Click chemistry to fluorescent hyperbranched polymers. 1 – Synthesis, characterization and spectroscopic properties

European Polymer Journal 59 (2014) 290–301

Contents lists available at ScienceDirect

European Polymer Journal journal homepage: www.elsevier.com/locate/europolj

Click chemistry to fluorescent hyperbranched polymers. 1 – Synthesis, characterization and spectroscopic properties Sandra Medel a, Paula Bosch a,⇑, Carmen de la Torre b, Pedro Ramírez b,1 a b

Instituto de Ciencia y Tecnología de Polímeros (ICTP-CSIC), Juan de la Cierva 3, 28006 Madrid, Spain Instituto de Química Orgánica General (IQOG-CSIC), Juan de la Cierva 3, 28006 Madrid, Spain

a r t i c l e

i n f o

Article history: Received 6 June 2014 Received in revised form 31 July 2014 Accepted 4 August 2014 Available online 17 August 2014 Keywords: Hyperbranched polymers Fluorescent probes Click functionalization Polarity sensitiveness Solvatochromism

a b s t r a c t The synthesis of two new fluorescent hyperbranched polymers with low molecular weight macromolecular skeleton is presented. An experimental approach to obtain fluorescent dansyl derivatives of commercial hydroxyl-terminated hyperbranched polymers (HBPs) have been obtained through click chemistry reactions. Photophysical characterization of the HBP probes compared with their low molecular weight reference compounds shows that the fluorophore in the polymer skeleton behaves as an individual molecule, showing similar emission properties in terms of absorption extinction coefficient, fluorescence quantum yields and position of the fluorescent band. Solvatochromic study of the HBPs in several solvents using ET30, p* and SPP scales demonstrates that there is not a loss of sensitivity towards polarity when the sensor moiety is included in the HBP structure. The absence of quenching deactivation processes in these multifunctional sensors has been attributed to their irregular hyperbranched skeleton, in contrast to what occurs in highly regular dendrimeric analogues. The advantage of the oligomeric probes versus the monomeric ones has been checked preparing transparent and sensitive films using the fluorophores as dopants. The swelling kinetics of the films has been monitored by fluorescence and the diffusion coefficient estimated. Negligible extraction of the hyperbranched probes was found for long time immersions of the films in acetonitrile, whereas low molecular weight derivatives showed continuous extraction. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction The fluorescent labeling of molecules is a powerful method for investigating events at a molecular level. It is a well-developed field in the area of biomolecules [1], whereas in the field of materials science it is increasing in importance [2]. The use of specific labels for proteins and biological molecules is very well developed, but in

⇑ Corresponding author. E-mail address: [email protected] (P. Bosch). Current address: Departamento de Química Orgánica, Universidad de Sevilla, C/Prof. García González 1, 41012 Sevilla, Spain. 1

http://dx.doi.org/10.1016/j.eurpolymj.2014.08.008 0014-3057/Ó 2014 Elsevier Ltd. All rights reserved.

general, the reactive labels are commercialized as expensive kits optimized for biological tests [3]. The preparation of fluorescent dendritic polymers has been the subject of deep research for the last 20 years. After the commercialization of amine-terminated dendrimers, fluorescent dendrimers have been used for many different purposes such as: gene carriers [4], nanoclusters [5], guest encapsulation [6], functionalization of surfaces [7], chemosensors [8], molecular recognition [9] and many others. However, the development of these molecules for a practical use has not been extended very much given the high price of the dendrimer starting material, the low availability of different chemical structures of the dendrimer

