Bulk Inclusions of Double Pyridazine Molecular Rotors in Hexagonal Tris(o-phenylene)cyclotriphosphazene

Article abstract of DOI:10.1021/acs.joc.9b00553

A new generation of double pyridazine molecular rotors differing in intramolecular dipole-dipole spacing was synthesized. All rotor molecules formed bulk inclusions in a tris(o-phenylenedioxy)cyclotriphosphazene (TPP) host. Results of dielectric spectroscopy were fitted to a pair of nine-state models that accounted for interactions of neighboring dipoles at either an aligned or opposed possible orientation of the local threefold dipole rotation potentials within a channel of the TPP host. The results indicate dipole-dipole interaction strengths at the 100 to 200 K scale that lead dipoles to preferentially populate a subset of low-energy configurations. They also reveal that pyridazines with ethynyl substituents in 3- and 6-positions have slightly higher rotational barriers (3.2-3.5 kcal/mol) than those carrying one ethynyl and one tert-butyl group (1.9-3.0 kcal/mol). Upon cooling, these barriers reduce the rate of thermal transitions between the potential wells so much that the inclusions cannot achieve ordered dipolar ground states.

Full text of DOI:10.1021/acs.joc.9b00553

Subscriber access provided by BUFFALO STATE
Article
Bulk Inclusions of Double Pyridazine Molecular Rotors in
Hexagonal Tris(o-phenylene)cyclotriphosphazene (TPP)
Ji#í Kaleta, Guillaume Bastien, Jin Wen, Martin Dracinsky, Edward Tortorici,
Ivana Cisarova, Paul D. Beale, Charles T. Rogers, and Josef Michl
J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.joc.9b00553 • Publication Date (Web): 27 May 2019
Downloaded from http://pubs.acs.org on May 29, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the
course of their duties.

Page 1 of 40
The Journal of Organic Chemistry
1
2
3
4
5
Bulk Inclusions of Double Pyridazine Molecular Rotors in Hexagonal
Tris(o-phenylene)cyclotriphosphazene (TPP)
6
7
8
9
Jiří Kaleta,a,b* Guillaume Bastien,a Jin Wen,a* Martin Dračínský,a Edward Tortorici,c Ivana
Císařová,d Paul D. Beale,c Charles T. Rogers,c and Josef Michla,b
a Institute of Organic Chemistry and Biochemistry AS CR, Flemingovo nám. 2, 166 10
Prague 6, Czech Republic.
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
b Department of Chemistry, University of Colorado, Boulder, CO 80309-0215, United States
c Department of Physics, University of Colorado, Boulder, CO 80309, United States
d Department of Inorganic Chemistry, Faculty of Science, Charles University in Prague,
Hlavova 2030, 12840 Prague 2, Czech Republic.
TOC Graphic
ABSTRACT
A new generation of double pyridazine molecular rotors differing in intramolecular
dipole-dipole spacing was synthesized. All rotor molecules formed bulk inclusions in tris(o-
phenylenedioxy)cyclotriphosphazene (TPP) host. Results of dielectric spectroscopy were
fitted to a pair of nine-state models that accounted for interactions of neighboring dipoles at
the either aligned or opposed possible orientation of the local three-fold dipole rotation
potentials within a channel of the TPP host. The results indicate dipole-dipole interaction
strengths at the 100 to 200 K scale that lead dipoles to preferentially populate a subset of low-
energy configurations. They also reveal that pyridazines with ethynyl substituents in
positions 3- and 6- have slightly higher rotational barriers (3.2 - 3.5 kcal/mol) than those
carrying one ethynyl and one tert.-Bu group (1.9 - 3.0 kcal/mol). Upon cooling, these
barriers reduce the rate of thermal transitions between the potential wells so much that the
inclusions cannot achieve ordered dipolar ground states.
INTRODUCTION
Preparation and study of regular two- and three-dimensional (2D and 3D) assemblies
of molecular level devices are attracting considerable attention as documented by many
original articles, reviews and book chapters published in recent years.1,2,3,4,5,6,7 In general, 2D
arrays of molecular devices are accessible using the Langmuir-Blodgett technique,8,9,10 self-
assembly on suitable (mostly metallic) substrates,11,12,13,14,15,16 or formation of surface
inclusions by host-guest interaction,17,18,19 where anchoring parts of molecular devices are
inserted inside pores in the host lattice, while their functional parts protrude outside the
surface.
The 3D materials are usually made by (co-)crystallization of individual
components20,21,22,23,24,25,26 or by sorption of guest molecules into a well organized porous
1
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 40
1
2
3
4
5
host matrix, such as a MOF27,28,29 or hexagonal tris(o-phenylene)cyclotriphosphazene (TPP,
Figure 1).30,31
We report results obtained with 3D assemblies of dipolar molecular rotors 1 - 5 (Chart
1) present as guests in the channels of TPP. The crystal structure of this host is composed of
~5 Å thick layers, with the neighboring layers rotated by 60°. The layers contain voids that
are superimposed to form channels oriented perpendicular to the layers and located ~11 Å
apart in a triangular arrangement. Although in each layer, the channel cross-section is
triangular, the staggering of the layers makes them appear hexagonal when viewed along
their axes.31 The internal diameter is 4.5 - 5 Å. Since the crystal is molecular, this diameter
and the channel-to-channel spacing are capable of expanding by as much as 10% when the
channels are filled with guest molecules.6,28
Our motivation is the desire to improve our understanding of dipole-dipole
interactions in these regular arrays and especially, of the effect of the locally trigonal
structure of the channels. The lateral distances between two neighboring dipoles are fixed by
the inter-channel distance and can be modified only slightly. However, the rotors 1 - 5 offer
a variety of intra-channel distances between neighbors, and we can investigate both those
situations in which neighboring dipoles are located in congruent triangular environments and
those in which their environments are mutually rotated by 60°. An understanding of dipole-
dipole interactions plays a key role in the design of potential collective behavior.
Ferroelectric phases32 would be particularly interesting since they are promising potential
applications in the field of (nano)electronics (memory devices, frequency filters, etc.).
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Chart 1. Structures of Double Pyridazine
Rotors 1 - 5 with Intramolecular (red) and
Intermolecular (purple) Pyridazine - Pyridazine
Center
to
Center
Distances
Shown.
Intermolecular distances are calculated for two
neighboring molecules located next to each
other within one TPP channel. Black italic
letters indicate the labeling of carbon atoms
used in Table S3.
2
ACS Paragon Plus Environment

Page 3 of 40
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Chemical formula of TPP (A), top
view of two successive layers of TPP (B), and
a idealized model of molecular rotors
anchored inside the TPP host (C).
The five rod-like molecular rotors 1 - 5 have four structural features in common: (i) a
bulky terminal tert.-butyl group, whose diameter exceeds that of a rotating pyridazine ring,
and which expands the interior of the TPP channel and thus reduces the barrier for dipole
rotation; (ii) two pyridazine rotators with a relatively small diameter (~5.80 Å) and high
dipole moment (4.1 Debye for pyridazine itself33); (iii) a bulky bicyclo[1.1.1]penta-1,3-diyl
group (except in 1) with diameter of ~6.30 Å, which holds the axle of the rotor molecule
centered in the channel, away from the channel walls, limiting possible wobbling motions,27
and also serves as a 13C NMR marker since its chemical shifts occur in an uncluttered spectral
region; (iv) a well-defined intramolecular pyridazine-pyridazine center to center distance.
Intermolecular rotator-rotator distances depend on guest loading. For saturated samples no
voids between rotor molecules within a channel are expected, and these distances will be the
sum of the lengths of terminal groups (Chart 1). If the strength of dipole-dipole interactions
is sufficient and the barriers are low enough, compounds with both inter- and intramolecular
dipole-dipole spacings within a channel smaller than inter-channel separation (1 - 4) would
yield an antiferroelectric phase, while 5 would produce a ferroelectric phase.
The pyridazine rotator also suffers from a drawback. Since the C-N and N-N bonds
are shorter (1.32 - 1.34 Å) than the remaining aromatic C-C bonds (1.37 - 1.40 Å), the
exocyclic C-C bonds in positions 3- and 6- are not exactly antiparallel (an averaged deviation
is ca 4°), giving the rotor molecule a banana shape. Such bending may increase rotational
barriers.
RESULTS
We describe the synthesis of double pyridazine molecular rotors 1  5, demonstrate
their ability to form bulk inclusions in TPP, and prove that all parts of the molecule and
especially the pyridazine units rotate inside TPP crystals. The inclusions were characterized
by 13C and 31P cross-polarization magic angle spinning (CP-MAS) and 31P single-pulse
excitation (SPE) solid-state NMR, powder X-ray diffraction (XRD), differential scanning
calorimetry (DSC), and dielectric spectroscopy. Pyridazine rotational barriers were
3
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 4 of 40
1
2
3
4
5
6
7
8
9
determined by dielectric spectroscopy and for 5 the lower limit of rotational rate was also
independently deduced from ssNMR. The results were accounted for by a nine-state model
that takes into account dipole-dipole interactions and is supported by MP6-D3H434 and B97-
D3/SVP35 calculations. The structures of six of the new compounds were confirmed by
single-crystal X-ray diffraction.
Synthesis. The shortest rotor 1 was synthesized in seven steps from 3,3-
dimethylbutanone (Scheme 1), whose condensation with glyoxylic acid monohydrate
followed by workup with aqueous ammonia provided the ammonium salt of 3-pivaloyllactic
acid (6)36 in the first step. Subsequent cyclization with hydrazine yielded the pyridazinone
7,36 which was aromatized and converted to the chloropyridazine 8 using POCl3. The
reactivity of the pyridazine ring was enhanced by substitution of the chlorine atom by iodine
using aromatic Finkelstein reaction, which provided 9 in nearly quantitative yield.
Subsequent Sonogashira coupling with TMSA afforded the silyl derivative 10, which was
treated with TBAF in THF to liberate the alkyne 11. Final Sonogashira coupling between 11
and 9 afforded the rotor 1 in 90% yield (Scheme 1).
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
O
1.
O
. H2O
OH
+
O
O- NH4
OH
O
O
POCl3
100 °C
2. NH3/H2O
100 °C
H2N-NH2.H2O
reflux
N
N
N
NH
O
Cl
, 92%
6
7
8
, 13%
NaI
57% aq. HI
100 °C
N
9
TMSA
Pd(PPh3)4, CuI
Et3N
N
1. TBAF
2. H2O
Pd(PPh3)4, CuI
Et3N
N
N
N
N
N
N
THF
25 °C
THF
25 °C
THF
25 °C
I
N
N
TMS
1
11
10
9
, 95%
, 90%
, 98%
, 92%
Scheme 1. Synthesis of Rotor 1.
The extended pyridazine derivative 12 was made in two steps from 3,6-
dichloropyridazine (13, Scheme 2). Reaction of cheap and commercially available 13 with
NaI in concentrated aqueous HI afforded the much more expensive 3,6-diiodopyridazine (14)
in excellent yield. Subsequent desymmetrization using Sonogashira coupling with 3,3-
dimethylbut-1-yne gave a mixture of pyridazines 12 and 15,28 easily separated by column
chromatography.
Pd(PPh3)4, CuI
NaI
57% aq. HI
Cl
I
I
Et3N
N
N
N
N
N
N
N
N
THF
70 °C
100 °C
Cl
I
13
14
12
15
, 11%
, 83%
, 42%
4
ACS Paragon Plus Environment

