The origin of bisignate circularly polarized luminescence (CPL) spectra from chiral polymer aggregates and molecular camphor: anti-Kasha's rule revealed by CPL excitation (CPLE) spectra

Sang Thi Duong and Michiya Fujiki *
Graduate School of Materials Science, Nara Institute of Science and Technology, 8916-5, Takayama, Ikoma, Nara 630-0192, Japan. E-mail: fujikim@ms.naist.jp

Received 10th June 2017 , Accepted 16th July 2017

First published on 17th July 2017


The first circularly polarized luminescence (CPL) excitation (CPLE) measurements of poly(fluorene-alt-bithiophene) (PF8T2) aggregates with bisignate CPL/circular dichroism (CD) signals and, for comparison, bisignate CPL signal camphor were achieved. Based on this finding, we propose that Kasha's rule can be broken in chiral luminophore systems. The |gem| values of PF8T2 hetero-aggregate with help of helical polysilanes (PSi-S/-R) as sacrificial scaffoldings boosted to 0.05–0.08 at ≈510 and ≈540 nm associated with a high quantum yield of 0.33. The gem values arose from hugely amplified |gabs| values of 0.15–0.25 at ≈500 nm and ≈510 nm. Moreover, PF8T2 homo-aggregate, maintaining the bisignate CPL and CD spectral profiles, was generated by PSi-S/-R selective photoscissoring reaction at 313 nm for 10 s.


Introduction

Kasha's rule is one of the most crucial guiding principles of how luminophores relax radiatively and non-radiatively to the ground state (S0) when they are in Sn (n = 1, 2, 3…) excited states.1 Several exceptions, as exemplified by photoluminescence (PL) and photochromism from the Sn (n = 2) state and dual PL phenomena, have received much attention.2 However, whether the rule is always valid for chiral luminophores remains an open question.

The circularly polarized luminescence (CPL) spectrum reflects the S1-state chirality, whereas the circular dichroism (CD) spectrum indicates the S0-state chirality. To date, several polymeric chiral luminophores as aggregates3 and thin films4 have been observed to provide bisignate CPL signals associated with bisignate CD signals. The origin of bisignate CD signals can be interpreted in terms of a coupled oscillator theory.5 This theory prompted an attempt to characterize chiral molecules with bisignate CD signals.6,7 Bicyclic camphor is of particular interest because of its bisignate CPL characteristics.8,9 Moreover, the signs of the CPL spectra of several molecules are definitively opposite to those of the CD signals at the first Cotton band.10

In previous papers, we assumed that the bisignate CPL characteristics of poly[(9,9-di-n-octylfluorenyl-2,7-diyl)-alt-bithiophene] (PF8T2) (Fig. 1) aggregate endowed with limonene chirality3a and photon chirality3c arise from the corresponding bisignate CD characteristics, while the CPL and CD signs are commonly of the same origin. To verify this hypothesis, herein, we measured the first CPL excitation (CPLE) spectra monitored at the (−)- and (+)-sign CPL signals of optically active PF8T2 aggregates and, for comparison, bisignate CPL-functional D-/L-camphor in cyclohexane (c-Hex) and ethanol (EtOH).


image file: c7py00958e-f1.tif
Fig. 1 Chemical structures of PF8T2, PSi-S, PSi-R and D-/L-camphor.

Optically active PF8T2 aggregates were efficiently obtained with the help of an enantiomeric pair of poly(n-hexyl-(S)-2-methylbutylsilane) (PSi-S) and poly(n-hexyl-(R)-2-methylbutylsilane) (PSi-R) (Fig. 1)11 using achiral PF8T2. Non-aggregate PF8T2 in chloroform did not show any CD signal (Fig. S2, ESI). The polysilane helicity transfer allows for the generation of CPL- and CD-active π-conjugated polymers during catalyst-free mirror symmetry breaking aggregation in achiral solvents.11

