Optimization of polycyclic electron-donors based on julolidinyl structure in push–pull chromophores for second order NLO effects

Dyes and Pigments 122 (2015) 74e84

Contents lists available at ScienceDirect

Dyes and Pigments journal homepage: www.elsevier.com/locate/dyepig

Review

Optimization of polycyclic electron-donors based on julolidinyl structure in pushepull chromophores for second order NLO effects Jialei Liu a, *, Wu Gao b, I.V. Kityk c, Xinhou Liu a, Zhen Zhen a a

Key Laboratory of Photochemical Conversion and Optoelectronic Materials, Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Beijing 100190, PR China b Xi'An Catalyst Chemical Co., LTD, Northwest Institute for Nonferrous Metal Research, Xi'an 710016, China c Faculty of Elecrical Engineering, Czestochowa University Technology, Armii Krajowej 17, Czestochowa 42201, Poland

a r t i c l e i n f o

a b s t r a c t

Article history: Received 6 March 2015 Received in revised form 3 June 2015 Accepted 8 June 2015 Available online 17 June 2015

Organic electro-optic (EO) materials have been widely explored and used in the fabrication of microwave photonic devices. Their principal advantages are the follows: lower half-wave voltage, wider bandwidth, lower cost and more convenience of integration. Organic second order nonlinear optical (NLO) chromophores, as the core of the organic EO devices, define the key technological parameters: EO efficiency, long term stability, solubility and optical losses. In the present review article, the use of DepeA chromophore molecules with polycyclic electron donors based on julolidinyl structure is reported and discussed. Significant improvement of electron donating ability and appropriate isolated effect are shown. Future perspectives of the application of such kind of NLO chromophores are considered. © 2015 Elsevier Ltd. All rights reserved.

Keywords: Nonlinear optics Electro optics Chromophore DepeA molecule Optical materials Electron donor

1. Introduction Due to low dielectric constants and huge EO (Pockels) coefficients, organic NLO materials have a great potential in wide bandwidth optoelectronic telecommunication systems and sensing such as electronic/photonic integrated circuits, phased array radar, terahertz spectroscopy, etc. [1e3] Recently researchers pay more attention to organic NLO materials possessing large first order hyperpolarizabilities and good optical transparency [4e6]. NLO chromophores can substantially enhance organic NLO parameters with respect to EO activity, solubility, machinability and optical loss [4e7]. In the past decades, a couple of excellent NLO chromophores have been designed and prepared. Some of them possessed large EO coefficients (above 300 pm/V at wavelength of 1310 nm), but their EO activities are far from a theoretical limit. And these materials present a key enabling platform for driving the rapid deployment of high-performance photonic devices [8,9]. Organic EO light modulators are important commercial devices for transmitting and transformation of ultrafast electrical signals into optical signals in fiber-optic telecommunications. They have made a breakthrough in

* Corresponding author. Tel./fax: þ861082543529. E-mail address: [email protected] (J. Liu). http://dx.doi.org/10.1016/j.dyepig.2015.06.007 0143-7208/© 2015 Elsevier Ltd. All rights reserved.

half wave voltage and bandwidth frequency. The bandwidth frequency has reached a magnitude equal to 150 GHz and the half wave voltage has been reduced to below 1 V, which is impossible using traditional inorganic crystals, like lithium niobate [10e13]. However, for practical application in commercial EO devices, it is necessary to achieve more higher first order hyperpolarizability coefficients for the chromophores and to optimize an adverse for strong inter-molecular electrostatic interaction among the polymer-chromophores [14,15], The solution of these problems could greatly improve the performances of organic NLO materials in EO efficiency machinability, long term stability, exploitation time and optical loss. Pushepull chromophores typically consist of a p-conjugated bridge end-capped with strong electron-donating and accepting groups. They form the well known class of compounds possessing second order NLO properties. In order to optimize their molecular hyperpolarizability described by third-order polar tensors, significant efforts have been focused on modifying of the p-conjugated bridge and the molecular structure of electron acceptors. At the same time, the donor units have remained relatively stable, mostly derived from 4-(dialkylamino)phenyl groups due to their relatively strong electron-donating ability and ease of synthesis and functionalization [16e21]. As the potential of electronic acceptors and

J. Liu et al. / Dyes and Pigments 122 (2015) 74e84

bridges reaches a peak, researchers concentrate their efforts to optimization of electron donors again. Recently, polycyclic donors attract an enhanced interest in fabrication of NLO chromophores, which have a great potential in enhancement of the first order hyperpolarizability and reduction of the inter-molecular electrostatic dipoleedipole interaction [22e32]. According to their strong contributions in improving of the EO coefficients and solubility, researchers from University of Washington and Chinese Academy of Sciences have designed and manufactured a couple of excellent NLO chromophores based on tetrahydroquinolinyl or julolidinyl polycyclic donor structures. Following the Table 1, most of the chromophores containing polycyclic donors possess relatively large EO coefficients, first order hyperpolarizability and good thermal stability. In this review, the principles and application of NLO effects in the titled organic materials will be considered and novel NLO chromophores with multi-cyclic donors will be presented based on their structures and modifying groups. 2. Principles and application of NLO effects in organic materials 2.1. Principles of the NLO effects The high frequency oscillating electric field of a light causes a space redistribution of the weakly bound valence electron clouds. Such redistribution involves a polarization process which creates an induced charge within the molecule. This microscopic polarization can be expressed as the following power series:

Pi ¼ aij Ej þ bijk Ej Ek þ gijk Ej Ek Ei þ /

(1)

Here a is the linear polarizability, b and g are the first and second molecular hyperpolarizabilities, respectively. The second- and third-order terms (b and g terms) only exist under the influence of intense electric fields such as a laser light. Furthermore, the even order tensor b, responsible for second-order NLO effects, vanishes in a centrosymmetric molecule. It is also well established that the macroscopic polarization of a bulk material under the applied strong electric field can be regarded as an averaged sum of the individual molecular polarizations:

75

P ¼ c1 E þ c2 EE þ c3 EEE þ /

(2)

where c2 and c3 are the second and third order susceptibility tensors, respectively. The macroscopic susceptibilities are related to the corresponding molecular hyperpolarizability terms b and g by local field Lorenz field factor and the density of molecules.