Page 5 of 40
The Journal of Organic Chemistry
1
2
3
4
Scheme 2. Preparation of Asymmetric Pyridazine 12.
5
6
7
8
Molecular rotors 2 and 3 were made in one-pot two-step syntheses from the bissilyl
derivative 1637,38 (Scheme 3). Addition of TBAF in THF caused immediate cleavage of both
TMS groups, and the free diyne 17 underwent Sonogashira coupling either with 9 to give 2 in
82% yield, or with 12 to provide 3 in 73% yield.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1. TBAF, THF, 25 °C
2. Pd(PPh3)4, CuI
Et3N, THF, 40 °C
N N
N N
N N
N N
9
2
3
, 82%
TMS
TMS
12
16
, 73%
Scheme 3. Synthesis of Rotors 2 and 3.
The next focus was the synthesis of the tetrayne 18 from the diyne 1737 (Scheme 4).
Stepwise construction of two additional triple bonds started with double lithiation of 1737
followed by reaction with solid paraformaldehyde, which resulted in the diol 19. Dess-
Martin periodinane oxidized 19 to the dialdehyde 20 (with a distinctive mushroom-like odor),
whose reaction with CBr4 in presence of PPh3 gave the tetrabromo derivative 21 in high
yield. Final dehydrobromination with n-BuLi in ether provided the volatile and highly
explosive tetrayne 18 as white crystalline solid with a characteristic acetylene-like odor
(Scheme 4).
Br
OH
O
Br
1. n-BuLi
2. (CH2O)n
3. H2O
CBr4
PPh3
Dess-Martin
periodinane
1. n-BuLi
2. NH4Cl/H2O
CH2Cl2
0 °C
Et2O
Hexane
-78 °C
CH2Cl2
25 °C
THF
Hexane
-78 °C 25 °C
Br
O
HO
Br
21
17
19
20
, 73%
18
, 82%
, 70%
, 86%
Scheme 4. Construction of Tetrayne 18.
Finally, long molecular rotors 4 and 5 were prepared from tetrayne 18 and iodo
derivatives 9 or 12 using standard Sonogashira coupling (Scheme 5). Both compounds were
obtained as white crystalline solids, which detonate at temperatures above 200 °C.
Pd(PPh3)4, CuI
Et3N, THF, 25 °C
N N
N N
N N
N N
9
4
5
, 52%
, 68%
18
12
Scheme 5. Assembly of Rotors 4 and 5.
5
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 40
1
2
3
4
5
6
7
8
9
Crystallography. Single crystals suitable for X-ray diffraction were obtained for two
molecular rotors (1 and 2) and four intermediates, 9, 18, 19, and 21. The basic
crystallographic parameters are summarized in Table S1 (Supporting Information).
Rotors 1 and 2 crystallized readily from CH2Cl2 solutions. In contrast, all attempts to
obtain suitable single crystals of the other three rotors 3 - 5 failed. This is in agreement with
previous observations where linear rigid 1,3-diethynylbicyclo[1.1.1]pentane (17) derivatives
longer than ~19 Å usually provided microcrystalline material, not suitable for standard X-ray
diffraction.
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
The highly explosive hydrocarbon 18 crystallized from CH2Cl2 solution in form of
colorless short thick needles which turned brown/black after less than one hour. Low
stability of these crystals significantly complicated their X-ray analysis and final structure
suffers from a relatively high R factor of 9.31.
Compounds 1, 2, 19, and 21 crystallized in monoclinic crystal system and P 21/n, C
2/c, P 21/n, and P 21/c space groups, respectively. Compound 9 formed orthorhombic crystals
with P 212121 space group, while 18 crystallized in tetragonal crystal system and the P 42bc
space group.
Selected interatomic distances for bicyclo[1.1.1]pentane (BCP) derivatives are
summarized in Table 1 (Figure 2). The length of bond α is dictated by the substituent R and
varies from 1.378 to 1.466 Å. As expected, the conjugated bonds in 2, 18, and 21 are shorter
(1.378 - 1.441 Å) than the one adjacent to the CH2OH group in 19 (1.466 Å). Interatomic
distances β, γ, and δ are almost constant and comparable to those in previously published
structures.38 The length of triple bond β varies from 1.194 to 1.198 Å. The length of the
exocyclic bridgehead C-C bond γ is 1.438 - 1.446 Å, and the inter-bridgehead distance δ lies
between 1.881 and 1.888 Å.
Table 1. Selected Interatomic Distances in BCP Derivatives 2, 18, 19, and 21.
Selected Interatomic Distances (Å)
Cmpd.
R
α
β
γ
δ
2
6-tBu-pyridazin-3-yl
CCH
1.441
1.378
1.466
1.419
1.195
1.198
1.194
1.194
1.446
1.438
1.446
1.445
1.884
1.881
1.886
1.888
18
19
21
CH2OH
CHCBr2
Figure 2. Labels of interatomic distances
used in Table 1.
Preparation and Characterization of Bulk Inclusions. Bulk inclusion compounds
are referred to as X%Y@TPP-d12, where X represents the molar percentage of rotor Y (1 - 5).
They were prepared by ball-milling neat guests 1 - 5 with freshly synthesized and solvent-
free hexagonal TPP-d12 and subsequent annealing at 70 °C for 24 h. Each guest compound
was used for the synthesis of two bulk inclusions differing in guest concentration (Table 2):
unsaturated (incompletely guest-loaded) and saturated (TPP channels are full and the surplus
of guest is located outside). Maximum guest loading was estimated as described in
6
ACS Paragon Plus Environment

Page 7 of 40
The Journal of Organic Chemistry
1
2
3
4
5
Supporting Information, page S2. It depends on the length of the guest molecules and is ~30
mol% for the shortest rotor 1 (16.7 Å), and only ~15 mol% for the longest one (5, 33.0 Å).
6
7
Table 2. Bulk Inclusion Compounds Prepared.
8
9
Estimated Maximum
Guest Length
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
Cmpd.
Guest Loading
(molar%)
Sample Namea
(Å)
30%1@TPP-d12 (unsaturated)
40%1@TPP-d12 (saturated)
15%2@TPP-d12 (unsaturated)
30%2@TPP-d12 (saturated)
20%3@TPP-d12 (unsaturated)
25%3@TPP-d12 (saturated)
15%4@TPP-d12 (unsaturated)
30%4@TPP-d12 (saturated)
15%5@TPP-d12 (unsaturated)
25%5@TPP-d12 (saturated)
1
2
3
4
5
16.7
31
22
18
18
15
22.7
27.9
27.8
33.0
a Degree of saturation was confirmed by 13C solid-state NMR.
Solid-state NMR was used to (i) characterize the bulk inclusions, (ii) study cross-
polarization dynamics to understand spectral irregularities, and (iii) demonstrate that all parts
of the rotor molecule rotate inside the channel and determine the lower limit to rotation rate
in (15%5@TPP-d12).
(i) Characterization of Bulk Inclusions. The NMR behavior of all inclusion
13
31
compounds was studied by C cross-polarization magic-angle spinning (CP MAS), P CP
31
MAS, and P single pulse excitation (SPE). In Table S3 (Supporting Information), solid-
state 13C NMR chemical shifts of neat rotors 1 - 5 and their unsaturated inclusions in TPP are
compared with chemical shifts of the rotors in CDCl3 solution, using atom labels shown in
Chart 1.
13
In general, solid-state NMR spectra of neat 1 - 5 resemble their solution C NMR
spectra, but some signals are slightly upfield or downfield shifted. The only exception is
13
compound 2 whose signals in C solid-state NMR spectrum are doubled due to presence of
two inequivalent shielding environments within the unit cell.
Spectra of all inclusion compounds contain three additional peaks due to hexagonal
TPP. All C resonances of included molecular rotors are upfield shifted with respect to
13
CDCl3 solutions. Saturated samples contain an additional set of peaks due to guest molecules
located outside TPP channels, with chemical shifts comparable to those of neat guests.
In all cases, both 31P CP MAS and 31P SPE NMR spectra contain a single signal with
a chemical shift ~32 ppm, characteristic of hexagonal TPP.
13
Rotor 1. The C NMR spectrum of 1 in CDCl3 solution (Figure 3A) consists of
seven well resolved peaks (Table S3 in Supporting Information and Chart 1). Two are due to
the tert.-Bu group (29.9 ppm for CH3 and 37.2 ppm for C), one at 88.8 ppm (C≡C), and four
at 122.5, 130.0, 144.9, and 169.2 ppm (pyridazine). Almost all of these peaks are slightly
shifted downfield in the 13C CP MAS NMR spectrum of neat solid 1 (Figure 3B).
7
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 8 of 40
1
2
3
4
5
6
7
8
9
The 13C CP MAS NMR spectrum of the unsaturated inclusion 30%1@TPP-d12
(Figure 3C) resembles some of the features of the spectrum of neat 1, but also contains three
additional peaks marked as TPP, which are due to the three crystallographically unequivalent
aromatic carbon atoms present in hexagonal TPP-d12. The triple bond carbon atom at 88.8
ppm in solution is not visible in the inclusion because of inefficient cross-polarization
dynamics caused by its longer distance from protons. All remaining carbon resonances of the
guest molecule are shifted upfield by 0.5 to 3.2 ppm relative to the isotropic spectrum in
CDCl3 solution.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
A slight excess of 1 in the saturated inclusion 40%1@TPP-d12 (Figure 3D) gives rise
to three additional peaks in the 13C CP MAS NMR spectrum, which represent all three
structural motifs in compound 1: the pyridazine core (C at 169.2 ppm), the triple bond (C at
89.6 ppm), and the tert.-Bu group (C at 37.5 ppm). All these peaks have identical or slightly
upfield shifted chemical shifts relative to 1 in solution. The remaining four resonances of
unincluded guest molecules are overlapped either with TPP signals or with peaks of included
1. The peaks marked with an asterisk in Figure 3 (panels B, E, F, G, and H) are spinning side
bands.
31
The P CP MAS NMR spectra of both 30%1@TPP-d12 and 40%1@TPP-d12 (Figure
3E and G) contain one singlet at ~32 ppm, which is visible because of polarization transfer
31
from included hydrogen atom carrying guest molecules. The corresponding P SPE NMR
spectrum (Figure 3F and H) also consists of one peak at ~32 ppm, which originates from
three crystallographically equivalent phosphorus nuclei present in hexagonal TPP-d12.
8
ACS Paragon Plus Environment

