Zinc-Ion-Stabilized Charge-Transfer Interactions Drive Self-Complementary or Complementary Molecular Recognition

Article abstract of DOI:10.1021/jacs.0c05940

Here, we show that charge-transfer interactions determine whether donor and acceptor ditopic ligands will associate in a complementary or self-complementary fashion upon metal-ion clipping. Anthracene-based (9,10LD and 1,5LD) and anthraquinone-based (1,5LA) ditopic ligands containing two imidazole side arms as zinc coordination sites were designed. The 9,10LD and 1,5LA systems associated in a complementary fashion (LA/LD/LA) upon clipping by two zinc ions (Zn2+) to form an alternating donor-acceptor assembly [(9,10LD)(1,5LA)2-(Zn2+)2]. However, once the charge-transfer interactions were perturbed by subtle modifications of the imidazole side arms (9,10LD′(S) and 1,5LA′(S)), self-complementary association (LD′/LD′/LD′/LD′ and LA′/LA′/LA′/LA′) between the donor (9,10LD′(S)) and acceptor (1,5LA′(S)) ligands predominantly occurred to form homoassemblies [(9,10LD′(S))4-(Zn2+)2 and (1,5LA′(S))4-(Zn2+)2]. As in the case of a homochiral pair (9,10LD′(S) and 1,5LA′(S)), self-complementary association (narcissistic self-sorting) occurred in the Zn2+ assembly with heterochiral combinations of the donor and acceptor ligands (9,10LD′(S)/1,5LA′(R) and 9,10LD′(S)/1,5LA′(R)/1,5LA′(R)). Narcissistic self-sorting also took place between the positional isomer of the donor ligands (9,10LD and 1,5LD) to form individual homoligand assemblies [(9,10LD)4-(Zn2+)2 and (1,5LD)4-(Zn2+)2]. Conversely, statistical association took place in the Zn2L4 assembly process of an isomorphous pair of the donor and acceptor ligands (1,5LD and 1,5LA).

Full text of DOI:10.1021/jacs.0c05940

Subscriber access provided by University of Wollongong Library
Article
Zinc Ion-Stabilized Charge-Transfer Interactions
Drive Self- or Complementary Molecular Recognition
Shuta Iseki, Kohei Nonomura, Sakura Kishida, Daiji Ogata, and Junpei Yuasa
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.0c05940 • Publication Date (Web): 12 Aug 2020
Downloaded from pubs.acs.org on August 13, 2020
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 10
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Zinc Ion-Stabilized Charge-Transfer Interactions Drive Self- or Com-
plementary Molecular Recognition
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
Shuta Iseki,† Kohei Nonomura,† Sakura Kishida,† Daiji Ogata,† and Junpei Yuasa*,†
†Department of Applied Chemistry, Tokyo University of Science, 1-3 Kagurazaka, Shinjuku-ku, Tokyo 162-8601, Japan
ABSTRACT: Here, we show that charge-transfer interactions determine whether donor and acceptor ditopic ligands will associate in a
complementary or self-complementary fashion upon metal ion clipping. Anthracene- (9,10LD and 1,5LD) and anthraquinone- (1,5LA) based
ditopic ligands containing two imidazole side arms as zinc coordination sites were designed. The 9,10LD and 1,5LA system associated in a
complementary fashion (LA/LD/LA) upon clipping by two zinc ions (Zn2+) to form an alternating donor-acceptor assembly
[(9,10LD)(1,5LA)2–(Zn2+)2]. However, once the charge-transfer interactions were perturbed by subtle modifications of the imidazole side
arms (9,10LD′(S) and 1,5LA′(S)), self-complementary association (LD′/LD′/LD′/LD′ and LA′/LA′/LA′/LA′) between the donor- (9,10LD′(S)) and
acceptor- (1,5LA′(S)) ligands predominantly occurred to form homo-assemblies [(9,10LD′(S))4–(Zn2+)2 and (1,5LA′(S))4–(Zn2+)2]. As in the case
of a homochiral pair (9,10LDʹ(S) and 1,5LAʹ(S)), self-complementary association (narcissistic self-sorting) occurred in the Zn2+ assembly
with heterochiral combinations of the donor and acceptor ligands (9,10LDʹ(S)/1,5LAʹ(R) and 9,10LDʹ(S)/1,5LAʹ(R)/1,5LAʹ(R)). Narcissistic self-
sorting also took place between the positional isomer of the donor ligands (9,10LD and 1,5LD) to form individual homoligand assemblies
[(9,10LD)4–(Zn2+)2 and (1,5LD)4–(Zn2+)2]. Conversely, statistical association took place in the Zn2L4 assembly process of an isomorphous
pair of the donor and acceptor ligands (1,5LD and 1,5LA).
l
INTRODUCTION
Molecular recognition is a specific interaction that can distinguish
one molecule from another by means of noncovalent interactions.
Such recognition has an important role in supramolecular chemis-
try. Hydrogen bonding is the most extensively used type of non-
covalent interaction in this context. Hydrogen bonding is highly di-
rectional and has high association constants that are effective in ei-
ther self-complementary or complementary molecular recogni-
tion.1 Alternatively, charge-transfer (CT) interactions are another
unique motif that drives alternative association between donor (D)
and acceptor (A) molecules owing to its intrinsic complementary
nature.2–5 However, association constants of D-A complexes are
typically very low (Scheme 1a),2,5 therefore CT-based molecular
recognition requires a third component (M) that combines D and A
through other noncovalent interactions such as solvophobic forces,
hydrogen bonding, and metal–ligand coordination.6–11 If the bind-
ing forces of M with D and with A are greater than the CT interac-
tion between D and A, a statistical mixture of the ternary complexes
results (Scheme 1b). It might also be possible to produce a driving
bias for alternating D-A association by design of D and A such that
their CT interactions are reinforced upon combination with M
(Scheme 1c). Conversely, if self-complementary interactions are
high enough, complementary D-D and A-A associations will dom-
inate (Scheme 1d). Such weakly attractive CT interactions have not
previously been considered to be a core molecular recognition mo-
tif that can control self-complementary and complementary molec-
ular recognition in supramolecular systems.
Scheme 1. (a) Relative Gibbs energy changes (DG) of a charge transfer complex (D–
A), (b-d) mediator (M)-assisted self-complementary (D–M–D and A–M–A) and com-
plementary (D–M–A) molecular recognitions in b) –DGAA = –DGDD = –DGDA, (c) –
DGAA, –DGDD < –DGDA, and (d) –DGAA, –DGDD < –DGDA. (e) Chemical structures of
anthracene- (9,10LD, 9,10LD′(R), and 9,10LD′(S)) and anthraquinone-based ditopic ligands
(
1,5LA, 1,5LA′(R), and 1,5LA′(S)).
Herein, for the first time, we successful demonstrated that the
weak attractive force of CT interactions effectively determines
whether D and A molecules associate in a self-complementary or
complementary fashion upon self-assembly with zinc ions (Zn2+).
We designed Dand A ditopic ligands (9,10LD and 1,5LA, respectively)
containing two imidazole side arms at 9,10-positions of anthracene
and 1,5-positions of anthraquinone through ethynyl spacers
(Scheme 1e) so that they could form an alternating A-D-A stack
[(9,10LD)(1,5LA)2–(Zn2+)2] upon clipping by Zn2+ with a tetrahedral
coordination preference (vide infra, Figure 3a).12–14 Notably, once
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
the CT interactions were perturbed by subtle modifications of the
Page 2 of 10
imidazole side arms, D and A ditopic ligands (9,10LDʹ(S) and 1,5LAʹ(S),
respectively) began to associate in a self-complementary fashion to
form homo-assemblies [(9,10LDʹ(S))4–(Zn2+)2 and (1,5LAʹ(S))4–(Zn2+)2].
This finding suggests a potential role of CT interactions as unique
molecular recognition motifs in fine-tuning of supramolecular sys-
tems. Because CT interactions have an important influence on pho-
tophysical properties, the present work will also provide a basis for
future design of CT-based functional materials.