ð2Þ cXXX ¼ NbXXX cos3 q ðconstÞ

(3)

where N ¼ chromophore number density (molecules/cc); b ¼ molecular first hyperpolarizability, which is determined by the structure of the chromophores; ¼ acentric order parameter, which is determined by the intermolecular interaction among the chromophores and the poling condition; the constant depends on the dielectric properties of the material lattice and is defined by ð2Þ local electric field. As shown in quation (3), cXXX should increase linearly by increasing N, b or the order parameter . Based on these susceptibilities, many special nonlinear optical effects were derived, such as: Pockels effect (EO effect) described by the third rank polar tensor:

. r33 ¼ 2c2 n4

(4)

The EO effect is a second order NLO process where an external DC or low AC frequency electric field couples with the optical field to change the refractive index of a material. The change of refractive index can be expressed as: Dn ¼ n3E(0)/2; where n is the refractive index with no external electric field, r is the EO coefficient of the material, and E(0) is the applied DC or low frequency electric field. Thus the resulting phase shift D4 can be expressed as: D4 ¼ n3 rELp=l; where L is the coupling length of the applied electric field with incident light. Correspondingly, a phase change of the incident light can be converted to an intensity change in output light using interference effect. As a result, the net outcome is the modulation of light through an externally applied electric field. A half wave modulation voltage can be obtained as: Vp ¼ ðlh=n3 rLÞ, where h is the distance between the electric field electrodes.

Table 1 Principal parameters of the chromophores considered in this review. Chromophore

r33 (pm/V)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

20 76 218 72 36 53 266 127 62 94 20 35 52 337 72 86 128 98

Td ( C)

202 244 242 247 219 238 276 239 229

251 236 202

Td: Thermal decomposition temperature of the NLO chromophores. Tg: Glass transition temperature of the EO polymers. Tp: poling temperature of the EO films.

Tg ( C)

Tp ( C)

bzzz (1030 esu)

Ref

110, 137 112, 137 106, 129 / 140 140 / /

105 105 135 10 above Tg 10 above Tg 120, 147 122, 147 116, 139 / 145 145 Around Tg Around Tg

937 848 831 870 124 934 916 925 831 800 1900 294 262 254 246 672 686 809

[22] [22] [22,26] [23] [24] [25] [25] [25] [25] [27] [27] [28] [28] [29,30] [29] [31] [31] [31]

Around Tg Around Tg Around Tg

76

J. Liu et al. / Dyes and Pigments 122 (2015) 74e84

2.2. Preparation of EO films Despite the advantages in dielectric constant and EO coefficient, magnitudes organic EO materials also have a technological advantages in the production of EO devices. The organic EO films are prepared by spin coating, which can reduce the cost in the process of devices preparation and confirm the sizes of the EO films as large as we need. In this review, all of the EO films are prepared as guest/ host system. And the thickness of the EO films are around 2 mm. The process of this kind of EO films prepared as follow: 1) the host polymers are dissolved in solvent with suitable boiling point with the concentration around 10 wt%e20 wt%; 2) after the polymers are completely dissolved in the solvents, chromophores are added to the solution with different concentration; 3) after the mixtures are stirred for several hours and all of the chromophores are dissolved, the solutions are filtered through syringe filters; 4) the EO films are spin-coated onto the substrates at different speeds according to the thickness of the films needed; 5) residual solvents are removed by heating the films in a vacuum oven at appropriate temperature.

2.3. Poling process The most common ways of poling are corona and contact poling. In both cases a dc electric field is applied to the polymer at a temperature where the chromophores' dipoles can be aligned. The largest possible electric field is used for poling to obtain the highest degree of polar order, i.e., the ground state dipole moments of chromophores are almost parallel. The thin polymer film is usually sandwiched between two parallel conducting plates, and the polar axis is perpendicular to the film's plane. In the case of contact poling the electric field is applied through an electrode directly on top of the polymer film. Contact poling usually provides a more aligned ordered sample than does corona poling. Corona poling have two electrodes. One of the electrodes with a shape causes a significantly greater electric field at its surface than that between the electrodes. Typically a sharp needle, wire, or grid is charged with several kV until an electrical discharge ionizes the surrounding atmosphere. Depending on the polarity of the corona needle, either positive or negative ions can be deposited on

the surface of the polymer film. With a corona needle the degree of poling decreases outward in a radial fashion. Efficient poling plays an important role in the resulting EO coefficient's magnitudes. If an electronic breakdown occurs in the material due to excessively high field strength, material impurity or electrode processing pinholes can be formed. The optimized poling method varies from material to material, and several approaches should be explored when characterizing different materials. In both corona and contact poling, the poling temperature, electric field strength and time are critical to the resulting degree of polar order. The time required to orient a sample is system dependent. In general, a guest host system can be poled fairly quickly. Slight poling temperature electric field strength and time changes can be significant. Poling at temperatures above or below the perfect poling temperature can decrease the EO coefficients. 2.4. Measurement technique of EO activity The EO coefficient in a bulk system can be measured by several techniques including attenuated total reflectance spectroscopy (ATR), constant bias, and simple reflection measurement. The method used in most of the references is simple reflection technique that was firstly described by Teng and Man [61]. The simple reflection technique has the advantages of easy sample preparation, relatively short measurement times, no measurement induced sample damage, and straightforward interpretation. The simple reflection technique converts a phase modulation into an intensity modulation. This process forms interference between the TE and TM polarizations. Light polarized at 45 to the plane of incidence is propagated through the ITO electrode and the poled polymer film. The light is subsequently reflected off the gold electrode and continues out of the sample. Then it propagates through a Soleil Babinet compensator, which adds a variable phase shift between the reflected TE and TM waves. Finally, the light travels through an analyzer set at negative 45 from the plane of incidence. 3. NLO chromophores with bicyclic and tricyclic donors As shown in Chart 1, chromophores with tetrahydroquinolinyl, and julolidinyl groups as electron donors have been prepared by

Chart 1. Structure of chromophore 1, 2, 3 and 4.