Page 9 of 40
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. 13C NMR spectrum of 1 in
solution (A), 13C CP MAS NMR of neat 1
(B), 30%1@TPP-d12 (C), and 40%1@TPP-
d12 (D). 31P SPE NMR: 30%1@TPP-d12 (E),
40%1@TPP-d12 (G), and 31P CP MAS
NMR: 30%1@TPP-d12 (F), 40%1@TPP-d12
(I). Asterisks label spinning side bands.
Rotor 2. Compound 2 consists of four characteristic units (tert.-Bu groups, pyridazine
cores, triple bonds, and BCP cage), all easily identifiable in the 13C NMR spectrum in CDCl3
solution (Figure 4A and Table S3 in Supporting Information).
In the 13C CP MAS NMR spectrum of neat 2, all carbon atom resonances are doubled,
except for signals of the BCP cage and the tert.-Bu group (Figure 4B), indicating that two
crystallographically different environments are present in the unit cell. The signal of the BCP
quaternary bridgehead carbon atoms is overlapped with three equivalent methyls of the tert.-
Bu group.
The 13C CP MAS NMR spectrum of the unsaturated inclusion 15%2@TPP-d12
contains three additional peaks due to TPP (Figure 4C). All four pyridazine carbon
resonances are shifted upfield (165.6, 143.2, 127.0, and 119.0 ppm) relative to the solution
NMR spectrum (168.4, 145.4, 129.4, and 122.2 ppm). Both C≡C signals are present in the
9
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 10 of 40
1
2
3
4
5
6
7
8
9
solution spectrum at 77.3 and 91.6 ppm, but are absent in the spectrum of 30%2@TPP-d12.
Three equivalent CH2 groups in BCP give one peak at 56.7 ppm, shifted upfield relative to
the solution spectrum (58.9 ppm). The same trend is also observed for both resonances of the
tert.-Bu group (C at 34.2 ppm and CH3 at 28.5 ppm), which are shielded and thus shifted
upfield relative to 13C NMR of 2 in CDCl3 solution (37.0 and 29.9 ppm).
13
The C CP MAS NMR spectrum of the saturated inclusion 30%2@TPP-d12 is more
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
complex than that of the unsaturated sample 15%2@TPP-d12 (Figure 4D), because it contains
additional four signals (one pyridazine C at 168.9 ppm, a weak C≡C signal at 77.3 ppm, three
equivalent CH2 groups of the BCP cage at 59.0 ppm, and C of the tert.-Bu group at 36.9
ppm). They arise from an excess of 2, located outside the TPP channels. The signals of the
remaining carbon atoms of unincluded 2 are overlapped with either TPP resonances or those
of included 2.
The P SPE and P CP MAS NMR of both 15%2@TPP-d12 and 30%2@TPP-d12
(Figures 4E - H) again contain only one singlet at ~32 ppm. This shows that both inclusion
compounds contain only the hexagonal form of TPP and that hydrogen atom carrying guest
molecules are located in the close proximity to the phosphorus nuclei.
31
31
Figure 4. 13C NMR spectrum of 2 in
solution (A), 13C CP MAS NMR of neat 2
(B), 15%2@TPP-d12 (C), and 30%2@TPP-
10
ACS Paragon Plus Environment

Page 11 of 40
The Journal of Organic Chemistry
1
2
3
4
5
6
d12 (D). 31P SPE NMR: 15%2@TPP-d12 (E),
30%2@TPP-d12 (G), and 31P CP MAS
NMR: 15%2@TPP-d12 (F), 30%2@TPP-d12
(H). Asterisks label spinning side bands.
7
8
9
Compounds 3, 4, and 5 differ from 2 only in the numbers and/or positions of triple
13
31
bonds. Their C and P NMR spectra and spectra of their inclusion compounds were
interpreted in an analogous manner and do not require additional comments. The spectra and
13C chemical shifts are listed in Table S3 and Figures S3, S4, and S5 (Supporting
Information).
(ii) Cross-polarization (CP) Dynamics. The CP dynamics was studied for 2 and its
inclusion complex 15%2@TPP-d12. The dependence of the individual signal intensities on
the length of CP contact time was measured in the range of 0.5 - 9 ms.
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
The signals of protonated carbon atoms a, f and g in the spectra of neat 2 have more
than 50% of their maximum intensity already after a contact time of 0.5 ms (Chart 1 and
Figure 5A). This behavior is expected because the molecules of 2 do not exhibit any sign of
dynamics in their NMR spectra and the carbon nuclei are therefore strongly coupled to the
directly attached hydrogen atoms. The signals of carbon atoms e, h and i, which are two
bonds away from a hydrogen atom, have low intensity after a 0.5 ms contact time, but this
intensity grows rapidly as the contact time increases. This observation fits slower CP
dynamics associated with the longer distance to the nearest hydrogen atom (~2.2 Å). The
signals of triple bond carbon atoms c and d, which are separated by three bonds from the
nearest hydrogen atoms, grow even more slowly with increasing contact time (Chart 1 and
Figure 5A). This is the result of inefficient CP given by even longer C-H distances. The c
and d carbon atoms are separated by 2.7 and 3.1 Å, respectively, from the nearest hydrogen
atom in the same molecule of 2; this distance is comparable with the closest intermolecular
contacts with hydrogen atoms (e.g. 2.9 Å between c and hydrogen atoms attached to carbon
atom a from a neighboring molecule) and in this case both intra- and inter-molecular CP
probably contribute to the signal. The slowest growth with increasing contact time is
observed for the signal of the tert-butyl carbon atoms j. It is known that rapid rotation of tert-
butyl groups attenuates the CP process.39
13
In the C CP MAS spectra of 15%2@TPP-d12, new signals corresponding to the
carbon atoms of TPP are present (Figure 4). Similarly as in neat 2, signals of the protonated
carbons a, f and g have high intensity already at short contact times (Chart 1 and Figure 5B).
Carbon atoms h and i separated by two bonds from the nearest hydrogen atom exhibit a weak
signal at short contact times, but this signal grows rapidly at longer contact times (the signal
of e, also separated by two bonds from a hydrogen atom, is hidden under one of the signals of
TPP). The signal of tert-butyl carbon atoms j also exhibits a slow increase of intensity due to
insufficient CP. The most remarkable difference in the variable contact time experiments on
neat 2 and 15%2@TPP-d12 is observed for the signals of the triple bonds c and d, whose
intensity increases very slowly with increasing contact time. This observation may be
attributed to insufficient intramolecular CP from the closest hydrogen nuclei, further
attenuated by fast rotation of the pyridazine rings that leads to smaller C-H coupling and,
furthermore, by the absence of intermolecular interactions with hydrogen atoms. The
attenuated signals of the triple bonds may thus serve as additional evidence for the formation
of an inclusion complex. The signals of TPP also grow at a similar slow rate and this
indicates that the shortest distance between TPP carbon atoms and the hydrogen atoms of 2 is
similar to that between the triple bond carbon atoms and the closest hydrogen atoms (ca
11
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 12 of 40
1
2
3
4
5
6
7
8
9
3.0 Å). Note also that the changes in signal intensities upon the formation of the inclusion
compound are not caused by different relaxation behavior of the signals. All the variable
contact time experiments were performed with the repetition time of 4 s, but a control
experiment with a repetition time of 30 s provided similar signal intensities as that with the
4 s repetition time for both 2 and 15%2@TPP-d12.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. The dependence of relative 13C CP MAS signal intensity (normalized to the
maximum intensity of a) on the contact time for crystalline 2 (A) and 15%2@TPP-d12
(B).
13
The triple-bond carbon signals can be observed in C NMR spectra of the inclusion
compounds acquired with direct carbon excitation without CP. This is documented in Figure
6 for compound 5 and its inclusion complex 15%5@TPP-d12. The triple bond signals are
missing in the CP MAS spectrum of 15%5@TPP-d12 with contact time of 4 ms (Figure 6C),
but are clearly present in the spectrum acquired with direct carbon excitation (Figure 6D).
Furthermore, the intensity of TPP signals is higher in the direct excitation spectrum because
the intensity is not driven by intermolecular C - H interactions any more.
12
ACS Paragon Plus Environment

