PCCP View Article Online
Published on 10 December 2013. Downloaded by Georgia Institute of Technology on 31/10/2014 09:07:52.
PAPER
Cite this: Phys. Chem. Chem. Phys., 2014, 16, 3108
View Journal | View Issue
Understanding the fast thermal isomerisation of azophenols in glassy and liquid-crystalline polymers Jaume Garcia-Amoro ´ s and Dolores Velasco* The good solubility of azophenols in low molar mass liquid crystals together with the ability of their related polymers to form homogeneous nematic and glassy thin films make such azoderivatives valuable chromophores to get a great variety of photoactivatable systems with fast switching speeds under ambient conditions. In fact, the final applicability of these systems is mainly determined by the thermal cis-to-trans isomerisation rate of the photoactive azophenol used, in other words, by the intimate mechanism the
Received 25th October 2013, Accepted 9th December 2013
reaction goes through. The kinetico-mechanistic study reported herein shows that the rate of the thermal
DOI: 10.1039/c3cp54519a
located, mainly to its capability to establish hydrogen bonding with its surroundings. With a proper
back reaction for azophenols is very sensitive to the local environment where the azo chromophore is design, azophenol-based polymers can exhibit thermal isomerisation rates as fast as those of the
www.rsc.org/pccp
monomers in solution even without the presence of any solvent.
Introduction Materials that sense their surroundings and tune their properties as a response to them are being researched extensively worldwide since the number of their potential applications is growing exponentially. Siloxane-based liquid crystal polymers (LCPs) can be very attractive materials for this purpose since they combine the molecular order characteristic of the liquid-crystalline state with the ability to form homogeneous thin films. The incorporation of suitable organic chromophores not only into LCPs but also into glassy polymers allows tuning many properties of the entire material by optical control, for instance, its transmission intensity, helical twisting power, fluorescent emission, molecular shape, etc.1 Indeed, light is one of the most attractive triggers to modulate the properties of materials since it is a free, renewable and environmentally friendly energy source which, moreover, can be remotely supplied over a very localised area of the probe. Azobenzenes, well-known light-sensitive molecules, have attracted great interest for this aim owing to the reversible photoisomerisation between their two isomers – trans and cis. In addition, the initial trans form can be regenerated thermally in the dark.2 This photochemical signature has been widely exploited for many applications within not only materials science3–10 but also biology and medicine.11–17 Main characteristic parameters and abilities of azobenzenebased photoactive materials are directly related to the rate of `nics, Departament de Quı´mica Orga `nica, Institut de Grup de Materials Orga Nanocie`ncia i Nanotecnologia (IN2UB), Universitat de Barcelona, Martı´ i Franque`s 1, E-08028, Barcelona, Spain. E-mail: [email protected]; Fax: +34 93 339 78 78; Tel: +34 93 403 92 60
3108 | Phys. Chem. Chem. Phys., 2014, 16, 3108--3114
the thermally-activated cis-to-trans isomerisation of the azo dye used which, in turn, is determined mainly by its molecular architecture.18 Azophenols are valuable chromophores to get actuating and switching materials with fast responses since they are endowed with a rapid thermal isomerisation process at room temperature. Nevertheless, the rate at which the thermal back reaction proceeds for such azo dyes depends dramatically on the environment where they are located.19,20 With this in mind, here we report a thorough and comprehensive study about the thermal cis-to-trans isomerisation of azophenols in different environments. First and prior to the introduction of the azo monomers AZO-Ac and AZO-OH (see Scheme 1) into macromolecular systems, their thermal isomerisation kinetics have been studied not only in conventional organic solvents of different polarity but also in macroscopically ordered nematic solutions. The results obtained in both isotropic and nematic media have allowed us to evaluate the effect of the liquid crystal order on the rate of the thermal back reaction as well as on the corresponding energetic parameters. Such magnitudes have been correlated further with the plausible acting mechanism for
Scheme 1 Chemical structure of the photoactive azo monomers AZO-Ac and AZO-OH.
This journal is © the Owner Societies 2014
View Article Online
Published on 10 December 2013. Downloaded by Georgia Institute of Technology on 31/10/2014 09:07:52.
Paper
Scheme 2 Chemical composition of the photoactive copolymers (CPAZO-Ac and CPAZO-OH) and homopolymers (PAZO-Ac and PAZO-OH). The degree of polymerisation, DP, of the main polyhydrogenomethylsiloxane (PHMS) backbone used is DP = 31.
the process. Finally, the kinetic study of the corresponding linear copolymers (CPAZO-Ac and CPAZO-OH, see Scheme 2) and homopolymers (PAZO-Ac and PAZO-OH) has been carried out in order to examine not only the effect of the linkage of the azo monomers on the polymer backbone but also the influence of the intermolecular interactions that can be established between the different monomers on the thermal relaxation of such macromolecular systems.
Experimental Materials and instrumentation All reagents were used as received without further purification. Anhydrous CH2Cl2 was dried over calcium hydride and freshly distilled before use. FT-IR spectra were registered on a Nicolet 6700 FT-IR spectrophotometer from Thermo Scientific. Electronic spectra were recorded on a Varian Cary 500E UV-Vis-NIR spectrophotometer. 1H NMR spectra were collected on a Varian Mercury spectrometer and they were processed further with the MestRec commercially available software. Differential scanning calorimetry (DSC) was performed using a Mettler-Toledo DSC821 apparatus under nitrogen flow at a scan rate of 5 K min1. Polarized optical microscopy (POM) was performed using a Nikon Eclipse polarizing microscope equipped with a Linkam THMS 600 hot stage and a Linkam CI93 programmable temperature controller at a scan rate of 5 K min1. Synthesis of the nematic mesogen and azo monomers The nematic liquid crystal 4-methoxyphenyl-4-(3-butenyloxy)benzoate (M4OMe) and both photoactive azo monomers 4-acetyl4 0 -(5-hexenyloxy)azobenzene (AZO-Ac) and 4-(5-hexenyloxy)-4 0 hydroxyazobenzene (AZO-OH) were prepared as reported elsewhere.21,22 Synthesis of the polymers Both acetylated linear polymers, poly[6-(4-(4-acetylphenyl)diazenylphenyloxy))hexylmethylsiloxane-co-(4-(4-((4-methoxyphenoxycarbonyl)phenyloxy)butylmethylsiloxane] (CPAZO-Ac) and poly[6-(4-(4acetylphenyl)diazenylphenyloxy))hexylmethylsiloxane] (PAZO-Ac), were prepared via the well-known Pt-catalysed hydrosilylation reaction between the different monomers, which contain a
This journal is © the Owner Societies 2014
PCCP
terminal reactive alkene, and polyhydrogenomethylsiloxane (PHMS-31) at ca. 70 1C for 2 days. After this time, the complete disappearance of the Si–H band at 2100 cm1 was detected by IR spectroscopy. The free hydroxyl-containing polymers, poly[6(4-(4-hydroxyphenyl)diazenylphenyloxy))hexylmethylsiloxane-co(4-(4-((4-methoxyphenoxycarbonyl)phenyloxy)butylmethylsiloxane] (CPAZO-OH) and poly[6-(4-(4-hydroxyphenyl)diazenylphenyloxy))hexylmethylsiloxane] (PAZO-OH), were prepared from their acetylated counterparts by dissolving them in a mixture of CH2Cl2 and MeOH with a catalytic amount of acetyl chloride at room temperature for 2 days. This reaction conditions allow a mild and chemoselective cleavage of the ester group of the acetylated side-chain group. The complete disappearance of the CQO stretching signal at ca. 1765 cm1 and the appearance of a band corresponding to the O–H stretching at around 3450 cm1 were observed as expected. No hydrolysis of the other ester group placed in the mesogen M4OMe was observed (CQO stretching at 1728 cm1). Synthesis of PAZO-Ac. AZO-Ac (201.4 mg) and PHMS-31 (36.7 mg) were dissolved in thiophene-free toluene (3 cm3). Next, a solution of Pt(COD)Cl2 in dry dichloromethane (1%, 20 mL) was added. The reaction mixture was magnetically stirred at 70 1C for 48 h under an inert atmosphere. Then, the solvent was removed under reduced pressure and the residue was dissolved in the minimum amount of THF. The resulting solution was dropped over methanol at 10 1C (200 cm3) with vigorous stirring in order to precipitate the polymer, which was isolated by decantation of the mixture. This process was repeated twice until no monomer was detected by TLC. PAZO-Ac was obtained as a yellow solid. Yield: 142 mg (60%). 1H NMR (CDCl3, 400 MHz): d 7.80 (ar H), 7.15 (ar H), 6.86 (ar H), 3.88 (CH3–O), 2.18 (CH2), 1.72 (CH2), 1.39 (CH2), 0.54 (Si–CH2) and 0.10 (Si–CH3) ppm. FT-IR (ATR): n 2924 and 2854 (C–H st), 1761 (CQO st), 1595 (CQC st), 1495 (NQN st) cm1. UV-Vis: lmax (toluene) = 350 nm (e = 20 050 M1 cm1). Synthesis of PAZO-OH. PAZO-Ac (80 mg) was dissolved in a mixture of CH2Cl2 and methanol (4 : 1 v/v, 3 cm3). Next, acetyl chloride (1 drop) was added. The reaction mixture was stirred at room temperature under an inert atmosphere for 48 h. Then, the solvent was removed under reduced pressure. PAZO-OH was obtained as a dark red solid polymer. Yield: 64 mg (90%). 1 H NMR (d6-acetone, 400 MHz): d 7.81 (ar H), 7.01 (ar H), 4.07 (CH2–O), 1.79 (CH2), 1.47 (CH2), 0.64 (Si–CH2) and 0.12 (Si–CH3) ppm. FT-IR (ATR): n 3473 (O–H st), 2922 and 2854 (C–H st), 1596 (CQC st), 1495 (NQN st) cm1. UV-Vis: lmax (acetonitrile) = 357 nm (e = 12 900 M1 cm1). Synthesis of CPAZO-Ac. AZO-Ac (45 mg, 10% mol), M4OMe (360 mg, 90% mol) and PHMS-31 (56 mg) were dissolved in thiophene-free toluene (3 cm3). Next, a solution of Pt(COD)Cl2 in dry dichloromethane (1%, 50 mL) was added. The reaction mixture was magnetically stirred at 75 1C for 48 h under an inert atmosphere. The purification of CPAZO-Ac was carried out identically to the one described above for PAZO-Ac. CPAZO-Ac was obtained as a yellow rubber. Yield: 129 mg (40%). 1H NMR (CDCl3, 400 MHz): d 8.02 (ar H), 7.83 (ar H), 7.02 (ar H), 6.84 (ar H), 3.91 (CH2–O), 3.75 (CH3–O), 1.77 (CH2), 1.58 (CH2), 0.60 (Si–CH2) and 0.12 (Si–CH3) ppm. FT-IR (ATR): n 2933 and 2871 (C–H st), 1765 (CQO st acetyl), 1728 (CQO st LC), 1605 (CQC st), 1466 (NQN st) cm1.
