Snapshotting the excited-state planarization of chemically locked N,N′-disubstituted dihydrodibenzo[a,c]phenazines

Article abstract of DOI:10.1021/jacs.6b11789

For deeper understanding of the coupling of electronic processes with conformational motions, we exploit a tailored strategy to harness the excited-state planarization of N,N′-disubstituted dihydrodibenzo[a,c]phenazines by halting the structural evolution via a macrocyclization process. In this new approach, 9,14-diphenyl-9,14-dihydrodibenzo[a,c]phenazine (DPAC) is used as a prototype, in which the para sites of 9,14-diphenyl are systematically enclosed by a dialkoxybenzene-alkyl-ester or-ether linkage with different chain lengths, imposing various degrees of constraint to impede the structural deformation. Accordingly, a series of DPAC-n (n = 1-8) derivatives were synthesized, in which n correlates with the alkyl length, such that the strength of the spatial constraint decreases as n increases. The structures of DPAC-1, DPAC-3, DPAC-4, and DPAC-8 were identified by the X-ray crystal analysis. As a result, despite nearly identical absorption spectra (onset ～400 nm) for DPAC-1-8, drastic chain-length dependent emission is observed, spanning from blue (n = 1, 2, ～400 nm) and blue-green (n = 3-5, 500-550 nm) to green-orange (n = 6) and red (n = 7, 8, ～610 nm) in various regular solvents. Comprehensive spectroscopic and dynamic studies, together with a computational approach, rationalized the associated excited-state structure responding to emission origin. Severing the linkage for DPAC-5 via lipase treatment releases the structural freedom and hence results in drastic changes of emission from blue-green (490 nm) to red (625 nm), showing the brightening prospect of these chemically locked DPAC-n in both fundamental studies and applications.

Full text of DOI:10.1021/jacs.6b11789

Subscriber access provided by University of Newcastle, Australia
Article
Snapshotting the Excited-State Planarization of Chemically-
Locked N,N#-Disubstituted-dihydrodibenzo[a,c]phenazines
Wei Chen, Chi-Lin Chen, Zhiyun Zhang, Yi-An Chen, Wei-Chih Chao, Jianhua Su, He Tian, and Pi-Tai Chou
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.6b11789 • Publication Date (Web): 10 Jan 2017
Downloaded from http://pubs.acs.org on January 10, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 10
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Snapshotting the Excited-State Planarization of Chemically-Locked
N,N′-Disubstituted-dihydrodibenzo[a,c]phenazines
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
§
§
§
Wei Chen,†, ChiꢀLin Chen,‡, Zhiyun Zhang,†,‡, YiꢀAn Chen,‡ WeiꢀChih Chao,‡ Jianhua Su,† He
Tian,*,†and PiꢀTai Chou*,‡
†Key Laboratory for Advanced Materials and Institute of Fine Chemicals, East China University of Science & Technology,
Shanghai 200237, P. R. China
‡Department of Chemistry, National Taiwan University, Taipei, 10617 Taiwan, R.O.C.
ABSTRACT: For deeper understanding of the coupling of electronic processes with conformational motions, we exploit a tailored
strategy to harness the excitedꢀstate planarization of N,N′ꢀdisubstitutedꢀdihydrodibenzo[a,c]phenazines by halting the structural
evolution via macrocyclization process. In this new approach, 9,14ꢀdiphenylꢀ9,14ꢀdihydrodibenzo[a,c]phenazine (DPAC) is used
as a prototype, in which the para sites of 9,14ꢀdiphenyl are systematically enclosed by a dialkoxybenzeneꢀalkylꢀester or ꢀether linkꢀ
age with different chain lengths, imposing various degrees of constraint to impede the structural deformation. Accordingly, a series
of DPAC-n (n = 1~8) derivatives were synthesized, in which n correlates with the alkyl length, such that the strength of the spatial
constraint decreases as n increases. The structures of DPAC-1, DPAC-3, DPAC-4 and DPAC-8 were identified by the Xꢀray crysꢀ
tal analysis. As a result, despite nearly identical absorption spectra (onset ~400 nm) for DPAC-1~8, drastic chainꢀlength dependent
emission is observed, spanning from blue (n = 1, 2, ~400 nm) and blueꢀgreen (n = 3~5, 500ꢀ550 nm) to greenꢀorange (n = 6) and
red (n = 7, 8, ~610 nm) in various regular solvents. Comprehensive spectroscopic and dynamic studies, together with a computaꢀ
tional approach, rationalized the associated excitedꢀstate structure responding to emission origin. Severing the linkage for DPAC-5
via lipase treatment releases the structural freedom and hence results in drastic changes of emission from blueꢀgreen (490 nm) to
red (625 nm), showing the brightening prospect of these chemically locked DPAC-n in both fundamental and application.
1. Introduction
planarization coupled with the elongation of the πꢀ
delocalization over the benzo[a,c]phenazines moiety. Along
Recently, specific fluorophores with anomalous electroniꢀ
cally excitedꢀstate phenomena, giving rise to e.g., large Stokes
shift or multiꢀcolor emission, have drawn considerable attenꢀ
the planarization, DPAC also encounters the steric hindrance
raised by the bulky N,N′ꢀdisubstitutes, leading to a local minꢀ
imum state, which then further relaxes to maximize the planarꢀ
ization. The overall structural relaxation process of DPAC can
be qualitatively described by a sequential kinetic pattern deꢀ
picted in Figure 1b below.8a
tion because of their exploitations in flexible fullꢀcolor disꢀ
plays and biosensors.1 To facilitate the rational design of apꢀ
plicable molecules with intriguing photophysical properties, a
great deal of inꢀdepth studies along with the excitedꢀstate dyꢀ
namics of the fluorophores have been conducted, among
which excitedꢀstate electron transfer,2ꢀ4 proton transfer,5 and
energy transfer6 are three essential processes.
Beyond the three processes described above, the coupling of
nuclear motions with electronic processes is also one of the
most fundamental issues in chemistry. Whereas the intramoꢀ
lecular rotations (or twisting) coupled with electron transfer
have been extensively studied,2,3 rare cases are those chromoꢀ
phores demonstrating elongation of the πꢀdelocalization due to
bentꢀtoꢀplanar motions, modulating dramatically the luminesꢀ
cence.7 In this regard, our attention was attracted by a series of
hole
transporting
materials,
N,N′ꢀdisubstitutedꢀ
dihydrodibenzo[a,c]phenazines (DHPs). We discovered the
unusual photophysical property that some DHPs reveal anomꢀ
alously large Stokesꢀshifted emission in regular solvents.8
Figure 1. (a) Chemical structure of DPAC as well as its crystal
structure from front, bottom and right view with the selected anꢀ
gles: Θb = ∠C1−N1−N2−C2, ΘS1 = ∠C3−N1−N2 and ΘS2
=
Using
9,14ꢀdiphenylꢀ9,14ꢀdihydrodibenzo[a,c]phenazine
∠C4−N2−N1. (b) The structural relaxation of DAPC and its deꢀ
rivatives in the electronically excited state. Using DAPC as the
prototype, R* specifies the initially prepared state and I* is an
intermediate with the local minimum energy, both of which posꢀ
sess certain charge transfer character. P* denotes the final planariꢀ
(DPAC, see Figure 1a) as a prototype, we found that, upon
excitation at the lowest lying absorption band at ~350 nm,
DPAC exhibits a ~610 nm red emission that is independent of
the solvent polarity. The spectroscopic and dynamic results
lead us to propose a mechanism of skeletal motion toward the
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 10
zation state with the global minimum energy along the structural
scissoring the linkage leads to structural relaxation and conseꢀ
quently drastic changes of the emission property, demonstratꢀ
ing their fruitful potential in molecular sensing.