Page 13 of 40
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6. 13C solid-state NMR spectra of 5
in CDCl3 solution (A), neat 5 (B), and
15%5@TPP-d12 measured with cross-
polarization (C) and direct excitation (D) of
carbon nuclei.
(iii) Lower Bound for Rotational Frequency. Fast molecular re-orientation in
solutions averages anisotropic chemical shielding to a single isotropic value. In solids
molecular reorientation is suppressed and NMR spectra of static powder samples consist of
broad lines that correspond to sums of signals from nuclei with all possible orientations
relative to the magnetic field direction. Rotation of the sample at the magic angle (54.7°)
with respect to this direction partially averages the molecular orientations and the broad
solid-state NMR signals break up into a set of lines separated by the spinning frequency.
Magic angle spinning (MAS) at high frequencies leaves only the isotropic line in the
spectrum accompanied by a few low-intensity spinning side-bands (SSBs). Molecular
mobility in the solid material leads to a suppression of the chemical shift anisotropy40 and
therefore, the number and intensity of SSBs may be used to diagnose internal motion.
In the room-temperature 13C NMR spectra of neat rotors acquired at 13 kHz MAS, the
dominant lines are those corresponding to isotropic chemical shifts and the few observed
SSBs belong mostly to the signals of triple bonds carbons, which have a very large chemical
shift anisotropy. However, when the spectra are acquired at 4.3 kHz MAS, SSBs associated
with all carbon signals appear. In many cases, the SSBs have even higher intensity than the
signals at isotropic chemical shifts, as illustrated for 5 in Figure 7. The lowest intensity SSBs
are those associated with the CH2 carbon atoms of the BCP cage and CH3 carbon atoms of
the tert-butyl groups, due to rotational mobility of these groups in the crystalline state of neat
5. On the other hand, many SSBs are present in the signals of the pyridazine carbon atoms,
which indicates that these aromatic rings do not undergo π jumps or any kind of rotation.
In the spectrum of the inclusion complex 15%5@TPP-d12 acquired at 13 kHz MAS,
only very few SSBs are present and they belong to the signals of TPP (black asterisks in
13
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 14 of 40
1
2
3
4
5
6
7
8
9
Figure 7C). In the spectrum acquired at 4.3 kHz, many more SSBs appear and are mostly due
to TPP (Figure 7D). The SSBs associated with 5 have very low intensity or are missing
altogether. This observation confirms the formation of an inclusion complex; the TPP host is
rigid and provides signals with many SSBs in slow MAS spectra, but the guest molecule is
relatively mobile and rapidly reorients all of its parts, thus suppressing the effect of chemical
shift anisotropy and reducing the number of SSBs. This suppression shows that the speed of
the molecular reorientation is at least comparable to chemical shift anisotropy expressed in
Hz or faster, thus setting the lower limit of the rotational rate at ~10 kHz at 300 K.40,41 It is
likely that the other rotors behave similarly since (i) the interior of the TPP channel is the
same in all cases, (ii) almost all rotors (except 1) shares the same building blocks (tert.-Bu,
BCP, pyridazine rings, triple bonds) and differ only in length and (iii) experimentally
obtained rotational barriers are distributed in quite narrow region.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 7. 13C CP-MAS spectra of 5 and
15%5@TPP-d12 measured with 13 kHz (A
and C) and 4.3 kHz MAS (B and D). The
blue dashed lines connect the isotropic lines
in the spectra of 5 and the asterisks indicate
the SSBs (black for TPP signals and red for
15%5@TPP-d12 signals).
Differential Scanning Calorimetry (DSC). The samples were analyzed in a cyclic
mode starting by heating from room temperature to 310 °C, followed by cooling to 40 °C,
and heating to 310 °C again. Since all samples decomposed before reaching 310 °C the
subsequent cooling and heating cycles resulted in featureless traces. Therefore, only the first
heating curves are shown in Figure S1 (Supporting Information).
The trace of neat hexagonal TPP-d12 agrees with previously published data.42 The
inconspicuous broad exotherm with a maximum at ~130 °C is due to a first-order transition
of hexagonal TPP-d12 into its more stable monoclinic form. The endotherm at 220 °C is
14
ACS Paragon Plus Environment

Page 15 of 40
The Journal of Organic Chemistry
1
2
3
4
5
associated with a transition of monoclinic TPP-d12 into its hexagonal form and the intense
endotherm at 247 °C is due to melting.
In general, DSC traces of the neat guests 1 - 5 (blue lines B, D, F, H, and J in Figure
S1) revealed that only the shortest rotor 1 melts. The BCP containing compounds 2 - 5
decomposed without melting as demonstrated by strong exotherms in their DSC traces. Not
surprisingly, extension of the guest length by addition of triple bonds decreases the stability
of the compounds: the shortest BCP based rotor 2 decomposes at ~290 °C and the longest
rotor 5 with four more triple bonds already at ~240 °C. Compounds 4 and 5 detonate during
DSC analysis.
The DSC traces of inclusion compounds usually consist of an endotherm attributed to
melting at ~245 °C, which is overlapped with a broad exotherm peak at 240-290 °C
corresponding to decomposition of the guest molecules. See Supporting Information for
detailed description of individual DSC traces.
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Powder X-Ray Diffraction. X-ray powder patterns were acquired for all ten
inclusion compounds (Table 2). All the samples showed patterns consistent with a hexagonal
bulk inclusion compound with the lattice parameters and uncertainties listed in the Table 3.
The hexagonal structure is characterized by two equal “in-plane” lattice parameters a (Å),
and a distance between planes c (Å). No unassigned peaks appear in any of the patterns with
the exception of 25%5@TPP-d12, which has a single broad peak of unknown origin located at
q = ~1.5 Å-1, consistent with a lattice spacing of roughly 4.2 Å (Figure S6 in Supporting
Information).
Table 3. Hexagonal lattice parameters for inclusion compounds.a
Cmpd.
1
Inclusion
a (Å)
c (Å)
30%1@TPP-d12
40%1@TPP-d12
15%2@TPP-d12
30%2@TPP-d12
20%3@TPP-d12
25%3@TPP-d12
15%4@TPP-d12
30%4@TPP-d12
15%5@TPP-d12
25%5@TPP-d12
11.770 ± 0.001
11.838 ± 0.001
11.811 ± 0.001
11.853 ± 0.001
11.693 ± 0.001
11.715 ± 0.001
11.736 ± 0.001
11.757 ± 0.001
11.673 ± 0.001
11.717 ± 0.001
9.948 ± 0.002
10.021 ± 0.002
2
3
4
5
9.908 ± 0.001
9.930 ± 0.001
9.980 ± 0.002
10.026 ± 0.002
9.960 ± 0.001
9.994 ± 0.002
9.926 ± 0.001
9.968 ± 0.002
a Lattice parameters for guest free hexagonal TPP: a = 11.454(5) Å, and c = 10.160(5) Å.43
Figure 8 represents a characteristic diffractogram of bulk inclusion in hexagonal TPP
(15%2@TPP-d12). All remaining diffractograms are shown in the Supporting Information
(page S10, Figure S6).
15
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 16 of 40
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 8. PXRD of 15%2@TPP-d12.
Dielectric Spectroscopy. We measure the motion of the dipole rotators in the
inclusion compounds directly by loading powdered inclusion material onto the surface of a
thin-film interdigital capacitor (see Supporting Information) and then measuring the
frequency- and temperature-dependent electric susceptibility as the difference between the
inclusion-loaded and the original empty capacitor. The “in-phase” or real part of the
capacitance measures the ability of the material to produce a dipole moment per unit volume
while the “out-of-phase” or imaginary part of the capacitance measures dielectric energy loss
processes.
All ten inclusion compounds (Table 2) show characteristic dielectric loss peaks and
changes in the capacitive response. Figure 9 shows the real part and imaginary part of the
capacitance for 30%1@TPP-d12. We found a single loss peak, an associated drop in the real
part, consistent with freezing out of the dipole reorientation, and frequency-dependent
differences in the response that are qualitatively consistent with Arrhenius thermal activation
over barriers. Simultaneous fitting of the data sets to the Debye model of independent
dipoles for populations of 9-state rotors (described in detail in the Discussion section) allows
us to extract a distribution of activation energy barriers to dipole rotation around two central
values of 2.4 ± 0.1 kcal/mol and 2.0 ± 0.1 kcal/mol, and a central attempt time of 0.5 × 10-13
s. The two populations both favor the anti-ferroelectric alignments of local dipoles. The two
populations differ in that one shows an asymmetry of roughly 106 K (listed as A in Table 4 to
indicate an anti-ferro aligned 9-state), while the other is indistinguishable from non-
interacting dipoles. Two populations of dipoles are found for all ten inclusion compounds.
Their barrier heights, attempt times, and asymmetries are listed in Table 4.
Table 4. Dipole rotation parameters - logarithm of the attempt times (log τ0), energy barriers
(EB), and barrier asymmetries (S). Letter “A” and “F” represents systems allowing antiferro-
and ferro-alignement.
EB1
EB2
a
Cmpd.
1
Inclusion
log τ0
S1 (K)c
S2 (K)c
(kcal/mol)b
(kcal/mol)b
30%1@TPP-d12
40%1@TPP-d12
15%2@TPP-d12
30%2@TPP-d12
-13.7
-13.5
-12.2
-13.7
2.4
2.5
2.7
3
105 (A)
118 (A)
147 (A)
167 (A)
2.0
2.1
2.1
2.4
0 (A/F)
0 (A/F)
32 ± 6 (F)
42 ± 20 (F)
2
16
ACS Paragon Plus Environment