J. Liu et al. / Dyes and Pigments 122 (2015) 74e84

the research groups of Alex Jen in USA and Zhen Zhen in China [22,23,26]. Compared with the traditional 4-(dialkylamino)phenyl donors or 4-(diaromaticamino)phenyl donors, the ring-fused aminophenyl structures in tetrahydroquinolinyl and julolidinyl donors facilitate the overlap of the p-orbital of the amino atom with the phenyl ring. As a consequence a good mechanism to gradually increase the electron-donating strength from diethylaminophenyl to julolidinyl groups is given [33,34]. Flexible convenient cyclic structures around the rigid NLO chromophores can enhance the distance between the chromophores. The inter-molecular interaction between the chromophores would be effectively reduced and the solubility of the chromophores would be substantially improved. By these principles, tetrahydroquinolinyl, and julolidinyl donors would make the organic NLO materials more effective. According to the results of theoretical quantum chemical simulations and intra-molecular charge-transfer (ICT) absorption band, chromophore 1 should have larger second order NLO activity. However, the r33 magnitude of EO films prepared in polymethyl methacrylate (PMMA) polymer with 10 wt% chromophore loading density is just about 20 pm/V at 1310 nm. And the solubility of chromophore 1 is very poor. These results are far from the researchers' expectation. To explain these phenomenons, the authors have done lots of works including theoretical simulations and spectral measurement. The results have shown that these phenomena are attributed to the extremely strong electron donating ability of double ring-fused amino-phenyl structure of julolidinyl donor. The latter is due to the chromophore with zwitterionic structure. Zwitterionic structure leads to decrease of b values and improvement of the inter-molecular interaction considerably. So how to avoid the chromophore to form zwitterionic structure has become a crucial problem to prepare an NLO chromophore with larger EO activity. Chromophore 2 using tetrahydroquinolinyl group as donor provides r33 value equal to about 76 pm/V at l ¼ 1310 nm with 10 wt% chromophore loading density using PMMA as the host polymer. The structure of chromophore 2 is similar to the traditional CLD chromophores with 4-(dialkylamino)phenyl as donors (TCLD, as shown in Chart 2). The degree of the orbital overlap between the nitrogen atom and benzene is suitable for the mobility of the electronic cloud. So the electron donating ability of tetrahydroquinolinyl donor is a little slighter than julolidinyl group and the NLO chromophores are hard to become zwitterionic structure.

Chart 2. Structure of chromophore FTC and TCLD.

77

Its r33 value at high chromophore loading density is not reported. However, It can be inferred that polymer EO films prepared by chromophore 2 would have higher r33 values than TCLD at high chromophore loading density, due to its solubility and intermolecular interaction. This result is confirmed by the EO efficiency of chromophore 3. Chromophore 2, as a promising alternative to TCLD, represents a new DepeA architecture that could be used for the development of next generation of high-performance EO materials. Chromophore 3 is designed by introducing of an appropriate substituent to provide a suitable shape modification to improve a solubility. Though it had the same conjugated structure and it should have the similar first-order hyperpolarizability with chromophore 2. A huge EO coefficients (up to 218 pm/V at 1310 nm) were achieved for a guestehost EO polymer of chromophore 3 in PMMA host polymer. It is almost 50% higher with respect to the highest values reported so far for the material systems based on TCLD chromophores. Obviously, this giant EO activity is attributed to the improvement of the electron donating ability of tetrahydroquinolinyl donor and the isolation effect of the substituent group. Optimizing the size and location of isolated groups, the EO coefficients for such chromophores would be improved further [35]. Cis,cis-1,7-diethoxy-3-isopropyljulolidine group is designed and synthesized as an electron donor in chromophore 4. The polyene structure of the p-conjugated electron bridge in chromophore 1e3 is replaced by thiophene. Additionally traditional tricyanofuran (TCF) acceptor is served as an electron acceptor. The doped EO films containing chromophore 4 show the largest r33 value equal to about 72 pm/V at 25 wt% chromophore loading density (using amorphous polycarbonate (APC) as the host polymer) at 1310 nm. This value is almost two times higher than the EO magnitude of the traditional (N,N-diethyl) aniline NLO chromophore FTC (as shown in Chart 2) [36]. High r33 values indicates that the julolidine donor can substantially improve the electron-donating ability and reduce intermolecular dipoleedipole electrostatic interactions leading to enhanced macroscopic EO response. Due to the poor electronic acceptor and p-conjugated electronic bridge, the r33 value is not as large as EO films prepared by chromophore 3. Traditional NLO chromophores are easy to aggregate, which is beneficial for self-assembly. But it is unfavourable for NLO applications, because the chromophores easily form anti-parallel aggregation compensating total ground state dipole moment. This phenomenon can be demonstrated by the absorption strength, wavelength and shape of the absorption band in UVeVis spectra.

Fig. 1. UVeVis absorption spectra at different concentrations for chromophore 4 in CHCl3.

78

J. Liu et al. / Dyes and Pigments 122 (2015) 74e84

When there is aggregation phenomenon among the chromophores, the absorption strength and band will be changed; different shoulder peaks will be appeared due to the aggregation types, such as H-aggregate and J-aggregate [37]. The UVeVis absorption of the chromophore 4 measured in CHCl3 solvent at different concentrations (ranging from 0.005 mM to 0.1 mM) is shown in Fig. 1. As the concentration increase, no obvious spectral blue-shift or red-shift is observed. Even at high concentrations, no obvious characteristic absorption band of aggregation is discovered. This result confirms that flexible cycle structure could substantially reduce the anti-parallel aggregation [38]. 4. NLO chromophores with benzyloxy julolidinyl donor 8-Hydroxy-1,1,7,7-tetramethyljulolidine-9-carboxaldehyde has been widely used in medicine. Recently, it is discovered as an excellent electron donor in NLO chromophores, due to its specific structure and reactive sites. Double ring-fused aminophenyl structure favoured stronger electron donating ability with respect to traditional N,N-dialkylaniline donors. The aldehyde group can reduce difficulty to couple the donor with most of the p-conjugated electronic bridge by Wittig reaction. The phenolic hydroxyl is also an important reactive site, which can assist to insert different isolated groups, including benzyl, alkyl halide and so on. As shown in Chart 3, chromophore 5 and 6 are designed and prepared based on 8-hydroxy-1,1,7,7-tetramethyljulolidine-9-carboxaldehyde donor [24,25]. Chromophore 5 is a simple NLO chromophore, which just had a short p conjugated electronic bridge of carbonecarbon double bond and a large isolated group of benzyl. The preparation of chromophore 5 is very easy and the yield is very high. So chromophore 5 is suitable for application during exploring the devices preparing process. EO films prepared by APC doped chromophore 5 display the r33 magnitude equal to about 36 pm/V at the saturated doping concentration of 40 wt%. This EO coefficient is two times higher with respect to its counterpart with 4-(N,N-diethyl)aniline donor (16 pm/V) at the saturated concentration of 25 wt%. High loading density and r33 value indicate that julolidinyl-based donor is helpful to improve the solubility and match ability of the NLO chromophores in polymers. Translating of large microscopic hyperpolarizability into large macroscopic EO response is still a challenge to the NLO research, which significantly depend on the inter-molecular interactions, for instance, pep stacking and charge transport between the chromophores. Moreover, weak non-covalent bond interactions such as CH … X (XLO, N, S, halogen), pep stacking and XH … p (X ¼ O, N, C) play crucial role in chromophore alignment and poling process [39e41]. The crystals of chromophore 5 are arranged in cross stacking and the introduction of the benzyloxy group prevents the non-classical hydrogen bonding interaction of C]CeH/NCeC. So the principal dihedral angle between the two planar parts (donor

Chart 3. Structure of chromophore 5 and 6.