Results and discussion

PF8T2 hetero-aggregates with PSi-S and -R

Firstly, to detect the CPLE spectra of PF8T2 aggregates, we performed optimization to boost the CD amplitudes by optofluidically tuning the refractive index (nD)3a,c,11,12 of the chloroform (CHCl3)–methanol (MeOH) cosolvent using values ranging from 1.375 to 1.426. The optofluidic effect is crucial to confine and store photon energy into μm-size particles. The particles act as an optical cavity due to whispering gallery and Rayleigh scattering modes to boost (chir)optically signals. For example, an ultralow-threshold droplet laser made of 20-to-40 μm size rhodamine-loaded benzyl alcohol (nD = 1.54) in a fluorinated ether (nD = 1.29),12b a gigantic enhanced CD signal (gabs = −0.35 at 325 nm) of 5 μm size helical polysilane aggregates (nD = 1.7) in achiral solvents (nD = 1.37)12c and circular differential scattering of 100 μm size PMMA particle (nD = 1.49) in limonene (nD = 1.47)12d are reported so far.

The gabs values at the first Cotton band (≈500 nm) of the PF8T2 hetero-aggregate with PSi-S (and PSi-R) in a 1-to-1 ratio of repeating units was resonantly magnified at a specific nD (≈1.4145) of the cosolvent (CHCl3/MeOH = 2.2/0.8 (v/v)) (Fig. S3(a) and (b), ESI). The raw absorbance at 500 nm was less than 0.2, thereby avoiding the self-absorption effect in the CPL/CPLE study (Fig. S4(a), ESI).

Fig. 2(a) shows the representative bisignate CD and corresponding UV-vis spectra of PF8T2 hetero-aggregates with PSi-S (and PSi-R). The aggregates exhibit hugely boosted gabs values at 504 nm (−0.19 [PSi-S] and +0.25 [PSi-R]) and weak gabs values at 405 nm (+0.046 [PSi-S] and −0.050 [PSi-R]). The aggregates clearly show bisignate CD bands (grey zones) near 321 nm due to PSi aggregation;11,12c the ∓CD (PSi-S) and ±CD (PSi-R) signs arise from the original CD profiles in CHCl3 (Fig. S2, ESI).


image file: c7py00958e-f2.tif
Fig. 2 (a) CD and ultraviolet-visible (UV-vis) and (b) CPL and PL spectra (excited at 420 nm) of PF8T2 hetero-aggregate with PSi-S (blue lines) (PSi-R, red lines) in a 1-to-1 ratio. CPLE and PLE spectra of the hetero-aggregates with PSi-S (blue lines) (PSi-R, red lines) collected at (c) 575 nm and (d) 490 nm (PSi-R) and 495 nm (PSi-S). [PF8T2]0 = [PSi-R]0 = 1.0 × 10−5 M in chloroform–methanol (2.2/0.8 (v/v)). For CPLE measurement monitored at 495 nm due to weakness of PF8T2 with PSi-S, [PF8T2]0 = [PSi-S]0 = 1.0 × 10−5 M in chloroform–methanol (2.1/0.9 (v/v)) was chosen.

Choosing the good and poor solvents appropriately is crucial to boosting the CD and CPL magnitudes. When EtOH is employed, the |gabs| value of the PF8T2 aggregates with PSi-S/-R in CHCl3–EtOH (2.1[thin space (1/6-em)]:[thin space (1/6-em)]0.9 (v/v)) decreases by approximately 25% but maintains bisignate CD profiles (Fig. S5, ESI). Nearly identical resonance behaviour was observed at a specific nD of 1.419 (Fig. S3(c) and (d) in ESI). Hereafter, the gabs and gem values are used to discuss the degree of circular polarization.3–11,12c,13

Fig. 2(b) depicts representative bisignate CPL at four wavelengths associated with PL spectra (excited at 420 nm) of PF8T2 aggregates with PSi-S and PSi-R. The greatly boosted gem values are −0.057 (PSi-S) and +0.065 (PSi-R) at 533 nm, −0.048 (PSi-S) and +0.054 (PSi-R) at 513 nm, −0.051 (PSi-S) and +0.069 (PSi-R) at 578 nm and +0.004 (PSi-S) and −0.006 (PSi-R) at 492 nm.

Fig. 2(c) displays the CPLE spectra collected at (+)-/(−)-sign CPL signals in the range of 575 nm from PF8T2 aggregates with PSi-R and PSi-S. The (−)-sign of the CPLE spectra in the range of 270–520 nm is identical to the (−)-sign of the CPL spectra; conversely, the (+)-sign of the CPLE spectra is identical to the (+)-sign of the CPL spectra.