Page 17 of 40
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
20%3@TPP-d12
25%3@TPP-d12
15%4@TPP-d12
30%4@TPP-d12
15%5@TPP-d12
25%5@TPP-d12
-13.0
-13.2
-13.0
-12.2
-12.6
-12.8
3.2
3.3
2.1
1.9
3.5
3.2
118 (A)
120 (A)
184 (A)
209 (A)
176 (A)
229 (A)
2.4
2.6
2.3
2.1
2.9
2.9
40 ± 40 (F)
0 (F)
3
4
5
57 (A)
72 (A)
38 (A)
74 (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
Standard error is: a ± 0.2, b ± 0.1 kcal/mol, c ±5 K.
Figure 9. Dielectric loss of 30%1@TPP-d12
between 10 and 250 K. Black, 100 Hz; red,
1 000 Hz; blue, 10 000 Hz. Real (A) and
imaginary (B) capacitance.
Relative dielectric susceptibility for the five saturated inclusions as a function of
temperature is displayed in Figure 10. For comparison with the TPP background, we also
show the dielectric susceptibility temperature dependence for an inclusion of benzene, a
nonpolar aromatic hydrocarbon (benzene@TPP-d12). All data sets are at 10 kHz.
17
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 18 of 40
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 10. Relative dielectric susceptibility
for five polar saturated inclusion compounds
(40%1@TPP-d12,
30%3@TPP-d12,
30%2@TPP-d12,
30%4@TPP-d12, and
25%5@TPP-d12), and nonpolar inclusion
compound of benzene@TPP-d12.
Calculations. To obtain a rough representation of the time-averaged rotational
potential in which the two pyridazine rings of a rotor molecule move and to minimize the
effects of TPP lattice truncation, a model was used in which a central TPP channel containing
the rotor under study (shown in orange in Figure S8 in Supporting Information) was
surrounded by six other channels and all seven channels are filled with molecules of 1.
Completely relaxed geometry optimization was performed for a slab consisting of four layers
of TPP by the MP6-D3H434 method, which includes van der Waals interactions. The
instantaneous geometry obtained in this manner does not have three-fold symmetry, whereas
the time-averaged system does. To obtain an approximation to the symmetric time-averaged
potential, the sixteen models shown in Figure S9 were averaged (in these models, the central
pyridazine is replaced with 1,2,4,5-tetrazine, similar in shape to pyridazine, but without a
dipole moment and structural distortion). The geometries of the six outer channels and their
guests (shown in blue) that were obtained in this fashion were then fixed at the averaged
values. The dihedral angles φ1 and φ2 of the two pyridazine rings of 1 were fixed in steps of
15° and the remainder of the central TPP channel was fully relaxed to locate the optimal
conformation of 1. The rotational potential was calculated as a function of φ1 and φ2. The
simulations were performed with the program MOPAC2016.44 The computational procedure
is crude compared with a possible more demanding molecular dynamics treatment but is
expected to provide an idea of the potential energy surface for pyridazine rotation.
Figure 11A shows the approximate contour lines of a such a surface calculated by the
B97-D3/SVP method35 for the two-dimensional subspace of geometries defined by the
rotational dihedral angles φ1 and φ2 (measured clockwise) that the two pyridazine rings of 1
inserted into the central channel form with respect to a line that bisects two opposite edges of
the hexagon formed by the channel walls when viewed along the channel axis (at φ1 = φ2, the
planes of the pyridazine rings are parallel and their dipoles point in the same direction). All
other geometrical degrees of freedom in the central channel are optimized.
The nine minima in the potential energy surface are surrounded by dark blue contour
lines and are located close to φ1 values of 90, 210, and 330, and φ2 values of 30, 150, and 270.
The maxima are surrounded by dark red lines. The shape of the surface exhibits three-fold
symmetry with respect to both φ1 and φ2. The expected (see Supporting Information) nine
low-energy conformations differing by ~60° between the two pyridazine rings located in
18
ACS Paragon Plus Environment

Page 19 of 40
The Journal of Organic Chemistry
1
2
3
4
5
6
adjacent layers of TPP are clearly present. To facilitate visualization, Figure 11 also shows
views along the channel direction of the dipoles associated with the two pyridazine rings (a
green and an orange arrow) at selected values of φ1 and φ2 (minima in black and saddle points
in green hexagons).
7
In measurements of dielectric losses, Debye peaks are expected for those jumps
between conformations that change the projection of the molecular dipole moment of 1 into
the direction of the electric field. The field direction was chosen to be perpendicular to the
channel axes and parallel to the reference direction from which φ1 and φ2 are measured. In
Figure 11B, we provide a map of the magnitude of this projection (in Debye) as a function of
φ1 and φ2. Those jumps that connect φ1 and φ2 values that differ in color should produce
signals. Examples are jumps from (330,30) to (330,150) and to (210,30).
Jumps connecting values of the same color should be silent in dielectric spectra. An
example is a jump from (210,150) to (210,270). Jumps from one minimum to another are
easiest along lines passing through saddle points. These lines are oriented roughly vertical or
horizontal, implying that only one of the pyridazines prefers to rotate at a time. The
transition states separating two energy minima are located in regions defined by light blue
and cyan contour lines and lie 4 - 5 kcal/mol above the minima; activation energies are
expected to be somewhat lower. For example, the saddle point at (140,150) interconnects
minima at (90,150) and (210,150).
A B97-D3/SVP calculation in a six-layer TPP model for the energy profile for sliding
a molecule of 1 from one position to the next within the TPP channel, with pyridazine rings
relaxed to their best orientations at each point along the channel axis, showed a relatively
high barrier of ~10 kcal/mol. It thus appears that the motion of 1 along the axis is much
slower than rotation of its rings, which only has a barrier of about 4 - 5 kcal/mol. As a result
of an unsymmetrical optimal location of 1 with respect to the center of the TPP layer, the two
pyridazine rings have different environments and the rotation angles φ1 and φ2 remain
inequivalent.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
19
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 20 of 40
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 11.
A: B97-D3/SVP calculated
potential energy surface (in kcal/mol) for
rotation of 1 inserted in the central channel of a
four-layer TPP slab surrounded by six other
channels filled with 1. The two-dimensional
subspace of geometries shown is defined by the
rotational angles φ1 and φ2 (see text). Above,
examples of dipole orientations at various
rotational angles φ1 and φ2, viewed along the
channel axis. B: Projection of the molecular
dipole moment of 1 in Debye into the direction
φ1 = φ2 = 0.
DISCUSSION
Synthesis. Preparation of rotor molecules 2 - 5 is an extension of the already well
developed chemistry of 1,3-diethynylbicyclo[1.1.1]pentane (17)18,37,38,45,46,47,48,49 and thus
does not require much comment. Two issues will be mentioned briefly.
(i) The nonvolatile and bench-stable silyl derivative 16 was identified as a very useful
source of the volatile diyne 17, which is explosive in crystalline state. Compound 17 can be
liberated from 16 in-situ using TBAF in THF. This completely avoids manipulation of the
20
ACS Paragon Plus Environment

Page 21 of 40
The Journal of Organic Chemistry
1
2
3
4
5
volatile and potentially dangerous diyne 17 and significantly simplifies further synthetic
operations.
(ii) The tetrayne 18, readily available from the diyne 16 in four synthetic steps using
Corey-Fuchs synthesis, was purified by chromatography on silica gel with pentane. This
purification significantly improved the yields of subsequent synthetic steps. It is advisable to
store the strained tetrayne 18 with a small amount of solvent, since it undoubtedly is a highly
energetic compound, and it is strongly recommended to perform all manipulations with it
only in solution. Solid microcrystalline 18 spontaneously detonates at temperatures above ca
85 °C (attempted determination of melting point of three tiny crystals with total weight <1
mg caused a detonation comparable with that of ca 30 mg of NI3). It is expected that
mechanical stress including crushing of larger crystals can initiate explosion even at much
lower temperatures.
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Inclusion Compounds. TPP is known to be available in two crystal structures.50
A
monoclinic form is the stable phase for pure TPP at ambient conditions up to ~150 °C and
has no free space for guest inclusion. A hexagonal form is the stable phase above ~150 °C
for pure TPP and is the structure adopted upon inclusion of guests at any temperature. The
included guest molecules stabilize it against collapse to the monoclinic form.51,52 The
observed powder X-ray patterns of the inclusion compounds show only the hexagonal
structure and no evidence for the presence of any monoclinic TPP. All samples show a 2 -
3% expansion of the in-plane lattice parameter a and a small contraction (1 - 2%) of
parameter c, the distance between layers. This agrees with the expected structures of bulk
inclusions, because guest molecules in general cause a small in-plane expansion of the TPP
host matrix.18,19,27,28 In cases where the guest mole fraction is higher than expected for
complete inclusion, peaks associated with crystals of the neat rotor molecules were not
detected. Perhaps the excess guest molecules remain amorphous.
The inclusions of compound 2 produce the largest in-plane lattice parameters (Table
3), which indicates more available space in the TPP channels. This may be an important
contribution to the lower rotational barriers also found for this case.
The TEM analysis revealed that the morphology of 30%1@TPP-d12, which was
chosen as representative example, is comparable with previously prepared bulk inclusions.28
The sample consists of disc-shaped particles with ~50 nm diameter and ~25 nm thickness.
One ideal and defect-free discoidal TPP particle with such size can accommodate ~24 000
molecules of rotor 1 or ~14 000 molecules of 5 (see Supporting Information for additional
detail).
Solid-State NMR. The methods discussed so far provided information regarding
basic properties of the inclusion compounds and yielded some hints about the surface vs. bulk
type of inclusion, but were not able to distinguish between unsaturated and saturated samples,
nor did they provide information about positions of the guest molecules inside TPP channels.
Solid-state 31P and 13C analysis of all samples answers these questions.
All solid-state 31P SPE NMR spectra contain only one singlet at ~32 ppm,
corresponding to three crystallographically equivalent phosphorus nuclei present in
hexagonal TPP, and thus reveal exclusively the hexagonal form of TPP. No traces of the
monoclinic polymorph of TPP were detected, in agreement with PXRD.
31
The P CP MAS NMR spectra of all samples again consist of a single peak at ~32
ppm, due to transfer of polarization from hydrogen atoms attached to guest molecules located
in close proximity to the phosphorus nuclei. 13C CP MAS spectra provide additional strong
evidence confirming our identification of the materials as bulk inclusions. In unsaturated
samples, the chemical shifts of all carbon atoms of the guest molecules were upfield shifted
21
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 22 of 40
1
2
3
4
5
6
7
8
9
relative to samples in CDCl3 solutions (Table S3 in Supporting Information), and this is
attributable to shielding by neighboring phenyl rings of TPP molecules. On the other hand,
saturated samples provided two sets of peaks: one upfield shifted and assigned to included
guest molecules, and the other with almost unshifted resonances with respect to CDCl3
solutions and assigned to rotor molecules located outside the TPP crystallites. All solid-state
13C CP MAS NMR spectra of inclusion compounds also contain three additional peaks due to
the TPP matrix, and this is an additional proof that hydrogen atom carrying rotor molecules
are inserted inside TPP channels, because during cross-polarization (CP) experiments
polarization is transferred not only to carbon atoms of guests, but also to carbon atoms of
neighboring perdeuteriated TPP molecules.
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
The 13C CP MAS NMR spectra of all BCP carrying unsaturated inclusions are
characterized by attenuation or complete absence of triple bond carbon signals located next to
the BCP cage. This phenomenon, which was already observed several times in the
past,18,19,27 but was thoroughly studied for the first time now, was attributed to insufficient
intramolecular CP from the closest hydrogen nuclei, additionally attenuated by fast rotation
of the pyridazine rings that leads to very small C-H coupling inside the TPP channel. The
significantly reduced intensity or complete absence of the triple bonds signals may thus serve
as additional evidence for the formation of inclusion compounds.
Pyridazine Rotators: Dynamics and Mutual Interactions. The evidence above
supports a picture where the double pyridazine molecules are included into the bulk of a
hexagonal TPP host. Therefore, the individual pyridazine dipole rotator moieties (i) are
constrained to rotate about the axis of the long TPP channels; (ii) are in rotational potential
energy environments that are locally three-fold, but where neighboring dipole environments
can be either aligned in the three preferred directions or anti-aligned depending on dipole-
dipole distance; (iii) have other pyridazine dipole rotators at varying separation along the
channels and in neighboring channels; and (iv) are incorporated into the final material with
dipole volumetric densities that differ for molecules 1 - 5. The dielectric behavior that we
observe shows the impact of these four points.
We first discuss the scaling of dielectric susceptibility with dipole volumetric density,
where models of independent dipoles, most appropriate for describing the high-temperature
para-electric phase, predict a direct proportionality between the dipole volume density and
the dielectric susceptibility. Fully saturated inclusion compounds of 1 have the most dipoles
per unit length and therefore per volume and these dipole density numbers then decrease in
sequence from 1 down to 5. Figure 10 (see discussion in Supporting Information) shows that
the relative susceptibility indeed sorts all the inclusion compounds with inclusions of 1
having the largest susceptibility and 5 the smallest. The inclusion compounds of 3 and 4 are
interesting cases as they have nearly equal dipole density but different dipole-dipole
separations and local structure. Nevertheless, their inclusion compounds have nearly equal
susceptibilities at higher temperatures. They also show interesting differences, as do the rest
of the inclusion compounds, at lower temperatures, where the local environments and dipole-
dipole interactions begin to play an important role.
Important aspects of the lower-temperature dielectric behavior shown by all the
inclusion compounds are the dielectric loss peaks in the imaginary component of the
dielectric response (Figure 10), the associated Arrhenius transition rates between potential
wells with barriers (Table 4), and the associated drop in dielectric susceptibility as the dipoles
freeze into local potential wells. Inclusion compounds of 3 and 5 show rotational energy
barriers in excess of 3 kcal/mol, whereas inclusions of 1, 2, and 4 have barriers below 3
kcal/mol. The higher rotation energy barriers for 3 and 5, compounds where the pyridazine
22
ACS Paragon Plus Environment