and bridge) is enlarged (33.6 ). Large dihedral angle modifies the electron mobility, which can prevent the chromophores forming zwitterionic structure to reduce the macroscopic NLO susceptibility. The donor and isolated groups of chromophore 6 have the same structures with chromophore 5. The p conjugated electron bridge is extended by the addition of thiophene as p-conjugated electronic bridge. Such modification can effectively improve the first-order hyperpolarizability and solubility of the NLO chromophores. In traditional chromophores, the insert of thiophene can improve the EO coefficients of the polymer EO films for several times [42,43]. Organic EO films prepared by APC doped by chromophore 6 have shown the largest EO coefficients magnitudes equal to 53 pm/V with 25 wt% chromophore loading density at 1310 nm. The EO activity is just improved up to 47%, however it is not still sufficiently large [44,45]. The addition of benzyl group was effective for the steric hindrance. But benzyl group is quite close to the conjugated plane, meaning that it is not able to cover enough volume for the site-isolation. Else, such rigid isolated groups also restrain the rotation of the chromophores in the poling process, and reduce the poling efficiency. When the molecular weight of the chromophore is small, this phenomenon is not very obvious. However, with the increase of molecular weight, this phenomenon is more and more obvious. This is also one of the reasons for chromophore 6 with little EO coefficient. The following chromophores with flexible isolated groups showing larger EO coefficients can confirm this.

5. Julolidinyl donor NLO chromophores containing thiophene electronic bridges Dipoleedipole interactions between the chromophores are additional key factors influencing the EO activity of organic EO films. According to London theory for modelling inter-molecular interactions between spherical shaped chromophores, the poling induced EO coefficient can be calculated in good approximation and it is proportional to:

    mb W 1  L2 Ep r33 f MW kT

(5)

where Ep is the poling field, k is the Boltzmann constant, T is the poling temperature, L is the first order Langevin function (L(x) ¼ coth(x)  1/x), MW is the molecular weight of chromophore and W is the potential energy of the chromophore dipolar moment that can be presented as the sum of three terms, namely, the orientation force, induction force and the dispersion force:

W¼

  1 2m4 3Ia2 2 þ 2m a þ 4 R6 3kT

(6)

Here R is the average distance between chromophores, a is the linear polarizability, and I is the ionization potential [62]. It is obviously that the varying distance between chromophore is the main way to improve the EO activities of the organic materials. In order to improve the average distance between the chromophores, large flexible modified groups are introduced into chromophore 7, 8 and 9, whose structures are presented in Chart 4. Except improving the average distance between NLO chromophores, large flexible modified groups also can modify the steric hindrance and improve the free mobility of NLO chromophores, which are positive for achieving large macroscopic EO coefficients [25].

J. Liu et al. / Dyes and Pigments 122 (2015) 74e84

79

Chart 4. Structure of chrmophore 7, 8, 9.

The crystal information of chromophore 7 is given by the authors to understand the molecular conformation. Chromophore 7 is crystallized in triclinic space group P1. The rigid 12-membered ring in donor bearing multi-methyl group effectively keeps off neighbour aromatic rings to prevent the close packing of chromophores. The huge steric hindrance siloxane, linked by flexible hexyl group, is expanded to the free volume near the conjugated bridge and acceptor, which perform the site-isolation of chromophore. The distances between siloxane and donor are equal to 5.7554(36) Å and between siloxane and acceptor are 15.4837(25) Å. So they are coincidently close to the length of conjugated backbone (16.4578(33) Å, from donor to acceptor). This coincidence may suggest that it enables the chromophores to rotate within the well fitting free volume formed by siloxane-hexyl group and simultaneously the site isolation is able to cover the conjugated bridge and acceptor. This molecular configuration generates the weak pep interactions with the centroidecentroid bond distance with length more than 5.0 Å between the antiparallel neighbour chromophores. At the same time higher distance of pep interactions for push pull chromophores is equal to about 4.0 Å. According to Eq. (3) the distance of pep interaction confirm that donor modification effectively isolating the chromophores and it is reasonable to improve the EO activity [46]. The crystal information for chromophore 8 is not reported. Obviously, the modification of chromophores 8 has less effective steric hindrance than chromophore 7. Without the siloxanetermination, the flexible hexyl alcohol played less effective role in the site-isolation. Due to the introduction of terminated siloxane, the siloxane-hexyl group may cover the free volume extended to the conjugated bridge. It is clear that chromophore 7 have the most effective steric hindrance for site isolation. For chromophore 6 and chromophore 8, it is assumed that chromophore 6 have the better free mobility with respect to 8, because chromophore 8 has the longer steric hindrance group hexanol restraining the molecular mobility. The EO activities of chromophore 7 and 8 are also studied in guest/host system using different APCs as the host polymers. Firstly, 25 wt% of chromophore 7 and 8 are embedded into APC with low glass transition temperature (APC-1, Tg: 150  C) to verify the influence of donor modification on the EO activities. EO films prepared by chromophore 8 show the highest r33 value (114 pm/V) after poling. But EO films prepared by chromophore 7 just show r33 value equal to about 89 pm/V. Then, the EO coefficients of the EO films by formulating chromophore 7 and 8 into APC possessing

higher glass temperature (190  C) are also studied. The results are completely different with respect to the obtained at lower glass temperature APC, though their r33 values are all improved. EO films prepared by chromophore 7 show highest r33 value of 266 pm/V, but EO films for chromophore 8 just demonstrate the highest r33 value of 127 pm/V. The reason for the observed effects is attributed to the molecular structures of the modified groups. Small modified group make chromophore 8 more mobile at lower temperature, but the short average distances between the chromophores restrains the improvement of the EO coefficients. For chromophore 7, large modified group favours larger average distances between the chromophores. However, there exists a poor mobility at lower poling temperature in APC with low Tg. When the poling temperature is raised at APC with higher Tg, the mobility is improved greatly and the poling efficiency is also improved significantly. It is quite interesting that chromophore 9 have almost the same length flexible chain with respect to chromophore 8. The only difference is the hydroxy group and chlorine atom at the end of the flexible chain. At the same time the r33 value for chromophore 9 is relatively small (just 37 pm/V). The authors attribute such small EO coefficient to less site isolators to reduce the dipoleedipole interaction and to form an antiparallel aggregation. According to the structures of chromophore 7 and 8, we don't expect to explain sufficiently this phenomenon. Maybe, there are many other reasons, such as optimal poling conditions, the interaction between chlorine atom and the conjugated groups.