Fig. 2(d) presents the CPLE spectra obtained at (−)-/(+)-sign CPL signals in the range of 490–495 nm from PF8T2 aggregates with PSi-R and PSi-S. The (−)-sign CPLE spectra of 250–440 nm with λext ≈ 445 nm is obvious and identical to the (−)-sign of the weak CPL spectra with λext 490 nm; similarly, the expected (+)-sign of the CPLE spectra of 250–440 nm with λext ≈ 440 nm is obvious and identical to the (+)-sign of the weak CPL spectra with λext 480 nm.

Two PLE spectral data sets (Fig. S6, ESI) of PF8T2 aggregates with PSi-R/-S monitored at 420–440 nm and 550–580 nm and the corresponding PL data sets (excited at 420 nm) provide direct evidence of the coexistence of two broad π–π* transitions at approximately 300–400 nm with a tail of ≈450 nm (peak: 375 nm) and 330–520 nm (peak: 450 nm). These results are consistent with the chiroptical analysis of CPL/CPLE spectral data sets (Fig. 2(c) and (d)).

These boosted CD/UV-Vis/CPL/CPLE spectral characteristics of the hetero-aggregates in the cosolvent are maintaining when the dried aggregates are prepared by removing the cosolvent in a stream of dry nitrogen gas flow at room temperature. The aggregates with a high nD can be dispersed in a high-vacuum silicone grease with a low nD (Dow-Corning-Toray). The grease was coated onto a quartz substrate.

Fig. 3(a) depicts the bisignate CD and UV-vis spectra of the grease on quartz containing PF8T2 aggregates with PSi-S and PSi-R. The specimen exhibits similarly hugely boosted gabs values at 510 nm (−0.16 [PSi-S] and +0.25 [PSi-R]) associated with weak gabs at 385 nm (+0.23 [PSi-S] and −0.45 [PSi-R]). The specimen clearly shows bisignate CD and UV bands (grey zones) near 325 nm due to PSi aggregation.11,12c


image file: c7py00958e-f3.tif
Fig. 3 (a) Raw CD and UV-vis and (b) normalised raw CPL and PL spectra excited at 390 nm of PF8T2 aggregate with PSi-S (blue lines) (PSi-R, red lines) in a 1-to-1 ratio in silicone grease onto quartz substrate. CPLE and PLE spectra of the aggregates with PSi-S (blue lines) (PSi-R, red lines) collected at (c) 575 nm and (d) 480–490 nm in the grease onto the substrate. The aggregates are generated that [PF8T2]0 = [PSi-R]0 = [PSi-S]0 = 1.0 × 10−5 M in chloroform–methanol (2.2/0.8 (v/v)).

Fig. 3(b) shows the bisignate CPL associated with PL spectra excited at 390 nm of PF8T2 aggregates with PSi-S and PSi-R. The greatly boosted gem values are −0.058 (PSi-S) and +0.074 (PSi-R) at 537 nm and −0.048 (PSi-S) and +0.076 (PSi-R) at 514 nm. Moreover, (+)-sign CPL signal (PSi-S) and (−)-sign CPL signal (PSi-R) at ≈470 nm could be seen.

Fig. 3(c) presents the CPLE spectra monitored at (+)-/(−)-sign CPL signals at 575 nm of PF8T2 aggregates with PSi-R and PSi-S. The (−)-sign of the CPLE spectrum in the range of 280–530 nm is identical to the (−)-sign of the CPL spectrum in the range of 450–600 nm and the (−)-sign of the CD spectrum in the range of 490–700 nm; conversely, the (+)-sign of the CPLE spectrum in the range of 280–530 nm is identical to the (+)-sign of the CPL spectrum in the range of 440–650 nm and the (+)-sign of the CD spectrum in the range of 440–650 nm.

Fig. 3(d) compares the CPLE spectra monitored at (−)-/(+)-sign, ultraweak CPL signals monitored at 480–490 nm of the PF8T2 aggregates with PSi-R and PSi-S. The (+)-sign of the CPLE spectrum with λext 383 nm in the range of 250–440 nm is obvious and identical to the (+)-sign of the ultraweak CPL spectrum at ≈470 nm; similarly, the (−)-sign of the CPLE spectrum in the range of 250–440 nm is obvious and identical to the (−)-sign of the ultraweak CPL spectrum at ≈460 nm.