S. Medel et al. / European Polymer Journal 59 (2014) 290–301

branches or the difficulty and time consuming synthetic steps. In addition, a strong quenching of the fluorescence emission is found with increasing generation for highly modified dendrimers [10]. In materials science, and looking for a practical use of the sensors, the large-scale preparation of the sensors should be possible. For most cases, hyperbranched polymers may be able to replace dendrimers in utility, because they can be prepared much more rapidly and economically [11]. Their structure gives them excellent flow and processing properties, and they are characterized by lower viscosity than those linear polymers of comparable molecular weight. The highly branched structures gives further access to a larger number of reactive end groups and thus, HBPs have found commercial applications, such as in coatings and nanotechnology [12]. The functionalization of hyperbranched structures with fluorescent groups has not attracted the same interest. A very few specific architectures have been described [13,14], but little effort has been devoted to develop a general methodology to obtain fluorescent hyperbranched polymers. Recently, the synthesis of different fluorescent HBPs have been described for different applications such as biological imaging [15], detection of explosives [16] or detection of chemicals (ascorbic acid [17], glucose [18]). In general, all of them involve de novo multi-step synthetic procedures. The azide–alkyne click reaction has been used as an efficient functionalization tool since the pioneering work of Sharpless et al. [19]. Its advantages of efficiency, mild reaction conditions, and function tolerance, etc. [20,21], make it an attractive and efficient strategy. It has been recently used to graft hyperbranched polymers to different substrates, such as surfaces [22], carbon nanotubes [23] or membranes [24], for the preparation of functional hyperbranched polymers [25–28] and for the functionalization of their peripheral groups [29]. Furthermore, the 1,2,3-triazole ring formed is resistant to hydrolysis, oxidation, reduction and other modes of cleavage. In this work we present a convenient general route for the functionalization of hydroxyl terminated commercial hyperbranched polymers with fluorescent groups, particularly centered on the preparation of two dansyl-functionalized hyperbranched polymers, Boltorn H20 and Hybrane P1000, and their study as fluorescent sensors for polarity changes in their microenvironment. Taking into account the versatility and the high yields obtained through click chemistry, the approach for the synthesis has been the synthesis of azide-modified hyperbranched polymers and subsequent reaction with a dansyl moiety functionalized with a triple bond. The advantages of the hyperbranched probes versus the monomeric ones are presented. The dansyl chromophoric group, 5-(dimethylamino)-1naphthalenesulfonamide, has been selected for labeling the hyperbranched polymers because it shows intense absorption bands in the near UV and a strong fluorescence in the visible region. These properties made this fluorophore to be extensively used for sensing or labeling purposes [30].

291

2. Experimental 2.1. Materials Propargylamine (98%), mesyl chloride (>99.7%), sodium bicarbonate (99.7%), anhydrous magnesium sulfate (>99.5%), dansyl chloride (>99%), triethylamine (>99%), hydrochloric acid (37%), sodium hydroxide (>98%), sodium chloride (>99%), sodium azide (>99%), (+)-sodium-L-ascorbate (>98%), copper(II) sulfate pentahydrate (>98%), ethylenediaminetetraacetic acid (>98.5%), ammonium hydroxide sol. (28.0–30.0% NH3 basis), anhydrous sodium sulfate (>99%), p-toluenesulfonic acid monohydrate (>98.5%), methyl 4-chlorobutyrate (>98%) were obtained from Aldrich and used without further purification. 2Hydroxyethyl methacrylate (HEMA) and 1,6-hexanediol dimethacrylate (HDDMA), both form Aldrich, were used as received. Phenyl bis(2,4,6-trimethyl benzoyl) phosphine oxide (Irgacure 819) from Ciba Specialty Chemicals was used without purification. Solvents used for the synthesis were: anhydrous dichloromethane (99.5%, Aldrich, dried over molecular sieves), pyridine (anhydrous, 99.8%, Aldrich), diethyl ether (Carlo Erba), chloroform (Scharlau), benzene (>99%, Aldrich), ethyl acetate (99.5%, Aldrich) and dimethylformamide (99%, Aldrich), all reagent grade and used as received. MilliQ water was obtained from water purification facility (Millipore Milli-U10). Hybrane P1000 (Mwtheor = 1186 g/mol, AOHtheor = 7) and Boltorn H20 (Mwtheor = 1706 g/mol, AOHtheor = 16), both from Polymer Factory, were used as received. Flash column chromatography was performed by using silica gel (60 Å pore size, 230–400 mesh particle size, Merck). Reactions were monitored using thin layer chromatography (TLC) on silica gel-coated plates (Merck 60 F254). Detection was performed with UV light and/or by insertion in an iodine chamber. Solvents for spectroscopic analysis (acetonitrile, methanol, ethyl acetate, chloroform, tetrahydrofuran, diethyl ether, cyclohexane and toluene, from Sigma–Aldrich) and ethanol (from Scharlau) were of spectroscopic grade and used as received.