6. Julolidinyl donor NLO chromophores containing thieno [3,2-b]thiophene bridges As shown in Chart 5, thieno[3,2-b]thiophene chromophore was used as conjugated bridge to cooperate with julolidinyl donor forming novel NLO chromophores. The solvatochromic behaviour of chromophores is strongly influenced by the strength of the donor and acceptor in combination with the p-bridge [27]. The UVeViseNIR absorption spectra of chromophore 10 and 11 were measured in six solvents with different dielectric constants and the corresponding optical absorption data are presented in Table 2. Both of them have shown an intense low-energy ICT absorption bands. With the increase of solvent polarity, both of the chromophores are spectrally red-shifted initially, reaching the maximum

80

J. Liu et al. / Dyes and Pigments 122 (2015) 74e84

Chart 5. Structure of chrmophore 10 and 11.

Table 2 Summary of low energy optical absorption data. Spectral positions of absorption maxima in different solvents lmax (nm) Chromophore

DO

TL

CF

DC

AC

AN

Dlmax

Polarity 10 11

4.8 628 640

2.4 668 666

4.4 700 712

3.4 703 696

5.4 688 644

6.2 693 643

75 72

DO: dixoane; TL: toluene; CF: CHCl3; DC: CH2Cl2; AC: acetone; AN: acetonitrile.

value in chloroform or dichloromethane, and then, chromophores 10, exhibiting the saturation behaviours. The cyanine-like characteristic (the neutral and charge-separated limiting resonance forms contributed equally to the ground state) band shape of them indicates that the two limiting resonance forms have approximately the same contribution to the ground state dipole moments [47]. Chromophores 11 is polarized distinctly beyond the cyanine limit into the zwitterionic regime in acetone and acetonitrile solvents. Furthermore, chromophores 10 showed the relatively small FWHM (full width at half maximum) of the low energy ICT absorption band and the most strong absorbance in high polar solvents due to the relatively short polyene p-bridge and the strong TCF acceptor [48,49]. APC guestehost polymer films with 20 wt% of chromophore loading density are prepared. The r33 values of the EO films pre-

pared by chromophore 10 and 11 are very promising. The r33 magnitudes for EO films prepared by chromophore 10 (94 pm/V at 1.31 mm) is four times higher than the EO films prepared by chromophore 11 as the guest (just 20 pm/V). At the same time the calculated first order hyperpolarizability of chromophore 11 is five times larger with respect to chromophore 10. The authors attribute this phenomenon to the poor solubility and antiparallel chromophore aggregation 10. Obviously, this phenomenon is the same as the results between chromophore 1 and chromophore 2. Super conjugated structures improve the electron mobility of chromophore 1. Due to the super strong electron mobility, zwitterionic structure is formed under the poling dc-electric field. So the firstorder hyperpolarizability and poling efficiency substantially reduce the EO in polymer films. The spectra before and after irradiation with 450 nm light in CHCl3 solutions for chromophore 10 and 11 are shown in Fig. 2. Though the authors just would like to compare the photochemical stability and the spectra for chromophore 11 obviously proved that zwitterionic structure was formed after the light irradiation [50]. This result also provides reliable evidence for the formation of zwitterionic structure in the aligned oriented dc-electric filed. As the rapid development of NLO chromophores, the first order hyperpolarizability has reached the limit of recent poling technology. How to arrange the NLO chromophores with higher first order

Fig. 2. The UVeViseNIR absorption spectra for the solutions irradiated with 450 nm light. The spectra before and after irradiation with 450 nm light of CHCl3 solutions for chromophore 10 (a) and chromophore 11 (b).

J. Liu et al. / Dyes and Pigments 122 (2015) 74e84

hyperpolarizability and prevent them to form zwitterionic structure will be a new topic in a future).

7. NLO chromophores containing tetracyclic donor NLO chromophore containing tetracyclic donor is also reported by M Zhang et al. [28]. This tetracyclic donor is prepared based on julolidinyl structure Furan cycle is introduced between the julolidinyl structure and the p electronic bridge, which is shown in Chart 6 for chromophore 13. The UVeVis spectral absorption for chromophore 13 with its reference to chromophore 12 were recorded in a series solvents with different dielectric constants as is displayed in Table 1. The two chromophores exhibit the similar typical pep* intra-molecular charge-transfer (ICT) band spectrum and the absorption maximum (lmax) is strongly dependent on the solvent polarity. Increasing the solvent polarity from 1,4-dioxane to dichloromethane a clear red spectral shift is observed for the lmax and the absorption bands shape is changed from broad to sharp gradually. Both of these results properly means that chromophores 12 and 13 possess neutral ground state (in 1,4-dioxane and toluene) and then approaching the cyanine limit (in dichloromethane and chloroform) [31]. The later can be compared with chromophore 12 with a lmax of 624 nm in 1,4-dioxane and 636 nm in toluene, the lmax of chromophore 13 with an additional furan ring at the donor end is red-shifted by 26 nme650 nm in 1,4-dioxane and by 34 nme670 nm in toluene, respectively. These facts reflect a lower energy gap of chromophore 13 with respect to chromophore 12. It is well known that low energy bands in UVeVis spectra indicate a correlated enhancement of p-electron delocalization and large first-order hyperpolarizability [51,52]. The lmax for the chromophore 13 and that of the chromophore 12 appear to have the same values at 711 nm in chloroform and 714 nm in dichloromethane, thereby suggesting their comparable ICT abilities in such kind of dielectric environment.

Chart 6. Structure of chrmophore 12 and 13.

81

Their photophysical properties are shown in Table 3. In more polar solvents like acetone, both chromophores reverse to a blue shift of lmax and are accompanied by a change of absorption bands from sharp to broad. A possible explanation for this behaviour was that both chromophores can be polarized beyond into the zwitterionic ground state in more polar solvents. In addition, chromophore 13 is more blue-shifted than chromophore 12 by comparison of their lmax values in acetone. Thus, we can expect that chromophore 13 is more zwitterionic than chromophore 12 [33]. A further increase of solvents polarity causes again the red shift of lmax of 14 nm and 20 nm from acetone to DMF, respectively. These interesting solvatochromism behaviours of chromophores 12 and 13 are ascribed to the use of strong julolidine-based electron donor, as is usually the case for julolidine derivatives [12]. Notably, the inclusion of an additional furan ring at the donor end of chromophore 13 causes an enhanced ground-state polarization. EO films prepared by APC as the host polymer and chromophore 13 as guest show the largest EO coefficients of 52 pm/V with the chromophore loading density of 20 wt% in APC at 1310 nm. It is about 1.5 times higher than the EO films prepared by chromophore 12. This is also an evidence to support the conclusion that polycyclic structure in the donor part can enhance the first-order hyperpolarizability and reduce the dipoleedipole interaction strength between the NLO chromophores. However, it is a surprising that the onset decomposition temperature is higher than 229  C, and is much higher than traditional NLO chromophores containing furan electronic bridges. The combined large EO activities and high thermal stability confirm an important role of the benzo[b]furan moiety in the design of new NLO chromophores in the future.