According to the energy migration or energy transfer scenarios1b,c associated with Kasha's rule,1a molecules, polymers and colloids that are photoexcited to Sn (n = 1, 2, 3…) states in homogeneous solution emit from the lowest photoexcited energy state (e.g., the S1 state). If the S1-state is forbidden or has an extremely small oscillator transition, S2-state emission occurs. This idea is applicable to hetero-aggregates and co-crystals consisting of multiple S1-states.

The two CPLE spectral sets of PF8T2 aggregates with PSi-R collected at 480–495 nm and 575 nm suggest the breaking of Kasha's rule and the following energy transfer scenarios: the dominant (+)-sign CPL signal at 500–700 nm arises from the (+)-sign CD and (+)-sign CPLE signals at 280–530 nm, whereas the weaker (−)-sign CPL signal in the 420–490 nm range arises from the (−)-sign CD and (−)-CPLE signals at 250–440 nm. Conversely, in PF8T2 with PSi-S, the dominant (−)-sign CPL signal at 500–700 nm arises from the (−)-sign CD and CPLE signals at 330–530 nm, while the ultraweak (+)-sign CPL signal at 430–490 nm is attributed to the (+)-sign CD and (+)-sign CPLE signals at 330–440 nm.

These CPLE/PLE signals do not reflect the bisignate CD (313 and 326 nm) and UV signals (322 nm) characteristic of PSi aggregates. Thus, no direct electronic interactions occur between σ-conjugated helical main chains and π-conjugated main chains, regardless of the CPL wavelengths monitored. Only attractive London dispersion forces between chiral alkyl groups/helices of PSi-S (PSi-R) and n-octyl groups of non-helical PF8T2 cooperatively allow for the emergence of optically active PF8T2 aggregates with the help of aggregation-inducing methanol as a commonly used poor solvent of PSi-S (PSi-R) and PF8T2.

Noted that quantum yields (ΦPL) of PF8T2 hetero-aggregate with PSi-S/PSi-R in CHCl3–MeOH (2.2/0.8 (v/v)) are 0.33–0.34 that are nearly unchanged compared to that of PF8T2 homo-aggregate (ΦPL = 0.30), but decrease half of PF8T2 in homogeneous CHCl3 solution (Fig. S7, ESI).

PF8T2 homo-aggregates endowed with sacrificial PSi-S and -R

Regarding the bisignate CPL-/CD-active PF8T2 aggregates, our recent works11 prompted us to further investigate whether PSi-S/-R acts as a photoscissable helix scaffolding that allows the emergence of helical PF8T2 homo-aggregates. The photoscissoring reaction at 313 nm for 10 s facilitates maintaining the original bisignate CD amplitudes of PF8T2 associated with bisignate CPL spectra with a slight decrease in gem values, as proven by the complete disappearance of bisignate Siσ–Siσ* bands at ≈320 nm (Fig. S8, ESI). PSi-S/-R begins decompose within 5 s and decomposes completely in 10 s (Fig. S9, ESI). The CPL-signs at 490 nm and 520–600 nm are identical to the CD signs at 320–420 nm and 450–600 nm, respectively. PSi-selective photoscissoring is, thus, a feasible strategy to produce bisignate CPL-/CD-PF8T2 homo-aggregates.

Furthermore, the obvious decrease in the aggregate size after photoscissoring indicates that PSi-R/-S decomposes after UV irradiation, as indicated by dynamic light scattering (DLS) data (Fig. S10, ESI).11 This phenomenon is explainable by the backward reaction of Ostwald ripening.11b The hydrodynamic diameters of PF8T2 hetero-aggregates (≈807 nm with PSi-S and ≈480 nm with PSi-R) decrease to ≈461 nm with PSi-S and ≈346 nm with PSi-R after photoscissoring at 313 nm for 600 s. Fluorescence optical microscopic (FOM) images (excited at 365 nm) of PF8T2 aggregates with PSi-R indicate the initial green-colour emitters survive after 600 s (Fig. S11, ESI). Atomic force microscopy (AFM) using Al-coated cantilever cannot provide clear helical mages of PF8T2-PSi-R aggregates on HOPG surface (Fig. S12, ESI). However, sub-μm-size doughnut-like and dot-like images (1.3 nm in height) are observed.