Phys. Chem. Chem. Phys., 2014, 16, 3108--3114 | 3109
View Article Online
PCCP
Paper
Published on 10 December 2013. Downloaded by Georgia Institute of Technology on 31/10/2014 09:07:52.
Table 1 Mesomorphic behaviour of the prepared polymers determined by DSC at a scan rate of 5 K min1
System
Tg (K)
DCP (J g1 K1)
CPAZO-Ac CPAZO-OH PAZO-Ac PAZO-OH
281 294
0.2 0.2
314
0.4
a
a
a
TS–N (K)
DHS–N (J g1)
TN–I (K)
DHN–I (J g1)
— — 371 —
— — 12.6 —
340 351 399 —
1.3 0.3 2.6 —
Not detected by DSC.
Synthesis of CPAZO-OH. CPAZO-Ac (80 mg) was dissolved in a mixture of CH2Cl2 and methanol (4 : 1 v/v, 3 cm3). Next, acetyl chloride (1 drop) was added. The reaction mixture was stirred at room temperature under an inert atmosphere for 48 h. Afterwards, the solvent was removed under reduced pressure. The purification of CPAZO-OH was carried out identically to the one described above for PAZO-Ac. CPAZO-OH was obtained as a yellow rubber. Yield: 67 mg (84%). 1H NMR (CDCl3, 400 MHz): d 8.01 (ar H), 7.77 (ar H), 7.00 (ar H), 6.83 (ar H), 3.89 (CH2–O), 3.74 (CH3–O), 1.78 (CH2), 1.61 (CH2), 0.60 (Si–CH2) and 0.12 (Si–CH3) ppm. FT-IR (ATR): n 3413 (O–H st phenol), 2933 and 2872 (C–H st), 1728 (CQO st LC), 1604 (CQC st), 1465 (NQN st) cm1. Mesomorphism of the polymers The characterization of the mesomorphic behaviour of the different polymers was carried out by means of DSC (Table 1). Both copolymers, CPAZO-Ac and CPAZO-OH, showed a broad nematic mesophase between their glass transition temperature at Tg = 281–294 K and their nematic-to-isotropic phase transition temperature at TN–I = 340–351 K (DHN–I = 1.3–0.3 J g1), respectively. This feature was also supported by POM experiments, where the characteristic Schlieren texture of the nematic mesophase was observed when a sample of both copolymers was analysed under crossed polarisers (see Fig. 1). Homopolymer PAZO-Ac exhibited a smectic phase between its glass transition temperature, which could not be detected by DSC, and its smectic-to-nematic phase
Fig. 1 POM microphotographs under crossed polarisers of the smectic phase of PAZO-Ac at 365 K (a), nematic phase of PAZO-Ac at 395 K (b), nematic phase of CPAZO-Ac at 298 K (c) and nematic phase of CPAZO-OH at 298 K (d).
3110 | Phys. Chem. Chem. Phys., 2014, 16, 3108--3114
transition temperature at TS–N = 371 K (DHS–N = 12.6 J g1). Moreover, this system showed a nematic phase up to TN–I = 399 K (DHN–I = 2.6 J g1). The presence of free-phenol groups in PAZO-OH caused the disappearance of all mesomorphism. In this way, PAZO-OH melted directly from its glassy state to the isotropic liquid phase at Tg = 314 K. Preparation of the samples and order analysis For the kinetic experiments in the liquid-crystalline state, ca. 4 mM solid solutions of the different azocompounds in the corresponding nematic mesogen were used. Samples were prepared by mixing the desired amounts of the mesogen and the corresponding azo dye followed by homogenization by magnetic stirring for 10 minutes in the isotropic state. Monodomain samples were prepared in 10 mm optical path quartz cells, their surface being rubbed with a piece of cloth in a single direction. Homogeneity of the samples was checked by local probe microscopy. POM experiments were run by rotation of the analyzer of the microscope with respect to the rubbing direction. Upon reaching 451, the expected change from darkness to brightness was observed, which together with the absence of any characteristic texture was indicative of a successful macroscopic orientation of the nematic director. Thin films of the polymers CPAZO-OH (nematic) and PAZO-OH (glassy) were prepared on quartz slides by putting a drop of a concentrated solution of the appropriate polymer in acetone (2 mg cm3) and allowing further evaporation of the solvent thereby applying simultaneously a roll-on-roll-like technique. Kinetic experiments For the experiments in isotropic solvents, 20 mM solutions of the azo dye in the corresponding solvent were measured in 1 cm optical path quartz cells. For the solid solutions, the concentration was ca. 4 mM and 10 mm quartz cells were used instead (see above). A population of cis-azobenzenes was generated by UV photolysis and their thermal relaxation was followed by time-resolved UV-Vis spectroscopy. For long-lived cis-azobenzenes, the samples were irradiated with a Philips high-pressure mercury lamp (total nominal power 500 W) for 30 minutes by using an aqueous Co(NO3)2 solution (0.5 M) as an optical filter. Time-resolved absorption spectroscopy was performed on a Varian Cary 500E UV-Vis-NIR spectrometer. For short-lived samples, the photoisomerisation of the chromophores was performed by means of the nanosecond laser flash-photolysis technique. In this case, the cis isomer was generated by applying a Q-switched Nd-YAG laser pulse (lirrad = 355 nm, 5 ns pulse width, 1–10 mJ per pulse) and the time evolution of the sample absorbance was monitored at 901 by a white-light beam produced using a PTI 75 W Xe lamp. The light transmitted by the sample was spectrally resolved using a monochromator and detected using a Hamamatsu R928 photomultiplier, whose output was fed into a digital oscilloscope through a 50 ohm resistor.23 In all instances, the solutions were thermostated in the dark at the desired temperature (0.1 K) and the thermal cis-to-trans isomerisation was monitored by the change in absorption at a determined observation wavelength,
This journal is © the Owner Societies 2014
View Article Online
Published on 10 December 2013. Downloaded by Georgia Institute of Technology on 31/10/2014 09:07:52.