1
2
3
4
5
6
7
8
9
l
RESULTS AND DISCUSSION
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
Zinc Ion-Stabilized Charge-Transfer Interactions. Figure
1a shows UV–vis absorption spectral changes during titration of a
9,10LD/1,5LA mixture ([9,10LD]0 = 6.7 × 10–6 M; [1,5LA]0 = 1.3 × 10–5
M) with Zn(OTf)2 (OTf = OSO2CF3) in acetonitrile, where the
9,10LD/1,5LA mixture shows successive spectral change with isosbes-
tic points at 281 and 486 nm and a characteristic CT band was
clearly visible at l = 520 nm. Noteworthy, theoretical UV–vis ab-
sorption spectrum calculated by time-dependent density function
theory (TD-DFT) indicates that the CT band appeared at l = 546
nm (Supporting Information Figure S1-S3). The isosbestic points
indicate the quantitative conversion of 9,10LD and 1,5LA into a single
self-assembly (Scheme 2a). A titration plot of the absorbance mon-
itored at l = 420 nm has a saturation point at [Zn2+]/([9,10LD]0 +
[1,5LA]0) = 0.67 (inset of Figure 1a), indicating that the three ditopic
ligands (1,5LA/9,10LD/1,5LA) were clipped by the two Zn2+ ions
(Scheme 2a).
Next, we subtly modified 9,10LD and 1,5LA by replacing the
methyl groups of the imidazole side arms with branched substitu-
ents (9,10LDʹ(S) and 1,5LAʹ(S), respectively) to reduce solvophobic ef-
fects, which should reduce the CT interactions.5,15 Unlike the
9,10LD/1,5LA mixture (Figure 1a), a 9,10LDʹ(S)/1,5LAʹ(S) mixture showed
biphasic UV–vis absorption spectral changes (Figure 1d; Support-
ing Information Figure S4) with individual isosbestic points in re-
sponse to the molar ratio of [Zn2+]/([9,10LDʹ(S)]0 + [1,5LAʹ(S)]0) = 0–
0.25 (first phase: red lines) and 0.25–0.58 (second phase: blue
lines). These results are attributed to self-complementary associa-
tions of 1,5LAʹ(S) (LAʹLAʹLAʹLAʹ) and of 9,10LDʹ(S) (LDʹLDʹLDʹLDʹ)
upon assembly with Zn2+ (Scheme 2b).16–18 The titration plot shows
the inflection and saturation points at [Zn2+]/([9,10LDʹ(S)]0
+
[1,5LAʹ(S)]0) = 0.25 and 0.50, respectively (Figure 1d inset), which
illustrates the stoichiometry of the self-complementary associations
as [9,10LDʹ(S)]:[Zn2+] = 2:1 and as [1,5LAʹ(S)]:[Zn2+] = 2:1 correspond-
ing to L4Zn2-assemblies. In this context, we have revealed that the
imidazole-based ditopic ligands (L) with Zn2+ mostly give rise to
L4Zn2-assemblies.14a,d,e
Next, differential UV–vis absorption spectral changes (Fig-
ure 2a and d) were produced from the Zn2+-titration data of the
9,10LD/1,5LA mixture and 9,10LDʹ(S)/1,5LAʹ(S) mixture (Figure 1a and b,
respectively) and compared with those obtained with individual ti-
tration experiments of the D or A ditopic ligand alone with Zn2+
(Figure 2b, c, e and f; Supporting Information Figure S6 and S7).
In the Zn2+-titration of the 9,10LDʹ(S)/1,5LAʹ(S) mixture, the correspond-
ing differential spectra at the first- and second-phase (Figure 2d)
appear to be identical to those obtained with the individual spectral
titration of 1,5LAʹ(S) with Zn2+ and that of 9,10LDʹ(S) with Zn2+ (Figure
2e and f). Conversely, the differential UV–vis absorption spectra in
the Zn2+-titration of the 9,10LD/1,5LA mixture (Figure 2a) disagreed
with those obtained from the individual Zn2+-titration of 9,10LD and
that of 1,5LA (Figure 2b and c).
Figure 1. (a and d) UV–vis absorption spectral changes during titration of (a)
9,10LD/1,5LA mixture ([9,10LD]0 = 6.7 × 10–6 M and [1,5LA]0 = 1.3 × 10–5 M) with Zn2+ [0
(green line)–2.0 × 10–5 M (orange line)], (d) 9,10LDʹ(S) 1,5LAʹ(S) mixture ([9,10LDʹ(S)]0 =
/
2.0 × 10–5 M and [1,5LAʹ(S)]0 = 2.0 × 10–5 M) with Zn2+ [0–1.0 × 10–5 M (red lines)–2.3
× 10–5 M (blue lines)] in acetonitrile at 298 K. Insets show plots of absorbance at (a) l
= 420 nm vs x = [Zn2+]/([9,10LD]0 + [1,5LA]0), (d) l = 500 nm vs x = [Zn2+]/([9,10LDʹ(S)
]
0
+ [1,5LAʹ(S)]0). (b and e) UV–vis absorption spectra of (b) blue line: (9,10LD)4–(Zn2+)2,
red line: (1,5LA)4–(Zn2+)2, (e) blue line: (9,10LDʹ(S))4–(Zn2+)2, red line: (1,5LAʹ(S))4–
(Zn2+)2.19 e values were calculated based on the ligand concentration. (c and f) Orange
solid lines show the observed UV–vis absorption spectra of (c) 9,10LD/1,5LA mixture
([9,10LD]0 = 6.6 × 10–6 M and [1,5LA]0 = 1.3 × 10–5 M) in the presence of Zn2+ (2.0 × 10–
5 M), (f) 9,10LDʹ(S) 1,5LAʹ(S) ([9,10LDʹ(S)]0 = 1.0 × 10–5 M and [1,5LAʹ(S)]0 = 1.0 × 10–5 M)
/
in the presence of Zn2+ (1.0 × 10–5 M) in acetonitrile. Dashed lines in Figure 1c and 1f
show calculated UV–vis absorption spectra obtained from the sum of the absorption
spectra of the homoligand-assemblies in the ratios: (c) (9,10LD)4–(Zn2+
) (33%) and
2
(
1,5LA)4–(Zn2+)2 (67%), (f) (9,10LD’(S))4–(Zn2+)2 (50%) and (1,5LA’(S))4–(Zn2+)2 (50%).
2
ACS Paragon Plus Environment

Page 3 of 10
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2. Schematic illustration for Zn2+-driven assembly formation between (a)
9,10LD and 1,5LA, (b) 9,10LDʹ(S) and 1,5LAʹ(S). Photograph shows the colors of acetonitrile
solutions of 9,10LD/1,5LA (right) mixture and 9,10LDʹ(S) 1,5LAʹ(S) (left) mixture containing
Zn2+
/
.
(1,5LAʹ(S))4–(Zn2+)2 (Figure 1e)19 was in complete agreement with the
observed absorption spectrum of the 9,10LDʹ(S)/1,5LAʹ(S) mixture with
Zn2+ (Figure 1f orange solid line), which indicates that Zn2+-as-
sisted self-complementary associations occurred in the
9,10LDʹ/1,5LAʹ(S) system (Scheme 2b). Conversely, the calculated ab-
sorption spectrum (Figure 1c dashed line) obtained from the sum
of the two absorption spectra [(9,10LD)4–(Zn2+)2 and (1,5LA)4–(Zn2+)2
(Figure 1b)]19 appeared to disagree with the observed absorption
spectrum of the 9,10LD/1,5LA mixture with Zn2+ showing the CT band
at l = 520 nm (Figure 1c orange solid line). The developing CT
band produced notable differences in the apparent colors of the
9,10LD/LA and 9,10LDʹ(S)/1,5LAʹ(S) upon addition of Zn2+ (Scheme 2
photograph).
We further verified the stoichiometry of the alternating A-D-
Aassembly by global titration analysis (Figure 2g), where the ratios
of the 9,10LD and 1,5LA were changed from [9,10LD]0/([9,10LD]0 +
[1,5LA]0) = 0 to 1.0 while the total concentrations of 9,10LD and 1,5LA
were kept constant ([9,10LD]0 + [1,5LA]0 = 2.0 × 10–5 M). The Job-
plot analysis based on the absorbance at l = 520 nm assigned to the
CT band suggests that the stoichiometry of [9,10LD]0/([9,10LD]0 +
[1,5LA]0) = 0.33 (Figure 2g inset), which clearly indicates an alter-
nating A-D-A assembly ((9,10LD)(1,5LA)2–(Zn2+)2). The structure of
(
9,10LD)(1,5LA)2–(Zn2+)2 was optimized at B3LYP/6-31G(d,p) [C, H,
N, O]; LANL2DZ (Zn) assuming the most symmetrical structure
of the alternating A-D-A association (Figure 3a and b), which
seems to be reasonable for situations in which CT interactions were
reinforced upon clipping by two Zn2+ ions.