1
2
3
4
relaxation process. R*, I* and P* are all assumed to be the emisꢀ
sive states. The DFT optimized geometries of DPAC in R*, I*
and P* state are depicted. Note the difference in Θb and ΘS1, S2
among R*, I* and P*.
2. Results and Discussion
2.1 Design, Synthesis and Crystal structures. A series of
DPAC-n derivatives (n = 1~8, see Scheme 1), where n correꢀ
lates with the different alkyl lengths, were synthesized and
characterized. In a qualitative manner, we expected a decrease
in constraint strength upon increasing n. As depicted in
Scheme 1, treatment of DPAC with Vilsmeier reagent introꢀ
duced two aldehyde groups in the para sites of 9,14ꢀdiphenyl
of DPAC and gave the dialdehyde DPAC-CHO, which was
subsequently reduced by sodium borohydride to give the key
intermediate dibenzyl alcohol DPAC-OH, a structure that
could be further cyclized by a dialkoxybenzeneꢀalkylꢀether or
ꢀester linkage with different chain lengths. Therefore, diol
DPAC-OH reacted with Linker-1 1,2ꢀbisbromomethylbenzen
(NaH, THF, reflux) to deliver the smallest macrocycle DPAC-
1 in this work. To tune the chain length, a series of compounds
Linker-n (n = 2~8, see Scheme 1) endꢀcapped with two carꢀ
boxyl groups were prepared by nucleophilic substitution of
1,2ꢀbenzenediol with bromoꢀalkylꢀacids bearing different alkyl
chain lengths (see Scheme S1 in the supporting information
(SI)). Finally, various macrocycles DPAC-n (n = 2~8) were
obtained from esterification of diol DPAC-OH with Linker-n
(n = 2~8) using EDCI/DMAP (see Scheme 1). Furthermore,
the molecular structures of DPAC-n (n = 1, 3, 4 and 8) were
further confirmed by singleꢀcrystal Xꢀray diffraction analysis.
5
6
7
8
As a result, at the early time domain (a few to few tens of ps)
after FranckꢀCondon excitation, the temporal evolution of the
emission of DPAC in the regular organic solvents reveals
timeꢀdependent panchromatic emission, consisting of initial
charge transfer (R*), intermediate (I*) and the planarized,
global minimized emission (P*), which is drastically different
from that of solely red emission (P*) observed in the roomꢀ
temperature steady state approach. The dynamics of structural
relaxation of DHPs were affected by the steric hindrance
raised by N,N'ꢀdisubstituted side chain motion. That is, the
more steric hindrance induced by the bulky side chain may
cause further retardation of structural relaxation along the plaꢀ
narization. This suggests the feasibility of “controlling” the
planarization process through constraining the N,N'ꢀ
disubstituted side chain motion.
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 1. The Synthetic Route and Chemical Structures
of
Chemically
Locked
9,14-Diphenyl-9,14-
dihydrodibenzo[a,c]phenazines DPAC-na
Figure 2. Single crystal structures of (a) DPAC-1, (b) DPAC-3,
(c) DPAC-4 and (d) DPAC-8 from their two angles of view with
the values of Θb, ΘS1 and ΘS2.
a
EDCI: 1ꢀethylꢀ3ꢀ(3ꢀdimethylaminopropyl)carbodiimidehydrochloride,
DMAP: 4ꢀdimethylaminopyridine
By slow evaporation from dichloromethane/ethanol mixture
solutions, we were able to grow single crystals of DPAC-n (n
= 1, 3, 4 and 8). As shown in Figure 2, all corresponding Xꢀ
ray structures reveal significant nonplanar distortions inherited
from DPAC. For the convenience of analysis, three angles (Θb,
ΘS1 and ΘS2) in the singleꢀcrystal structures were put forward
and calculated. Using DPAC for illustration (see Figure 1a),
Θb (∠C1−N1−N2−C2) represents the bending angle between
planes 1 (C1, N1 and N2) and 2 (N1, N2 and C2), while the
angle between axes N1−N2 and N1−C3 or N2−C4 is denoted
as ΘS1 (∠C3−N1−N2) or ΘS2 (∠C4−N2−N1). As a result, Θb
of 134° for DPAC-1 signifies the sharp decrease of the bent
angle compared with those of DPAC-3, -4 and ꢀ8 (138ꢀ143º).
In comparison to the Θb of ~137° for DPAC, a decrease of up
Aiming to understand deeply the coupling of excitedꢀstate
electronic processes with conformational motions, we thus
conceived a series of annulated structures based on DPAC as a
core, where para sites of the 9,14ꢀdiphenyl moieties are chemꢀ
ically threaded by dialkoxybenzeneꢀalkylꢀester (or ꢀether) with
different chain lengths, forming a series of new DHPs, DPAC-
n (n = 1~8, see Scheme 1). Our aim was to impose various
degrees of constraint to impede the excitedꢀstate structural
deformation in hope of trapping a specific conformation so
that we would be able to take a snapshot of the structural evoꢀ
lution and the associated photoluminescence. As elaborated in
the following sections, the proof of concept is rendered by the
constraintꢀdependent spectral evolution of DPAC-n spanning
much of the panchromatic region. We also demonstrate that
2
ACS Paragon Plus Environment

Page 3 of 10
Journal of the American Chemical Society
to ~5o in Θb for DPAC-1 may signify its further nonplanar
DPAC-2~8 have a core chromophore similar to that of the
1
2
3
4
5
6
7
8
distortion upon imposing the small alkyl chain constraint (vide
infra). Meanwhile, ΘS1 and ΘS2 obviously rise from ~109° to
~118° and ~107° to ~123°, respectively. Note that in contrast
to DPAC, the ΘS1 of which is in close proximity to ΘS2, there
is an obvious disparity between ΘS1 and ΘS2 in DPAC-3, -4
and -8, which might be attributed to the flexible chain motion.