Paper
lobs. Relaxation times (t = 1/k) were independent of the observation wavelength and they were determined with an associated error lower than 10%. The volumes of activation, DVa, were determined by analyzing the pressure-dependence of the thermal isomerisation rate constant, k. For runs at variable pressure, a previously described pressurizing system and a pillbox cell were used, which were connected to a J&M TIDAS spectrophotometer.24
Results and discussion Both trans-AZO-Ac and trans-AZO-OH exhibit a strong symmetryallowed p–p* transition at 355 nm and a weak symmetry-forbidden n–p* transition at ca. 450 nm. Upon UV-illumination, a decrease in the intensity of the 350 nm band accompanied by an increase in the 450 nm signal is observed as a consequence of the trans-to-cis photoisomerisation of the azo dye. When the illumination is ceased, the thermal cis-to-trans back reaction occurs recovering thereby the thermodynamically stable trans isomer (Fig. 2). The thermal cis-to-trans isomerisation kinetics of both AZO-Ac and AZO-OH has been studied in three different isotropic solvents: acetonitrile, ethanol and toluene. In all of them, cis-AZO-Ac exhibited a very slow thermal cis-to-trans isomerisation process, presenting relaxation times of more than a day at room temperature (Table 2). This feature suggests that the transition state of the thermal back reaction has a non-polar character.25,26 The volumes of activation for the thermal isomerisation of AZO-Ac were determined in both toluene and ethanol. In both solvents, the DVa values obtained were close to zero and similar to the ones reported for 3,30 -[1,10-diaza-4,7,13,16-tetraoxa-18-crown-6]biscarbonylazobenzene, an azobenzene-bridged crown-ether, which is normally taken as a defined standard for the inversion mechanism since it cannot thermally-isomerise via the rotational pathway due to structural restrictions.27,28 These results confirmed that the operating mechanism for the thermal isomerisation of cis-AZO-Ac is the inversional one. cis-AZO-OH exhibited relaxation times of 311 ms and 405 ms in ethanol and acetonitrile at 298 K, respectively (Table 2 and Fig. 3a).
Fig. 2 Changes in the electronic spectrum of a thermally-isomerising AZO-Ac toluene solution at 328 K ([AZO-Ac] = 20 mM, lobs = 355 nm).
This journal is © the Owner Societies 2014
PCCP Table 2 Relaxation time, t, for the thermal cis-to-trans isomerisation process of the monomers AZO-Ac and AZO-OH and their related polymers at 298 K
Solvent
t (h)
AZO-Ac
Acetonitrile Ethanol Toluene ZLI-1695
63 49 35 20
AZO-OH
CPAZO-Ac
Toluene
33
CPAZO-OH
PAZO-Ac
Toluene
33
PAZO-OH
a
Solvent
t (ms)
Acetonitrile Ethanol Toluene ZLI-1695 5CB MC5CHX Toluene Nematic Ethanol Acetonitrile Film
405 311 1.4 106 261 287 1370a 254 333 17 283 125
Value extrapolated from Eyring’s plot.
Otherwise, the isomerisation rate was more than 1000-fold slower when cis-AZO-OH was dissolved in toluene at this temperature (t = 23 min, Table 2). Accordingly, the presence of a free phenol group in the azo dye clearly alters the intimate mechanism that the isomerisation process goes through. In both ethanol and acetonitrile, the thermal back reaction proceeds through a solvent-assisted tautomerisation to yield a hydrazone-like intermediate. Subsequently, a rotation around the N–N bond of the hydrazone-like intermediate can occur regenerating thereby the thermodynamically stable trans isomer (see Scheme 3).19,20,29,30 Such an intermediate cannot be promoted when the azo dye is dissolved in toluene and, therefore, it should be reached via dimerization or self-aggregation of the azo dye. For the acetylated monomer, AZO-Ac, the hydrazonelike tautomer is not enabled. The operation of a different isomerisation mechanism for both AZO-Ac and AZO-OH is also reflected on the thermal activation parameters of the process (DHa and DSa, see Table 3). Not only a remarkably lower enthalpy of activation (94 kJ mol1 vs. 15 kJ mol1) but also a considerably more negative entropy of activation (29 J K1 mol1 vs. 181 J K1 mol1) was registered for the thermal isomerisation of cis-AZO-OH. This feature evidences a more easily accessible and highly organized transition state due to the establishment of intermolecular hydrogen bonding, which enables the rotational pathway for the isomerisation process. cis-AZO-Ac exhibits a relaxation time of 20 hours when it is introduced as a doped guest into the ordered nematic mesophase of mesogen ZLI-1695 (see Scheme 4). Indeed, the rate of the thermal cis-to-trans isomerisation of AZO-Ac is up to 3.1-fold faster than the one registered in conventional isotropic solvents. This feature is explained in terms of the nematic potential obliging to both mesogen and azo dye molecules to recover their initial alignment along the director direction since it was dramatically disrupted previously due to the photoinduced trans-to-cis isomerisation of the azo dye. This cooperation between both liquid crystal and azo dye molecules is reflected in a less energetic (86 kJ mol1 vs. 94 kJ mol1, see Table 3) but more organized transition state (51 J K1 mol1 vs. 29 J K1 mol1) for the thermal isomerisation process in the nematic mesophase.31–33
Phys. Chem. Chem. Phys., 2014, 16, 3108--3114 | 3111
View Article Online
Published on 10 December 2013. Downloaded by Georgia Institute of Technology on 31/10/2014 09:07:52.
PCCP
Paper
Fig. 3 Transient absorption generated by laser pulse irradiation with UV light (lirrad = 355 nm, lobs = 370 nm) for AZO-OH in ethanol (a), a nematic film of CPAZO-OH (b) and PAZO-OH in ethanol (c) at 298 K ([AZO-OH] = 20 mM).
Scheme 4 Chemical structure and mesomorphic behaviour of nematogens 5CB, ZLI-1695 and MC5CHX. Phase transition temperatures are indicated in Celsius.
Scheme 3 Proposed isomerisation mechanism for the thermal cis-totrans back reaction of the hydroxyl-substituted azoderivative AZO-OH.
Table 3 Thermal activation parameters (DHa and DSa) for the thermal cis-to-trans isomerisation of AZO-Ac and AZO-OH in different isotropic solvents and host nematic mesogens
AZO-Ac
AZO-OH
Solvent
DHa (kJ mol1)
Toluene Ethanol Acetonitrile ZLI-1695 Ethanol Acetonitrile ZLI-1695 MC5CHX
93 96 96 86 19 13 11 37
1 1 1 1 1 1 1 1
DSa (J K1 mol1) 34 25 27 51 184 179 207 126
3 2 1 1 1 1 1 1
Due to the clear dependence of the thermal isomerisation rate of cis-AZO-OH on the solvent nature (see above), the kinetics of its thermal back reaction was investigated in several host nematogens (Scheme 4). cis-AZO-OH showed relaxation times of 287 ms and 261 ms in the oriented nematic phase of the cyano-containing mesogens 5CB and ZLI-1695, respectively. These values are close to the ones obtained in ethanol and acetonitrile isotropic solutions (311 ms and 405 ms, respectively, Table 2), which evidences that, in the nematic mesophase,
3112 | Phys. Chem. Chem. Phys., 2014, 16, 3108--3114
cis-AZO-OH thermally isomerises through the same mechanism than it does in isotropic solutions, that is, the rotational one. In contrast to that observed for AZO-Ac, only a very slight enhancement of the rate of the thermal back reaction was observed for cis-AZO-OH due to the existence of the nematic mean field. The thermal isomerisation rate depends more dramatically on the capability of the mesogen molecules to establish hydrogen bonding with the azo chromophore than on the reestablishment of the nematic ordering. Indeed, a much lower isomerisation rate was detected for cis-AZO-OH when it was doped in the nematic phase of the non-cyano-containing mesogen MC5CHX (t = 1.4 s, Table 2), where no such strong hydrogen bonding between the mesogens and the azo moieties can be established. This differential behaviour is also clearly evidenced in the thermal activation parameters registered for the cis-totrans isomerisation process of AZO-OH in this media. The capability to establish hydrogen bonding of mesogen ZLI-1695 allows the system to reach easier the hydrazone-like intermediate and, therefore, a lower DHa value is registered (11 kJ mol1 vs. 37 kJ mol1 in MC5CHX, see Table 3). Accordingly, a harder molecular reorganization is required for cis-AZO-OH to evolve to the corresponding transition state in ZLI-1695 than in MC5CHX. This feature accounts for the more negative DSa value registered in the former mesogen (207 J K1 mol1 vs. 126 J K1 mol1 for ZLI-1695 and MC5CHX, respectively). Both acetylated linear polymers, the copolymer CPAZO-Ac and the homopolymer PAZO-Ac, exhibited identical relaxation
This journal is © the Owner Societies 2014
View Article Online
Published on 10 December 2013. Downloaded by Georgia Institute of Technology on 31/10/2014 09:07:52.