Thus, the positive ESI-MS of the 9,10LD/1,5LA mixture was
measured in acetonitrile containing Zn2+ (Figure 3c) and the results
compared with those of their homo-assemblies (Figure 3d, e) as
well as the 9,10LDʹ(S)/1,5LAʹ(S) mixture with Zn2+ (Supporting Infor-
mation Figure S8). The 9,10LD/1,5LA mixture showed a mass peak
corresponding to the alternating A-D-A assembly at m/z = 1794.13
and fragment mass signals.20 Conversely, we could not find any
mass signal corresponding to the alternating A-D-A assembly in
the 9,10LDʹ(S)/1,5LAʹ(S) mixture with Zn2+ (Supporting Information
Figure S8).
Furthermore 1H NMR spectra of the 9,10LD/1,5LA mixture and
9,10LDʹ(S)/1,5LAʹ(S) mixture with Zn2+ (Figure 4a and d) were com-
pared with those of their homo-assemblies (Figure 4b, c and 4e,
f).19 Among the homo-assemblies studied here, NMR-based struc-
ture determination of (9,10LDʹ(S))4–(Zn2+)2 was achieved in our recent
report.14d In the other cases, the structures of (9,10LD)4–(Zn2+)2 and
(1,5LA)4–(Zn2+)2 were determined with the use of 1H,1H-COSY,
Figure 2. (a–f) Left: differential UV–vis absorption spectra obtained from titration of
(a) 9,10LD/1,5LA mixture ([9,10LD]0 = 6.7 × 10–6 M and [1,5LA]0 = 1.3 × 10–5 M) with Zn2+
(0–2.0 × 10–5 M), (b) 1,5LA (2.0 × 10–5 M) alone with Zn2+ (0–1.2 × 10–5 M), (c) 9,10LD
(2.0 × 10–5 M) alone with Zn2+ (0–1.0 × 10–5 M), (d) 9,10LDʹ(S) 1,5LAʹ(S) mixture
/
([9,10LDʹ(S)]0 = 2.0 × 10–5 M and [1,5LAʹ(S)]0 = 2.0 × 10–5 M) with Zn2+ [1st phase: 0–1.0
× 10–5 M (red lines), 2nd phase: 1.0 × 10–5 M–2.3 × 10–5 M (blue lines)], (e) 1,5LAʹ(S)
(2.0 × 10–5 M) alone with Zn2+ (0–1.2 × 10–5 M), and (f) 9,10LDʹ(S) (2.0 × 10–5 M) alone
with Zn2+ (0–1.2 × 10–5 M) in acetonitrile at 298 K. Right: scheme for complex for-
mation corresponding to the UV–vis absorption spectral changes. (g) Stacked plot for
UV–vis absorption spectra for titration of 9,10LD/1,5LA mixture (total concentration:
[
9,10LD]0 + [1,5LA]0 = 2.0 × 10–5 M) with Zn2+ [0 (red line)–2.0 × 10–5 M (blue line)] in
acetonitrile at 298 K, where the molar ratio changes from [9,10LD]0/([9,10LD]0 + [1,5LA]0)
= 0 to 1.0. Dashed line shows l = 520 nm. Inset shows plot of absorbance at l = 520
nm in the presence of 2.0 × 10–5 M of Zn2+ in each titration plot, vs [9,10LD]0/([9,10LD]0
+ [1,5LA]0).
In the same manner, a calculated absorption spectrum (Figure
1f black dashed line) obtained from the sum of the absorption spec-
tra of the individual homo-assemblies [(9,10LDʹ(S))4–(Zn2+)2 and
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 10
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. (a–c) 1H NMR spectra of (a) 9,10LD/1,5LA mixture ([9,10LD]0 = 1.0 × 10–3
M
and [1,5LA]0 = 2.0 × 10–3 M) with Zn2+ (2.0 × 10–3 M), (b) 9,10LD (2.0 × 10–3 M) alone
with Zn2+ (1.0 × 10–3 M), and (c) 1,5LA (2.0 × 10–3 M) alone with Zn2+ (1.0 × 10–3 M),
in CD3CN at 298 K. Insets: Structures of (a) (9,10LD)(1,5LA)2–(Zn2+), (b) (9,10LD)4–
(Zn2+)2, and (c) (1,5LA)4–(Zn2+)2 determined by NMR analysis. Color code, 9,10LD: blue,
Figure 3. (a) Equation and (b) the structure of (9,10LD)( 1,5LA)2–(Zn2+
) optimized at
2
1,5LA: red, Zn2+: yellow, H: white. (d–f) 1H NMR spectra of (d) 9,10LDʹ(S) 1,5LAʹ(S) mix-
/
B3LYP/6-31+G(d,p) [C, H, N, O]; LANL2DZ (Zn). Color code, 9,10LD: blue, 1,5LA: red,
Zn2+: yellow. (c–e) Positive ESI-MS spectrometry of (c) 9,10LD/1,5LA mixture, (d) 1,5LA
alone, and (e) 9,10LD alone in acetonitrile with the presence of Zn2+. Intensity is nor-
malized by their most intense fragment. Mass spectrum of the whole region is given
in Supporting Information Figure S9.
ture ([9,10LDʹ(S)]0 = 2.0 × 10–3 M and [1,5LAʹ(S)]0 = 2.0 × 10–3 M) with Zn2+ (4.0 × 10–3
M), (e) 1,5LAʹ(S) (2.0 × 10–3 M) alone with Zn2+ (1.0 × 10–3 M), (f) 9,10LDʹ(S) (2.0 × 10–3
M) alone with Zn2+ (1.0 × 10–3 M) in CD3CN at 298 K.
observation suggests that both the anthracene and anthraquinone
aromatic protons exist outside of the aromatic rings, which is con-
sistent with an A-D-A stack having three aromatic rings in an al-
most complete overlapping arrangement (Figure 4a inset). Con-
versely, the 1H NMR spectrum of (9,10LD)(1,5LA)2–(Zn2+)2 contained
both broadened and well-resolved signals. We used NMR tech-
niques, including 1H,1H-COSY, and ROESY, in combination with
ligand exchange experiments to determine the structure of
ROESY, and NOESY NMR experiments in this work (see details
in the Supporting Information Figure S10-S16). Although
(1,5LAʹ(S))4–(Zn2+)2 underwent line broadening, likely because of its
dissociation and reassociation on the NMR timescale (Figure 4e),
we assumed the same L4Zn2 structure as (1,5LA)4–(Zn2+)2 for
(1,5LAʹ(S))4–(Zn2+)2 judging from the similarity of their UV–vis titra-
tion results (Figure 2b vs e). The homo-assemblies, (9,10LD)4–
(Zn2+)2 and (1,5LA)4–(Zn2+)2, exhibited upfield-shifted protons at
5.24 and 6.14 ppm, respectively (Figure 4b and c), which are at-
tributed to shielding of aromatic protons owing to stacking of aro-
matic rings inside the homo-assemblies (insets of Figure 4b and
c).14d,21 Conversely, the 9,10LD/1,5LA mixture with Zn2+ had an NMR
spectrum with almost no signal from the homoligand assemblies
[(9,10LD)4–(Zn2+)2 and (1,5LA)4–(Zn2+)2] (Figure 4a). This result
clearly indicates the formation of the A-D-A assembly
[(9,10LD)(1,5LA)2–(Zn2+)2]. In contrast to the upfield-shifted protons
observed for (9,10LD)4–(Zn2+)2 and (1,5LA)4–(Zn2+)2 (Figure 4b and c),
no such signals were observed for (9,10LD)(1,5LA)2–(Zn2+)2. This
(
9,10LD)(1,5LA)2–(Zn2+)2 (see details in the Supporting Information
Figure S18-S24). These results suggested that the broadened and
resolved signals likely came from 1,5LA and 9,10LD inside the A-D-
A assembly, respectively. The broadening of the NMR signals is
likely caused by rapid dissociation and reassociation of the 1,5LA
ligands on the NMR timescale.22,23 Conversely, the central 9,10LD
ligand inside the A-D-A assembly should be fixed by strong poly-
aromatic CT interactions with the two 1,5LA ligands, and therefore
no dissociation and reassociation took place on the NMR timescale.