Theoretically, upon electronic excitation of N,N'ꢀdisubstitutedꢀ
constraintꢀfree DPAC in the ground state. On the other hand,
it also affirms the above structural analyses of a more crooked
structure for DPAC-1 (cf., DPAC-2~8), probably due to its
tight structure constraint (vide supra, see Figure 2), rendering
less πꢀdelocalization.
dihydrodibenzo[a,c]phenazines,
a
favorable bentꢀtoꢀplanar
motion takes place to elongate the
πꢀconjugation. In this sense,
9
conformational information derived from single crystal analyꢀ
sis is consistent with the explanation, as structural strain deꢀ
creases upon increase of the length of the alkyl chain.
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)
Toluene (Abs)
400 nm
DPAC-1
1.0
0.5
Cyclohexane (em)
Toluene (em)
Dichlomethane (em)
Acetonitrile (em)
340 nm
0.0
1.0
420 nm
DPAC-2
(b)
(c)
(d)
345 nm
0.5
0.0
1.0
460 nm
DPAC-3
345 nm
0.5
0.0
1.0
Figure 4. The chromaticity coordinates (CIE) of DPAC-n (n =
1~8) in toluene at room temperature. (λex = 360 nm). Insert: the
fluorescence images of DPAC-n (n = 1~8) in toluene under irraꢀ
diation of 365 nm UV light.
DPAC-4
478 nm
347 nm
0.5
0.0
1.0
DPAC-5
492 nm
(e)
Despite the similarity in absorption spectra, DPAC-n exhibꢀ
it remarkable differences in the emission properties. For
DPAC-1 and DPAC-2, the peak wavelength of fluorescence is
located at 400 nm and 420 nm, respectively, in toluene (Figure
3a and b), followed by a trend of red shifted emission of
DPAC-3 (460 nm), DPAC-4 (478 nm), DPAC-5 (492 nm),
DPAC-6 (590 nm), DPAC-7 (600 nm), and DPAC-8 (607 nm)
(Figure 3cꢀh). Accordingly, the emission peak wavelength is
distinctly chainꢀlength (i.e., n) dependent, being red shifted
upon increasing n from 1 to 8. Corresponding to emission
spectra, fluorescence images of DPAC-n (n = 1~8) clearly
exhibit the red shift of emission with increment of chainꢀ
length (see Figure 4). From DPAC-1 to DPAC-8 in toluene,
the emission color changes in an order of violetꢀblue (0.16,
0.03), palatinate blue (0.15, 0.07), maya blue (0.16, 0.18),
turquoise (0.18, 0.28), aquamarine (0.23, 0.39), yellow (0.45,
0.45), orange (0.53, 0.44), and dark orange (0.55, 0.44). Apꢀ
parently, the drastic chainꢀlength dependent emission spans
much of the visible range. For DPAC-6, in addition to the 590
nm emission, there appears to be another emission band maxꢀ
imized at ~490 nm, showing dual or even multiple emission
properties. Careful analysis also indicates that the blue and
green emission for DPAC-1~5 is dependent on solvent polariꢀ
ty, being red shifted as much as ~40 nm in peak wavelength
from cyclohexane to acetonitrile (Figure 3aꢀe), whereas the
red emission for DPAC-6~8 is only slightly dependent on
solvent polarity (Figure 3fꢀh). This is also well demonstrated
by the fluorescence images of DPAC-n (n = 1~8) in various
solvents (see Figure S2). For DPAC-4, for example, with the
solvent polarity increasing from nonpolar cyclohexane to
highly polar acetonitrile, the emission color tunes from blue
(0.17, 0.21) to green (0.21, 0.38), exhibiting remarkable solvaꢀ
tochromism. The results manifest the difference in the transiꢀ
tion characters of these nꢀdependent, multiple color emissions
with conformational motions.
347 nm
0.5
0.0
1.0
590 nm
DPAC-6
(f)
350 nm
0.5
0.0
1.0
600 nm
DPAC-7
(g)
350 nm
0.5
0.0
1.0
DPAC-8
607 nm
613 nm
(h)
(i)
350 nm
0.5
0.0
1.0
DPAC
351 nm
0.5
0.0
300
400
500
600
700
Wavelength (nm)
Figure 3. Steadyꢀstate absorption (in toluene, dashed line) and
photoluminescence (solid line) spectra of DPAC-1~8 (a) ~ (h) and
DPAC (i) in various solvents at room temperature (λex = 360 nm).
2.2 Steady State Photophysical Properties. The normalꢀ
ized absorption and emission spectra of DPAC-n (n = 1~8) in
various solvents from nonpolar cyclohexane to highly polar
acetonitrile are shown in Figure 3 and Figure S1 (absorption in
different solvents). The absorption peak wavelengths of 345ꢀ
350 nm (onset at ~400 nm) for DPAC-2~8 in various studied
solvents are about the same as that of DPAC (see Figure 3i).
In comparison, distinctly, both the absorption onset (~385 nm)
and peak wavelength (~340 nm) of DPAC-1 are slightly blue
shifted by ~5ꢀ15 nm. On the one hand, this result indicates that
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 10
It has been established that the nonplanar, Vꢀshaped DPAC emissions. This viewpoint is firmly supported in the later timeꢀ
1
2
3
4
5
6
7
8
and its derivatives undergo structural bending motion between
plane 1 and plane 2, i.e., along Θb (see Figure 1b) in the elecꢀ
tronically excited state, reaching a more planar configuration
to extend the πꢀconjugation and hence an energy minimum.8a
The bending process along Θb is accompanied by the simultaꢀ
neous optimization of ΘS1 and ΘS2 and hence encounters the
steric hindrance introduced by the N,N′ꢀdisubstituted sideꢀ
chains, which may evolve to a local minimum energy. As for
the local minimum states, the nonplanar structure may localize
the N,N′ꢀdisubstituents and dibenzo[a,c]phenazine moieties
such that the transition possesses charge transfer character.
Thus, previous timeꢀresolved measurement was able to acquire
various transient emissions at different time domains ascribed
to initial charge transfer (R*), and intermediate (I*) and plaꢀ
narization (P*) emission (see Figure 1b).
resolved approach. Finally, independent of DPAC-1~8, norꢀ
mal Stokes shifted blue (~400ꢀ430 nm) emission was observed
for all titled compounds in solid (see Figure S4). The result
firmly supports that the planarization process of DPAC-1~8
required largeꢀamplitude structural deformation, which is proꢀ
hibited in solid due to the lattice constraint.
2.3 Time Resolved Emission Spectroscopy. To gain inꢀ
depth insight into the underlying relaxation mechanism, we
then further performed timeꢀresolved measurement. Two repꢀ
resentative compounds, DPAC-5 and DPAC-6, will receive
particular focus. DPAC-6 is of prime concern due to its obviꢀ
ous two emissions maximized at 490 and 590 nm, which posꢀ
sibly originate from intermediate and planarization emission,
respectively. DPAC-5 is selected as the other prototype beꢀ
cause it has a proximal alkyl chain length with respect to
DPAC-6 but exhibits a very different emission spectrum.