Paper
times to that of its monomeric counterpart AZO-Ac in toluene solutions (t = 33 h, see Table 2). This result evidences that the covalent attachment of inversion-isomerising azo dyes, like AZO-Ac, to the main polysiloxane backbone as a side-chain group does not overly influence their thermal isomerisation kinetics. In contrast, a toluene-isomerising solution of the linear copolymer CPAZO-OH showed a relaxation time of the same order (t = 254 ms) than the ones registered for the azo-monomer AZO-OH in ethanol and acetonitrile (311 ms and 405 ms, respectively, see Table 2) and very far away from the one registered for cisAZO-OH in toluene (23 min). It should be noticed that this fast thermal isomerisation rate was observed even in the nematic state of the very same copolymer CPAZO-OH (t = 333 ms at 298 K, Fig. 3b), where no solvent was present. Hence, it can be stated from these experimental results that when azophenol monomers are covalently bonded to the polysiloxane backbone, although they are present at a low concentration (10% mol, see Scheme 2), they still are able to interact efficiently with each other through intermolecular hydrogen bonding due to their forced spatial proximity. In order to gain deeper insight into this point, the thermal cis-to-trans isomerisation kinetics of the linear homopolymer PAZO-OH was studied in both ethanol and acetonitrile. PAZO-OH showed a relaxation time of only 17 ms in ethanol at 298 K (see Fig. 3c). This value is ca. 20 times lower than the one registered for the monomer AZO-OH in the same solvent (311 ms, see Table 2). Such a feature evidences that, besides the possibility of the azophenolic monomers to establish hydrogen bonding with the surrounding solvent molecules, a cooperative and efficient interaction between neighbouring azophenol moieties occurs. In acetonitrile, a relaxation time of the same order as the one for cisAZO-OH was detected (283 ms vs. 405 ms, respectively). In this solvent, the intermolecular interaction between the azo dye and the surrounding solvent molecules (solvation effect) counteracts the one occurring between neighbouring azophenol moieties. Moreover, it is highly remarkable that such dye–dye interaction was even observed in the glassy state of the homopolymer PAZO-OH, registering thereby a relaxation time of 125 ms. In this way, the low thermal isomerisation rate of the hydroxylsubstituted azo monomer AZO-OH in ethanol and acetonitrile solutions has been successfully transferred to both nematic and glassy polymers (CPAZO-OH and PAZO-OH, respectively), generating thereby fast responding solvent-free photoactive systems.
Conclusions The kinetic results presented herein demonstrate that both AZO-Ac and AZO-OH clearly thermally isomerise through different mechanisms in all the environments analysed. For AZO-Ac, which isomerises following the inversional pathway, the nematic liquid-crystalline order plays a crucial role in its thermal isomerisation kinetics enhancing it up to 3.1-fold with respect to the one in conventional organic solvents. On the other hand, for AZO-OH, which undergoes a rotational mechanism, its capability to establish hydrogen bonding with its surroundings, either with the solvent or with other azophenol
This journal is © the Owner Societies 2014
PCCP
molecules, is the main factor to be fulfilled for getting a fast thermal back reaction. In this way, for rotationally-isomerising azocompounds, the nematic field has no significant effect on the rate of their thermal back reaction. Following this strategy the fast isomerisation rate observed for azophenols in isotropic solvents can be successfully transferred to nematic or even glassy polymers, where no solvent is present.
Acknowledgements Financial support from the Ministerio de Ciencia e Innovacio´n (CTQ-2012-36074, Spain) is gratefully acknowledged. J. Garcia´s is grateful for a Beatriu de Pino ´s post-doctoral grant Amoro from the Generalitat de Catalunya (2011 BP-A-00270, Spain). Authors thank Prof. Santi Nonell and Dr Walter A. Massad for their help in the laser flash-photolysis measurements. We also acknowledge Prof. Manuel Martı´nez and Dr Carlos Rodrı´guez del Rı´o for the measurements of the volumes of activation.
Notes and references 1 Smart light-responsive materials: azobenzene-containing polymers and liquid crystals, ed. Y. Zhao and T. Ikeda, WILEY-VCH, New Jersey, 2009. 2 H. Rau, Photochemistry and Photophysics, CRC Press, Stockholm, 1990, ch. 4, vol. 2. ´pez, H. Finkelmann, P. Palffy-Muhoray and 3 M. Camacho-Lo M. Shelley, Nat. Mater., 2004, 3, 307. 4 M. Yamada, M. Kondo, J.-I. Mamiya, Y. Yu, M. Kinoshita, C. J. Barrett and T. Ikeda, Angew. Chem., Int. Ed., 2008, 47, 4986. 5 T. Ikeda and T. Ube, Mater. Today, 2011, 14, 480. 6 V. Ferri, M. Elbing, G. Pace, M. D. Dickey, M. Kharnikov, P. Samorı`, M. Mayor and M. A. Rampi, Angew. Chem., Int. Ed., 2008, 47, 3407. ´s, E. Go ´mez, E. Valle ´s and D. Velasco, Chem. 7 J. Garcia-Amoro Commun., 2012, 48, 9080. ´s, S. Nonell and D. Velasco, Chem. Commun., 8 J. Garcia-Amoro 2011, 47, 4022. ´s, H. Finkelmann and D. Velasco, J. Mater. 9 J. Garcia-Amoro Chem., 2011, 21, 1094. ´s, H. Finkelmann and D. Velasco, 10 J. Garcia-Amoro Chem.–Eur. J., 2011, 17, 6518. 11 D. L. Fortin, M. R. Banghart, T. W. Dunn, K. Borges, D. A. Wagenaar, Q. Gaudry, M. H. Karakossian, T. S. Otis, W. B. Kristan, D. Trauner and R. H. Kramer, Nat. Methods, 2008, 4, 331. 12 P. Stawski, M. Sumser and D. Trauner, Angew. Chem., Int. Ed., 2012, 51, 5748. 13 A. Mourot, T. Fehrentz, Y. Le Freuvre, C. M. Smith, C. Herold, D. Dalkara, F. Nagy, D. Trauner and R. H. Kramer, Nat. Methods, 2012, 9, 396. 14 F. Zhang, A. Zarrine-Afsar, M. Sameer Al-Abdul-Wahid, R. S. Prosser, A. R. Davidson and A. G. Woolley, J. Am. Chem. Soc., 2009, 131, 2283.
Phys. Chem. Chem. Phys., 2014, 16, 3108--3114 | 3113
View Article Online
Published on 10 December 2013. Downloaded by Georgia Institute of Technology on 31/10/2014 09:07:52.
PCCP
15 F. Zhang, K. A. Timm, K. M. Arndt and G. A. Woolley, Angew. Chem., Int. Ed., 2010, 49, 3943. 16 A. A. Beharry and G. A. Woolley, Chem. Soc. Rev., 2011, 40, 3022. ´s, M. Dı´az-Lobo, S. Nonell and D. Velasco, 17 J. Garcia-Amoro Angew. Chem., Int. Ed., 2012, 51, 12820. ´s and D. Velasco, in Responsive Materials and 18 J. Garcia-Amoro Methods (Advanced Materials Series), ed. A. Tiwari and H. Kobayashi, WILEY-Scrivener Publishing LLC, New Jersey, 2013, ch. 2. ´s, A. Sa ´nchez-Ferrer, W. A. Massad, S. Nonell 19 J. Garcia-Amoro and D. Velasco, Phys. Chem. Chem. Phys., 2010, 12, 13238. 20 M. Kojima, S. Nebashi, K. Okawa and N. Kurita, J. Phys. Org. Chem., 2005, 18, 994. 21 G. W. Gray and B. Jones, J. Chem. Soc., 1953, 4179. ´s, A. Pin ˜ol, H. Finkelmann and D. Velasco, 22 J. Garcia-Amoro Org. Lett., 2011, 13, 2282. ´nez-Banzo, T. Torres and S. Nonell, 23 N. Rubio, A. Jime J. Photochem. Photobiol., A, 2007, 185, 214.
3114 | Phys. Chem. Chem. Phys., 2014, 16, 3108--3114
Paper
24 C. Gallego, M. Martı´nez and V. S. Safont, Organometallics, 2007, 26, 527. 25 E. R. Talaty and J. C. Fargo, Chem. Commun., 1967, 65. 26 See, for example, J. Binenboym, A. Burcat, A. Lifshitz and J. Shamir, J. Am. Chem. Soc., 1966, 88, 5039, and references therein. 27 T. Asano, T. Okada, S. Shinkai, K. Shigematsu, Y. Kusano and O. Manabe, J. Am. Chem. Soc., 1981, 103, 5161. 28 S. Shinkai, Y. Kusano, K. Shigematsu and O. Manabe, Chem. Lett., 1980, 1303. 29 O. Tsutsumi and T. Ikeda, Mol. Cryst. Liq. Cryst., 2001, 368, 411. 30 G. Gabor, Y. F. Frei and E. Fischer, J. Phys. Chem., 1968, 72, 3266. ´s, M. Martı´nez, H. Finkelmann and 31 J. Garcia-Amoro D. Velasco, J. Phys. Chem. B, 2010, 114, 1287. ´s, H. Finkelmann and D. Velasco, Phys. 32 J. Garcia-Amoro Chem. Chem. Phys., 2011, 13, 11233. 33 P. Beyer and R. Zentel, Macromol. Rapid Commun., 2005, 26, 874.