2.2. Synthesis 2.2.1. Synthesis of 5-(dimethylamino)-N-(2-propynyl)-1naphthalenesulfonamide (DANSyne) According to previously published methods [31–33] propargylamine (360 lL, 5.55 mmol, 3 equiv.) was added to a solution of dansyl chloride (500 mg, 1.85 mmol, 1 equiv.) and triethylamine (260 lL, 1.85 mmol, 1 equiv.) in anhydrous dichloromethane (15 mL) at 0 °C under nitrogen. After stirring for 1 h, the reaction mixture was warmed to room temperature and stirred for an additional hour (TLC monitoring). The volatiles were removed under vacuum and the crude product was purified by column chromatography on silica gel (CH2Cl2:Et2O, 1:1 v/v) giving DANSyne (460 mg, 86%) as a pure yellow powder. ESI-MS (m/z) Calcd for C15H16N2O2S [M]+ 288; found [M+H]+ 289.

292

S. Medel et al. / European Polymer Journal 59 (2014) 290–301

Anal. Calcd for C15H16N2O2S: C, 62.48; H, 5.59; N, 9.71; S, 11.12. Found: C, 62.77; H, 5.49; N, 9.64; S, 11.03. 1 H NMR (300 MHz, CDCl3, d ppm): 8.56 (d, J = 8.6 Hz, 1H), 8.27 (d, J = 7.3 Hz, 2H), 7.57 (t, J = 7.7, 1H), 7.52 (t, J = 8.5 Hz, 1H), 7.19 (d, J = 7.6 Hz, 1H), 4.99 (t, J = 5.7 Hz, 1H), 3.77 (dd, J = 2.6, 5.7 Hz, 2H), 2.88 (s, 6H), 1.91 (t, J = 2.6 Hz, 1H). 13 C NMR (75 MHz, CDCl3, d ppm): 152.0, 134.2, 130.8, 129.9, 129.8, 128.5, 123.1, 118.6, 115.2, 77.8, 72.6, 45.4, 33.0. 2.2.2. Synthesis of N-(2-azidoethyl)benzamide To a solution of N-(2-hydroxyethyl)benzamide [34] (500 mg, 3.03 mmol, 1 equiv.) in dry pyridine (10 mL) cooled at 0 °C (350 lL, 4.54 mmol, 1.5 equiv.) mesyl chloride was added and the mixture was allowed to warm to room temperature. After 24 h of stirring, the reaction mixture was diluted with dichloromethane (50 mL) and washed consecutively with aqueous HCl (3  20 mL), saturated aqueous NaHCO3 (40 mL) and brine (40 mL). The organic layer was dried over MgSO4, filtered and concentrated under vacuum. The crude product was dissolved in DMF (20 mL) and sodium azide (120 mg, 1.85 mmol, 1.5 equiv.) was added. The mixture was stirred at 90 °C for 18 h after which it was quenched with saturated aqueous NaHCO3 (30 mL). The mixture was extracted with CH2Cl2 (4  30 mL). The combined organic phases were washed with brine (70 mL), dried over Na2SO4, filtered and concentrated to afford N-(2-azidoethyl)benzamide (210 mg, 54%) as a colorless oil. 1 H NMR (300 MHz, CDCl3, d ppm): 7.77 (d, 2H, HAr), 7.45 (m, 3H, HAr), 6.47 (br s, 1H, ANH), 3.63 (m, 4H, ACH2A). 13 C NMR (75 MHz, CDCl3, d ppm): 168.0, 133.9, 131.4, 128.3, 126.9, 50.4, 39.3. 2.2.3. Synthesis of dansyl model (DANSmod) DANSyne (138 mg, 0.5 mmol), N-(2-azidoethyl)benzamide (91 mg, 0.5 mmol) and (+)-sodium L-ascorbate (19 mg, 0.1 mmol) in CH2Cl2 (10 mL) and MilliQ water (10 mL) were stirred under nitrogen for 30 min. Then, CuSO45H2O (12 mg, 0.05 mmol) was added and the reaction mixture was vigorously stirred at room temperature for 3 h. The mixture was extracted with CH2Cl2 (3  15 mL) and the combined organic layers were washed with brine (20 mL), dried over Na2SO4, filtered and concentrated to afford a yellow powder (109 mg, 50%). 1 H NMR (300 MHz, CDCl3, d ppm): 8.53 (d, J = 8.5 Hz, 1H), 8.23 (d, J = 8.4 Hz, 2H), 7.72 (d, J = 7.3 Hz, 2H), 7.49 (m, 3H), 7.40 (m, 2H), 7.29 (s, 1H), 7.15 (d, J = 7.6 Hz, 1H), 6.75 (br s, 1H), 5.53 (t, J = 5.8 Hz, 1H), 4.44 (t, J = 5.4, 1H), 4.19 (d, J = 6.1 Hz, 1H), 3.86 (dt, J = 5.4 Hz, 5.8 Hz, 2H), 2.88 (s, 6H). 13 C NMR (75 MHz, CDCl3, d ppm): 167.8, 152.0, 144.0, 137.6, 134.4, 133.6, 131.9, 130.7, 129.8, 129.7, 129.5, 128.7, 128.6, 127.0, 123.1, 118.6, 115.3, 49.6, 45.4, 39.7, 38.6. 2.2.4. Synthesis of P1000-N3 To a solution of HybraneÒ P1000 (500 mg, 0.42 mmol, 1 equiv.) in dry pyridine (10 mL) cooled at 0 °C mesyl chloride was added (340 lL, 4.4 mmol, 1.5 equiv.) and the mixture was allowed to warm to room temperature. After