Jablonski diagram and Kasha's rule of PF8T2 aggregates

In Fig. 2, 3 and Fig. S4 (ESI), S5 (ESI), two or three distinct π–π* transitions (green zones) attributable to several π–π* transitions of PF8T2 aggregates can be characterized in terms of the couplet oscillator concept between chirally assorted chromophores.3,5,6 The bisignate CD signals do not originate from a mixture of several conformational isomers of PF8T2,3a,c in analogy of bisignate CD spectral profiles of aggregates made of rigid rod-like helical polysilanes.12c Clearly, the second Cotton CD band exists at approximately 400 nm, although a weaker broadened tail band is masked among multiple π–π* transitions (three green zones) in the range of 350–600 nm, regardless of the presence or absence of PSi-S/-R.

Substantially boosted CD, CPL, CPLE and PLE spectroscopic data in the optically tuned fluidic medium inspired the proposal of a modified Jablonski diagram. Bisignate CD, bisignate CPL and bisignate CPLE spectral characteristics should not obey Kasha's rule.1 Exciton coupling could be responsible for the bisignate signal of chiral polymers emitting from high-lying excited states (>S1).5,6

A revisited Jablonski diagram of PF8T2 aggregates that do not obey Kasha's rule is proposed. As illustrated in Fig. 4 and Fig. S13 in the ESI, bisignate CD profiles arise from two S1 states (expressed as α- and β-states) according to the coupled oscillator theory of chirally oriented chromophores.3–7,12c If Kasha's rule is applied to the coupled oscillator systems, the CPL-sign and profile of the α-state (i.e., the lowest S1 state) have the same CD sign, whereas when (+)-sign and (−)-sign CD profiles at different energy levels are coexcited to states above the Sn-states (n = 1, 2…), the photoexcited energy relaxes differently from the S0-states; as a result, the (+)-sign α- and (−)-sign β-states produce (+)-sign and (−)-sign CPL spectra, respectively.


image file: c7py00958e-f4.tif
Fig. 4 A revisited Jablonski diagram explaining the bisignate CD, CPL and CPLE spectra of PF8T2 aggregates.

Molecular camphor in solutions

The CPLE spectra collected here allow us to gain additional insight into the origin of the bisignate CPL but apparent monosignate CD spectra of camphor in dilute solution.8,9 Although D-/L-camphor is used as a chiroptical standard in CD, CPL and vibrational CD spectroscopy, its CPLE spectrum has not been reported. A recent theoretical and experimental study suggested that photoexcited camphor in a double well adopts global (planar form C[double bond, length as m-dash]O, Fig. 1) and local (bent form C[double bond, length as m-dash]O, Fig. 1) minima, although the planar one may be dominant in the ground state.9

Fig. 5(a) and (b) show the CD and CPL spectra of camphor in nonpolar cyclohexane (c-Hex). From Fig. 5(a), the D-form clearly reveals (+)-sign vibronic CD bands at 293 nm, 302 nm and 313 nm (shoulder), whereas the L-form has (−)-sign vibronic CD bands in the same wavelengths: for D-form, gabs = +0.078 at 302 nm and +0.064 at 293 nm, whereas for L-form, gabs = −0.079 at 302 nm and −0.064 at 293 nm.


image file: c7py00958e-f5.tif
Fig. 5 (a) The CD and UV-vis and (b) CPL and PL spectra (excited at 290 nm) of D-/L-camphor in c-Hex (1.0 × 10–2 M). The CPLE and PLE spectra monitored at 370 nm and 480 nm of (c) D-camphor and (d) L-camphor in c-Hex (1.0 × 10−2 M).

From Fig. 5(b), camphor reveals clearly bisignate CPL spectra at ≈370 nm and ≈460 nm: for the D-form, gem = +0.0015 at 370 nm and gem = −0.0023 at ≈460 nm, whereas for the L-form, gem = −0.0013 at ≈370 nm and gem = +0.0021 at ≈480 nm.