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
In this study, via inserting an alkyl chain to impose the
strain energy and halt the planarization midway, we were able
to resolve different statuses of the structural evolution of the
DPAC core chromophore even in a steady state manner,
which is tentatively attributed to the initially prepared state
emission (400ꢀ420 nm, DPAC-1 and DPAC-2), the intermeꢀ
diate emission (460ꢀ520 nm, DPAC-3~5) and the planarizaꢀ
tion emission (600ꢀ620 nm, DPAC-7 and DPAC-8). Despite
the solvent independent absorption spectra, the shortꢀ
wavelength emission bands (400ꢀ520 nm) are red shifted sigꢀ
nificantly upon increasing the solvent polarity, indicating their
excitedꢀstate charge transfer property and hence the occurꢀ
rence of solvatochormism. Conversely, the red emission band
originating from planarized structure is much less subject to
the solvent polarity, confirming its delocalized ππ* character.
As for DPAC-5, we first carried out a femtosecond fluoresꢀ
cence upꢀconversion study in toluene to probe the reaction
dynamics in the early time domain. The results are shown in
Figure 5a, and pertinent data are listed in Table 1. Upon 370
nm excitation and monitoring at the blue side of the emission
(450 nm), the relaxation dynamics for DPAC-5 exhibited an
instant rise (< system response of 120 fs), followed by a very
fast decay component of 23 ± 2 ps decay components. Upon
monitoring at 500 nm, 550 nm, 600 nm, and 650 nm, shown in
Figure 5a, a finite rise component (negative preꢀexponential
value) gradually appeared, which was fitted to be within 23 ±
2 ps. Also, independent of the monitored emission wavelength,
there existed a very long population decay component that
could not be completely acquired by the upꢀconversion techꢀ
nique. Alternatively, the long population decay kinetics was
resolved by the picoꢀnanosecond time correlated single photon
counting (TCSPC) measurement. The results depicted in Figꢀ
ure 5b reveal a single exponential decay component of 10.2 ns
for all monitored emission wavelengths, indicating only a sinꢀ
gle population decay component.
To better understand the solvatochromism for short waveꢀ
length emissions of DPAC-1~6, the solvent effect on the
emission properties can be further described by the Lippertꢀ
Mataga equation, which is and expressed as
ꢁv = vabs – vem = 2ꢁꢂ2ꢁf/(hca3) + constant
where ꢁv denotes the Stokes shift, vabs and vem correspond to
the absorption and the emission maxima in terms of waveꢀ
numbers (cm)ꢀ1, respectively, ꢁꢂ is the change of dipole moꢀ
ment between ground and excited states, h is Planck’s conꢀ
stant, c is the speed of light, ꢁƒ is the orientational polarizabilꢀ
ity of the solvent, and a is the Onsager solvent cavity radius
which is obtained by B3LYP/6ꢀ31+g(d,p) method in this study,
assuming a spherical shape of the molecule. Figure S3 showꢀ
cases the LippertꢀMataga plots of DPAC-1~6, and ꢁꢂ is then
calculated from the slope of the LippertꢀMataga plots and
listed in Table S1. The deduced ꢁꢂ values for DPAC-1~6 in
shortꢀwavelength emission bands are large (> 7.7 Debye),
indicating that the R* and I* have a much larger dipolar moꢀ
ment than their corresponding ground states, confirming their
excitedꢀstate charge transfer character. We also found that the
ꢁꢂ values for DPAC-1~6 decrease upon increase of the length
of the alkyl chain, due perhaps to the released constraint via
the longer alkylꢀester linkage. Later we will elaborate that the
intermediate corresponds to the skeleton motion of the side
chain without significant planarization.
Figure 5. (a) Femtosecond and (b) nanosecond fluorescence tranꢀ
sients of DPAC-5 in toluene at room temperature observed at
various wavelengths as depicted. See Table 1 for the fitting results.
(λex = 370 nm). Note: For clarity, a logarithm plot is used in (b).
Upon planarization, the emission is dominated by the ππ*
delocalized transition of dibenzo[a,c]phenazine moiety, which
has a negligible charge transfer character. Interestingly,
DPAC-6 in the excited state reaches a critical stage where the
intermediate emission decay rate is competitive with its rate of
planarization, resulting in both intermediate and planarization
4
ACS Paragon Plus Environment

Page 5 of 10
Journal of the American Chemical Society
Table 1. Fitting Results of TCSPC and Femtosecond Fluorescence Up-conversion Measurements for DPAC-5 and DPAC-6
in Toluene at Room Temperature
1
2
DPAC-5
DPAC-6
3
4
5
λprobe
(nm)
obs (preꢀexp. factor)b,c,d
τ
obs (ns)a
τ
obs (ns) (preꢀexp. factor)a,c
τ
obs (preꢀexp. factor) b,c,d
τ
6
7
8
9
τ1: 25 ps (0.94); τP: 10.2 ns (0.06)d
τ1: 17 ps (ꢀ0.40); τP: 10.2 ns (0.60)d
τ1: 17 ps (ꢀ0.47); τP: 10.2 ns (0.53)d
τ1: 20 ps (ꢀ0.42); τP: 10.2 ns (0.58)d
τ1: 25 ps (ꢀ0.52); τP: 10.2 ns (0.48)d
τ1: 0.33 (0.89), τ2: 9.01 (0.11)
τ1: 0.33 (0.84), τ2: 9.15 (0.16)
τ1: 0.33 (0.52), τ2: 9.33 (0.48)
τ1: 0.33 (ꢀ0.64), τ2: 9.47 (0.36)
τ1: 0.33 (ꢀ0.62), τ2: 9.61 (0.38)
τ1: 17 ps (0.978); τ2: 330 ps (0.007) d; τp: 9 ns (0.015)d
τ1: 9 ps (ꢀ0.310); τ2: 330 ps (0.590) d; τp: 9 ns (0.100)d
τ1: 13 ps (ꢀ0.472); τ2: 330 ps (0.526)d; τp: 9 ns (0.002)d
τ1: 17 ps (ꢀ0.443); τ2: 330 ps (ꢀ0.094)d; τp: 9 ns (0.463)d
τ2: 330 ps (ꢀ0.460) d; τp: 9 ns (0.540)d
450
500
550
600
650
τ: 10.0
τ: 10.7
τ: 10.2
τ: 10.3
τ: 10.2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a Lifetime was measured by using a TCSPC system with a pulsed hydrogenꢀfilled lamp as the excitation source. (λex = 370 nm)
b Lifetime was measured by using an upꢀconversion system with femtosecond excitation pulses. (λex = 370nm)
c Numbers in parenthesis are normalized preꢀexponential factors of the decay in percentage.
d The time constant τp was determined from the TCSPC result and used for the fitting of upꢀconversion signals.
500 to 550 nm, which correlated with the appearance of a long
rise component shown in the upꢀconversion study (see Figure
6b and Table 1). Extending the monitored wavelength to 650
nm, as shown in the TCSPC results (Figure 6b, lowermost),
the 0.33 ns decay component was not resolvable anymore.