3 days of stirring, the reaction mixture was diluted with dichloromethane (200 mL) and washed with aqueous HCl (3  30 mL), saturated aqueous NaHCO3 (80 mL) and brine (100 mL) successively. The organic layer was dried over Na2SO4, filtered and the solvent was evaporated under vacuum, giving a light brown solid (490 mg, 67%). The mesylated polymer was redissolved in DMF (20 mL) and sodium azide (193 mg, 3.0 mmol, 1.5 equiv.) was added. The mixture was stirred at 90 °C for 24 h after which it was quenched with saturated aqueous NaHCO3 (30 mL). The mixture was extracted with CH2Cl2 (4  30 mL). The combined organic phases were washed with brine (70 mL), dried over Na2SO4, filtered and concentrated to afford P1000-N3 (425 mg, 85%) as a cream powder. Anal. Calcd for C62H77N23O12: C, 55.72; H, 5.81; N, 24.11. Found: C, 54.32; H, 5.75; N, 23.59. FTIR–ATR: 2.112 cm1 (AN3), 1.717 cm1, 1.637 cm1 (C@O). 1 H NMR (400 MHz, CDCl3, d ppm): 8.2–7.2 (HAr, 16H), 5.6–5.0 (ACH2CHOA, 4H), 4.4–2.8 (ACH2A, ACH2CHN3, 32H), 1.6–0.9 (ACH3, 30H). 13 C NMR (100 MHz, CDCl3, d ppm): 171.0, 164.8, 138.7– 126.5, 70.5, 57.5–50.0, 17.9, 17.3. 2.2.5. Synthesis of P1000-DANS P1000-N3 (124 mg, 0.09 mmol), DANSyne (210 mg, 0.73 mmol) and (+)-sodium L-ascorbate (29 mg, 0.15 mmol) in CH2Cl2 (10 mL) and MilliQ water (10 mL) were stirred under argon for 30 min. Then CuSO45H2O (19 mg, 0.07 mmol) was added and the reaction mixture was vigorously stirred at room temperature for 24 h. The course of the reaction was monitored by FTIR–ATR, following the disappearance of the azide peak at 2100 cm1. The reaction mixture was diluted with CH2Cl2 (20 mL) and washed with a saturated solution of ethylenediaminetetraacetic acid (EDTA) in 17.5% aqueous NH3 (20 mL). The aqueous phase was extracted with CH2Cl2 (3  15 mL) and the combined organic layers were washed with brine (20 mL), dried over Na2SO4, filtered and the solvent was evaporated under vacuum. The crude product was purified by column chromatography on silica gel (CHCl3:Et2O, 1:1) to afford P1000-DANS (227 mg, 70%) as a yellow powder. Anal. Calcd for C152H173N35O24S6: C, 59.53; H, 5.69; N, 15.99; S, 6.27. Found: C, 54.23; H, 5.29; N, 15.30; S, 5.71. 1 H NMR (500 MHz, CDCl3, d ppm): 8.6–7.0 (m, HAr), 5.5– 3.8 (m, ACH2A), 2.8 (br s, ACH3, 36HDANS), 1.8–0.7 (m, ACH3, 30Hpolym). 