In contrast to the 9,10LD/1,5LA mixture with Zn2+, the 9,10LDʹ(S)/1,5LAʹ(S)
mixture with Zn2+ exhibited an NMR spectrum (Figure 4d) that
comprised individual donor and acceptor homoligand assemblies
4
ACS Paragon Plus Environment

Page 5 of 10
Journal of the American Chemical Society
(blue circles and red triangles, respectively), where the broadened
NMR signals of (1,5LAʹ(S))4–(Zn2+)2 were shifted (see Supporting In-
1
2
3
4
5
formation S25). This finding also underlines that the
9,10LDʹ(S)/1,5LAʹ(S) system favored self-complementary association
(D/D/D/D or A/A/A/A) rather than complementary A-D-A stack-
ing upon Zn2+ binding (Scheme 2b).24
6
7
8
9
Self-Sorting According to Chirality Along with the CT In-
teractions. Next, we examined whether self-complementary as-
sociation (narcissistic self-sorting) as observed in the
9,10LDʹ(S)/1,5LAʹ(S) system (homochiral pair) can be switched to com-
plementary association (social self-sorting) by altering the combi-
nation of chirality of the donor and acceptor ligands. To this end,
we tested two kinds of heterochiral pairs, 9,10LDʹ(S)/1,5LAʹ(R) and
9,10LDʹ(S)/1,5LAʹ(R)/1,5LAʹ(S). These combinations open up the possibil-
ity of new alternating A-D-A assemblies [(9,10LDʹ(S))(1,5LAʹ(R))2–
(Zn2+)2 and (9,10LDʹ(S))(1,5LAʹ(R))(1,5LAʹ(S))–(Zn2+)2] (Scheme 3). If
these heterochiral A-D-A assemblies are markedly lower in energy
than the homochiral one [(9,10LDʹ(S))(1,5LAʹ(S))2–(Zn2+)2], comple-
mentary association along with heteroligand assembly will take
place. The 9,10LDʹ(S)/1,5LAʹ(R) mixture with Zn2+ gave a UV–vis ab-
sorption spectrum (Figure 5a orange solid line) that agreed well
with a calculated absorption spectrum obtained from the sum of the
absorption spectra of the individual homoligand-assemblies
[(9,10LDʹ(S))4–(Zn2+)2 and (1,5LAʹ(R))4–(Zn2+)2] (Figure 5a dashed line),
whereas no CT band was detected (for the corresponding UV–vis
titration analysis, see Supporting Information Figure S27). In the
same manner, almost complete agreement was obtained between
the UV–vis absorption spectrum of the 9,10LDʹ(S)/1,5LAʹ(R)/1,5LAʹ(S)
mixture with Zn2+ and a calculated absorption spectrum obtained
from the sum of the absorption spectra of their individual homolig-
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
Scheme 3. Summary of Zn2+-driven assembly in (a) 9,10LD’(S)
/
1,5LA’(R) mixture (b)
9,10LD’(S) 1,5LA’(R) 1,5LA’(S) mixture.
/
/
1
and-assemblies (Figure 5b). Notably, the H NMR spectra of the
9,10LDʹ(S)/1,5LAʹ(R) mixture with Zn2+ and that of the
9,10LDʹ(S)/1,5LAʹ(R)/1,5LAʹ(S) mixture with Zn2+ comprised their individ-
ual homoligand-assemblies (Supporting Information Figure S28).
These results suggest that self-complementary associations also oc-
curred
in
the
heterochiral
9,10LDʹ(S)/1,5LAʹ(R)
and
9,10LDʹ(S)/1,5LAʹ(R)/1,5LAʹ(S) systems (Scheme 3). All possible isomers
for the A/D/A-type heteroligand assemblies and A/A/A/A-type
homoligand assemblies from the hetero chiral pair
(
9,10LDʹ(S)/1,5LAʹ(R)/1,5LAʹ(S)) were optimized by DFT to check the rel-
ative energy of the isomers (Supporting Information Figure S29).
The DFT study indicated that the overall energy gain by the change
from a homochiral pair to a hetero chiral pair is at most only 1.00
kcal/mol, which is not a sufficient driving bias to change over com-
plementary association (A-D-A) from self-complementary associ-
ation (D/D/D/D and A/A/A/A). Conversely, in the
9,10LDʹ(S)/1,5LAʹ(R)/1,5LAʹ(S) system, a maximum of nine isomers were
envisioned for (1,5LAʹ(R or S))4–(Zn2+)2 depending on the arrangement
of LAʹ(R) and LAʹ(S). Although we could not determine the major iso-
mer because of NMR line broadening of (1,5LAʹ(R or S))4–(Zn2+)2
(Supporting Information Figure S28), the DFT calculations sug-
gested a small energy difference (DE < 8.03 kcal/mol) between the
possible isomers of (1,5LAʹ(R or S))4–(Zn2+)2 (Supporting Information
Figure S29). Hence, the (1,5LAʹ(R or S))4–(Zn2+)2 assembly formation
should be statistical in nature.
Figure 5. Orange solid lines show the observed UV–vis absorption spectra of (a)
9,10LD’(S) 1,5LA’(R) mixture ([9,10LD’(S)]0 = 1.0 × 10–5 M and [1,5LA’(R)]0 = 1.0 × 10–5 M) in
the presence of Zn2+ (6.0 × 10–5 M), (b) 9,10LD
/
’(S)
’(S)
0
/
1,5LA’(R) 1,5LA’(S) mixture ([9,10LD
/ ]
= 1.0 × 10–5 M, [1,5LA’(R)]0 = 5.0 × 10–6 M, and [1,5LA’(S)]0 = 5.0 × 10–6 M) in the presence
of Zn2+ (6.0 × 10–5 M) in acetonitrile. Dashed lines show calculated UV–vis absorption
spectra obtained from the sum of the absorption spectra of the homoligand-assemblies
in the ratios: (a)
(
9,10LD’(S))4–(Zn2+
)
(50%) and
(
1,5LA’(R))4–(Zn2+
)
(50%), (b)
2
2
(
9,10LD’(S))4–(Zn2+)2 (50%), (1,5LA’(R))4–(Zn2+)2 (25%), and (1,5LA’(S))4–(Zn2+)2 (25%).
and (1,5LD) and used 1,5LD as an isomorphous 1,5-substituted accep-
tor ligand (1,5LA). DFT-based modeling of a (9,10LD)(1,5LD)2–(Zn2+)2
structure indicated that the combination of 9,10LD and 1,5LD could
have three anthracene rings in an almost complete overlapping po-
sition (Scheme 4). To investigate this possibility, we conducted
Effects of Substituent Positions in Donor and Acceptor
Ligands on Self-Sorting. Finally, we investigated how posi-
tions of the imidazole side arms in the donor and acceptor ligands
affect Zn2+-assisted self-complementary or complementary associ-
ation.25–30 To this end, we synthesized a 1,5-substituted donor lig
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 10
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 4. Summary of Zn2+-driven assembly in 9,10LD/1,5LD mixture. Optimized
structure of (9,10LD)(1,5LD)2–Zn2+ calculated by DFT/B3LYP/6-31G(d) [C, H, N, O];
LANL2DZ (Zn).
1
UV–vis titration and H NMR studies of a 9,10LD/1,5LD mixture in
the presence of Zn2+ (Figure 6a,b and 7d, respectively). Prior to
these experiments, we used NMR to determine the homoligand as-
sembly of 1,5LD in combination with ESI-mass spectrometry and the
UV–vis titration of 1,5LD with Zn2+ to confirm the stoichiometry of
the self-complementary associations as [1,5LD]:[Zn2+] = 2:1 (see de-
tails in the Supporting Information Figure S30-34). On the basis of
these results we conclude that 1,5LD also formed an open-type Zn2L4
assembly [(1,5LD)4–(Zn2+)2] as in the case of 1,5LA. Figure 6a shows
the UV–vis absorption spectral changes during titration of the
9,10LD/1,5LD mixture with Zn2+ in acetonitrile, where essentially two
sets of spectral characteristics were observed in response to the mo-
lar ratio of [Zn2+]/([9,10LD]0 + [1,5LD]0) = 0.00–0.25 (first phase: right
blue lines) and 0.25–0.50 (second phase: dark blue lines). Quanti-
tative analysis of the differential absorption spectra obtained from
the titration experiments indicates that the observed spectral
changes at the first and second phase mainly arose from homolig-
and assembly formation of 1,5LD and 9,10LD, respectively (Support-
ing Information Figure S34). The resulting UV–vis absorption
spectrum of the 9,10LD/1,5LD mixture with Zn2+ (Figure 6b blue solid
line) agreed well with a calculated absorption spectrum (Figure 6b
black dashed line) obtained from the sum of the absorption spectra
of the individual homo-assemblies [(9,10LD)4–(Zn2+)2 and (1,5LD)4–
(Zn2+)2], thus suggesting self-complementary association (narcis-
sistic self-sorting) occurred in the 9,10LD/1,5LD system (Scheme 4).