Instead, a 0.33 ns rise component was resolved, followed by a
~9 ns population decay. This 0.33 ns long rise component can
be seen from the upꢀconversion study as well (Figure 6a, lowꢀ
ermost), although the fluorescence upꢀconverted data could
not be precisely fitted due to the small acquisition time range
of 50 ps.
Scheme 2. The Proposed Excited-state Planarization Pro-
cess of DPAC-n.a
Figure 6. (a) Femtosecond and (b) nanosecond fluorescence tranꢀ
sients of DPAC-6 in toluene at room temperature observed at
various wavelengths as depicted. See Table 1 for the fitting results.
(λex = 370 nm). Note: A logarithm plot is used in (b) for clarity to
show biexponential decay for 450 and 500 nm emissions.
In comparison to DPAC-5, much different relaxation dyꢀ
namics were resolved for DPAC-6. Figure 6a reveals the early
relaxation dynamics for DPAC-6 in toluene, and pertinent
fitted data are listed in Table 1. The relaxation dynamics at the
blue edge of emission (450 nm) consisted of an instant rise (<
120 fs), followed by a fast decay of 17 ps and rather small
amplitude, but much longerꢀlived, component that remained
constant at the acquisition range of 50 ps. This long decay
component was further probed by the nanosecond TCSPC.
The result shown in Figure 6b (uppermost) revealed biꢀ
exponential decay behavior with time constants fitted to be
0.33 ns and ~9 ns.
a For Θb, ΘS1 and ΘS2, please see Figure 1b for DPAC.
The above timeꢀresolved measurements clearly support the
concept that the degree of excitedꢀstate planarization of DPAC
can be controlled by chemically imposing an alkyl chain to act
as a brake. As shown in Scheme 2, upon excitation at the lowꢀ
est lying state (S1), DPAC-5 undergoes a structural relaxation
process from initially prepared state (R*) to an intermediate
(I*), i.e., R* → I* with a time constant of 23 ± 2 ps. The inꢀ
termediate I* is then subject to the strength imposed by the n =
5 alkyl chain, such that further structural evolution is prohibitꢀ
ed, resulting in a long population decay time of 10.2 ns. For
DPAC-6, we reasonably assign the time constants of the R*
→ I* and I* → P* processes to be 17 ps (τ1) and 0.33 ns (τ2),
respectively, followed by a population decay of the P* state of
~9 ns (τP). Clearly, the R* → I* processes for DPAC-5 (24 ps)
and DPAC-6 (17 ps) are longer than that of DPAC (2.8 ps,
Upon increasing the monitored emission wavelengths to
500 nm and 550 nm, as shown in Figure 6a for femtosecond
upꢀconversion data, the early 17 ps decay component observed
at 450 nm disappeared and became a 17 ps rise component.
Note that in Figure 6a (for 500 and 550 nm) the partial overlap
between 17 ps rise and 17 ps decay components cancelled out,
giving an instant rise component. Also, the long decay comꢀ
ponent increased the weighting ratio (% in preꢀexponential
factor) from 500 to 550 nm. The results of corresponding naꢀ
nosecond TCSPC (see Figure 6b) also clearly indicated that
the 0.33 ns decay component reduced its weighting ratio from
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 10
τ1),8a manifesting that the dialkoxybenzeneꢀalkylꢀester linkagꢀ
Debye and 2.4 Debye for DPAC-5 and DPAC-6, respectively)
1
2
3
4
5
6
7
8
es n = 5 (DPAC-5) and n = 6 (DPAC-6) increase the strain
during the R* → I* process. The increase in n releases strain
and hence results in a faster R* → I* process for DPAC-6
than that for DPAC-5. The prohibition of the I* → P* process
for DPAC-5 implies that the I* → P* process requires a greatꢀ
er degree of motion, such that the strain introduced by the
alkylꢀester linkage plays a more significant role (cf. R* → I*
process). This viewpoint is also supported by the I* → P*
process of 330 ps in DPAC-6, which is slower than that of 38
ps in DPAC8a by more than an order of magnitude. As a result,
the intermediate emission DPAC-6 in toluene can be clearly
observed in a steady state manner.
supports the excitedꢀstate charge transfer character. Unfortuꢀ
nately, the geometry optimized R* cannot be located due to
rather small R* → I* energy barrier. We also calculated the
vertical electronic transition from the ground state (R), which
is 384 nm and 390 nm for DPAC-5 and DPAC-6, respectively.
These values are similar to those calculated in other solvents
such as in toluene (384 nm for DPAC-5 and 390 nm for
DPAC-6). In other words, the absorption spectrum is much
less changed by the solvent polarity, reaffirming the occurꢀ
rence of charge transfer in the excited state, followed by solꢀ
vent relaxation and then planarization to account for the emisꢀ
sion spectral changes.
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
We also performed the timeꢀresolved measurement for
DPAC-1 and DPAC-8. DPAC-1 was expected to impose the
largest strain, as indicated indirectly by the smallest Θb angle
among the titled compounds (vide supra). Evidently, indeꢀ
pendent of the monitored emission wavelength (400ꢀ500 nm),
solely a single emission decay of 2.0 ns was resolved, which is
unambiguously ascribed to the population decay of the R*
state. Apparently, the R* → I* process for DPAC-1 is subject
to the high energy barrier due to the strain imposed by the
alkylꢀether chain. On the other hand, the timeꢀresolved results
shown in Figure S5 and Table S2 of SI for DPAC-8 revealed
nearly identical results with those of DPAC,8a indicating that
the alkylꢀester chain of n = 8 imposes negligible strain on the
overall R* → I* → P* process for DPAC-8.
At the ground state, both geometry optimized DPAC-5
(Figure S6) and DPAC-6 (Figure 7) have similar Θb of 136°.
The
optimized
ground
state
and
corresponding
Franck−Condon excited state are denoted as R and R*, respecꢀ
tively. The vertical electronic transition from R → R* is calcuꢀ
lated to be 384 nm for DPAC-5 and 390 nm for DPAC-6 in
toluene (see Table 2). Note that the toluene environment was
chosen because it has been widely exploited in the fluoresꢀ
cence upꢀconversion study. For DPAC-5, the computation
result showed that in addition to R*, there was only a global
minimum in the S1 potential energy surface (PES) located at
Θb = 134° (see Figure S6), which is denoted as I* to correlate
with the experimental result (vide supra). In stark contrast, and
importantly, DPAC-6 was found to have two energy minima,
i.e., a local minimum at Θb =134° denoted as I* and a global
minimum at Θb = 158° represented as P* (Figure 7).