This journal is © the Owner Societies 2014
Published on 10 December 2013. Downloaded by Georgia Institute of Technology on 31/10/2014 09:07:52.
PAPER
Cite this: Phys. Chem. Chem. Phys., 2014, 16, 3108
View Journal | View Issue
Understanding the fast thermal isomerisation of azophenols in glassy and liquid-crystalline polymers Jaume Garcia-Amoro ´ s and Dolores Velasco* The good solubility of azophenols in low molar mass liquid crystals together with the ability of their related polymers to form homogeneous nematic and glassy thin films make such azoderivatives valuable chromophores to get a great variety of photoactivatable systems with fast switching speeds under ambient conditions. In fact, the final applicability of these systems is mainly determined by the thermal cis-to-trans isomerisation rate of the photoactive azophenol used, in other words, by the intimate mechanism the
Received 25th October 2013, Accepted 9th December 2013
reaction goes through. The kinetico-mechanistic study reported herein shows that the rate of the thermal
DOI: 10.1039/c3cp54519a
located, mainly to its capability to establish hydrogen bonding with its surroundings. With a proper
back reaction for azophenols is very sensitive to the local environment where the azo chromophore is design, azophenol-based polymers can exhibit thermal isomerisation rates as fast as those of the
www.rsc.org/pccp
monomers in solution even without the presence of any solvent.
Introduction Materials that sense their surroundings and tune their properties as a response to them are being researched extensively worldwide since the number of their potential applications is growing exponentially. Siloxane-based liquid crystal polymers (LCPs) can be very attractive materials for this purpose since they combine the molecular order characteristic of the liquid-crystalline state with the ability to form homogeneous thin films. The incorporation of suitable organic chromophores not only into LCPs but also into glassy polymers allows tuning many properties of the entire material by optical control, for instance, its transmission intensity, helical twisting power, fluorescent emission, molecular shape, etc.1 Indeed, light is one of the most attractive triggers to modulate the properties of materials since it is a free, renewable and environmentally friendly energy source which, moreover, can be remotely supplied over a very localised area of the probe. Azobenzenes, well-known light-sensitive molecules, have attracted great interest for this aim owing to the reversible photoisomerisation between their two isomers – trans and cis. In addition, the initial trans form can be regenerated thermally in the dark.2 This photochemical signature has been widely exploited for many applications within not only materials science3–10 but also biology and medicine.11–17 Main characteristic parameters and abilities of azobenzenebased photoactive materials are directly related to the rate of `nics, Departament de Quı´mica Orga `nica, Institut de Grup de Materials Orga Nanocie`ncia i Nanotecnologia (IN2UB), Universitat de Barcelona, Martı´ i Franque`s 1, E-08028, Barcelona, Spain. E-mail: [email protected]; Fax: +34 93 339 78 78; Tel: +34 93 403 92 60
3108 | Phys. Chem. Chem. Phys., 2014, 16, 3108--3114
the thermally-activated cis-to-trans isomerisation of the azo dye used which, in turn, is determined mainly by its molecular architecture.18 Azophenols are valuable chromophores to get actuating and switching materials with fast responses since they are endowed with a rapid thermal isomerisation process at room temperature. Nevertheless, the rate at which the thermal back reaction proceeds for such azo dyes depends dramatically on the environment where they are located.19,20 With this in mind, here we report a thorough and comprehensive study about the thermal cis-to-trans isomerisation of azophenols in different environments. First and prior to the introduction of the azo monomers AZO-Ac and AZO-OH (see Scheme 1) into macromolecular systems, their thermal isomerisation kinetics have been studied not only in conventional organic solvents of different polarity but also in macroscopically ordered nematic solutions. The results obtained in both isotropic and nematic media have allowed us to evaluate the effect of the liquid crystal order on the rate of the thermal back reaction as well as on the corresponding energetic parameters. Such magnitudes have been correlated further with the plausible acting mechanism for
Scheme 1 Chemical structure of the photoactive azo monomers AZO-Ac and AZO-OH.
This journal is © the Owner Societies 2014
View Article Online
Published on 10 December 2013. Downloaded by Georgia Institute of Technology on 31/10/2014 09:07:52.
Paper
Scheme 2 Chemical composition of the photoactive copolymers (CPAZO-Ac and CPAZO-OH) and homopolymers (PAZO-Ac and PAZO-OH). The degree of polymerisation, DP, of the main polyhydrogenomethylsiloxane (PHMS) backbone used is DP = 31.
the process. Finally, the kinetic study of the corresponding linear copolymers (CPAZO-Ac and CPAZO-OH, see Scheme 2) and homopolymers (PAZO-Ac and PAZO-OH) has been carried out in order to examine not only the effect of the linkage of the azo monomers on the polymer backbone but also the influence of the intermolecular interactions that can be established between the different monomers on the thermal relaxation of such macromolecular systems.
Experimental Materials and instrumentation All reagents were used as received without further purification. Anhydrous CH2Cl2 was dried over calcium hydride and freshly distilled before use. FT-IR spectra were registered on a Nicolet 6700 FT-IR spectrophotometer from Thermo Scientific. Electronic spectra were recorded on a Varian Cary 500E UV-Vis-NIR spectrophotometer. 1H NMR spectra were collected on a Varian Mercury spectrometer and they were processed further with the MestRec commercially available software. Differential scanning calorimetry (DSC) was performed using a Mettler-Toledo DSC821 apparatus under nitrogen flow at a scan rate of 5 K min1. Polarized optical microscopy (POM) was performed using a Nikon Eclipse polarizing microscope equipped with a Linkam THMS 600 hot stage and a Linkam CI93 programmable temperature controller at a scan rate of 5 K min1. Synthesis of the nematic mesogen and azo monomers The nematic liquid crystal 4-methoxyphenyl-4-(3-butenyloxy)benzoate (M4OMe) and both photoactive azo monomers 4-acetyl4 0 -(5-hexenyloxy)azobenzene (AZO-Ac) and 4-(5-hexenyloxy)-4 0 hydroxyazobenzene (AZO-OH) were prepared as reported elsewhere.21,22 Synthesis of the polymers Both acetylated linear polymers, poly[6-(4-(4-acetylphenyl)diazenylphenyloxy))hexylmethylsiloxane-co-(4-(4-((4-methoxyphenoxycarbonyl)phenyloxy)butylmethylsiloxane] (CPAZO-Ac) and poly[6-(4-(4acetylphenyl)diazenylphenyloxy))hexylmethylsiloxane] (PAZO-Ac), were prepared via the well-known Pt-catalysed hydrosilylation reaction between the different monomers, which contain a
This journal is © the Owner Societies 2014
PCCP
terminal reactive alkene, and polyhydrogenomethylsiloxane (PHMS-31) at ca. 70 1C for 2 days. After this time, the complete disappearance of the Si–H band at 2100 cm1 was detected by IR spectroscopy. The free hydroxyl-containing polymers, poly[6(4-(4-hydroxyphenyl)diazenylphenyloxy))hexylmethylsiloxane-co(4-(4-((4-methoxyphenoxycarbonyl)phenyloxy)butylmethylsiloxane] (CPAZO-OH) and poly[6-(4-(4-hydroxyphenyl)diazenylphenyloxy))hexylmethylsiloxane] (PAZO-OH), were prepared from their acetylated counterparts by dissolving them in a mixture of CH2Cl2 and MeOH with a catalytic amount of acetyl chloride at room temperature for 2 days. This reaction conditions allow a mild and chemoselective cleavage of the ester group of the acetylated side-chain group. The complete disappearance of the CQO stretching signal at ca. 1765 cm1 and the appearance of a band corresponding to the O–H stretching at around 3450 cm1 were observed as expected. No hydrolysis of the other ester group placed in the mesogen M4OMe was observed (CQO stretching at 1728 cm1). Synthesis of PAZO-Ac. AZO-Ac (201.4 mg) and PHMS-31 (36.7 mg) were dissolved in thiophene-free toluene (3 cm3). Next, a solution of Pt(COD)Cl2 in dry dichloromethane (1%, 20 mL) was added. The reaction mixture was magnetically stirred at 70 1C for 48 h under an inert atmosphere. Then, the solvent was removed under reduced pressure and the residue was dissolved in the minimum amount of THF. The resulting solution was dropped over methanol at 10 1C (200 cm3) with vigorous stirring in order to precipitate the polymer, which was isolated by decantation of the mixture. This process was repeated twice until no monomer was detected by TLC. PAZO-Ac was obtained as a yellow solid. Yield: 142 mg (60%). 1H NMR (CDCl3, 400 MHz): d 7.80 (ar H), 7.15 (ar H), 6.86 (ar H), 3.88 (CH3–O), 2.18 (CH2), 1.72 (CH2), 1.39 (CH2), 0.54 (Si–CH2) and 0.10 (Si–CH3) ppm. FT-IR (ATR): n 2924 and 2854 (C–H st), 1761 (CQO st), 1595 (CQC st), 1495 (NQN st) cm1. UV-Vis: lmax (toluene) = 350 nm (e = 20 050 M1 cm1). Synthesis of PAZO-OH. PAZO-Ac (80 mg) was dissolved in a mixture of CH2Cl2 and methanol (4 : 1 v/v, 3 cm3). Next, acetyl chloride (1 drop) was added. The reaction mixture was stirred at room temperature under an inert atmosphere for 48 h. Then, the solvent was removed under reduced pressure. PAZO-OH was obtained as a dark red solid polymer. Yield: 64 mg (90%). 1 H NMR (d6-acetone, 400 MHz): d 7.81 (ar H), 7.01 (ar H), 4.07 (CH2–O), 1.79 (CH2), 1.47 (CH2), 0.64 (Si–CH2) and 0.12 (Si–CH3) ppm. FT-IR (ATR): n 3473 (O–H st), 2922 and 2854 (C–H st), 1596 (CQC st), 1495 (NQN st) cm1. UV-Vis: lmax (acetonitrile) = 357 nm (e = 12 900 M1 cm1). Synthesis of CPAZO-Ac. AZO-Ac (45 mg, 10% mol), M4OMe (360 mg, 90% mol) and PHMS-31 (56 mg) were dissolved in thiophene-free toluene (3 cm3). Next, a solution of Pt(COD)Cl2 in dry dichloromethane (1%, 50 mL) was added. The reaction mixture was magnetically stirred at 75 1C for 48 h under an inert atmosphere. The purification of CPAZO-Ac was carried out identically to the one described above for PAZO-Ac. CPAZO-Ac was obtained as a yellow rubber. Yield: 129 mg (40%). 1H NMR (CDCl3, 400 MHz): d 8.02 (ar H), 7.83 (ar H), 7.02 (ar H), 6.84 (ar H), 3.91 (CH2–O), 3.75 (CH3–O), 1.77 (CH2), 1.58 (CH2), 0.60 (Si–CH2) and 0.12 (Si–CH3) ppm. FT-IR (ATR): n 2933 and 2871 (C–H st), 1765 (CQO st acetyl), 1728 (CQO st LC), 1605 (CQC st), 1466 (NQN st) cm1.