Page 23 of 40
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
moieties are between carbon triple bonds, are consistent with previous observations on rotors
of similar structure but carrying a single pyridazine.28 Rotation about the triple bonds is
expected to have very low barriers in the gas phase, but they increase flexibility and
accentuate non-linearity that apparently allow the dipolar rotators to interact more strongly
with the TPP channel walls. Similarly, the observed barriers near 2 kcal/mol for pyridazine
rotators adjoining tert.-Bu end groups are consistent with observations on rotors carrying a
single pyridazine.28 Taken together, the observed rotation barrier heights and attempt times
all appear to be consistent with the local structure surrounding the pyridazine sections in TPP.
Finally, we turn to a more subtle behavior that appears in the overall temperature
dependence of the dielectric susceptibility, and which we suggest arises from the correlated
alignment of the local three-fold potentials experienced by different dipolar rotators in TPP.
This behavior allows us to measure the impact of dipole-dipole interactions on the dielectric
susceptibility. The behavior we observe appears in the real part of the dielectric
susceptibility (Figures 9A and 10) and is most obvious in the inclusions of 1, 2, and 4. We
observe a systematic tendency of the dielectric susceptibility to slowly arc downwards away
from the typical 1/T Curie-like behavior predicted for independent dipoles in symmetric well
structures. This softening of the initially increasing susceptibility with decreasing
temperature takes place over a broad range of temperatures beginning at perhaps 200 K, leads
to a maximum in the susceptibility between 100 and 150 K depending upon the particular
inclusion, and then decreasing susceptibility at the lower temperatures until dipole rotation is
frozen out by the Arrhenius rates. Modeling this type of softening in dipolar susceptibility
typically proceeds in terms of potential well structures for the dipolar rotation with one or
more lower energy wells for each dipole, wells that eventually capture the dipole population,
leading to a decreased equilibrium population of dipoles that can respond to the applied
electric field.53,54,55 The issue in these cases is to understand how particular low energy wells
arise naturally in a given system. For the double pyridazine molecules in TPP, the
calculations above suggest that the individual dipole rotators move in a locally three-fold set
of potential wells. A natural source of local asymmetry in the well depths is provided by the
dipole-dipole interactions between either intra- or inter-molecule pairs. In dilute systems, the
dominant interaction would be between the two dipoles on one double pyridazine molecule.
For denser inclusions, both intra- and inter-molecular interactions and summations of
energies over many molecules would be necessary. In all cases, the softening of the
dielectric susceptibility in the 100 to 200 K region likely signals that dipole-dipole
interactions are leading to asymmetry in the rotational potential wells and thereby suggest
that the dipole interaction energies are themselves on the scale of roughly 100 K.
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
As we discuss further in the Supporting Information, a given dipole rotator and its
neighbor, each in its respective three-fold potential, can be modeled as a nine-state system.
The TPP environment provides for two distinct types of nine-state situations, involving
dipole pairs that are separated along the channel by a distance equal to either an even or an
odd multiple of the thickness of a TPP layer. For dipole pairs separated by a distance equal
to an even number of TPP layers, the respective three-fold potentials are aligned and allow
the dipoles to have aligned orientations for three of the nine states, or to be oriented at 120
degrees to each other in the remaining six states. The three aligned states have high and
positive-valued dipole-dipole interaction energy, so are high-energy states overall, with a net
dipole moment of twice the single dipole value. The six non-aligned states, each with a net
dipole moment equal to the single dipole value, form a polar ground-state manifold. At low
temperatures, these six polar states allow for a dipole contribution to the dielectric
susceptibility. By comparison, for dipole pairs separated by a distance equal to an odd
23
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 24 of 40
1
2
3
4
5
6
7
8
9
number of TPP layers, the respective three-fold potentials for the dipole pair are rotated 60°
relative to each other. That orientation allows for three of the nine states to have anti-aligned
dipoles, no net dipole moment, and a negative dipole-dipole interaction energy so that these
three states form a non-polar ground state manifold. The six non-anti-aligned states have
equal energy that is higher due to the dipole-dipole interactions, have a net dipole moment of
√3 of the single dipole moments, and can contribute to the dielectric susceptibility at higher
temperatures. The full inclusion system will, of course, be composed of large numbers of
dipole pairs of various separations. However, the main message is that for dipole pairs along
the channels with neighbors an odd number of TPP layers apart (e.g., the intra-molecular
separation for inclusions of 1), dipoles can fall into a low-energy non-polar anti-aligned
configurations that will reduce the dielectric susceptibility. Even pairs with even layer
separation (e.g. neighboring dipoles in saturated inclusions of 2) will show a decreased
number of available polar states at low temperatures.
We have developed two simple statistical mechanics models for these two different
nine-state local models to demonstrate the behavior described above under the assumption
that only the dipole-dipole interactions contribute to the energy asymmetries of the various
configurations. Assuming that the asymmetry is due to dipolar interactions between dipoles
some distance a apart the models yield predicted dipolar contributions to the dielectric
susceptibility for odd number of layers separation, χodd, and even number of layers separation,
χeven, according to Equations 1.
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
χodd ≈ (N/V)(p2/3kBT){3(eS/2kBT)/[2(eS/2kBT) + e(+S/kBT)]}
χeven ≈ (N/V)(p2/3kBT){[(e+S/2kBT) + 2(eS/2kBT)]/[2(e+S/2kBT) + e(S/kBT)]}
S ≈ 3p2/8πεε0a3
(1)
Each shows the expected proportionality to dipole density, the typical Curie-like 1/T
increase with decreasing temperature, and a quotient that describes the equilibrium
populations of the nine states. The odd-layer quotient approaches zero at low temperatures as
dipoles populate the three-state non-polar ground state manifold. The even-layer quotient
approaches a finite value of one-half, as dipoles populate the six-state partly polar ground
state manifold. These susceptibilities obtain until low enough temperatures where the
kinetics of barrier crossing freeze out dipole motion entirely. The susceptibility functions are
identical for zero asymmetry, where they agree with the classical results of Debye56 and
Fröhlich.57 Finite asymmetry leads to a suppression of susceptibility for the odd layer case,
while the even case retains the Curie-like behavior.
Table 5 lists the expected intra- and inter-molecular dipole-dipole separations for 1 
5 for saturated inclusions. None of the intramolecular separations are perfect integer
multiples of the TPP layer spacing. Still, our calculations for the energy surface of 1 (Figures
11) show that the picture of three-fold wells separated by 120° is accurate. Under the
assumption that fractional deviations from perfect integer layer separation are similarly
accommodated for the longer double pyridazine molecules, Table 5 lists the type of local
dipole-dipole ground state configuration we would expect for both the intra- and inter-
molecular dipole pairs. The result clearly shows that we expect a softening of the dielectric
susceptibility below the Curie-like 1/T very much as we observe. We fitted the real and
imaginary parts of the capacitance data to a model that is a superposition of the two
susceptibility functions above. This model takes into account the possibility of both local
24
ACS Paragon Plus Environment