13 C NMR (125 MHz, CDCl3, d ppm): 171.0, 164.8, 152.0, 144.4, 138.2, 134.8, 133.3, 130.3, 129.7, 129.5, 128.3, 123.1, 118.9, 115.2, 55.4, 53.4, 45.3, 38.5, 18.0. 2.2.6. Synthesis of H20-N3 The azide substituted acid, 4-azidobutanoic acid, was prepared by a slight modification of a literature procedure [35]. Methyl 4-chlorobutyrate (10 mL, 82 mmol, 1 equiv.) and NaN3 (8 g, 123 mmol, 1.5 equiv.) were dissolved in DMF (140 mL) in a 250 mL round bottom flask. The mixture solution was heated to 110 °C and stirred for 24 h. After the mixture was cooled down to room temperature, 150 mL distilled water was added and the mixture was

293

S. Medel et al. / European Polymer Journal 59 (2014) 290–301

extracted with diethyl ether (3  100 mL). The combined organic layers were dried over anhydrous Na2SO4, filtered and concentrated on a rotary evaporator to give a colorless liquid. The liquid was dissolved in 150 mL of NaOH (12.6 g, 315 mmol, 4 equiv.) MeOH:H2O (1:1, v/v) solution. The mixture was stirred at room temperature for 24 h and then acidified with aqueous HCl to pH 1. The compound was extracted with ethyl acetate and the combined organic fractions were dried over MgSO4, filtered and concentrated to afford 4-azidobutanoic acid (6.82 g, 65%) as a transparent colorless liquid. A mixture of Boltorn H20 (2.025 g, 1.19 mmol, 1 equiv.), 4-azidobutanoic acid (2.988 g, 23 mmol, 1.2 equiv. of the total functionality) and p-toluenesulfonic acid monohydrate (10% p/p) in benzene (80 mL) was refluxed using a Dean Stark apparatus for 48 h, as described before [36]. The course of the reaction was monitored by GPC. Then ethyl acetate (100 mL) was added and the solution was neutralized with saturated sodium bicarbonate (100 mL) and washed with brine (2  80 mL). The organic layer was dried over anhydrous Na2SO4, filtered and the solvent was concentrated on a rotary evaporator to give Boltorn H20-N3 (3.2 g, 90%) as a pale yellow oil. Anal. Calcd for C121H184N36O56: C, 47.82; H, 6.10; N, 16.59. Found: C, 47.77; H, 6.80; N, 16.14. 1 H NMR (300 MHz, CDCl3, d ppm): 4.5–4.0 (m, ACH2A, 48H), 3.8–3.5 (m, ACH2A, 24H), 3.4 (t, ACH2A, 28H), 2.4 (t, ACH2A, 28H), 1.9 (m, ACH2A, 28H), 1.2 (m, ACH3, 36H). 13 C NMR (75 MHz, CDCl3, d ppm): 173.0–171.0, 71.0– 62.0, 50.5, 50.4, 48.6, 46.5, 46.4, 31.0, 30.8, 24.2, 24.1, 17.7, 17.6.

