.5 Hz, 4H, β),
a
b
5
.33 (s, 24H, z), 4.80 (m, 4H, h), 4.44−4.39 (m, 8H, c and d), 4.24 (d, J = 15.2 Hz, 12H, y ), 4.11
b
(
d, J = 15.2 Hz, 12H, y ), 3.85 (s, 12H, OMe), 3.97−0.85 (m, 28H, f + g+ k). Conformational
a
1
isomer 3C4R ꢁ4PF : H NMR (600 MHz, CD CN) δ: 8.74 (d, J = 6.5 Hz, 4H, α), 8.16 (bs, 4H,
D
6
3
NH ), 8.19 (s, 2H, OH), 8.19 (s, 2H, OH), 7.95 (s, 2H, OH), 7.64 (s, 2H, OH), 7.60 (s, 2H, OH),
2
7
2
4
.14 (s, 2H, ArH), 6.93 (m, 4H, b), 6.90 (s, 2H, ArH), 6.56 (m, 2H, a), 6.50 (s, 2H, e), 6.40 (s,
H, ArH), 6.26 (s, 2H, ArH), 6.26 (s, 2H, ArH), 5.95−5.71 (m, 24H, x + x ), 5.54 (d, J = 6.5 Hz,
a
b
H, β), 5.33 (s, 24H, z), 4.80 (m, 4H, h), 4.44−4.39 (m, 8H, c and d), 4.24 (d, J = 15.2 Hz, 12H,
y ), 4.11 (d, J = 15.2Hz, 12H, y ), 3.85 (s, 12H, OMe), 3.97−0.85 (m, 28H, f + g+ k).
b
a
4
C5RAA ·4TFA: 4CVꢁ2PF (25 mg, 39 µmol, 1.0 equiv) and P (60 mg, 98 µmol, 2.5 equiv) were
6
o
mixed in MeCN (30 mL) and stirred at 55 C for 30 min before [1⊂CB] (140 mg, 98 µmol, 2.5
equiv) was added. The reaction was monitored by TLC (SiO : Me CO with 2% NH PF (m/v), I
2
2
2
4
6
stain, R = 0.28). After all the 4CV·2PF had been consumed, the solvent was removed and the
f
6
residue was purified by column chromatography (SiO : Me CO with 0.3% NH PF and then
2
2
4
6
Me CO with 2 % NH PF (m/v)).b The fraction with R = 0.28 (TLC / SiO : Me CO with 2 %
2
4
6
f
2
2
NH PF (m/v)) was collected and the solvent was removed under the reduced pressure. H O (20
4
6
2
mL) was added to the residue to remove excess of NH PF , whereupon the product ‘crashes out’
4
6
as an orange precipitate. The solid was collected by filtration, washed with an excess of H O to
2
2
6
ACS Paragon Plus Environment

Page 27 of 47
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
remove NH PF and dried under high vacuum to afford 4C5R·4PF as a light orange powder
4
6
6
(
yield 59%). In the case of the hetero[5]rotaxane 4C5R·4PF , while several isomers are observed,
6
it is nigh impossible assign them to each of the four different conformational isomers of P rings
on the dumbbell. When the separation of the conformational isomers was attempted using
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
recycling reverse phase HPLC (H O / MeCN (0.1 % TFA) = 80 : 20 to 60: 40 in 40 mins, flow
2
rate: 17 mL/min), three major fractions were collected (see the Supporting Information). Only
one fraction was obtained pure (Fraction 2, yield 80 %) and the conformation was assigned to
one in which both P rings adopt a C conformation, namely 4C5R •4TFA, resulting from an
5
AA
1
1
1
1
13
1
1
analysis of the H NMR spectrum together with the 2D Hꢀ H COSY, Hꢀ C HSQC and Hꢀ H
1
NOESY spectra (see the Supporting Information) . H NMR (600 MHz, (CD ) SO, 298 K): δ:
3 2
8.96 (s, 10H), 8.80 (s, 10H), 7.53 – 7.38 (m, 4H), 7.34 (s, 4H), 7.25 (d, J = 6.0 Hz, 4H), 7.08 (d,