Page 25 of 40
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
non-polar configurations and residual partially polar configurations. We find a satisfactory
description of the overall temperature-dependent shape of both the real and the imaginary
parts as shown e.g., in the fitting of Figure 9. The fitting results listed in Table 4 are in very
good qualitative agreement with dipole-dipole interaction energies for the nearest neighbor
dipoles in the 100 to 200 K range (S1 in Table 4). These systems settle into the anti-aligned
configurations with decreasing temperature and lead to the drop in dielectric susceptibility.
The fitting results suggest that they then leave behind a population of more widely separated
dipoles (lower observed anisotropy energy S2 in Table 4) that then contribute to dielectric
susceptibility until the residual rotation barriers prevent dipole motion at the lowest
temperatures.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 5. Estimated intra-molecular (within one double pyridazine molecule) dipole-dipole
spacing and inter-molecular dipole-dipole spacing (between nearest neighbor double
pyridazine molecules at saturation of the inclusion) in units of the single TPP layer spacing of
5 Å. Included is an estimate listed as either (Aligned) or (Anti-aligned) about whether the
local intra-molecule and neighboring inter-molecule dipoles are in three-fold potential wells
that allow for dipole alignment (an even number of TPP layers) or allow for dipoles to be
anti-aligned (an odd number of TPP layers).
Cmpd.
1
Intra (layers)
Inter (layers)
1.3 (Anti-aligned)
2 (Aligned)
2
3
4
5
2.5 (Aligned)
2.5 (Aligned)
2 (Aligned)
3 (Anti-aligned)
2 (Aligned)
3.5 (Anti-aligned)
3.5 (Anti-aligned) 3 (Anti-aligned)
SUMMARY
Five double pyridazine molecular rotors differing in inter- and intramolecular
pyridazine-pyridazine spacing were synthesized and included as guests in the TPP host to
obtain bulk inclusion compounds. Their formation was confirmed using solid-state NMR and
powder X-ray diffraction. It was proven that the absence of triple bond carbon resonances
located next to the BCP cage is due to inefficient cross-polarization in the spectra of inclusion
compounds and could be used as an additional proof of inclusion. The rotation of pyridazine
rotators was independently studied by dielectric spectroscopy and solid-state NMR.
Dielectric spectroscopy results are well described by a simple model where the pyridazine
dipole moments move in local three-fold rotational potentials. Interestingly, neighboring
dipoles can occupy two different types of correlated three-fold potentials depending upon the
separation of neighbors: For dipoles located approximately an even number of TPP layers
apart, the three-fold potentials are identical and therefore have wells that are at identical
dipole orientations. Dipoles located an odd number of layers apart experience three-fold
potentials that are rotated by 180°, allowing for anti-aligned dipole orientations. The data
sets are well fitted by two different 9-state model susceptibilities, with dipole-dipole
interactions of 100 to 200 K in rough agreement with those predicted for these two distinct
types of system, and which lead to different energies and equilibrium population differences
that match the dielectric susceptibilities observed in the various inclusion compounds. The
25
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 26 of 40
1
2
3
4
5
6
7
8
9
pyridazines substituted by ethynyl units in both 3- and 6- positions have a slightly higher
rotational barrier (3.2 - 3.5 kcal/mol) than those carrying one ethynyl and one tert.-Bu group
(1.9 - 3.0 kcal/mol). These values seem to be still too high for achieving a ferroelectric
ground state. Calculations of rotational barriers for 1@TPP (<4.5 kcal/mol) reveal the
expected three-fold approximate symmetry and correlate reasonably with experimentally
obtained value (2.5 kcal/mol).
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
Overall, these results indicate that our understanding of these inclusion systems is
reaching the point where fairly simple design rules can be followed to synthesize pyridazine-
based rotor molecules with low rotation barriers and where the remaining potential landscape
within the TPP channels might be used to enforce a preference for ferroelectric alignments.
EXPERIMENTAL SECTION
Materials. All reactions were carried out under argon atmosphere with dry solvents
freshly distilled under anhydrous conditions. Standard Schlenk and vacuum line techniques
were employed for all manipulations of air- or moisture-sensitive compounds. Yields refer to
isolated, chromatographically and spectroscopically homogenous materials.
1,3-Diethynylbicyclo[1.1.1]pentane (17),37,38 1,3-bis((trimethylsilyl)ethynyl)bicyclo-
18
[1.1.1]pentane (16),37,38 and TPP-d12 were synthesized according to published procedures.
THF and ether were dried over sodium with benzophenone and distilled under argon prior to
use. Triethylamine and CH2Cl2 were dried over CaH2 and distilled under argon prior to use.
All other reagents were used as supplied unless otherwise stated.
The code for designating an inclusion compound is X%Y@TPP-d12, where X is the
molar percentage of rotor Y.
The published procedure19 for preparing the inclusion compounds from components
was modified as follows. The neat rotors 1 - 5 (amounts of guest used for preparation of
inclusion compounds are in Table S2), and neat solvent-free hexagonal TPP-d12 (100 mg,
0.212 mmol) were mixed, and ball milled five times in 5 min. intervals at room temperature
using stainless steel ball mill. Between each of those periods, the sample was thoroughly
scratched from the walls. Finally, the white fine powders were recovered and placed into a
small vial wrapped in the aluminum foil. This vial was inserted into a small flask and
aubjected to three vacuum/argon cycles. The flask was then tightly closed and annealed
under argon for 24 hours at 70 °C.
Procedures. Analytical thin-layer chromatography (TLC) was performed using
precoated TLC aluminum sheets (Silica gel 60 F254). TLC spots were visualized using either
UV light (254 nm) or a 5% solution of phosphomolybdic acid in ethanol, and heat (200 °C)
as a developing agent. Flash chromatography was performed using silica gel (high purity
grade, pore size 60 Å, 70 - 230 mesh). Melting points are reported uncorrected. Chemical
1
13
shifts in H, and C spectra are reported in ppm on the δ scale relative to CHCl3 (δ = 7.26
1
13
1
ppm for H and δ = 77.0 ppm for C), acetone-d6 (δ = 2.05 ppm for H, and δ = 29.8 and
206.3 ppm for 13C), THF-d8 (δ = 1.72 and 3.58 ppm for 1H, and δ = 25.40 and 67.50 ppm for
13C), as internal references. The volume/volume (v/v) ratios of solvents were used to prepare
mobile phases for column chromatographies.
High-resolution mass spectra (HRMS) using atmospheric-pressure chemical
ionization (APCI) and electrospray ionization (ESI) were obtained on a mass analyzer
combining linear ion trap and the Orbitrap, and those using chemical ionization (CI) mode
were taken on a time-of-flight mass spectrometer. Assignment of the ions in the mass spectra
was based on simulation of the isotope envelopes using software provided by the
manufacturer of mass spetrometers.
26
ACS Paragon Plus Environment