Fig. 5(c) and (d) show the first CPLE and PLE spectra of camphor collected at 370 nm and 480 nm. The D-form, (+)-sign CPL monitored at 370 nm clearly produces (+)-sign a CPLE spectrum at 240–320 nm, whereas the (−)-sign CPL monitored at 480 nm gives a (−)-sign CPLE spectrum at 240–320 nm. L-/D-Camphor exhibits mirror-image relationships between the CPLE, CPL and CD spectra. Notably, although the PLE peak wavelength observed at 480 nm was 303 nm, whilst the PLE peak wavelength monitored at 370 nm was 313 nm, regardless of whether the L- and D-form was studied, the corresponding CPLE spectral shapes monitored at 370 nm and 480 nm are nearly mirror images. However, the corresponding PLE spectra at ≈300 nm are slightly different, possibly, due to different conformers, as discussed later.

The CD, UV-vis, CPL, PL, CPLE and CPL spectral characteristics of camphor in polar EtOH show similar but structureless and broader CD and UV spectra due to hydrogen bonding with EtOH (Fig. S14(a) and S17(a), ESI); additionally, the bisignate broad CPL profile (Fig. S14(b), ESI) is in agreement with the characteristics reported previously.8,9 The CPLE spectra in EtOH are similar to those in c-Hex (Fig. S14(c) and (d), ESI).

The recent theoretical results9 and present CPLE, CPL and CD spectra led us to suggest that D-camphor in the ground state exists as a mixture of the most-stable/most-polar planar and less-stable/less-polar bent geometries. The planar form, which exhibits (+)-sign CD at 313 nm, emits (+)-sign CPL at 370 nm with a Stokes’ shift of 4950 cm−1, whilst the bent forms that show (−)-sign CD at approximately 303 nm emit (−)-sign CPL at 480 nm with a large Stokes’ shift of ≈11[thin space (1/6-em)]120 cm−1 due to their substantially reorganized frameworks. The C[double bond, length as m-dash]O moiety in the ground or photoexcited states is rather more bent9 than the rigid one expected. To prove this hypothesis, we calculated the potential energy surface, dipole moment and CD/UV spectra of D-camphor by varying the C(7)–C(4)[double bond, length as m-dash]O(16) angle (Fig. 6(a), (b) and Fig. S18, S19 in ESI).


image file: c7py00958e-f6.tif
Fig. 6 (a) Potential energy curve and dipole moment and (b) UV/CD wavelengths and gCD values of D-camphor as a function of the C(7)–C(4)[double bond, length as m-dash]O angle (density functional theory [DFT] and time-dependent DFT [TD-DFT] with B3LYP and the aug-ccpvDZ basis set). Red and blue zones indicate (+)- and (−)-sign CD bands, respectively, at 300–350 nm.

The most stable geometry consists of a planar form of C(1)–(C3)–C(4)[double bond, length as m-dash]O(16) with an angle of 153° and 3.27 Debye (Fig. 6a), resulting in a (+)-gCD value at 300 nm (Fig. 6b). When the angles vary between 140° and 160° (red zones in Fig. 5a and b), D-camphor has a gCD ≈+0.15 in the range of 300–310 nm. However, when the angles are strongly distorted (i.e., between 168° and 195° [blue zones in Fig. 6a and b]), D-camphor reveals a weaker (−)-sign CD on the order of gCD ≈ 0.003–0.050 at approximately 310–352 nm. A local minimum is observed at a bent angle of 190°. Indeed, alternative high-sensitivity PL spectroscopy using a JASCO FP6500 with a high-gain photomultiplier tube (PMT) allowed the detection of three well-resolved vibronic PLE bands in the range of 355–395 nm with spacing of 1300–1400 cm−1 in c-Hex, which exist in addition to the Raman anti-Stokes lines of c-Hex (Fig. S15(a) and (b), ESI), regardless of the detection wavelengths (480 and 500 nm), as seen in Fig. S15(c) and (d) (ESI). However, CPLE bands are not obvious when monitoring is performed at 500 nm using the CPL-200 spectrometer with narrow bandwidths of 10 nm (excitation and monitoring) (Fig. S16(b) and (d), ESI). These three vibronic PLE bands at 355–395 nm could be evidence of distorted camphor with the substantially bent C[double bond, length as m-dash]O geometries.