J = 2.3 Hz, 4H), 6.73 (s, 10H), 6.54 (t, J = 2.3 Hz, 2H), 6.49 (s, 2H), 5.82 (d, J = 15.2 Hz, 12H),
5.63 (d, J = 14.9 Hz, 12H), 5.54 (s, 24H), 4.67 (d, J = 5.9 Hz, 4H), 4.52 (d, J = 15.1 Hz, 12H),
4.39 (d, J = 14.0 Hz, 12H), 4.36 – 4.21 (m, 8H), 3.88 (m, 4H), 3.85 (s, 12H), 3.63 (bs, 20H), 3.03
13
(
m, 4H), 2.33 – 2.14 (m, 4H), 1.05 (m, 4H). C NMR (125 MHz, (CD ) SO, 298 K) δ = 160.0,
3
2
157.6, 155.8, 155.5, 147.9, 145.0, 143.0, 136.2, 133.7, 127.7, 126.2, 124.3, 118.7, 118.3, 117.3,
07.9, 101.0, 69.4, 69.3, 61.9, 55.2, 52.8, 50.8, 50.4, 50.0, 43.2, 28.4, 26.9, 23.3, 22.5. HRꢀESIꢀ
1
4
+
3+
MS: calcd for [M – 4PF + 4H] m/z = 994.6014, found m/z = 994.6042, [M – 4PF +3H] m/z
6
6
= 1325.4651, found m/z = 1325.4667
ASSOCIATED CONTENT
Supporting Information
Full details of instrumentation and analytical techniques; synthesis and characterization data for
1
1
1
the [4]ꢀ and [5]rotaxanes 2C4R – 5C5Rꢁ4PF and their precursors; H– H COSY, NOESY, H–
6
2
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 28 of 47
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
13
1
C HSQC and variableꢀtemperature H NMR spectroscopic investigation of the [4]rotaxane
2C4R – 3C4Rꢁ4PF ; UVꢀVis absorption spectra of 2C4R – 5C5Rꢁ4PF and 2C3Rꢁ4PF ; highꢀ
6
6
6
resolution mass spectra of [4]ꢀ and [5]rotaxanes 2C4R – 5C5Rꢁ4PF and their precursors;
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
partially resolved crystal data and molecular modeling of the [4]ꢀ and [5]rotaxane using
molecular dynamics and energy minimization. This material is available free of charge via the
Internet at http://pubs.acs.org.
Author Information
Corresponding Author
Stoddart@northwestern.edu
ACKNOWLEDGEMENTS
This work was performed under the auspices of the King Abdulaziz City of Science and
Technology (KACST) and Northwestern University (NU) Joint Center of Excellence for
Integrated NanoSystems (JCEINS). We thank Drs Turki S. Saud and Nezar H. Khdary for their
interest in the research. C. K. thanks the Royal Society in the UK for support as a Newton
Fellow Alumnus (Followꢀon Scheme 2012ꢀ2013). N.L.S. thanks NSF for a graduate Research
Fellowship. P. R. M thanks the USꢀUK Fulbright Commission for an All Disciplines Scholar
Award.
References
(1) a) Villa, J.; Warshel, A. J. Phys. Chem. B 2001, 105, 7887; b) Olsson, M. H. M.; Parson,
W. W.; Warshel, A. Chem. Rev. 2006, 106, 1737.
(
2) For a review, see: Dong, Z. Y.; Luo, Q.; Liu, J. Q. Chem. Soc. Rev. 2012, 41, 7890.
3) a) Chevallier, F.; Breit, B. Angew. Chem. Int. Ed. 2006, 45, 1599; b) Waloch, C.;
Wieland, J.; Keller, M.; Breit, B. Angew. Chem. Int. Ed. 2007, 46, 3037.
(
2
8
ACS Paragon Plus Environment

Page 29 of 47
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
(
4) a) Anslyn, E.; Breslow, R. J. Am. Chem. Soc. 1989, 111, 8931; b) Breslow, R. Acc.