8. Julolidinyl donor NLO chromophores containing auxiliary donor Auxiliary donors have attracted interest in NLO chromophores due to their strong electron donating ability and significant isolation effect, which can effectively favour large first order hyperpolarizability of the chromophores and the adverse strong intermolecular electrostatic interaction among the chromophores molecules. Due to its aromatic structure and electron mobility, thiophene is widely used in organic semiconductors [53]. Conjugated structure and sulphur atoms favour an occurrence of donating ability. NLO chromophores 14 and 15 are synthesized by a facile route: their structures were given in Chart 7 [29,30]. In chromophore 14, the thiophene ring is designed as a dual-function structure playing the role of electron donor and providing steric hindrance. Chromophore 15 was designed to confirm the electron donor ability of the thiophene.

Table 3 Photophysical parameters of chromophores 12 and 13. (104M1cm1)

Chromophore

Solvent

lmax (nm)

3

12

Dioxane Toluene Chloroform Dichloromethane Acetone DMF

624 636 711 714 699 719

3.51 4.03 7.09 6.92 5.54 5.58

145 141 114 121 154 143

13

Dioxane Toluene Chloroform Dichloromethane Acetone DMF

650 670 711 714 686 700

1.63 1.77 2.59 2.42 1.87 1.98

128 116 107 118 147 147

FWHM (nm)

82

J. Liu et al. / Dyes and Pigments 122 (2015) 74e84 Table 4 Energy level positions (E), ground state dipole moment(m), electronic polarizability (a), hyperpolarizability at l ¼ 1900 nm; (b) and bond-length alternation (BLA) for the chromophores 9, 14 and 15.

Chart 7. Structure of chrmophore 14 and 15.

The chlorohexyloxy group and thiophene ring are situated on both sides of the conjugated plane, which effectively protected the p-conjugated bridge from an inter-molecular DielseAlder (DA) cycloaddition reaction [54]. The length of the chlorohexyloxy group is close to the distance from electron to acceptor. This arrangement acts to site-isolate the chromophore within the internal free volume created by the chlorohexyloxy group, which is favourable for dipole orientation during the poling process. This molecular conformation effectively isolates the chromophores and suppresses the dipoleedipole interactions. The crystal packing data showed ring centroid to ring-centroid separations equal to 4.9 Å to 5.7 Å, This fact which confirms better site isolation than that of the 4.2 Å reported previously [55,56]. The crystal analysis shows that chromophore 14 displays excellent site-isolation and effectively reduces the inter-molecular electrostatic interactions in the EO response. Chromophore 14 is doped into APC with the loading density of 40 wt% for the preparation of EO films. The poled films show an exceptionally large EO coefficient (r33 ¼ 337 pm/V at 1310 nm), which confirms that the microscopic molecular nonlinearity of chromophores can be effectively translated into macroscopic electro-optic properties. This value is exceptionally large with respect to those reported (TCLD: 70e110 pm/V at 1310 nm) for guest-host EO polymers. Such a result confirms the importance of dual-function structure auxiliary donor. Thiophene-2-carbaldehyde structure was introduced to chromophore 15, instead of the thiophene group in chromophore 14. The addition of aldehyde group in thiophene could restrict the electron mobility on thiophene and suppresses the donor ability. To obtain further understanding of the p-conjugated electron-bridge dependent microscopic properties of these chromophores, Additional density functional theory (DFT) calculations were carried out and chromophores were rotated into frame such that the x axis was aligned along the dipole axis. The relevant theoretical parameters, including HOMO and LUMO energy position of the levels, ground state dipole moment, polarizability, zero-frequency molecular, first order hyperpolarizability and bond length alternation (BLA), are displayed in Table 4. Because of different conjugated electron-bridge structure, chromophore 9, 14 and 15 demonstrated the distinct HOMO and LUMO energy position of the levels. Hence, thiophene and formylthiophene can be defined as the electron-isolator because of which chromophores are distracted to form antiparallel dimer. In addition, the energy gaps (Eg) of HOMO and LUMO are estimated to be 2.02, 2.06 and 1.34 eV, respectively, which is in accordance with the lmax of intra-molecular charge-transfer absorption.

EHOMO(eV) ELUMO(eV) DE(eV) mx (Debye) mtotal (Debye) atotal(esu) bx (1030 esu) btotal(1030 esu) Mb (1030 esu  D) BLA (Å)

9

14

15

0.402 0.938 1.341 18.54 22.05 250.35 825.64 831.2 18,328 0.03761

1.967 0.0558 2.0228 21.1 22.63 192.77 252.98 254.66 5763 0.04812

1.697 0.358 2.055 18.8 19.03 196.45 241.81 246.87 4698 0.04876

The ground state dipole moments for all the three chromophores are estimated to be 22.63, 19.03 and 22.05 D, respectively, showing a noticeable change by replacing different conjugated electron bridges. Due to the high electron density on the conjugated plane, chromophore 14 showed the largest dipole moment (21.10 D) on dipolar x axis. For the 15, the electron-withdrawing of formyl group diminished the electron density on x axis and the ground state dipole moments decreased to 18.80 D, close to 18.54 D for 9. In terms of the polarizability, chromophores with larger dipole values are more sensitive to the higher voltage electrical field. This property may contribute to the effective acentric ordering of chromophores in the applied poling dc electrical field, but it is also likely to induce the dipoleedipole interactions to form the antiparallel dimers. It has been also demonstrated that chromophore 9 with the strongest polarizability is more sensitive to the environment in strong polar solvent and in solid state, which cause the distinct change of lmax in solution and in EO film. All the three chromophores show a decaying trend with respect to molecular hyperpolarizability as the increasing trend of BLA values from 9 to 14 and 15. In this regard, we may conclude that divinylthiophene-conjugated chromophore 9 possesses considerably better microscopic nonlinearity than 14 and 15. However, in the case of site-isolation and electron isolation, 14 and 15 are assumed to have the better performance in the attenuation of dipoleedipole interactions than 9. 9. Julolidinyl donor NLO chromophores containing auxiliary acceptor It is well known that due to the stability of pyrrol and furan, they can't be used in production of NLO chromophores [57]. The introduction of electron withdrawing group can improve their stability, but the electron mobility would be simultaneously reduced and the b value also would reduced. Strong electron donating ability of julolidinyl donors can offset the decrease of electron mobility in electronic bridge. As shown in Chart 8, pyrrole p-conjugation electronic bridge was introduced to cooperate with Julolidinyl donor to form NLO chromophores. The pyrrole moiety bridge has been modified with the electron withdrawing group (eBr, eNO2) substituting benzene ring. The addition of side phenyl groups to chromophores 16 and 17 can enhance the thermal and chemical stability and expand the average distance between the chromophores. So their first order hyperpolarizability contribution to bulk EO performance will be more effective than chromophore 18 [31].