Phys. Chem. Chem. Phys., 2014, 16, 3108--3114 | 3109
View Article Online
PCCP
Paper
Published on 10 December 2013. Downloaded by Georgia Institute of Technology on 31/10/2014 09:07:52.
Table 1 Mesomorphic behaviour of the prepared polymers determined by DSC at a scan rate of 5 K min1
System
Tg (K)
DCP (J g1 K1)
CPAZO-Ac CPAZO-OH PAZO-Ac PAZO-OH
281 294
0.2 0.2
314
0.4
a
a
a
TS–N (K)
DHS–N (J g1)
TN–I (K)
DHN–I (J g1)
— — 371 —
— — 12.6 —
340 351 399 —
1.3 0.3 2.6 —
Not detected by DSC.
Synthesis of CPAZO-OH. CPAZO-Ac (80 mg) was dissolved in a mixture of CH2Cl2 and methanol (4 : 1 v/v, 3 cm3). Next, acetyl chloride (1 drop) was added. The reaction mixture was stirred at room temperature under an inert atmosphere for 48 h. Afterwards, the solvent was removed under reduced pressure. The purification of CPAZO-OH was carried out identically to the one described above for PAZO-Ac. CPAZO-OH was obtained as a yellow rubber. Yield: 67 mg (84%). 1H NMR (CDCl3, 400 MHz): d 8.01 (ar H), 7.77 (ar H), 7.00 (ar H), 6.83 (ar H), 3.89 (CH2–O), 3.74 (CH3–O), 1.78 (CH2), 1.61 (CH2), 0.60 (Si–CH2) and 0.12 (Si–CH3) ppm. FT-IR (ATR): n 3413 (O–H st phenol), 2933 and 2872 (C–H st), 1728 (CQO st LC), 1604 (CQC st), 1465 (NQN st) cm1. Mesomorphism of the polymers The characterization of the mesomorphic behaviour of the different polymers was carried out by means of DSC (Table 1). Both copolymers, CPAZO-Ac and CPAZO-OH, showed a broad nematic mesophase between their glass transition temperature at Tg = 281–294 K and their nematic-to-isotropic phase transition temperature at TN–I = 340–351 K (DHN–I = 1.3–0.3 J g1), respectively. This feature was also supported by POM experiments, where the characteristic Schlieren texture of the nematic mesophase was observed when a sample of both copolymers was analysed under crossed polarisers (see Fig. 1). Homopolymer PAZO-Ac exhibited a smectic phase between its glass transition temperature, which could not be detected by DSC, and its smectic-to-nematic phase
Fig. 1 POM microphotographs under crossed polarisers of the smectic phase of PAZO-Ac at 365 K (a), nematic phase of PAZO-Ac at 395 K (b), nematic phase of CPAZO-Ac at 298 K (c) and nematic phase of CPAZO-OH at 298 K (d).
3110 | Phys. Chem. Chem. Phys., 2014, 16, 3108--3114
transition temperature at TS–N = 371 K (DHS–N = 12.6 J g1). Moreover, this system showed a nematic phase up to TN–I = 399 K (DHN–I = 2.6 J g1). The presence of free-phenol groups in PAZO-OH caused the disappearance of all mesomorphism. In this way, PAZO-OH melted directly from its glassy state to the isotropic liquid phase at Tg = 314 K. Preparation of the samples and order analysis For the kinetic experiments in the liquid-crystalline state, ca. 4 mM solid solutions of the different azocompounds in the corresponding nematic mesogen were used. Samples were prepared by mixing the desired amounts of the mesogen and the corresponding azo dye followed by homogenization by magnetic stirring for 10 minutes in the isotropic state. Monodomain samples were prepared in 10 mm optical path quartz cells, their surface being rubbed with a piece of cloth in a single direction. Homogeneity of the samples was checked by local probe microscopy. POM experiments were run by rotation of the analyzer of the microscope with respect to the rubbing direction. Upon reaching 451, the expected change from darkness to brightness was observed, which together with the absence of any characteristic texture was indicative of a successful macroscopic orientation of the nematic director. Thin films of the polymers CPAZO-OH (nematic) and PAZO-OH (glassy) were prepared on quartz slides by putting a drop of a concentrated solution of the appropriate polymer in acetone (2 mg cm3) and allowing further evaporation of the solvent thereby applying simultaneously a roll-on-roll-like technique. Kinetic experiments For the experiments in isotropic solvents, 20 mM solutions of the azo dye in the corresponding solvent were measured in 1 cm optical path quartz cells. For the solid solutions, the concentration was ca. 4 mM and 10 mm quartz cells were used instead (see above). A population of cis-azobenzenes was generated by UV photolysis and their thermal relaxation was followed by time-resolved UV-Vis spectroscopy. For long-lived cis-azobenzenes, the samples were irradiated with a Philips high-pressure mercury lamp (total nominal power 500 W) for 30 minutes by using an aqueous Co(NO3)2 solution (0.5 M) as an optical filter. Time-resolved absorption spectroscopy was performed on a Varian Cary 500E UV-Vis-NIR spectrometer. For short-lived samples, the photoisomerisation of the chromophores was performed by means of the nanosecond laser flash-photolysis technique. In this case, the cis isomer was generated by applying a Q-switched Nd-YAG laser pulse (lirrad = 355 nm, 5 ns pulse width, 1–10 mJ per pulse) and the time evolution of the sample absorbance was monitored at 901 by a white-light beam produced using a PTI 75 W Xe lamp. The light transmitted by the sample was spectrally resolved using a monochromator and detected using a Hamamatsu R928 photomultiplier, whose output was fed into a digital oscilloscope through a 50 ohm resistor.23 In all instances, the solutions were thermostated in the dark at the desired temperature (0.1 K) and the thermal cis-to-trans isomerisation was monitored by the change in absorption at a determined observation wavelength,
This journal is © the Owner Societies 2014
View Article Online
Published on 10 December 2013. Downloaded by Georgia Institute of Technology on 31/10/2014 09:07:52.