2.4 Computational Approaches. We then developed theoꢀ
retical support for the strainꢀimposed mechanism of this
unique class of constrained Vꢀshaped compounds. In this
computational approach, we selected DPAC-5 and DPAC-6
as the prototypes and applied the timeꢀdependent density funcꢀ
tional theory (TDꢀDFT, see SI) to access the excited state
properties, in which the S0 → S1 absorption was calculated
directly from the FranckꢀCondon excitation of the geometry
optimized ground state. Then the vertical S1 → S0 transition
was calculated from the structurally optimized S1 state to obꢀ
tain the emission energy gap. The calculated absorption and
emission data are gathered in Table 2, including the experiꢀ
mental results in toluene for comparison.
For both DPAC-5 and DPAC-6, as shown in Table 3, the
calculated HOMOs are greatly localized in the center moiety
N,N′ꢀdiphenylꢀdihydropyrazine, while the LUMOs are all
mainly localized at the fused phenanthrene ring. The results of
computation clearly demonstrate that the lowest lying excited
states for DPAC-5 and DPAC-6 are in part associated with the
chargeꢀtransfer character (Table 3). We then computed the S0
→ S1 transition for the other DPAC-n compounds and found
that the HOMO/LUMO characters were all similar (see Table
S3 and S4 of SI), which is also the same as that of DPAC.8a
This, together with the negligible contribution of the alkylꢀ
ether or alkylꢀester chain to the frontier orbitals, rationalizes
similar absorption spectra for all the studied compounds.
Figure 7. The (a) Θb and (b) ΘS1, ΘS2 of the DFT optimized
geometry of DPAC-6 in the S1 excited state at the initially
prepared state (R*), the local minimum (I*) and the global minꢀ
imum (P*).
The vertical transition from I* → I for DPAC-5 was calcuꢀ
lated to be 492 nm, while the I* → I and P* → P transitions
for DPAC-6 were calculated to be 491 and 543 nm, respecꢀ
tively. The I* → I energy gap of ~490 nm for DPAC-5 and
DPAC-6 matches well the steadyꢀstate intermediate emission
around 490 nm (see Figure 3), while the calculated 543 nm for
the P* → P transition is consistent with the onset of freely
planarized emission for e.g. DPAC (or DPAC-8) in toluene
(see Figure 3).
We also performed the calculation of dipole moment for
both ground (B3LYP/6ꢀ31+g(d,p) method) and first excited
state (TDꢀB3LYP/6ꢀ31+g(d,p) method) based on their geomeꢀ
try optimized structures. As shown in Table S1, in a qualitaꢀ
tive manner, the calculated large dipole moment for I* in the
S1 state for DPAC-5 (22.0 Debye) and DPAC-6 (24.5 Debye,
see Table S1) versus the rather small dipole moment in S0 (6.6
6
ACS Paragon Plus Environment

Page 7 of 10
Journal of the American Chemical Society
Table 2. The Experimental and Calculated Optical Characteristics for the DPAC-n Compoundsa
1
2
3
4
5
6
7
8
Stokes shift (cmꢀ1)
Absorption S0 → S1
exp(nm) (ε/Mꢀ1cmꢀ1)
calc(nm)
Emission S1 → S0
Q.Y.
τ
obs (ns)b
f
f
λ
λ
λ
exp(nm)
400
420
460
478
492
490
590
600
607
λ
calc(nm)
435
ꢀEexp
4411
5175
7246
7898
8493
8163
11622
11904
12097
ꢀEcalc
0.20
0.43
0.59
0.54
0.55
2.01
4.64
7.53
8.95
10.20
DPACꢀ1
DPACꢀ2
DPACꢀ3
DPACꢀ4
DPACꢀ5
340 (6300)
345 (8600)
345 (14800)
347 (11900)
347 (10800)
375
380
382
387
384
0.0595
0.093
0.1009
0.1135
0.1282
0.132
3678
9
492
491
543
0.057
0.048
0.068
5716
5274
7224
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
0.33, 9.01 (450nm)
0.33, 9.61(650nm)
8.18
DPACꢀ6
350 (21200)
390
0.1332
0.34
DPACꢀ7
DPACꢀ8
350 (5000)
350 (5400)
389
390
0.1406
0.1349
0.24
0.26
8.32
a
Data were recorded in toluene at room temperature. λexp = experimental absorption/emission wavelengths, λcalc = calculated absorpꢀ
tion/emission wavelengths, ε = molar extinction coefficient, f = oscillator strengths, ꢀEexp = experimental Stokes shift, ꢀEcalc = calculated
b
Stokes shift, Q.Y. = emission, quantum yield. τobs = experimental emission lifetimes, which are measured by the timeꢀcorrelated single
photon counting (TCSPC, Edinburgh FL 900) method.
Table 3. The Molecular Frontier Orbitals for DPAC-5 and DPAC-6 in Their Optimized Ground-state Geometries and the
Relaxed Structures in the S1 Excited State at the Local Minima and Global Minima
DPAC-5
136° (R and R*)
DPAC-6
134° (I*)
158° (P*)
Bending Angle (Θb)
134° (I*)
136° (R and R*)
LUMO
HOMO
For fair comparison, we also attempted to locate the energy
minimum of any optimized structure for DPAC-1 in the excitꢀ
ed state and found only one global minimum in the S1 state at
Θb = 134°, which is the same as Θb = 134° for DPAC-1 in the
ground state. Therefore, we assign this state to be the optiꢀ
mized structure of the R* state (in S1) reached by the Franckꢀ
Condon excitation. The estimated 435 nm emission for
DPAC-1 correlates well with its steadyꢀstate emission band
maximized at 405 nm in toluene.
hindrance raised by the N,N′ꢀdisubstituted side chain. As for
the I* → P* process for DPAC-6, in addition to the changes in
Θb from 134° (I*) to 158° (P*), an increase of ΘS1/ΘS2 from
126°/122° to 138°/139° and a decrease of ∠C5−C3−N1−N2
from 87° to 22° were also obtained. Similar structural changes
of the R* → I* process were obtained for DPAC-5, evidenced
by an increase of ΘS1/ΘS2 from 120°/119° to 129°/127° and a
decrease of ∠C5−C3−N1−N2 from 106° to 92° (see Figure
S6).
In addition to the difference in Θb among R*, I* and P*, alꢀ
so noted is the changes of space/orientation on the N,N′ꢀ
disubstituted side chain among the three states. For example,
from an R* state at Θb = 136° to the local minimum I* at Θb =
134°, the results of calculation for DPAC-6, shown in Figure 7,
indicate the conformational change at the N,N′ꢀ
dialkoxybenzeneꢀalkylꢀester (or either) linkage via swinging
and rotating round the N,N′ꢀdiaryls site, i.e., an increase of
ΘS1/ΘS2 from 122°/122° to 126°/122° and a decrease of ∠
C5−C3−N1−N2 from 103° to 87°. The results indicate that
DPAC-6, along the structural relaxation, encounters steric
Apparently, the dynamics of structural relaxation should be
affected by the N,N′ꢀdialkoxybenzeneꢀalkylꢀester (or ether)
linkage. Unlike with DPAC, in addition to the steric hindrance
of N,N′ꢀdialkoxybenzene, the structure constraint imposed by
the cyclic alkylꢀester chain (Scheme 1 and Figure 2) adds an
additional barrier to hinder the planarization for DPAC-1~6
and hence results in the distinct steadyꢀstate emission spectra
and the corresponding relaxation dynamics (vide supra).