Further conclusive evidence for narcissistic self-sorting was ob-
Figure 6. (a,c) UV–vis absorption spectral changes during titration of (a) 9,10LD/1,5LD
mixture ([9,10LD]0 = 2.0 × 10–5 M and [1,5LD]0 = 2.0 × 10–5 M) with Zn2+ [0–1.0 × 10–5
M (right blue lines)–6.0 × 10–5 M (dark blue lines)], (c) 1,5LD/1,5LA mixture ([1,5LD]0 =
2.0 × 10–5 M and [1,5LA]0 = 2.0 × 10–5 M) with Zn2+ (0––6.0 × 10–5 M) in acetonitrile
at 298 K. Insets show plots of absorbance at (a) l = 500 nm vs [Zn2+]/([9,10LD]0
+
[
1,5LD]0), (c) l = 460 nm vs [Zn2+]/([1,5LD]0 + [1,5LA]0). (b,d) Solid lines show the ob-
served UV–vis absorption spectra of (b) 9,10LD/1,5LD mixture ([9,10LD]0 = 2.0 × 10–5
M
and [1,5LD]0 = 2.0 × 10–5 M) with Zn2+ (2.0 × 10–5 M), (d) 1,5LD/1,5LA mixture ([1,5LD]0
= 2.0 × 10–5 M and [1,5LA]0 = 2.0 × 10–5 M) with Zn2+ (2.0 × 10–5 M), in acetonitrile.
Dashed lines show calculated UV–vis absorption spectra obtained from the sum of the
absorption spectra of the homoligand-assemblies in the ratios: (b) (9,10LD)4–(Zn2+
)
2
(50%) and (1,5LD)4–(Zn2+
) ) )
(50%), (d) (1,5LD)4–(Zn2+ (50%) and (1,5LA)4–(Zn2+
2
2
2
(50%).
As a further comparative experiment, we also examined the
possibility of complementary association (social self-sorting) be-
tween 1,5LD and 1,5LA (isomorphous pair) by the same procedure as
that used for the 9,10LD/1,5LD system. Prior to the experiments, we
performed DFT-based modeling of a A-D-A stack [(1,5LD)(1,5LA)2–
(Zn2+)2] and a D-A-D stack [(1,5LD)2(1,5LA)–(Zn2+)2]. Hence, we pre-
dicted three aromatic rings in a partial overlapping fashion for both
A-D-A and D-A-D stacking (Scheme 5). The 1,5LD/1,5LA mixture
underwent UV–vis absorption spectral changes during the titration
with Zn2+ (Figure 6c). The spectral changes were completed when
0.5 eq of Zn2+ was added relative to the total concentration of 1,5LD
and 1,5LA (Figure 6c inset), whereas no clear isosbestic point was
observed. On the basis of the obtained stoichiometry of
1
tained from H NMR spectroscopy. The 9,10LD/1,5LD mixture with
Zn2+ had a NMR spectrum (Figure 7d; Supporting Information Fig-
ure S35) that comprised individual homoligand assemblies (blue
closed and open circles), which was apparent when the spectrum
was compared with that of (9,10LD)4–(Zn2+)2 and (1,5LD)4–(Zn2+)2
(Figure 7f and e, respectively). The negative results obtained with
1,5LD (isomorphous of 1,5LA) underlines the importance of CT inter-
actions for complementary association (social self-sorting) be-
tween the donor (9,10LD) and acceptor (1,5LA) ligands with Zn2+.
6
ACS Paragon Plus Environment

Page 7 of 10
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 7. (a) Possible Zn2L4 isomers obtained from 1,5LD/1,5LA mixture. The structures
were optimized by DFT/B3LYP/6-31G(d) [C, H, N, O]; LANL2DZ (Zn), which were
modeled considering quasi symmetry and the vacant sites of Zn2+ were occupied by
acetonitrile molecules. (b) Positive ESI-MS spectrometry of 1,5LD/1,5LA mixture in ac-
etonitrile with the presence of Zn2+. Simulated ESI-MS spectra of [(1,5LD)n(1,5LA)(4 –
n)Zn2(OSO2CF3)2]2+ (n = 1–4). Mass spectrum of the whole region is given in Support-
ing Information Figure S37. (c-f) 1H NMR spectra of (c) 1,5LD/1,5LA mixture ([1,5LD]0
= 2.0 × 10–3 M and [1,5LA]0 = 2.0 × 10–3 M) with Zn2+ (4.0 × 10–3 M), (d) 9,10LD/1,5LD
mixture ([9,10LD]0 = 2.0 × 10–3 M and [1,5LD]0 = 2.0 × 10–3 M) with Zn2+ (4.0 × 10–3 M),
(e) (1,5LD)4–(Zn2+)2, and (f) (9,10LD)4–(Zn2+)2 in CD3CN at 298 K. Blue closed and open
circles indicate signals those corresponding (9,10LD)4–(Zn2+)2 and (1,5LD)4–(Zn2+)2, re-
spectively. Inset: (e) Structure of (1,5LD)4–(Zn2+)2 determined by NMR analysis. Color
code, 9,10LD: blue, 1,5LA: red, Zn2+: yellow, H: white.
Scheme 5. Summary of Zn2+-driven assembly in 1,5LD/1,5LA mixture. Optimized struc-
tures of (1,5LD)(1,5LA)2–Zn2+ and (1,5LD)2(1,5LA)–Zn2+ calculated by DFT/B3LYP/6-
31G(d) [C, H, N, O]; LANL2DZ (Zn). Color code, 1,5LD: blue, 1,5LA: red, Zn2+: yellow.
[Zn2+]/([1,5LD]0 + [1,5LA]0) = 0.50, Zn2L4 assemblies were envisioned
for the 1,5LD/1,5LA system with Zn2+. The resulting UV–vis absorp-
tion spectrum of the 1,5LD/1,5LA mixture with Zn2+ (Figure 6d orange
solid line) showed no CT band and a subtle difference from that of
a calculated absorption spectrum (Figure 6d black dashed line) of
the sum of the absorption spectra of the individual homo-assem-
blies [(1,5LD)4–(Zn2+)2 and (1,5LA)4–(Zn2+)2]. A 1H NMR spectrum of
the 1,5LD/1,5LA mixture with Zn2+ (Figure 7c) had a large number of
weak signals over a wide chemical shift range. The resulting fea-
tureless NMR spectrum (Figure 7c) likely arises from statistical as-
sociation of 1,5LD and 1,5LA in the Zn2L4 assemblies owing to their
structural similarity (Scheme 5). There are nine possible isomers
depending on the arrangement of 1,5LD and 1,5LA in Zn2L4. DFT-
based structures of the homoligand Zn2L4 assemblies are geometri-
cally similar to their homoligand assemblies (Figure 7a). ESI-mass
(positive) spectrum of the 1,5LD/1,5LA mixture with Zn2+ (Figure 7b)
provided the signals due to Zn2L4 assemblies containing 1,5LD and
conducted systematic structural modeling of A-D-A and D-A-D
stacking structures of donor and acceptor ligands with different
substitution positions through the use of DFT methods, where only
the combination of 9,10LD and 1,5LA leads to three aromatic rings in
an almost complete overlapping position (Supporting Information
Figure S36). Thus, these data further verified the proposed A-D-A
assembly between 9,10LD and 1,5LA and their CT interactions.
l
SUMMARYAND CONCLUSIONS
In conclusion, we have successfully demonstrated that CT in-
teractions effectively determine whether a combination of D and A
ditopic ligands will associate in self-complementary (D/D/D/D and
A/A/A/A) or complementary (A/D/A) fashion upon metal-ion clip-
ping. The D and A ditopic ligands (9,10LD and 1,5LA, respectively)
consisted of two imidazole side arms through ethynyl spacers,
which were pre-designed to form an alternating A-D-A stack upon
clipping by two Zn2+ ions. Hence, 9,10LD and 1,5LA favor the for-
mation of an alternating A-D-Aassembly [(9,10LD)(1,5LA)2–(Zn2+)2]
with Zn2+. However, once the CT interactions were weakened by
1,5LA ligands ([(1,5LD)n(1,5LA)(4 n)Zn2(OSO2CF3)2]2+ (n = 1–4)),
–
whose relative intensity ratio was essentially statistical in nature.