Paper
lobs. Relaxation times (t = 1/k) were independent of the observation wavelength and they were determined with an associated error lower than 10%. The volumes of activation, DVa, were determined by analyzing the pressure-dependence of the thermal isomerisation rate constant, k. For runs at variable pressure, a previously described pressurizing system and a pillbox cell were used, which were connected to a J&M TIDAS spectrophotometer.24
Results and discussion Both trans-AZO-Ac and trans-AZO-OH exhibit a strong symmetryallowed p–p* transition at 355 nm and a weak symmetry-forbidden n–p* transition at ca. 450 nm. Upon UV-illumination, a decrease in the intensity of the 350 nm band accompanied by an increase in the 450 nm signal is observed as a consequence of the trans-to-cis photoisomerisation of the azo dye. When the illumination is ceased, the thermal cis-to-trans back reaction occurs recovering thereby the thermodynamically stable trans isomer (Fig. 2). The thermal cis-to-trans isomerisation kinetics of both AZO-Ac and AZO-OH has been studied in three different isotropic solvents: acetonitrile, ethanol and toluene. In all of them, cis-AZO-Ac exhibited a very slow thermal cis-to-trans isomerisation process, presenting relaxation times of more than a day at room temperature (Table 2). This feature suggests that the transition state of the thermal back reaction has a non-polar character.25,26 The volumes of activation for the thermal isomerisation of AZO-Ac were determined in both toluene and ethanol. In both solvents, the DVa values obtained were close to zero and similar to the ones reported for 3,30 -[1,10-diaza-4,7,13,16-tetraoxa-18-crown-6]biscarbonylazobenzene, an azobenzene-bridged crown-ether, which is normally taken as a defined standard for the inversion mechanism since it cannot thermally-isomerise via the rotational pathway due to structural restrictions.27,28 These results confirmed that the operating mechanism for the thermal isomerisation of cis-AZO-Ac is the inversional one. cis-AZO-OH exhibited relaxation times of 311 ms and 405 ms in ethanol and acetonitrile at 298 K, respectively (Table 2 and Fig. 3a).
Fig. 2 Changes in the electronic spectrum of a thermally-isomerising AZO-Ac toluene solution at 328 K ([AZO-Ac] = 20 mM, lobs = 355 nm).
This journal is © the Owner Societies 2014
PCCP Table 2 Relaxation time, t, for the thermal cis-to-trans isomerisation process of the monomers AZO-Ac and AZO-OH and their related polymers at 298 K
Solvent
t (h)
AZO-Ac
Acetonitrile Ethanol Toluene ZLI-1695
63 49 35 20
AZO-OH
CPAZO-Ac
Toluene
33
CPAZO-OH
PAZO-Ac
Toluene
33
PAZO-OH
a
Solvent
t (ms)
Acetonitrile Ethanol Toluene ZLI-1695 5CB MC5CHX Toluene Nematic Ethanol Acetonitrile Film
405 311 1.4 106 261 287 1370a 254 333 17 283 125
Value extrapolated from Eyring’s plot.
Otherwise, the isomerisation rate was more than 1000-fold slower when cis-AZO-OH was dissolved in toluene at this temperature (t = 23 min, Table 2). Accordingly, the presence of a free phenol group in the azo dye clearly alters the intimate mechanism that the isomerisation process goes through. In both ethanol and acetonitrile, the thermal back reaction proceeds through a solvent-assisted tautomerisation to yield a hydrazone-like intermediate. Subsequently, a rotation around the N–N bond of the hydrazone-like intermediate can occur regenerating thereby the thermodynamically stable trans isomer (see Scheme 3).19,20,29,30 Such an intermediate cannot be promoted when the azo dye is dissolved in toluene and, therefore, it should be reached via dimerization or self-aggregation of the azo dye. For the acetylated monomer, AZO-Ac, the hydrazonelike tautomer is not enabled. The operation of a different isomerisation mechanism for both AZO-Ac and AZO-OH is also reflected on the thermal activation parameters of the process (DHa and DSa, see Table 3). Not only a remarkably lower enthalpy of activation (94 kJ mol1 vs. 15 kJ mol1) but also a considerably more negative entropy of activation (29 J K1 mol1 vs. 181 J K1 mol1) was registered for the thermal isomerisation of cis-AZO-OH. This feature evidences a more easily accessible and highly organized transition state due to the establishment of intermolecular hydrogen bonding, which enables the rotational pathway for the isomerisation process. cis-AZO-Ac exhibits a relaxation time of 20 hours when it is introduced as a doped guest into the ordered nematic mesophase of mesogen ZLI-1695 (see Scheme 4). Indeed, the rate of the thermal cis-to-trans isomerisation of AZO-Ac is up to 3.1-fold faster than the one registered in conventional isotropic solvents. This feature is explained in terms of the nematic potential obliging to both mesogen and azo dye molecules to recover their initial alignment along the director direction since it was dramatically disrupted previously due to the photoinduced trans-to-cis isomerisation of the azo dye. This cooperation between both liquid crystal and azo dye molecules is reflected in a less energetic (86 kJ mol1 vs. 94 kJ mol1, see Table 3) but more organized transition state (51 J K1 mol1 vs. 29 J K1 mol1) for the thermal isomerisation process in the nematic mesophase.31–33
Phys. Chem. Chem. Phys., 2014, 16, 3108--3114 | 3111
View Article Online
Published on 10 December 2013. Downloaded by Georgia Institute of Technology on 31/10/2014 09:07:52.
PCCP
Paper
Fig. 3 Transient absorption generated by laser pulse irradiation with UV light (lirrad = 355 nm, lobs = 370 nm) for AZO-OH in ethanol (a), a nematic film of CPAZO-OH (b) and PAZO-OH in ethanol (c) at 298 K ([AZO-OH] = 20 mM).
Scheme 4 Chemical structure and mesomorphic behaviour of nematogens 5CB, ZLI-1695 and MC5CHX. Phase transition temperatures are indicated in Celsius.
Scheme 3 Proposed isomerisation mechanism for the thermal cis-totrans back reaction of the hydroxyl-substituted azoderivative AZO-OH.
Table 3 Thermal activation parameters (DHa and DSa) for the thermal cis-to-trans isomerisation of AZO-Ac and AZO-OH in different isotropic solvents and host nematic mesogens
AZO-Ac
AZO-OH
Solvent
DHa (kJ mol1)
Toluene Ethanol Acetonitrile ZLI-1695 Ethanol Acetonitrile ZLI-1695 MC5CHX
93 96 96 86 19 13 11 37
1 1 1 1 1 1 1 1
DSa (J K1 mol1) 34 25 27 51 184 179 207 126
3 2 1 1 1 1 1 1
Due to the clear dependence of the thermal isomerisation rate of cis-AZO-OH on the solvent nature (see above), the kinetics of its thermal back reaction was investigated in several host nematogens (Scheme 4). cis-AZO-OH showed relaxation times of 287 ms and 261 ms in the oriented nematic phase of the cyano-containing mesogens 5CB and ZLI-1695, respectively. These values are close to the ones obtained in ethanol and acetonitrile isotropic solutions (311 ms and 405 ms, respectively, Table 2), which evidences that, in the nematic mesophase,
3112 | Phys. Chem. Chem. Phys., 2014, 16, 3108--3114
cis-AZO-OH thermally isomerises through the same mechanism than it does in isotropic solutions, that is, the rotational one. In contrast to that observed for AZO-Ac, only a very slight enhancement of the rate of the thermal back reaction was observed for cis-AZO-OH due to the existence of the nematic mean field. The thermal isomerisation rate depends more dramatically on the capability of the mesogen molecules to establish hydrogen bonding with the azo chromophore than on the reestablishment of the nematic ordering. Indeed, a much lower isomerisation rate was detected for cis-AZO-OH when it was doped in the nematic phase of the non-cyano-containing mesogen MC5CHX (t = 1.4 s, Table 2), where no such strong hydrogen bonding between the mesogens and the azo moieties can be established. This differential behaviour is also clearly evidenced in the thermal activation parameters registered for the cis-totrans isomerisation process of AZO-OH in this media. The capability to establish hydrogen bonding of mesogen ZLI-1695 allows the system to reach easier the hydrazone-like intermediate and, therefore, a lower DHa value is registered (11 kJ mol1 vs. 37 kJ mol1 in MC5CHX, see Table 3). Accordingly, a harder molecular reorganization is required for cis-AZO-OH to evolve to the corresponding transition state in ZLI-1695 than in MC5CHX. This feature accounts for the more negative DSa value registered in the former mesogen (207 J K1 mol1 vs. 126 J K1 mol1 for ZLI-1695 and MC5CHX, respectively). Both acetylated linear polymers, the copolymer CPAZO-Ac and the homopolymer PAZO-Ac, exhibited identical relaxation
This journal is © the Owner Societies 2014
View Article Online
Published on 10 December 2013. Downloaded by Georgia Institute of Technology on 31/10/2014 09:07:52.