J. Liu et al. / Dyes and Pigments 122 (2015) 74e84

83

conjugated p-system can make inter-chromophore electrostatic interactions less favourable. Chromophores 16 and 17 containing a modified pyrrole bridge can substantially contribute with their large microscopic b values into macroscopic bulk EO activities and more effectively than chromophore 18. This is due to the localization of modified benzene ring groups. However, the different r33 values for film-16/APC and film 17/APC prove that the different auxiliary acceptor groups (eBr, eNO2) show different influences on molecular properties. Moreover, the large r33 values for the chromophores show that the modified pyrrole bridges are stable enough to withstand high temperatures encountered in electric field poling and subsequent processing of chromophore/polymer materials. 10. Conclusions Chart 8. Structure of chrmophore 16, 17 and 18.

The EO efficiency of chromophore 16, 17 and 18 doped in APC is shown in Fig. 3. 10 wt% and may be considered like a typical guestehost EO polymers. The poled films of 16/APC, 17/APC and 18/ APC afford r33 values of 40, 53 and 60 pm/V, respectively. The r33 values of films 16/APC, 17/APC, and 18/APC are measured at different loading densities. The r33 values of film-18/APC are gradually improved from 60 pm/V (10 wt%) to 98 pm/V (25 wt%), while the r33 values drop to 90 pm/V as the loading density increased from 25 wt% to 30 wt%. The r33 values of film-17/APC are enhanced from 53 pm/V (10 wt%) to 128 pm/V (30 wt%). A similar trend is also observed for film-16/APC whose r33 values are enhanced from 40 pm/V (10 wt%) to 86 pm/V (30 wt%). When the concentration of the chromophore in APC is low, film 18/APC displays a larger r33 value than that of the film 17/APC. As the chromophore loading increase, this tendency is opposite. This can be explained by a fact that in a low-density range, the inter-molecular dipolar interactions are relatively weak. The inter-molecular dipoleedipole interactions would become stronger and stronger, what is accompanied by the increasing concentration of NLO chromophore moieties in the polymer. This fact would finally lead to molecular aggregation and to a decrease of the NLO coefficient [58e60]. The introduction of side-groups attached to the

Fig. 3. Behaviour of r33 magnitudes thin films as a function of chromophore loading densities.

Review of NLO chromophores with polycyclic electron donors which are widely designed and prepared in the past years is presented. Most of them have shown strong electron donating ability and large isolated effect, which could substantially improve the EO coefficients for the organic NLO materials. The EO coefficients for some NLO chromophores using polycyclic donors have been improved up to 50 percent. They demonstrate the highest values reported for the traditional dialkyl aniline donors. But the electron donating ability of some polycyclic electron donors are too strong to improve the EO coefficients. Their EO coefficients just reach only 20 pm/V, which is much lower than the traditional chromophores. Else, the result that introduction of polycyclic structures could also improve the stability of the NLO chromophores has been confirmed. Julolidinyl donor cooperated with thiophene bridge offered a novel serial of NLO chromophores having the similar EO coefficients as traditional CLD chromophores. However, their stabilities were greatly improved. Most of the chromophores prepared by polycyclic donors have shown the Tg equal to above 220  C, which is enough for the process of electric poling and waveguide preparation. Due to the strong electron donating ability, substantially more aromatic structures could be used as electronic bridge. Some of polycyclic donor chromophores cooperated with pyrrole and furan p electronic bridges have shown higher EO coefficients and thermal stability with respect to the well known chromophores possessing thiophene electronic bridges. Large electron density of polycyclic donors allows much easier to insert auxiliary donors, which could improve the electron donating ability and isolated effect. After the addition of thiophene auxiliary donor, the EO coefficients have been improved up to 5 times. Auxiliary donors could also keep the large EO coefficients, while UVeVIS absorption was shifted to blue. This could reduce the absorption losses of the EO materials in spectral range including communication wavelength. To solve the problems hindering their application in commercial application, such as large EO coefficients, good solubility, low absorption loss and so on, polycyclic structure will be an important development direction for organic NLO chromophores. Besides the polycyclic donors, polycyclic structures in electronic bridges, acceptors, even the isolated groups might also improve the comprehensive performance of the NLO chromophores. Due to the improvement in first order hyperpolarizability and stability, the design and fabrication of novel p electronic bridges and acceptors which have good match ability with the polycyclic donors would attract the larger researchers' attentions. Auxiliary donors which could adjust the electron donating ability and reduce the intramolecular interaction would be a crucial to get super large EO coefficients. Else, temporal stability of their EO polymers is also an important character for the application of organic EO materials in devices;

84

J. Liu et al. / Dyes and Pigments 122 (2015) 74e84

which could decide the stability and life of the EO devices. To improve the temporal stability, crosslinkable groups will be introduced to these novel chromophores for their attachment or crosslinkage in EO polymers. Acknowledgements We are grateful to the Directional Program of the Chinese Academy of Sciences (KJCX2.YW.H02) and Innovation Fund of the Chinese Academy of Sciences (CXJJ-11-M035) for financial support. References [1] Chang C, Chen C, Chou C, Kuo W, Jeng R. J Macromol Sci Polym Rev 2005;45: 125. [2] Gheorma I, Gopalakrishnan G. IEEE Photon Technol Lett 2007;19:1011e3. [3] Korotky S, Alfernessi R. In: Hutcheson LD, editor. Integrated optical circuits and components. New York: Marcel Dekker; 1987. p. 203. [4] Sullivan P, Dalton L. Accounts Chem Res 2010;43:10e8. [5] Cho M, Choi D, Sullivan P, Akelaitis A, Dalton L. Prog Polym Sci 2008;33: 1013e58. [6] Duncan T, Song K, Hung S, Miloradovic I, Nayak A, Persoons A, et al. Angew Chem Int Ed 2008;47:2978e81. [7] Huang S, Luo J, Yip H, Ayazi A, Zhou X, Gould M, et al. Adv Mater 2012;24: OP42e7. [8] Dalton L, Benight S, Johnson L, Knorr D, Kosilkin I, Eichinger B, et al. Chem Mat 2011;23:430e45. [9] Kim T, Kang J, Luo J, Jang S, Ka J, Tucker N, et al. J Am Chem Soc 2007;129: 488e9. [10] Kityk I, Makowska-Janusik M, Gondek E, Krzeminska L, Danel A, Plucinski K, et al. J Phys Conden Matter 2004;16:231. [11] Jungerman R, Johnsen C, McQuate D, Salomaa K, Zurakowski M, Bray R, et al. J Lightwave Technol 1990;8:1363. [12] Delwin L, Stephanie J, Song J, Robinson B, Dalton L. Chem Mat 2014;26:872. [13] Rao S, Coppola G, Gioffre M, Della Corte F. Opt Express 2012;20:9351. [14] Huang H, Liu J, Zhen Z, Qiu L, Liu X, Lakshminarayana G. Mater Lett 2012;75: 233e5. [15] Liu J, Wang L, Zhen Z, Liu X. Colloid Polym Sci 2012;290:1215e20. [16] Rainundo J, Blanchard P, Gallego-Planas N, Mercier N, Ledoux-Rak I, Hierle R, et al. J Org Chem 2002;67:205. [17] He M, Leslie T, Garner S, Derosa M, Cites J. J Phys Chem B 2004;108:8731. [18] Andreu R, Blesa M, Carrasquer L, Varın J, Orduna J, Villacampa B, et al. J Am Chem Soc 2005;127:8735. [19] Beaudin A, Song N, Bai Y, Men L, Gao JP, Wang Z, et al. Chem Mat 2006;18: 1079. [20] Hoang M, Kim M, Cho M, Kim K, Kim K, Jin J, et al. J Polym Sci Pol Chem 2008;46:5064. [21] Bures F. RSC Adv 2014;4:58826e51. [22] Zhou X, Luo J, Davies J, Huang S, Jen A. J Mater Chem 2012;22:16390. [23] Yang Y, Liu F, Wang H, Zhang M, Xu H, Bo S, et al. Phys Chem Chem Phys 2014;16:20209. [24] Wu J, Liu J, Zhou T, Bo S, Qiu L, Zhen Z, et al. RSC Adv 2012;2:1416e23. [25] Wu J, Peng C, Xiao H, Bo S, Qiu L, Zhen Z, et al. Dyes Pigments 2014;104: 15e23.