2.3. Film preparation The acrylic formulation was prepared by solution of the fluorescent probe and the photoinitiator (1% w/w) in a 1:1 (v/v) mixture of HEMA and HDDMA. The amount of the fluorescent probe was adjusted in each sample to have a chromophore concentration of 105 M. The formulation was placed in a 1.3  40  0.1 mm3 mold, covered by a polypropylene film and irradiated 9 min in a Biolink-BLX equipment, working at k = 365 nm. Specimens obtained fit in the diagonal direction of a fluorescence cuvette. 2.4. Characterization The NMR spectroscopic data were recorded with Varian Unity-500, Inova-400, Mercury-400 and Bruker-300 instruments with samples dissolved in CDCl3 at room temperature. Chemical shifts were assigned using the residual undeuterated solvent signal as an internal reference.

(a) P1000 HO N

O

OH

O O O O HO

OH

N

N O O

O

N

O O O

2.2.7. Synthesis of H20-DANS Boltorn azide, H20-N3 (411 mg, 0.135 mmol, 1 equiv.), DANSyne (474 mg, 1.64 mmol, 1 equiv.) and (+)-sodium L-ascorbate (65 mg, 0.33 mmol) in CH2Cl2 (10 mL) and MilliQ water (10 mL) were stirred under argon for 30 min. Then CuSO45H2O (41 mg, 0.16 mmol) was added and the reaction mixture was vigorously stirred at room temperature for 24 h. The course of the reaction was monitored by FTIR–ATR. The reaction mixture was diluted with CH2Cl2 (15 mL) and washed with a saturated solution of EDTA in 17.5% aqueous NH3 (20 mL). The aqueous phase was extracted with CH2Cl2 (3  15 mL), the combined organic layers were washed with brine (20 mL), dried over Na2SO4, filtered and the solvent was evaporated under vacuum to afford the triazole hyperbranched polymer, H20-DANS (820 mg, 93%), as a yellow powder. Anal. Calcd for C301H376N60O80S12: C, 55.62; H, 5.83; N, 12.93; S, 5.92. Found: C, 53.89; H, 6.28; N, 12.19; S, 6.20. 1 H NMR (500 MHz, CDCl3, d ppm): 8.48 (s, ACH, 12H), 8.26, 8.24, 8.19 (m, ACH, 24H), 7.40 (m, ACH, 36H), 7.10 (s, ACH, 12H), 4.18 (m, 86H), 3.65–3.25 (m, ACH2A, 32H), 2.83 (s, ACH3, 72H), 2.22 (m, ACH2A, 26H), 2.00 (m, ACH2A, 26H), 1.50–0.70 (m, ACH3, 36H). 13 C NMR (125 MHz, CDCl3, d ppm): 172.2, 172.0, 161.0, 160.6, 151.6, 144.2, 134.9, 130.4, 129.7, 129.5, 128.4, 123.3, 122.8, 119.1, 115.4, 71.0–60.5, 53.6, 49.0, 46.6, 45.3, 41.7, 38.7, 30.5, 30.4, 29.7, 25.2, 25.1, 23.1, 17.9, 17.7.