Due to the structural complexity, the computational apꢀ
proach to assess the activation energies along R* → I* and I*
→ P* PES for DPAC-6 (also, DPAC-7 and DPAC-8) unforꢀ
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 10
tunately failed. Experimentally, due to the use of a rotating or formation of lipids. It has been extensively exploited for
1
2
3
4
5
6
7
8
cell (clipped by two quartz plates) to contain the sample in the
fluorescence upꢀconversion measurement, the temperature
dependent study could not be performed. Alternatively, the
much longer, subꢀnanosecond I* → P* process (vide supra)
for DPAC-6 made possible temperatureꢀdependent study usꢀ
ing the TCSPC technique. Accordingly, the experiment was
performed in the temperature range of 283ꢀ243 K, where the
viscosity of toluene varies only slightly (0.66ꢀ1.29 cp), to
avoid environmental perturbation.9 We then assumed that I*
→ P* is the dominant decay for I*, which follows the Arrheꢀ
nius type of thermally activated process.
catalytic chiral resolution because of the growing demand for
enantiopure drugs.10 The underlying mechanism of triacylꢀ
glycerol lipase catalyzed DPAC-5 hydrolysis and the associatꢀ
ed signal transduction are illustrated in Scheme 3a.
(a)
DPAC-5 in ACN
1.0
Catalyzed after 5 hr
Catalyzed after 10 hr
Catalyzed after 15 hr
Catalyzed after 20 hr
Catalyzed after 25 hr
0.5
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
0.0
1.0
a / RT
kobs = Ae−ꢀE
0.8
0.6
0.4
0.2
0.0
(b)
420nm
where kobs is the I* emission decay rate constant, and A and Ea
denote the frequency factor and activation energy of the I* →
P* planarization process. Accordingly, a logarithm of kobs vs.
the reciprocal of absolute temperature gave a straight line, and
Ea and A were calculated to be 2.3 kcal/mol and 1.59 × 1011 sꢀ1,
respectively (see Figure 8). The relatively low frequency facꢀ
tor implies that the I* → P* process is associated with a large
amplitude motion, consistent with the appreciable changes of
Θb and ΘS1/ΘS2 calculated along the planarization process.
610nm
25
0.5
0.0
5
10
15
Time (hr)
20
400
500
600
700
Wavelength (nm)
Figure 9. (a) The emission spectra of DPAC-5 in acetonitrile
(black solid square) and DPAC-5 catalyzed by lipase with time
(solid circles of various colors). (b) The difference in the emission
spectra between DPAC-5 in acetonitrile and after catalysis by
lipase. (λex = 360 nm). Inset: The time dependent intensity changꢀ
es of 420 nm and 610 nm emission bands.
21.8
21.6
21.4
k
Theoretically, upon catalysis by lipase, the dual ester linkꢀ
ages of DPAC-5 will be hydrolyzed, forming a DPAC-OH
and a carboxylic acid (or carboxylate, depending on pH) prodꢀ
uct by rupturing the cyclic chain. Therefore, upon electronic
excitation, planarization can take place for DPAC-OH, free
from the constraint previously imposed by the alkylꢀester
chain. The result should lead to drastic changes of the emisꢀ
sion that can be exploited as an indicator of lipase activity.
Experimentally, Figure 9a shows the emission spectra of
DPAC-5 in acetonitrile (ACN) and after catalysis by lipase
with time in ACN:H2O = 99:1 (normalized to the same intensiꢀ
ty of DPAC-5 (peak wavelength) in ACN). The corresponding
emission changed from a single 500 nm band to multiple
emission bands. We then took the difference in these two
emission spectra, and the resulting spectrum is shown in Figꢀ
ure 9b, in which two emission band maxima at 420 nm and
610 nm are observed, and the timeꢀdependent intensity changꢀ
es of these two emission bands are inserted in Figure 9b.
While the 610 nm emission is unambiguously assigned to plaꢀ
narized P* emission of DPAC-OH, the origin of the 420 nm
emission band has to be further explored. We then performed
a controlled experiment by hydrolyzing DPAC-5 in a basic
aqueous solution (pH = 11), followed by extraction of the hyꢀ
drolyzed products by nonpolar solvents such as toluene or
hexane. Lipase was then added to the products in nonpolar
solvents. Due to the presence of trace water, the Le Chatelier
principle drove the reaction mainly toward esterification. The
emission spectra are shown in Figure S7 and Figure S8 of SI
for nꢀhexane and toluene, respectively. Clearly, in addition to
the minor 500 nm emission restored by the DPAC-5 product,
the 420 nm band is the dominant component, which may be
ascribed to a cluster of esterification product from different
assemblies (see Scheme 3 (a)+(b) processes). Consequently,
the rigid internal environment of this cluster prohibits the
21.2
E
= 2.30 (kcal/mol)
a
A = 1.59x1011
(
s-1
)
21.0
0.0034
0.0036
0.0038
1/T (K-1)
0.0040
0.0042
Figure 8. Arrhenius plot of the logarithm of rate constant (kobs
versus reciprocal of the absolute temperature of DPAC-6 in toluꢀ
ene from 283 to 243 K.
)
Scheme 3. (a) The Proposed Hydrolyzed Reaction of
DPAC-5 by Triacylglycerol Lipase (3.1.1.3). (a)+(b) The
Overall Possible Hydrolysis Mechanism Deduced from
Experimental Results
2.5 Ring Opening and Its Potential Application. As opꢀ
posed to the imposition of constraint the by alkylꢀester linkagꢀ
es, releasing the structural freedom by severing the alkylꢀester
linkages for DPAC-2~6 should lead to drastic ratiometric
changes of emission. We herein demonstrate a prototypical
experiment for sensing triacylglycerol lipase (3.1.1.3) via its
catalytic reaction to hydrolyze DPAC-5. Lipase is one of the
most important industrial enzymes that catalyze the hydrolysis
8
ACS Paragon Plus Environment

Page 9 of 10
Journal of the American Chemical Society
Mataga, N.; Chosrowjan, H.; Taniguchi, S., J Photoch Photobio C
structural planarization, giving rise to the 420 nm emission
1
2
3
4
5
6
7
8
2005, 6, 37; (e) Cho, D. W.; Fujitsuka, M.; Choi, K. H.; Park, M.
J.; Yoon, U. C.; Majima, T., J Phys Chem B 2006, 110, 4576.
(3) Grabowski, Z. R.; Rotkiewicz, K.; Rettig, W., Chem. Rev.
2003, 103, 3899.