These results suggest that the substitution position of the imidazole
side arms in donor and acceptor ligands also determines whether
donor/acceptor systems follow self-complementary association
(narcissistic self-sorting) or complementary association (social
self-sorting). In addition to these experimental results, we
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 10
(5) Das, A.; Ghosh, S. Supramolecular Assemblies by Charge-
subtle modifications of the imidazole side arms, the corresponding
reference ligands 9,10LDʹ(S) and 1,5LAʹ(S) associated into self-comple-
mentary structures (D/D/D/D or A/A/A/A), giving rise to their
homoligand-assemblies [(9,10LDʹ(S))4–(Zn2+)2 and (1,5LAʹ(S))4–
(Zn2+)2]. Self-complementary association (narcissistic self-sorting)
observed between 9,10LDʹ(S) and 1,5LAʹ(S) (homochiral pair) was pre-
1
2
3
4
5
6
7
8
Transfer Interactions between Donor and Acceptor Chromo-
phores. Angew. Chem., Int. Ed. 2014, 53, 2038-2054.
(6) (a) Kim, H.-J.; Heo, J.; Jeon, W. S.; Lee, E.; Kim, J.; Sa-
kamoto, S.; Yamaguchi, K.; Kim, K. Selective Inclusion of a
Hetero-Guest Pair in a Molecular Host: Formation of Stable
Charge Transfer Complexes in Cucurbit[8]uril. Angew.
Chem., Int. Ed. 2001, 40, 1526-1529; (b) Liu, Y.; Yu, Y.; Gao,
J.; Wang, Z.; Zhang, X. Water-Soluble Supramolecular
Polymerization Driven by Multiple Host-Stabilized Charge-
Transfer Interactions. Angew. Chem., Int. Ed. 2010, 49, 6576-
6579; (c) Ko, Y. H.; Kim, K.; Kang, J.-K.; Chun, H.; Lee, J.
W.; Sakamoto, S.; Yamaguchi, K.; Fettinger, J. C.; Kim, K.
Designed Self-Assembly of Molecular Necklaces Using
Host-Stabilized Charge-Transfer Interactions. J. Am. Chem.
Soc. 2004, 126, 1932-1933.
(7) Ghosh, S.; Ramakrishnan, S. Small-Molecule-Induced Fold-
ing of a Synthetic Polymer. Angew. Chem., Int. Ed. 2005, 44,
5441– 5447.
(8) Zhao, J.; Zhang, Y.-M.; Sun, H.-L.; Chang, X.-Y.; Liu, Y.
Multistimuli-Responsive Supramolecular Assembly of Cu-
curbituril/Cyclodextrin Pairs with an Azobenzene-Contain-
ing Bispyridinium Guest. Chem. Eur. J. 2014, 20, 15108-
15115.
(9) (a) Ghosh, S.; Ramakrishnan, S. Aromatic Donor–Acceptor
Charge-Transfer and Metal-Ion-Complexation-Assisted
Folding of a Synthetic Polymer. Angew. Chem., Int. Ed. 2004,
43, 3264-3268. (b) Chen, L.; Wang, H.; Zhang, D.-W.; Zhou,
Y.; Li, Z.-T. Quadruple Switching of Pleated Foldamers of
Tetrathiafulvalene– Bipyridinium Alternating Dynamic Co-
valent Polymers. Angew. Chem., Int. Ed. 2015, 54, 4028-
4031.
(10) Liang, J.; Soucie, L. N.; Blechschmidt, D. R.; Yoder, A.;
Gustafson, A.; Liu, Y. Aromatic Donor−Acceptor Interac-
tion-Based Co(III)-salen Self-Assemblies and Their Applica-
tions in Asymmetric Ring Opening of Epoxides. Org. Lett.
2019, 21, 513-518.
served even in heterochiral pairs
(
9,10LDʹ(S)/1,5LAʹ(R) and
9,10LDʹ(S)/1,5LAʹ(R)/1,5LAʹ(R)). In the same manner, narcissistic self-
sorting, assisted by Zn2+-binding, took place between the positional
isomer of the donor ligands (9,10LD and 1,5LD), giving rise to their
homoligand assemblies, (9,10LD)4–(Zn2+)2 and (1,5LD)4–(Zn2+)2.
Conversely, an isomorphous pair of the donor and acceptor ligands
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
(
1,5LD and 1,5LA) with Zn2+ resulted in statistical formation of Zn2L4
assemblies. Thus, the present work reveals that the weak attractive
forces of CT interactions have considerable potential as a useful
molecular recognition motif that can control self-complementary
and complementary molecular associations in supramolecular sys-
tems, which will be informative for CT-based functional materi-
als.31-33
l
ASSOCIATED CONTENT
Supporting Information. Experimental details, TD DFT of A-D-A
assembly, titration analysis, ESI mass spectra, NMR assignment of
the assemblies using 1H,1H COSY and 2D-ROESY NMR spectra,
optimized structures of the assemblies and association constant of
A-D-A assembly. This material is available free of charge via the
Internet at http://pubs.acs.org.
l
AUTHOR INFORMATION
Corresponding Author
yuasaj@rs.tus.ac.jp
l
ACKNOWLEDGMENT
This work was partly supported by JSPS KAKENHI Grant Num-
bers 19H02693, 19K22207, and 19H04596 in Scientific Research
on Innovative Areas “Coordination Asymmetry”, Grant-in-Aid for
the ASAHI Glass Foundation. We thank the Edanz Group
(www.edanzediting.com/ac) for editing a draft of this manuscript.
(11) Frank, M.; Ahrens, J.; Bejenke, I.; Krick, M.; Schwarzer,D.;
Clever, G. H. Light-Induced Charge Separation in Densely
Packed Donor−Acceptor Coordination Cages. J. Am. Chem.
Soc. 2016, 138, 8279-8287.
l
REFERENCES
(12) For the quinone-based metal complexes; (a) Yuasa, J.;
Yamada, S.; Fukuzumi, S. A Mechanistic Dichotomy in
Scandium Ion-Promoted Hydride Transfer of an NADH An-
alogue:ꢀ Delicate Balance between One-Step Hydride-Trans-
fer and Electron-Transfer Pathways. J. Am. Chem. Soc. 2006,
128, 14938-14948. (b) Yuasa, J.; Fukuzumi, S. Reversible
Formation and Dispersion of Chiral Assemblies Responding
to Electron Transfer. J. Am. Chem. Soc. 2007, 129, 12912-
12913; (c) Šalitroš, I.; Fuhr, O.; Gál, M.; Valášek, M.; Ruben,
M. Photoisomerization of Bis(tridentate)-2,6-Bis(1H-pyra-
zol-1yl)pyridine Ligands Exhibiting a Multi-Anthracene
Skeleton. Chem. Eur. J. 2017, 23, 10100-10109.
(13) For the selected examples of redox active metal complex, see:
(a) Mahata, K.; Frischmann, P. D.; Würthner, F. Giant Elec-
troactive M4L6 Tetrahedral Host Self-Assembled with Fe(II)
Vertices and Perylene Bisimide Dye Edges. J. Am. Chem. Soc.
2013, 135, 15656-15661. (b) Bivaud, S.; Balandier, J.-Y.;
(1) (a) Prins, L. J.; Reinhoudt, D. N.; Timmerman, P. Noncova-
lent Synthesis Using Hydrogen Bonding. Angew. Chem., Int.
Ed. 2001, 40, 2382-2426. (b) Sijbesma, R. P.; Meijer, E. W.
Quadruple Hydrogen Bonded Systems. Chem. Commun.
2003, 5-16. (c) Cooke, G.; Rotello, V. M. Methods of Modu-
lating Hydrogen Bonded Interactions in Synthetic Host–
Guest Systems. Chem. Soc. Rev. 2002, 31, 275-286.