OH OH

N

HO

(b) H20 OH OH

OH OH

O

O

O O

O

O

O

HO

O

O

O

O O

OH

O

OH

O

OH

O O

O

HO

O

O

O

HO

OH

O O

HO

O

O

O

OH OH

O O

O

OH OH

Scheme 1. Idealized structures of the hyperbranched polymers: (a) Hybrane P1000 and (b) Boltorn H20. Data provided by Polymer Factory.

294

S. Medel et al. / European Polymer Journal 59 (2014) 290–301

Table 1 Molecular weight, hydroxyl number and end-functionalities of the HBPs.

a b

HBP

Trade namea

Avg Mw theoreticalb (g/mol)

Branch functionalitya

Hydroxyl groupsa

Avg number of hydroxyl groupsa

P1000 H20

Hybrane P1000 Boltorn H20

1186 1706

Polyester amide Polyester

Secondary Primary

7 16

Data provided by the company (http://www.polymerfactory.com) for the corresponding HBP precursors. Data calculated from the ideal structure.

Mass spectra were recorded using an HPLC–MSD 1100 mass spectrometer (MS) with an atmospheric pressure ionization source or an electrospray ionization (ESI) source. Fourier transform infrared (FTIR) spectra were recorded using a PerkinElmer Spectrum One spectrophotometer with the attenuated total reflectance (FTIR–ATR) accessory and with a resolution of 8 cm1. Elemental analyses were made with a LECO CHNS-932 apparatus. Gel Permeation Chromatography (GPC) was performed at room temperature using a Waters 1515 isocratic HPLC pump equipped with a Waters 2414 refractive index detector and two Varian PLgel 5 lm MIXED-D columns (300  7.5 mm). The mobile phase was HPLC grade THF at a flow rate of 1.0 mL/min. Injection volumes of 250 lL were made at 5–10 mg/mL sample concentration. Polystyrene (PS) standards were used for the calibration. UV–Vis spectra were recorded on a Perkin Elmer Lambda 35 spectrophotometer using a standard cuvette. Fluorescence measurements were done on a Perkin Elmer LS-55 spectrophotometer using the wavelength of maximum absorption (kabs) as the excitation wavelength

O 2S

and varying the slits in order to achieve a better spectrum for the different probes. Fluorescence quantum yields were determined on the basis of the absorption and fluorescence spectra taken at wavelength excitation at the absorption maximum. Quinine sulfate in 0.1 M H2SO4 was used as standard (U = 0.577 for excitation at 350 nm). The optical densities of all the probes were in the range 0.03–0.05 at the absorption maxima. Permeation experiments were done placing the films in the diagonal direction of a fluorescence cuvette completely filled with acetonitrile. Fluorescence spectra were recorded at regular intervals of time, and diffusion coefficient was calculated according to Eq. (1) as previously described [37]:



M  M0 M eq  M0



 ¼

I  I0 Ieq  I0

 ¼

 1=2 2 Dt L p

ð1Þ

Extraction experiments were performed by immersion of the corresponding film in a 10 mL volumetric flask, provided with magnetic stirring, and recording the emission spectra of the solvent at the given time. Both in permeation and extraction experiments, prior to immersion the surface of the films was cleaned with a cotton pad soaked in acetonitrile.

Cl +

O 2S

CH2Cl2, Et3N

H 2N

NH

3. Results and discussion

(a) T = 0 ºC, 1h (b) r.t., 1h

N

3.1. Synthesis and characterization N

The starting hyperbranched polymers used, P1000 and H20, are commercial products obtained through chain reaction [38], so a number of different possible structures coexist. They have been chosen as starting materials

DANSyne yield = 86%

Scheme 2. Synthesis of DANSyne chromophore.