Page 27 of 40
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
X-Ray Diffraction. Crystallographic data were collected either on Nonius Kappa
CCD diffractometer (compounds 9, 19, and 21) equipped with Bruker APEX-II CCD detector
by monochromatized MoKα radiation (λ = 0.71073 Å) or on Bruker D8 VENTURE Kappa
Duo PHOTON100 (compounds 1, 2, and 18) by IμS micro-focus sealed tube CuKα (λ =
1.54178 Å) at a temperature of 150(2) K. The structures were solved by direct methods
(XP)58 and refined by full matrix least squares based on F2 (SHELXL2014).59 The hydrogen
atoms on carbon were fixed into idealized positions (riding model) and assigned temperature
factors either Hiso(H) = 1.2 Ueq (pivot atom) or Hiso (H) = 1.5 Ueq (pivot atom) for methyl
moiety.
Powder X-ray Diffraction. The X-ray powder patterns were taken using two
different systems: (i) Bruker D8 Discoverer powder diffractometer using Cu Kα radiation at
a wavelength of λ = 0.15418 nm and line detector LYNXEYE XE. Powder samples were
loaded on silicon holder (6° off cut from (111)). The measurement was performed in the
range of 5 - 80° 2θ with a step 0.02° 2θ and 1 s/step measurement time. Powder samples,
loaded into 1.0 mm diameter borosilicate glass capillaries with a wall thickness of 10 μm,
were mounted on a Huber four-circle goniometer. The scattered X-rays were measured with
a NaI scintillator point detector that was moved in a horizontal plane by an angle 2θ with
respect to the direction of the incident X-rays to scan the Bragg scattering profile. The
resolution of the instrument, in its usual configuration, is qres ≈ 0.003 Å-1. (ii) Diffractometer
based on a Rigaku MiniFlex 600 using Cu Kα radiation at an average wavelength of λ =
0.15418 nm, and a D/Tex Ultra 1D silicon strip detector. Powder samples were loaded into
single-crystal quartz substrates, which were cut to prevent unwanted quartz diffraction peaks.
Transmission Electron Microscopy. The sample was prepared by dropping an
aqueous suspension of 30%1@TPP-d12 onto a carbon-coated copper grid and drying in air.
Images were recorded with a JEOL JEM-1011 transmission electron microscope operated at
60 kV.
DSC Analysis was performed on a DSC Q20 (TA Instruments). Analysis of all
samples (neat 1 - 5, TPP-d12, and unsaturated and saturated bulk inclusion compounds)
consisted of three parts at 10 °C/min. The first part was heating from room temperature to
310 °C, the second was cooling to 40 °C, and the third was warming to 310 °C. Standard
loading was 4.0 - 7.5 mg.
Dielectric Spectroscopy. The powder samples were loaded onto the surface of
coplanar interdigitated electrode capacitators fabricated by gold metallization on fused silica
substrates (each with 50 fingers separated by ~10 μm gaps). Nominal capacitor values for the
empty capacitors were near 1 pF and dielectric loss tangents were at or below 10-5 level. An
Andeen-Hagerling capacitance bridge was used to measure the dielectric loss and capacitance
with sinusoidal alternating voltage (1.8 - 5 V and 0.10 - 20 kHz) applied on the capacitor.
Data were taken continuously with the temperature slowly changing between 7 K and room
temperature.
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
Calculations. The guest molecule 1 was included in a four-layer model of the TPP
lattice, with one 1 in each of its seven channels. The optimal positions of all guests in the
lattice were optimized at the MP6-D3H460 level with the whole lattice fully relaxed in the
first step. The resulting structure had only approximately a threefold symmetry, because the
centrally located 1 itself has a lower symmetry. To achieve the threefold symmetry expected
for the time-averaged potential, the pyridazine in the rotor molecule located in the central
channel was replaced with a non-polar 1,2,4,5-tetrazine and the TPP lattice was averaged
over the 16 symmetry-related models shown in Figure S9. Then the six channels surrounding
the central channel were fixed to reduce the effects of truncation of the solid in the simulation
27
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 28 of 40
1
2
3
4
5
6
7
8
9
of the rotational barrier of 1 (Figure S8 in Supporting Information). The dihedral angles φ1
and φ2 of the two pyridazine rings of 1 were fixed in steps of 15° and the remainder of the
central TPP channel was fully relaxed to locate the optimal conformation of 1. The rotational
potential was calculated at the B97-D3/SVP35 level as a function of φ1 and φ2. The
simulations were performed with the programs MOPAC201661 and TURBOMOLE V7.2.62
The NMR shieldings of 1 were calculated by the CASTEP program,63 version 8.0,
which is a DFT-based code that uses pseudopotentials to model the effects of core electrons,
and plane waves to describe the valence electrons. Positions of all atoms were optimized
prior to the NMR calculation; the unit cell parameters were fixed. Electron-correlation
effects were modeled using the generalized gradient approximation of Perdew, Burke, and
Ernzerhof.64 A plane wave basis set energy cutoff of 600 eV, default “on the fly generation”
pseudopotentials, and a k-point spacing of 0.05 Å-1 over the Brillouin zone via a Monkhorst-
Pack grid65 was used. The NMR calculations were performed using the GIPAW
approach.66,67
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
Solid-State NMR. High-resolution 13C and 31P solid-state NMR spectra were
obtained using a Bruker Avance II spectrometer operating at 125.7 MHz for 13C, 202.4 MHz
31
1
for P, and 499.9 MHz for H. Samples were packed into 3.2 mm magic angle spinning
(MAS) rotors and measurements taken using a MAS rate of 4-20 kHz using cross polarization
or direct excitation. The typical CP conditions used were: recycle delay 4 s, contact time
13
2 ms, acquisition time 30 ms. The C spectra were referenced with respect to external neat
tetramethylsilane by setting the carbonyl signal from a replacement sample of α-glycine to
176.03 ppm. Phosphorus spectra were referenced against (NH4)2HPO4 (δ = 1.37 ppm). The
assignment of the 13C SS-NMR signals was based on comparison with solution-state
spectrum, DFT calculated chemical shifts (see Calculations section for more details), and
short CP-MAS experiment (with the contact time of 40 μs), where only signals of carbon
atoms directly attached to a hydrogen are observable.
Synthesis. 1,2-Bis(6-tert-butylpyridazin-3-yl)ethyne (1). A flame-dried Schlenk flask was
charged with alkyne 11 (250 mg, 1.560 mmol, 1.0 equiv.), 9 (491 mg, 1.872 mmol,
1.2 equiv.), Pd(PPh3)4 (72 mg, 0.062 mmol, 4 mol %), and CuI (9 mg, 0.047 mmol, 3 mol %).
After three successive vacuum/argon cycles, degassed dry THF (10 mL) and triethylamine
(5 mL) were added from a syringe. The slightly yellowish solution was stirred for 18 h at
room temperature. A dense white solid precipitated. The crude reaction mixture was diluted
with CH2Cl2 (150 mL) and washed with saturated aqueous NH4Cl (2 × 30 mL). A clear
yellowish organic phase was dried over MgSO4 and volatiles were removed under reduced
pressure. Column chromatography on silica gel (hexane/EtOAc - 1:1 to remove unreacted
starting materials followed by hexane/EtOAc/CH2Cl2 - 1:1:1 to elute product) provided 1 as a
white crystalline solid (415 mg, 1.410 mmol, 90%).
1
Mp 246.4 - 247.9 °C. H NMR (400 MHz, CDCl3): δ 1.47 (s, 9H), 7.54 (d, J = 8.84
Hz, 2H), 7.68 (d, J = 8.83 Hz, 2H). 13C {1H} NMR (100 MHz, CDCl3): δ 29.9, 37.2, 88.8,
122.5, 130.0, 144.9, 169.2. IR (KBr): 3078, 3050, 2968, 2931, 2903, 2869, 2126, 1582,
1539, 1475, 1462, 1425, 1370, 1365, 1336, 1299, 1271, 1202, 1163, 1151, 1078, 944, 930,
867, 694, 508 cm-1. MS (EI), m/z (%): 294.2 (22, M), 279.2 (57, M - CH3), 252.1 (100),
+
237.1 (10, M – t-Bu), 210.1 (18). HRMS (EI) (m/z): [M]+ Calcd for C18H22N4 , 294.1844;
found, 294.1843. Anal. Calcd. for C18H22N4: C, 73.44; H, 7.53; N, 19.03. Found: C, 73.24;
H, 7.53; N, 18.73.
28
ACS Paragon Plus Environment

Page 29 of 40
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
1,3-Bis((6-tert-butylpyridazin-3-yl)ethynyl)bicyclo[1.1.1]pentane (2).
A flame-dried
Schlenk flask was charged with 1635 (300 mg, 1.152 mmol), 9 (755 mg, 2.880 mmol,
2.5 equiv.), Pd(PPh3)4 (53 mg, 0.046 mmol, 4 mol %), and CuI (7 mg, 0.035 mmol, 3 mol %).
After three successive vacuum/argon cycles, dry and degassed THF (8 mL) and
triethylamine (10 mL) were added from a syringe. Subsequently, solution of TBAF in THF
(1.0 M, 2.42 mL, 2.419 mmol, 2.1 equiv.) was added to the reaction mixture. The yellow
solution was stirred in an oil bath for 18 h at 40 °C. A dense white solid precipitated. The
crude reaction mixture was diluted with CH2Cl2 (150 mL), washed with saturated aqueous
NH4Cl (2 × 40 mL) and dried over MgSO4. Column chromatography on silica gel
(hexane/EtOAc - 1:1 to elute impurities and hexane/EtOAc/CH2Cl2 - 1:1:1 to elute product)
provided a brownish solid, which was subsequently triturated with EtOAc (8 × 1 mL).
Compound 2 was obtained as a white crystalline solid (362 mg, 0.941 mmol, 82%).
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
Mp > 250 °C (dec.). H NMR (400 MHz, CDCl3): δ 1.42 (s, 18H), 2.56 (s, 6H), 7.40
- 7.44 (m, 4H). 13C {1H} NMR (100 MHz, CDCl3): δ 29.9, 30.7, 37.0, 58.9, 77.3, 91.6,
122.2, 129.4, 145.4, 168.4. IR (KBr): 3042, 2959, 2916, 2877, 2236, 2217, 1578, 1536,
1473, 1461, 1420, 1366, 1329, 1301, 1272, 1203, 1148, 1105, 1047, 850, 803, 690, 531 cm-1.
MS (APCI), m/z (%): 385.2 (100, M + H). HRMS (APCI) (m/z): [M + H]+ Calcd for
+
C25H29N4 , 385.2387; found, 385.2387. Anal. Calcd. for C25H28N4: C, 78.09; H, 7.34; N,
14.57. Found: C, 77.98; H, 7.32; N, 14.40.
1,3-Bis((6-(3,3-dimethylbut-1-ynyl)pyridazin-3-yl)ethynyl)bicyclo[1.1.1]pentane (3). A
flame-dried Schlenk flask was charged with 1637 (350 mg, 1.343 mmol), 12 (845 mg,
2.955 mmol, 2.2 equiv.), Pd(PPh3)4 (62 mg, 0.054 mmol, 4 mol %), and CuI (8 mg,
0.040 mmol, 3 mol %). After three successive vacuum/argon cycles, dry and degassed THF
(8 mL) and triethylamine (10 mL) were added from a syringe. Subsequently, solution of
TBAF in THF (1.0 M, 2.82 mL, 2.820 mmol, 2.1 equiv.) was added to the reaction mixture.
The yellow solution was stirred in an oil bath for 18 h at 40 °C. A dense white solid
precipitated. The crude reaction mixture was diluted with CH2Cl2 (150 mL), washed with
saturated aqueous NH4Cl (2 × 40 mL) and dried over MgSO4. Column chromatography on
silica gel (EtOAc/CH2Cl2 - 1:1) provided a brownish solid, which was subsequently washed
on frit with methanol (4 × 2 mL), and EtOAc (4 × 2 mL). Compound 3 was obtained as a
white crystalline solid (425 mg, 0.983 mmol, 73%).
1
Mp > 260 °C (dec.). H NMR (400 MHz, CDCl3): δ 1.36 (s, 18H), 2.57 (s, 6H), 7.41
(s, 4H). 13C {1H} NMR (100 MHz, CDCl3): δ 28.2, 30.5, 30.8, 58.9, 76.4, 77.2, 93.6, 105.6,
128.7, 128.8, 145.1, 146.3. IR (KBr): 3108, 3072, 3042, 2968, 2924, 2900, 2879, 2238,
1568, 1526, 1475, 1457, 1396, 1364, 1327, 1295, 1277, 1204, 1120, 1104, 1059, 1033, 919,
869, 820, 662 cm-1. MS (APCI), m/z (%): 433.2 (100, M + H). HRMS (APCI) (m/z): [M +
+
H]+ Calcd for C29H29N4 , 433.2387; found, 433.2387. Anal. Calcd. for C29H28N4: C, 80.52;
H, 6.52; N, 12.95. Found: C, 80.39; H, 6.44; N, 12.85.
1,3-Bis((6-tert-butylpyridazin-3-yl)buta-1,3-diynyl)bicyclo[1.1.1]pentane (4). Warning:
This compound is explosive and detonates spontaneously at temperatures above 200 °C. It
should be handled with caution! A flame-dried Schlenk flask was charged with 9 (472 mg,
1.800 mmol, 3.0 equiv.), Pd(PPh3)4 (28 mg, 0.024 mmol, 4 mol %), and CuI (3 mg,
0.018 mmol, 3 mol %). After three successive vacuum/argon cycles, a solution of tetrayne 18
(98 mg, 0.600 mmol) in dry and degassed THF (6 mL) followed by triethylamine (5 mL) was
added from a syringe. The slightly yellowish solution was stirred for 18 h at room
temperature. A dense white solid precipitated. The crude reaction mixture was diluted with
29
ACS Paragon Plus Environment