(4) (a) Chien, Y. Y.; Wong, K. T.; Chou, P. T.; Cheng, Y. M., Chem
Commun 2002, 2874; (b) Terenziani, F.; Painelli, A.; Katan, C.;
Charlot, M.; BlanchardꢀDesce, M., J. Am. Chem. Soc 2006, 128,
15742; (c) Wong, K. T.; Ku, S. Y.; Cheng, Y. M.; Lin, X. Y.; Hung,
Y. Y.; Pu, S. C.; Chou, P. T.; Lee, G. H.; Peng, S. M., J. Org.
Chem. 2006, 71, 456; (d) Liu, C.; Tang, K. C.; Zhang, H.; Pan, H.
A.; Hua, J. L.; Li, B.; Chou, P. T., J. Phys. Chem. A 2012, 116,
12339.
(5) (a) Demchenko, A. P.; Tang, K.ꢀC.; Chou, P.ꢀT., Chem. Soc.
Rev. 2013, 42, 1379; (b) Padalkar, V. S.; Seki, S., Chem. Soc. Rev.
2016, 45, 169; (c) McDonald, L.; Wang, J.; Alexander, N.; Li, H.;
Liu, T.; Pang, Y., J. Phys. Chem. B 2016, 120, 766.
(6) (a) Clegg, R. M., Curr. Opin. Biotechnol. 1995, 6, 103; (b) Jia,
X.; Chen, Q.; Yang, Y.; Tang, Y.; Wang, R.; Xu, Y.; Zhu, W.; Qian,
X., J. Am. Chem. Soc. 2016, 138, 10778; (c) Sasmal, D. K.; Yadav,
R.; Lu, H. P., J. Am. Chem. Soc. 2016, 138, 8789.
(7) (a) Shukla, D.; Wan, P., J. Am. Chem. Soc. 1993, 115, 2990; (b)
Vollmer, F.; Rettig, W.; Birckner, E., J. Fluoresc. 1994, 4, 65; (c)
Doroshenko, A. O., Theor. Exp. Chem. 2002, 38, 135; (d) Chen, Y.;
Zhao, J.; Guo, H.; Xie, L., J. Org. Chem. 2012, 77, 2192; (e) Yuan,
C. X.; Saito, S.; Camacho, C.; Irle, S.; Hisaki, I.; Yamaguchi, S., J.
Am. Chem. Soc 2013, 135, 8842; (f) Yuan, C. X.; Saito, S.;
Camacho, C.; Kowalczyk, T.; Irle, S.; Yamaguchi, S., Chem-Eur J
2014, 20, 2193; (g) Saito, S.; Nobusue, S.; Tsuzaka, E.; Yuan, C.;
Mori, C.; Hara, M.; Seki, T.; Camacho, C.; Irle, S.; Yamaguchi, S.,
Nat. Commun. 2016, 7.
(8) (a) Zhang, Z. Y.; Wu, Y. S.; Tang, K. C.; Chen, C. L.; Ho, J. W.;
Su, J. H.; Tian, H.; Chou, P. T., J. Am. Chem. Soc 2015, 137, 8509;
(b) Chen, J. W.; Wu, Y. S.; Wang, X. D.; Yu, Z. Y.; Tian, H.; Yao, J.
N.; Fu, H. B., Phys. Chem. Chem. Phys. 2015, 17, 27658.
(9) Assael, M. J.; Dalaouti, N. K.; Dymond, J. H., Int. J. Thermo-
phys. 2000, 21, 291.
(R*) from the initially prepared state.
3. Conclusion
In summary, we present a novel concept for modulating the
excitedꢀstate process by systematically imposing various
strengths of the constraint to the N,N′ꢀdisubstitutedꢀ
dihydrodibenzo[a,c]phenazine donorꢀacceptor dyads DPAC-n
(n = 1~8), in which n correlates with the alkyl length, such that
the strength of the spatial constraint decreases as n increases,
in an aim to harness the conformational flexibility. Among
DPAC-n, the structures of DPAC-1, DPAC-3, DPAC-4 and
DPAC-8 were identified by Xꢀray crystal analysis. In stark
contrast to the large Stokesꢀshifted 610 nm red emission for
DPAC (absorption onset ~400 nm), drastic chainꢀlength n
dependent emission spanning from blue (n = 1, 2, ~400 nm)
and blueꢀgreen (n = 3~5) to greenꢀred (n = 6) and red (n = 7, 8,
~610 nm) was observed and has been verified to originate
from the charge transfer and planarization species, respectiveꢀ
ly. Scissoring the linkage releases the structural constraint and
hence results in drastic changes of the emission; this approach
was successfully exploited for sensing lipase activity using
DPAC-5. The results thus demonstrate for the first time the
systematic control of excitedꢀstate planarization of DPAC-n.
Based on the formation and release of the constraint and hence
the on and off of the planarization, future development of
molecule machines driven by light may be feasible in N,N′ꢀ
disubstitutedꢀdihydrodibenzo[a,c]phenazines, which deserves
much pursuit in both fundamental and applications.
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
ASSOCIATED CONTENT
Supporting Information. Additional synthetic detail, crystalloꢀ
graphic (including CIF), computational, and spectroscopic data
along with complete references are provided. This material is
available free of charge via the Internet at http://pubs.acs.org.
(10) (a) Tang, L. H.; Xia, L. M.; Min, S.; Guo, H. Y., Appl Bio-
chem Biotech 2007, 142, 194; (b) Kurtovic, I.; Marshall, S. N.;
Zhao, X.; Simpson, B. K., Fish Physiol Biochem 2010, 36, 1041.
AUTHOR INFORMATION
Corresponding Author
Email: tianhe@ecust.edu.cn (H. Tian).
Email: chop@ntu.edu.tw (P.ꢀT. Chou).
Author Contributions
§These three authors made equal contributions.
Notes
The authors declare no competing financial interests.
ACKNOWLEDGMENT
P.ꢀT. Chou thanks the Ministry of Science and Technology, Taiꢀ
wan, for financial support. H. Tian thanks National 973 Program
(2013CB733700) and NSFC/China.
REFERENCES
(1) (a) He, G.; Guo, D.; He, C.; Zhang, X.; Zhao, X.; Duan, C.,
Angew. Chem. Int. Ed 2009, 48, 6132; (b) Zhang, X.; Rehm, S.;
SafontꢀSempere, M. M.; Würthner, F., Nat Chem 2009, 1, 623; (c)
Mo, H.ꢀW.; Tsuchiya, Y.; Geng, Y.; Sagawa, T.; Kikuchi, C.;
Nakanotani, H.; Ito, F.; Adachi, C., Adv. Funct. Mater. 2016, 26,
6703ꢀ6710.
(2) (a) Rotkiewicz, K.; Grellmann, K. H.; Grabowski, Z. R., Chem.
Phys. Lett. 1973, 19, 315; (b) Birks, J. B., Rep Prog Phys 1975, 38,
903; (c) Rettig, W., Angew Chem Int Edit 1986, 25, 971; (d)
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 10
Table of Contents artwork
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
10
ACS Paragon Plus Environment