Chem. Res. 1995, 28, 146; c) Kirby, A. J. Angew. Chem. Int. Ed. 1996, 35, 707.
(5) a) Kang, J. M.; Rebek, J. Nature 1997, 385, 50; b) Yoshizawa, M.; Tamura, M.; Fujita,
M. Science 2006, 312, 251; c) Wang, Z. J.; Clary, K. N.; Bergman, R. G.; Raymond, K.
N.; Toste, F. D. Nat. Chem. 2013, 5, 100.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(
6) a) Yang, C.; Nakamura, A.; Fukuhara, G.; Origane, Y.; Mori, T.; Wada, T.; Inoue, Y. J.
Org. Chem. 2006, 71, 3126; b) Yang, C.; Mori, T.; Inoue, Y. J. Org. Chem. 2008, 73,
5786; c) Yang, C.; Mori, T.; Origane, Y.; Ko, Y. H.; Selvapalam, N.; Kim, K.; Inoue, Y.
J. Am. Chem. Soc. 2008, 130, 8574; d) Ke, C. F.; Yang, C.; Mori, T.; Wada, T.; Liu, Y.;
Inoue, Y. Angew. Chem. Int. Ed. 2009, 48, 6675; e) Yang, C.; Ke, C. F.; Liang, W. T.;
Fukuhara, G.; Mori, T.; Liu, Y.; Inoue, Y. J. Am. Chem. Soc. 2011, 133, 13786.
(7) Amirsakis, D. G.; GarciaꢀGaribay, M. A.; Rowan, S. J.; Stoddart, J. F.; White, A. J. P.;
Williams, D. J. Angew. Chem. Int. Ed. 2001, 40, 4256.
(8) a) Fiedler, D.; Leung, D. H.; Bergman, R. G.; Raymond, K. N. Acc. Chem. Res. 2005, 38,
3
49; b) Fujita, M.; Tominaga, M.; Hori, A.; Therrien, B. Acc. Chem. Res. 2005, 38, 369;
c) Pluth, M. D.; Bergman, R. G.; Raymond, K. N. Science 2007, 316, 85; d) Nishioka, Y.;
Yamaguchi, T.; Yoshizawa, M.; Fujita, M. J. Am. Chem. Soc. 2007, 129, 7000; e)
Nishioka, Y.; Yamaguchi, T.; Kawano, M.; Fujita, M. J. Am. Chem. Soc. 2008, 130, 8160;
f) Hastings, C. J.; Backlund, M. P.; Bergman, R. G.; Raymond, K. N. Angew. Chem. Int.
Ed. 2011, 50, 10570.
(9) a) Bakirci, H.; Koner, A. L.; Dickman, M. H.; Kortz, U.; Nau, W. M. Angew. Chem. Int.
Ed. Engl 2006, 45, 7400; b) Cacciapaglia, R.; Di Stefano, S.; Kelderman, E.; Mandolini,
L. Angew. Chem. Int. Ed. 1999, 38, 348; c) Komiyama, M.; Kina, S.; Matsumura, K.;
Sumaoka, J.; Tobey, S.; Lynch, V. M.; Anslyn, E. J. Am. Chem. Soc. 2002, 124, 13731;
d) Cacciapaglia, R.; Casnati, A.; Mandolini, L.; Reinhoudt, D. N.; Salvio, R.; Sartori, A.;
Ungaro, R. J. Am. Chem. Soc. 2006, 128, 12322; e) Nwe, K.; Andolina, C. M.; Morrow,
J. R. J. Am. Chem. Soc. 2008, 130, 14861.
(
10) a) Jon, S. Y.; Ko, Y. H.; Park, S. H.; Kim, H. J.; Kim, K. Chem. Commun. 2001, 1938; b)
Pattabiraman, M.; Natarajan, A.; Kaliappan, R.; Mague, J. T.; Ramamurthy, V. Chem.
Commun. 2005, 4542; c) Wang, R. B.; Yuan, L. N.; Macartney, D. H. J. Org. Chem.
2
9
ACS Paragon Plus Environment