(2) Foster, R. Organic Charge-Transfer Complex, Academic
Press, London, 1969.
(3) Ko, Y. H.; Kim, E.; Hwang, I.; Kim, K. Supramolecular As-
semblies Built with Host-Stabilized Charge-Transfer Interac-
tions. Chem. Commun. 2007, 1305-1315.
(4) Kumar, M..; Rao, K. V.; George, S. J. Supramolecular
Charge Transfer Nanostructures. Phys. Chem. Chem. Phys.
2014, 16, 1300-1313.
8
ACS Paragon Plus Environment

Page 9 of 10
Journal of the American Chemical Society
Chas, M.; Allain, M.; Goeb, S.; Sallé, M. A Metal-Directed
(18) Reiné, P.; Ortuño, A. M.; Resa, S.; Álvarez de Cienfuegos,
L.; Blanco, V.; Ruedas-Rama, M. J.; Mazzeo, G.; Abbate, S.;
Lucotti, A.; Tommasini, M.; Guisán-Ceinos, S.; Ribagorda,
M.; Campaña, A. G.; Mota, A.; Longhi, G.; Miguel, D.;
Cuerva, J. M. OFF/ON Switching of Circularly Polarized Lu-
minescence by Oxophilic Interaction of Homochiral Sulfox-
ide-Containing o-OPEs with Metal Cations. Chem. Commun.
2018, 54, 13985-13988.
(19) The absorption and 1H NMR spectra of the individual homo-
assemblies (Figure 1b,e, 4b,c, 4e,f) were obtained by addition
of 0.50 equiv. of Zn2+ into 9,10LD (or 9,10LD′(S)) alone or into
1,5LA (or 1,5LA′(S)) alone.
1
2
3
4
5
6
7
8
9
10
Self-Assembled Electroactive Cage with Bis(pyr-
rolo)tetrathiafulvalene (BPTTF) Side Walls. J. Am. Chem.
Soc. 2012, 134, 11968-11970. (c) Freye, S.; Hey, J.; Torras-
Galán, A.; Stalke, D.; Herbst-Irmer, R.; John, M.; Clever, G.
H. Allosteric Binding of Halide Anions by a New Dimeric
Interpenetrated Coordination Cage. Angew. Chem., Int. Ed.
2012, 51, 2191-2194. (d) Croué, V.; Goeb, S.; Szalóki, G.;
Allain, M.; Salle, M. Reversible Guest Uptake/Release by
Redox-Controlled Assembly/ Disassembly of a Coordination
Cage. Angew. Chem., Int. Ed. 2016, 55, 1746-1750.
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
(14) For the supramolecular system we term "metal-ion clipping",
see: (a) Imai, Y.; Nakano, Y.; Kawai, T.; Yuasa, J. A Smart
Sensing Method for Object Identification Using Circularly
Polarized Luminescence from Coordination-Driven Self-As-
sembly. Angew. Chem., Int. Ed. 2018, 57, 8973-8978. (b)
Imai, Y.; Yuasa, J. Supramolecular Chirality Transformation
Driven by Monodentate Ligand Binding to a Coordinatively
Unsaturated Self-Assembly Based on C3-Symmetric Ligands.
Chem. Sci. 2019, 10, 4236-4245. (c) Imai, Y.; Yuasa, J. Off–
Off–On Chiroptical Property Switching of a Pyrene Lumino-
phore by Stepwise Helicate Formation. Chem. Commun.
2019, 55, 4095-4098. (d) Nonomura, K.; Yuasa, J. Compet-
ing Allosteric Mechanisms for Coordination-Directed Con-
formational Changes of Chiral Stacking Structures with Aro-
matic Rings. Inorg. Chem. 2019, 58, 6474-6484. (e) Inukai,
N.; Kawai, T.; Yuasa, J. One-Step Versus Multistep Equilib-
rium of Carbazole-Bridged Dinuclear Zinc(II) Complex For-
mation: Metal-Assisted p-Association and -Dissociation Pro-
cesses. Chem. Eur. J. 2014, 20, 15159-15168. (f) Imai, Y.;
Kawai, T.; Yuasa, J. Metal Ion Clip: Fine-Tuning Aromatic
Stacking Interactions in the Multistep Formation of Carba-
zole-Bridged Zinc(II) Complexes. Chem. Commun. 2015, 51,
10103-10106. (g) Ogata, D.; Yuasa, J. Dynamic Open Coor-
dination Cage from Nonsymmetrical Imidazole–Pyridine Di-
topic Ligands for Turn-On/Off Anion Binding. Angew.
Chem., Int. Ed. 2019, 58, 18424-18428.
(15) (a) Smithrud, D. B.; Diederich, F. Strength of Molecular
Complexation of Apolar Solutes in Water and in Organic Sol-
vents Is Predictable by Linear Free Energy Relationships: A
General Model for Solvation Effects on Apolar Binding. J.
Am. Chem. Soc. 1990, 112, 339-343. (b) Chen, Z.; Fimmela,
B.; Würthner, F. Solvent and Substituent Effects on Aggre-
gation Constants of Perylene Bisimide π-Stacks – a Linear
Free Energy Relationship Analysis. Org. Biomol. Chem.
2012, 10, 5845-5855.
(16) Notably, such two-step spectral changes were also observed
in the titration in a 1:2 mixture of 9,10LD′(S) and 1,5LA′(S) with
Zn2+ (Supporting Information Figure S3). Hence the initial
molar ratio between the donor and acceptor had no effect on
whether complementary or self-complementary takes place.
(17) 1,5LA′(S) first formed the assembly with Zn2+ prior to forming
the Zn2+-assembly with 9,10LD′(S) (Scheme 2b). The higher af-
finity of 1,5LA′(S) than that of 9,10LD′(S) for Zn2+ might be at-
tributed to the positions of the imidazole side arm in 1,5LA′(S),
which provides appropriate coordination vectors for building
the L4Zn2-assembly in a tetrahedral coordination.
(20) The mass peak at m/z
= 1794.13 corresponds to
{(9,10LD)(1,5LA)2(Zn)2(OSO2CF3)3 + H}+, where one of the
two the anthraquinone molecules could undergo one-electron
reduction (1,5LA + e– → 1,5LA•–) during the ionization process.
(21) Closed-type complexes might offer another possible geome-
try for the L4Zn2 structure; however, such a coordination ge-
ometry causes strain in the ditopic ligands. A DFT study in-
dicated that the proposed open-type L4Zn2 structures (insets
of Figure 4b and c) are 91.8 and 59.9 kcal/mol lower in en-
ergy than those of the closed-type complexes (Supporting In-
formation Figure S17).
(22) For NMR line broadening owing to restricted motion of aro-
matic rings on the NMR timescale caused by strong polyaro-
matic interactions, see Nishioka, T.; Kuroda, K.; Akita, M.;
Yoshizawa, M. A Polyaromatic Gemini Amphiphile That As-
sembles into a Well-Defined Aromatic Micelle with Higher
Stability and Host Functions. Angew. Chem., Int. Ed. 2019,
58, 6579-6583.
(23) NMR line broadening and signal merging have also been re-
ported in other supramolecular CT systems; see: Chakraborty,
S.; Ray, D.; Aswal, V. K.; Ghosh, S. Multi-Stimuli-Respon-
sive Directional Assembly of an Amphiphilic Donor–Accep-
tor Alternating Supramolecular Copolymer. Chem. Eur. J.
2018, 24, 16379-16387.
(24) The titration analysis in Figure 1a suggests that the formation
of the A-D-A assembly [(9,10LD)( 1,5LA)2–(Zn2+)2] was almost
stoichiometric. The non-linear data fitting analysis (Support-
ing Information Figure S26) indicates that a stoichiometric
binding occurs with an association constant greater than KADA
~1033 M–4 (corresponding to –GADA0 ~45 kcal/mol). The large
0
GADA value might be attributed to zinc-imidazole coordina-
tion bonds associated with the A-D-A CT interactions inside
the assembly.
(25) (a) Liu, M.; Zhang, L.; Wang, T. Supramolecular Chirality in
Self-Assembled Systems. Chem. Rev. 2015, 115, 7304–7397.