Paper
times to that of its monomeric counterpart AZO-Ac in toluene solutions (t = 33 h, see Table 2). This result evidences that the covalent attachment of inversion-isomerising azo dyes, like AZO-Ac, to the main polysiloxane backbone as a side-chain group does not overly influence their thermal isomerisation kinetics. In contrast, a toluene-isomerising solution of the linear copolymer CPAZO-OH showed a relaxation time of the same order (t = 254 ms) than the ones registered for the azo-monomer AZO-OH in ethanol and acetonitrile (311 ms and 405 ms, respectively, see Table 2) and very far away from the one registered for cisAZO-OH in toluene (23 min). It should be noticed that this fast thermal isomerisation rate was observed even in the nematic state of the very same copolymer CPAZO-OH (t = 333 ms at 298 K, Fig. 3b), where no solvent was present. Hence, it can be stated from these experimental results that when azophenol monomers are covalently bonded to the polysiloxane backbone, although they are present at a low concentration (10% mol, see Scheme 2), they still are able to interact efficiently with each other through intermolecular hydrogen bonding due to their forced spatial proximity. In order to gain deeper insight into this point, the thermal cis-to-trans isomerisation kinetics of the linear homopolymer PAZO-OH was studied in both ethanol and acetonitrile. PAZO-OH showed a relaxation time of only 17 ms in ethanol at 298 K (see Fig. 3c). This value is ca. 20 times lower than the one registered for the monomer AZO-OH in the same solvent (311 ms, see Table 2). Such a feature evidences that, besides the possibility of the azophenolic monomers to establish hydrogen bonding with the surrounding solvent molecules, a cooperative and efficient interaction between neighbouring azophenol moieties occurs. In acetonitrile, a relaxation time of the same order as the one for cisAZO-OH was detected (283 ms vs. 405 ms, respectively). In this solvent, the intermolecular interaction between the azo dye and the surrounding solvent molecules (solvation effect) counteracts the one occurring between neighbouring azophenol moieties. Moreover, it is highly remarkable that such dye–dye interaction was even observed in the glassy state of the homopolymer PAZO-OH, registering thereby a relaxation time of 125 ms. In this way, the low thermal isomerisation rate of the hydroxylsubstituted azo monomer AZO-OH in ethanol and acetonitrile solutions has been successfully transferred to both nematic and glassy polymers (CPAZO-OH and PAZO-OH, respectively), generating thereby fast responding solvent-free photoactive systems.
Conclusions The kinetic results presented herein demonstrate that both AZO-Ac and AZO-OH clearly thermally isomerise through different mechanisms in all the environments analysed. For AZO-Ac, which isomerises following the inversional pathway, the nematic liquid-crystalline order plays a crucial role in its thermal isomerisation kinetics enhancing it up to 3.1-fold with respect to the one in conventional organic solvents. On the other hand, for AZO-OH, which undergoes a rotational mechanism, its capability to establish hydrogen bonding with its surroundings, either with the solvent or with other azophenol
This journal is © the Owner Societies 2014
PCCP
molecules, is the main factor to be fulfilled for getting a fast thermal back reaction. In this way, for rotationally-isomerising azocompounds, the nematic field has no significant effect on the rate of their thermal back reaction. Following this strategy the fast isomerisation rate observed for azophenols in isotropic solvents can be successfully transferred to nematic or even glassy polymers, where no solvent is present.
Acknowledgements Financial support from the Ministerio de Ciencia e Innovacio´n (CTQ-2012-36074, Spain) is gratefully acknowledged. J. Garcia´s is grateful for a Beatriu de Pino ´s post-doctoral grant Amoro from the Generalitat de Catalunya (2011 BP-A-00270, Spain). Authors thank Prof. Santi Nonell and Dr Walter A. Massad for their help in the laser flash-photolysis measurements. We also acknowledge Prof. Manuel Martı´nez and Dr Carlos Rodrı´guez del Rı´o for the measurements of the volumes of activation.
Notes and references 1 Smart light-responsive materials: azobenzene-containing polymers and liquid crystals, ed. Y. Zhao and T. Ikeda, WILEY-VCH, New Jersey, 2009. 2 H. Rau, Photochemistry and Photophysics, CRC Press, Stockholm, 1990, ch. 4, vol. 2. ´pez, H. Finkelmann, P. Palffy-Muhoray and 3 M. Camacho-Lo M. Shelley, Nat. Mater., 2004, 3, 307. 4 M. Yamada, M. Kondo, J.-I. Mamiya, Y. Yu, M. Kinoshita, C. J. Barrett and T. Ikeda, Angew. Chem., Int. Ed., 2008, 47, 4986. 5 T. Ikeda and T. Ube, Mater. Today, 2011, 14, 480. 6 V. Ferri, M. Elbing, G. Pace, M. D. Dickey, M. Kharnikov, P. Samorı`, M. Mayor and M. A. Rampi, Angew. Chem., Int. Ed., 2008, 47, 3407. ´s, E. Go ´mez, E. Valle ´s and D. Velasco, Chem. 7 J. Garcia-Amoro Commun., 2012, 48, 9080. ´s, S. Nonell and D. Velasco, Chem. Commun., 8 J. Garcia-Amoro 2011, 47, 4022. ´s, H. Finkelmann and D. Velasco, J. Mater. 9 J. Garcia-Amoro Chem., 2011, 21, 1094. ´s, H. Finkelmann and D. Velasco, 10 J. Garcia-Amoro Chem.–Eur. J., 2011, 17, 6518. 11 D. L. Fortin, M. R. Banghart, T. W. Dunn, K. Borges, D. A. Wagenaar, Q. Gaudry, M. H. Karakossian, T. S. Otis, W. B. Kristan, D. Trauner and R. H. Kramer, Nat. Methods, 2008, 4, 331. 12 P. Stawski, M. Sumser and D. Trauner, Angew. Chem., Int. Ed., 2012, 51, 5748. 13 A. Mourot, T. Fehrentz, Y. Le Freuvre, C. M. Smith, C. Herold, D. Dalkara, F. Nagy, D. Trauner and R. H. Kramer, Nat. Methods, 2012, 9, 396. 14 F. Zhang, A. Zarrine-Afsar, M. Sameer Al-Abdul-Wahid, R. S. Prosser, A. R. Davidson and A. G. Woolley, J. Am. Chem. Soc., 2009, 131, 2283.
Phys. Chem. Chem. Phys., 2014, 16, 3108--3114 | 3113
View Article Online
Published on 10 December 2013. Downloaded by Georgia Institute of Technology on 31/10/2014 09:07:52.
PCCP
15 F. Zhang, K. A. Timm, K. M. Arndt and G. A. Woolley, Angew. Chem., Int. Ed., 2010, 49, 3943. 16 A. A. Beharry and G. A. Woolley, Chem. Soc. Rev., 2011, 40, 3022. ´s, M. Dı´az-Lobo, S. Nonell and D. Velasco, 17 J. Garcia-Amoro Angew. Chem., Int. Ed., 2012, 51, 12820. ´s and D. Velasco, in Responsive Materials and 18 J. Garcia-Amoro Methods (Advanced Materials Series), ed. A. Tiwari and H. Kobayashi, WILEY-Scrivener Publishing LLC, New Jersey, 2013, ch. 2. ´s, A. Sa ´nchez-Ferrer, W. A. Massad, S. Nonell 19 J. Garcia-Amoro and D. Velasco, Phys. Chem. Chem. Phys., 2010, 12, 13238. 20 M. Kojima, S. Nebashi, K. Okawa and N. Kurita, J. Phys. Org. Chem., 2005, 18, 994. 21 G. W. Gray and B. Jones, J. Chem. Soc., 1953, 4179. ´s, A. Pin ˜ol, H. Finkelmann and D. Velasco, 22 J. Garcia-Amoro Org. Lett., 2011, 13, 2282. ´nez-Banzo, T. Torres and S. Nonell, 23 N. Rubio, A. Jime J. Photochem. Photobiol., A, 2007, 185, 214.
3114 | Phys. Chem. Chem. Phys., 2014, 16, 3108--3114
Paper
24 C. Gallego, M. Martı´nez and V. S. Safont, Organometallics, 2007, 26, 527. 25 E. R. Talaty and J. C. Fargo, Chem. Commun., 1967, 65. 26 See, for example, J. Binenboym, A. Burcat, A. Lifshitz and J. Shamir, J. Am. Chem. Soc., 1966, 88, 5039, and references therein. 27 T. Asano, T. Okada, S. Shinkai, K. Shigematsu, Y. Kusano and O. Manabe, J. Am. Chem. Soc., 1981, 103, 5161. 28 S. Shinkai, Y. Kusano, K. Shigematsu and O. Manabe, Chem. Lett., 1980, 1303. 29 O. Tsutsumi and T. Ikeda, Mol. Cryst. Liq. Cryst., 2001, 368, 411. 30 G. Gabor, Y. F. Frei and E. Fischer, J. Phys. Chem., 1968, 72, 3266. ´s, M. Martı´nez, H. Finkelmann and 31 J. Garcia-Amoro D. Velasco, J. Phys. Chem. B, 2010, 114, 1287. ´s, H. Finkelmann and D. Velasco, Phys. 32 J. Garcia-Amoro Chem. Chem. Phys., 2011, 13, 11233. 33 P. Beyer and R. Zentel, Macromol. Rapid Commun., 2005, 26, 874.
This journal is © the Owner Societies 2014