Experimental section

Detailed experimental and theoretical procedures, and data sets are elaborated in the ESI. The data sets include UV-CD-PL-PLE-CPL-CPLE spectra of PF8T2 aggregates, photoimages of PF8T2 aggregates before and after UV scissoring reactions, quantum yields of PF8T2 in CHCl3, AFM image of PF8T2 aggregate with PSi-R and UV-CD-CPL-CPLE spectra of D-/L-camphor and simulated UV-CD spectra of D-camphor.

Conclusions

In summary, we report the first detection of CPLE spectra and discuss the origin of bisignate CPL and bisignate CD spectra attributable to the π–π* transitions of PF8T2 hetero-aggregates induced by the helicity of PSi-S/-R and, for comparison, D-/L-camphor, which exhibits bisignate CPL and monosignate CD spectra. By choosing appropriate detection wavelengths when collecting bisignate CPL spectra, the bisignate CPLE spectra could be resolved; however, the CPLE sign depends critically on whether the detection wavelength has a (+) or (−) sign. The |gem| values of PF8T2 hetero-aggregates with PSi-S/-R as sacrificial scaffoldings boosted to 0.05–0.08 at 514–578 nm with ΦPL = 0.33, reflecting from huge |gabs| of 0.15–0.25 at 504–510 nm. Moreover, the bisignate CPL- and CD-active PF8T2 homo-aggregate was produced by PSi-S/-R-selective photoscissoring reaction.

Acknowledgements

We are grateful for the financial support from JSPS KAKEN-HI (16H04155). STD wishes to thank Prof. Hiroko Yamada, Prof. Hiroharu Ajiro, Dr Nor Azura Abdul Rahim, Shosei Yoshimoto and Jalil Binti Abd Jallilah for fruitful discussions and anonymous reviewers for valuable comments. We also thank Kazuki Yamazaki (GPC), Kazuhiro Miyake (AFM) and Yoshiko Nishikawa (DLS) at NAIST for technical assistance throughout this work.