(b) Jędrzejewska, H.; Szumna, A. Making a Right or Left
Choice: Chiral Self-Sorting as a Tool for the Formation of
Discrete Complex Structures. Chem. Rev. 2017, 117, 4863–
4899. (c) Greenaway, R. L.; Santolini, V.; Pulido, A.; Little,
M. A.; Alston, B. M.; Briggs, M. E.; Day, G. M.; Cooper, A.
I.; Jelfs, K. E. From Concept to Crystals via Prediction:
Multi-Component Organic Cage Pots by Social Self-Sorting.
Angew. Chem., Int. Ed. 2019, 58, 16275–16281. (d) Safont-
Sempere, M. M.; Fernández, G.; Würthner, F. Self-Sorting
Phenomena in Complex Supramolecular Systems. Chem. Rev.
2011, 111, 5784–5814.
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 10
(26) (a) Zhong, J.; Zhang, L.; August, D. P.; Whitehead, G. F. S.; sanen, K.; Lützen, A. Subcomponent Self-Assembly of a Cy-
clic Tetranuclear FeII Helicate in a Highly Diastereoselective
Self-Sorting Manner. Chem.–Eur. J. 2019, 25, 12294–12297.
(c) Schulte, T. R.; Holstein, J. J.; Clever, G. H. Chiral Self-
Discrimination and Guest Recognition in Helicene-Based
Coordination Cages. Angew. Chem., Int. Ed. 2019, 58, 5562–
5566.
1
2
3
4
5
6
7
8
Leigh, D. A. Self-Sorting Assembly of Molecular Trefoil
Knots of Single Handedness. J. Am. Chem. Soc., 2019, 141,
14249–14256. (b) Kai, S.; Kojima, T.; Thorp-Greenwood, F.
L.; Hardie, M. J.; Hiraoka, S. How Does Chiral Self-Sorting
Take Place in the Formation of Homochiral Pd6L8 Capsules
Consisting of Cyclotriveratrylene-Based Chiral Tritopic Lig-
ands?. Chem. Sci. 2018, 9, 4104–4108. (c) Chinnaraja, E.;
Arunachalam, R.; Suresh, E.; Sen, S. K.; Natarajan, R.;
Subramanian, P. S. Binuclear Double-Stranded Helicates and
Their Catalytic Applications in Desymmetrization of Meso-
diols. Inorg. Chem. 2019, 58, 4465–4479. (d) Pritchard, V.
E.; Rota Martir, D.; Oldknow, S.; Kai, S.; Hiraoka, S.; Cook-
son, N. J.; Zysman-Colman, E.; Hardie, M. J. Homochiral
Self-Sorted and Emissive IrIII Metallo-Cryptophanes. Chem.
–Eur. J. 2017, 23, 6290–6294. (e) Boer, S. A.; Turner, D. R.
Self-Selecting Homochiral Quadruple-Stranded Helicates
and Control of Supramolecular Chirality. Chem. Commun.
2015, 51, 17375–17378. (f) Gidron, O.; Jirásek, M.; Trapp,
N.; Ebert, M. O.; Zhang, X.; Diederich, F. Homochiral
[2]Catenane and Bis[2]Catenane from Alleno-Acetylenic
Helicates - A Highly Selective Narcissistic Self-Sorting Pro-
cess. J. Am. Chem. Soc. 2015, 137, 12502–12505.
(29) (a) Beaudoin, D.; Rominger, F.; Mastalerz, M. Chiral Self-
Sorting of [2+3] Salicylimine Cage Compounds. Angew.
Chem., Int. Ed. 2017, 56, 1244–1248. (b) Yao, L. Y.; Lee, T.
K. M.; Yam, V. W. W. Thermodynamic-Driven Self-Assem-
bly: Heterochiral Self-Sorting and Structural Reconfiguration
in Gold(I)-Sulfido Cluster System. J. Am. Chem. Soc. 2016,
138, 7260–7263. (c) Yan, L. L.; Tan, C. H.; Zhang, G. L.;
Zhou, L. P.; Bünzli, J. C.; Sun, Q. F. Stereocontrolled Self-
Assembly and Self-Sorting of Luminescent Europium Tetra-
hedral Cages. J. Am. Chem. Soc. 2015, 137, 8550–8555. (d)
Claessens, C. G.; Torres, T. Chiral Self-Discrimination in a
M3L2 Subphthalocyanine Cage. J. Am. Chem. Soc. 2002, 124,
14522–14523.
(30) (a) Bing, T. Y.; Kawai, T.; Yuasa, J. Ligand-to-Ligand Inter-
actions That Direct Formation of D2-Symmetrical Alternat-
ing Circular Helicate. J. Am. Chem. Soc., 2018, 140, 3683–
3689. (b) Ogata, D.; Yuasa, J. Remarkable Self-Sorting Se-
lectivity in Covalently Linked Homochiral and Heterochiral
Pairs Driven by Pd2L4 Helicate Formation. Chem. Commun.,
2020, DOI: 10.1039/d0cc03539d.
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
(27) (a) Kaes C.; Hosseini. M. W.; Rickard. C. E. F.; Skelton. B.
W.; White, A. H. Synthesis and Structural Analysis of a Hel-
ical Coordination Polymer Formed by the Self-Assembly of
a 2,2'-Bipyridine-Based exo-Ditopic Macrocyclic Ligand and
Silver Cations. Angew. Chem., Int. Ed. 1998, 37, 920–922.
(b) Masood, M. A.; Enemark, E. J.; Stack, T. D. P. Ligand
Self-Recognition in the Self-Assembly. Angew. Chem., Int.
Ed. 1998, 37, 928–932. (c) Telfer, S. G.; Sato, T.; Kuroda,
R.; Lefebvre, J.; Leznoff, D. B. Dinuclear Complexes of Chi-
ral Tetradentate Pyridylimine Ligands: Diastereoselectivity,
Positive Cooperativity, Anion Selectivity, Ligand Self-Sort-
ing Based on Chirality, and Magnetism. Inorg. Chem. 2004,
43, 421–429. (d) Kim, T. W.; Hong, J.-I.; Lah, M. S. Self-
Discrimination of the Racemic Ligands in the Self-Assembly
of [{(Dppp)Pt(L)}2]4+. Chem. Commun. 2001, 743–744. (e)
Rowland, J. M.; Olmstead, M. M.; Mascharak, P. K. Chiral
Monomeric and Homochiral Dimeric Copper(II) Complexes
of a New Chiral Ligand, N-(1,2-Bis(2-Pyridyl)Ethyl)Pyri-
dine-2-Carboxamide: An Example of Molecular Self-Recog-
nition. Inorg. Chem. 2002, 41, 1545–1549.
(31) Nacci, C.; Ample, F.; Bleger, D.; Hecht, S.; Joachim, C.; Grill,
L. Conductance of a Single Flexible Molecular Wire Com-
posed of Alternating Donor and Acceptor Units. Nat. Com-
mun. 2015, 6, 7397.
(32) In the present system, the original anthracene fluorescence of
the donor ligand (9,10LD) was markedly quenched upon for-
mation of the A-D-A stack (Supporting Information Figure
S38), indicating a photoinduced electron transfer from the an-
thracene to the anthraquinone moiety within the
(
9,10LD)(1,5LA)2–(Zn2+)2 assembly. Conversely, we did not ob-
serve the charge-separated state through the use of a nanosec-
ond laser flash photolysis system probably because of the
rapid charge recombination process.
(33) For future applications of the A-D-A system, we confirmed
the stability of the one-electron reduced species of the accep-
•–
tor ligand by reversible cyclic voltammetry of the 1,5LA
(28) (a) Volbach, L.; Struch, N.; Bohle, F.; Topić, F.; Schnaken-
burg, G.; Schneider, A.; Rissanen, K.; Grimme, S.; Lützen,
A. Influencing the Self-Sorting Behavior of [2.2]Paracyclo-
phane-Based Ligands by Introducing Isostructural Binding
Motifs. Chem.–Eur. J. 2020, 26, 3335–3347. (b) Anhäuser,
J.; Puttreddy, R.; Glanz, L.; Schneider, A.; Engeser, M.; Ris-
/
1,5LA redox system and direct detection of 1,5LA•– by electron
spin resonance (ESR) spectroscopy (Supporting Information
•+
Figure S39). Conversely, the 9,10LD/9,10LD redox system is
irreversible, suggesting that the one electron oxidized species
of the donor ligand is unstable. Further synthetic modifica-
tion of the donor ligand should be necessary to use the present
A-D-A system as a CT-based material.
TOC
10
ACS Paragon Plus Environment