[26] Shi Z, Cui Y, Huang S, Li Z, Luo J, Jen A. ACS Macro Lett 2012;1:793e6. [27] Zhang A, Xiao H, Cong S, Zhang M, Zhang H, Bo S, et al. J Mater Chem C doi:10. 1039/c4tc01896f. [28] Zhang M, Deng G, Zhang A, Xu H, Huang H, Peng C, et al. RSC Adv 2014;4: 33312. [29] Wu J, Xiao H, Qiu L, Zhen Z, Liu X, Bo S. RSC Adv 2014;4:49737. [30] Wu J, Bo S, Liu J, Zhou T, Xiao H, Qiu L, et al. Chem Commun 2012;48:9637e9. [31] Liu F, Wang H, Yang Y, Xu H, Zhang M, Zhang A, et al. J Mater Chem C 2014;2: 7785. [32] Wang H, Liu F, Yang Y, Xu H, Peng C, Bo S, et al. Dyes Pigments 2015;112: 42e9. [33] Hendrickx E, Steenwinckel DV, Persoons A, Samyn C, Beljonne D, Bredas J. J Chem Phys 2000;113:5439. [34] Wang H, Lu Z, Lord SJ, Willets KA, Bertke J, Bunge S, et al. Tetrahedron 2007;63:103. [35] Cheng Y, Luo J, Huang S, Zhou X, Shi Z, Kim T, et al. Chem Mat 2008;20:5047. [36] Wu W, Qin J, Li Z. Polymer 2013;54:4351e82. [37] Xiong Y, Tang H, Zhang J, Wang Z, Campo J, Wenseleers W, et al. Chem Mat 2008;20:7465e73. [38] Liu J, Bo S, Zhen Z, Liu X. J Incl Phenom Macrocycl Chem 2010;68:253e60. [39] Davies J, Elangovan A, Sullivan P, Olbricht B, Bale D, Ewy T, et al. J Am Chem Soc 2008;130:10565. [40] Li S, Li M, Qin J, Tong M, Chen X, Liu T, et al. CrystEngComm 2009;11:589. [41] Le Fur Y, Masse R, Nicoud J. New J Chem 1998;22:159. [42] Perez-Moreno J, Clays K, Kuzyk M, Zhao Y, Shen Y, Qiu L, et al. In: Proc. of SPIEvol. 6653; 2007. 66530w. [43] Liu J, Hou W, Feng S, Qiu L, Liu X, Zhen Z. J Phys Org Chem 2011;24:439e44. [44] Zhang X, Aoki I, Piao X, Inoue S, Tazawa H, Yokoyama S, et al. Tetrahedron Lett 2010;51:5873. [45] Pereverzev Y, Gunnerson K, Prezhdo O, Sullivan P, Liao Y, Olbricht B, et al. J Phys Chem C 2008;112:4355. [46] Wurthner G, Schmidt J, Stolte M, Wortmann R. Angew Chem Int Ed 2006;45: 3842. [47] Blanchard-Desce M, Alain V, Bedworth P, Marder S, Fort A, Runser C, et al. Chem Eur J 1997;3:1091e104. [48] Ma X, Ma F, Zhao Z, Song N, Zhang J. J Mater Chem 2010;20:2369e80. [49] Zhou X-H, Davies J, Huang S, Luo J, Shi Z, Polishak B, et al. J Mater Chem 2011;21:4437e44. [50] Liptay W. Angew Chem Int Ed 1969;8:177. [51] Luo J, Cheng Y, Kim T, Hau S, Jang S, Shi Z, et al. Org Lett 2006;8:1387. [52] Bouit P, Aronica C, Toupet L, Le Guennic B, Andraud C, Maury O. J Am Chem Soc 2010;132:4328. [53] Roncali J. Macromol Rapid Commun 2007;28:1761e75. [54] Shi Z, Luo J, Huang S, Zhou X, Kim T, Cheng Y, et al. Chem Mat 2008;20: 6372e7. [55] Kwon O, Kwon S, Jazbinsek M, Choubey A, Gramlich V, Gunter P. Adv Funct Mater 2007;17:1750e6. [56] Liao Y, Eichinger B, Firestone K, Haller M, Luo J, Kaminsky W, et al. J Am Chem Soc 2005;127:2758e66. [57] Li Q, Lu C, Zhu J, Fu E, Zhong C, Li S, et al. J Phys Chem B 2008;112:4545e51. [58] Briers D, De Cremer L, Koeckelberghs G, Foerier S, Verbiest T, Samyn C. Macromol Rapid Commun 2007;28:942e7. [59] Huang H, Deng G, Liu J, Wu J, Si P, Xu H, et al. Dyes Pigments 2013;99:753e8. [60] Lee J, Lee J. Dyes Pigments 2015;115:197e203. [61] Teng C, Man H. Appl Phys Lett 1990;56:1734. [62] Reyes-Esquedaa J, Darracq B, Garcia-Macedo J, Canva M, Blanchard-Desce M, Chaput F, et al. Opt Commun 2001;198:207e15.