Notes and references

  1. (a) M. Kasha, Discuss. Faraday Soc., 1950, 9, 14–19 RSC; (b) J. G. Calvert and J. N. Pitts Jr., Photochemistry, John Wiley and Sons, New York, NY, US, 1966 Search PubMed; (c) N. J. Turro, Modern Molecular Photochemistry, University Science Books, Sausalito, CA, US, 1991 Search PubMed.
  2. (a) M. A. Loi, J. Gao, F. Cordella, P. Blondeau, E. Menna, B. Bártová, C. Hébert, S. Lazar, G. A. Botton, M. Milko and C. Ambrosch-Draxl, Adv. Mater., 2010, 22, 1635–1639 CrossRef CAS PubMed; (b) C. L. Choi, H. Li, A. C. K. Olson, P. K. Jain, S. Sivasankar and A. P. Alivisatos, Nano Lett., 2011, 11, 2358–2362 CrossRef CAS PubMed; (c) S. Scuppa, L. Orian, A. Donoli, S. Santi and M. Meneghetti, J. Phys. Chem. A, 2011, 115, 8344–8349 CrossRef CAS PubMed; (d) T. Itoh, Chem. Rev., 2012, 112, 4541–4568 CrossRef CAS PubMed; (e) H.-W. Tseng, J.-Y. Shen, T.-Y. Kuo, T.-S. Tu, Y.-A. Chen, A. P. Demchenko and P.-T. Chou, Chem. Sci., 2016, 7, 655–665 RSC; (f) J. Roose, B. Z. Tang and K. S. Wong, Small, 2016, 12, 6495–6512 CrossRef CAS PubMed; (g) H. Qian, M. E. Cousins, E. H. Horak, A. Wakefiled, M. D. Liptak and I. Aprahamian, Nat. Chem., 2017, 9, 83–87 CAS; (h) A. Cembran, F. Bernardi, M. Garavelli, L. Gagliardi and G. Orlandi, J. Am. Chem. Soc., 2004, 126, 3234–3243 CrossRef CAS PubMed; (i) R. Kanazawa, T. Nakashima and T. Kawai, J. Phys. Chem. A, 2017, 121, 1638–1646 CrossRef CAS PubMed.
  3. (a) Y. Kawagoe, M. Fujiki and Y. Nakano, New J. Chem., 2010, 34, 637–647 RSC; (b) K. Watanabe, I. Osaka and K. Akagi, Adv. Mater., 2012, 24, 6451–6456 CrossRef CAS PubMed; (c) M. Fujiki, Y. Donguri, Y. Zhao, A. Nakao, N. Suzuki, K. Yoshida and W. Zhang, Polym. Chem., 2015, 6, 1627–1638 RSC.
  4. (a) K. Watanabe, H. Iida, S. Yorozuya and K. Akagi, Chem. Mater., 2012, 24, 1011–1024 CrossRef CAS; (b) B. A. San Jose, S. Matsushita and K. Akagi, J. Am. Chem. Soc., 2012, 134, 19795–19807 CrossRef CAS PubMed; (c) M. Gon, R. Sawada, Y. Morisaki and Y. Chujo, Macromolecules, 2017, 50, 1790–1802 CrossRef CAS.
  5. W. Kuhn, Trans. Faraday Soc., 1930, 26, 293–308 RSC.
  6. (a) S. F. Mason and G. W. Vane, J. Chem. Soc. B, 1966, 370–374 RSC; (b) S. F. Mason, R. H. Seal and R. H. Roberts, Tetrahedron, 1974, 30, 1671–1682 CrossRef CAS.
  7. H. G. Brittain and F. S. Richardson, J. Am. Chem. Soc., 1976, 98, 5858–5863 CrossRef CAS.
  8. (a) H. P. J. M. Dekkers and L. E. Closs, J. Am. Chem. Soc., 1976, 98, 2210–2219 CrossRef CAS; (b) P. H. Schippers, J. P. M. van der Ploeg and H. P. J. M. Dekkers, J. Am. Chem. Soc., 1983, 105, 84–89 CrossRef CAS.
  9. G. Longhi, E. Castigilioni, S. Abbate, F. Lebon and D. A. Lightner, Chirality, 2013, 25, 589–599 CrossRef CAS PubMed.
  10. (a) T. Kinuta, N. Tajima, M. Fujiki, M. Miyazawa and Y. Imai, Tetrahedron, 2012, 68, 4791–4796 CrossRef CAS; (b) K. Nakabayashi, S. Kitamura, N. Suzuki, S. Guo, M. Fujiki and Y. Imai, Eur. J. Org. Chem., 2016, 64–69 CrossRef CAS.
  11. (a) A. A. N. Rahim and M. Fujiki, Polym. Chem., 2016, 7, 4618–4629 RSC; (b) M. Fujiki and S. Yoshimoto, Mater. Chem. Front., 2017 10.1039/C7QM00096K.
  12. (a) Optofluidics–Fundamentals, Devices, and Applications, ed. Y. Fainman, L. P. Lee, D. Psaltis and C. Yang, McGraw-Hill, 2010 Search PubMed; (b) S. K. Y. Tang, Z. Li, A. R. Abate, J. J. Agresti, D. A. Weitz, D. Psaltis and G. M. Whitesides, Lab Chip, 2009, 9, 2767–2771 RSC; (c) Y. Nakano and M. Fujiki, Macromolecules, 2011, 44, 7511–7519 CrossRef CAS; (d) J. Crassous, A. Amon and J. Crassous, Phys. Rev. A, 2012, 85, 023806 CrossRef , (4 pages).
  13. E. L. Eliel, S. H. Wilen and L. N. Mander, Chiroptical Properties, in Stereochemistry of Organic Compounds, Wiley-Interscience, New York, NY, 1994, ch. 13 Search PubMed.

Footnote

Electronic supplementary information (ESI) available: Experimental and theoretical procedures, GPC charts, UV-CD-PL-PLE-CPL-CPLE spectra and photoimages of PF8T2 aggregates before and after UV scissoring reactions, quantum yields of PF8T2 in CHCl3 and aggregates, fluorescent optical and atomic force microscopic images of PF8T2 aggregates, UV-CD-CPL-CPLE spectra of camphor and simulated UV-CD spectra with potential energy and dipole moments of D-camphor. See DOI: 10.1039/c7py00958e

This journal is © The Royal Society of Chemistry 2017