OH

Ms

OH

HO

(a)

OH HO

OH

HO

Ms Ms

Ms

OH

Ms

HO

(b)

N3

N3 N3

N3

Ms

N3

(c)

DANS

DANS DANS

DANS

N3

P1000-N3

yield = 67%

HO

DANS

DANS

P1000-DANS yield = 70%

yield = 85%

Scheme 3. Synthetic route of P1000-DANS. (a) MsCl, py, r. t., 72 h; (b) NaN3, DMF, 90 °C, 24 h; (c) DANSyne, CuSO4, (+) Na L-ascorbate, CH2Cl2/H2O (1:1), Ar, r. t., 24 h.

(OH) 16

O

+ HO

O

(a) N3

(HO) 4

O H20-N 3 yield = 90%

(b)

N3

(HO) 4

DANS

12

12

H20-DANS yield = 93%

Scheme 4. Synthetic route of H20-DANS. (a) TsOH (cat.), benzene, 48 h; (b) DANSyne, CuSO4, (+) Na L-ascorbate, CH2Cl2/H2O (1:1), Ar, r. t., 24 h.

295

S. Medel et al. / European Polymer Journal 59 (2014) 290–301

because of their different molecular weight, chemical structure and number of reacting groups. We have selected a low molecular weight polymer from each family because of their better solubility properties, and, in addition, because their relatively low number of functional groups. We have used as the ideal structures and number of reactive groups those provided by the companies [39] (Scheme 1 and Table 1). Molecular weights have been calculated on the basis of the ideal structure. Our strategy for the synthesis has been to use the Huisgen 1,3-dipolar cycloaddition of acetylenic-functionalized fluorophores with azido-functionalized hyperbranched polymers. In this way, the link between the fluorescent probe and the structure is a NAC bond, which preserves the photophysical properties of the dansyl group [40]. Functionalization of the dansyl chromophore with the acetylenic group was easily done starting from dansyl chloride and propargylamine under very mild conditions, as described in Section 2, and DANSyne was isolated in 86% yield (Scheme 2). Then, the transformation of the starting AOH groups in P1000 has been easily done using mesylates as intermediates (Scheme 3). With this synthetic approach, we obtained the mesylated P1000 derivative in a good yield (67%) under mild conditions, and then the P1000-N3 after 24 h of reaction with sodium azide at 90 °C in DMF (yield = 85%). Click reaction between P1000-N3 and DANSyne occurred easily at room temperature after 24 h, and was followed by TLC. Complete disappearance of DANSyne

H20 H20-N3

(b) 1.0

P1000 P1000-N3

Normalized intensity

Normalized intensity

(a) 1.0

fluorophore starting material as well as the appearance of a new fluorescent spot in the TLC plate was accomplished after 24 h of reaction. The fluorescent polymer was isolated in 70% yield, and the complete absence of azido moieties was assured by FTIR–ATR spectroscopy. Absolute absence of remaining Cu2+ cations from the catalyst was assured by successive extractions of organic layer with EDTA, as traces of metal cations will destroy the photochemical properties of dansyl group. The degree of functionalization has been calculated by the data provided by 1H NMR spectrum and elemental analysis, and an average value of 6 dansyl chromophore units per molecule is obtained. It has not been possible to obtain the complete functionalization of the starting HBP, and considering the ideal structure of P1000 (Scheme 1), it is a feasible assumption than the internal AOH group does not react under the mild conditions employed, due to steric hindrance caused by the polymer branches. Taking into account the characterization data of the intermediates and final compounds, the incomplete functionalization of the starting AOH groups occurs in the first step of the synthetic route, and the P1000-N3 modified macromonomer contains an average number of one hydroxyl and six azide groups per molecule. The subsequent click reaction with DANSyne takes place in a quantitative yield. The synthesis of the H20-fluorescent derivative needed a different synthetic strategy (Scheme 4). It is reported that the starting H20 compound did not form mesylates easily [36] so we prepared the azide derivative slightly modifying

P1000-DANS

0.5

H20-DANS

0.5

0.0

0.0 12

14

16

12

18

14

16

18

t (min)

t (min)

Fig. 1. Gel permeation chromatograms for the starting HBP and their derivatives: (a) P1000, P1000-N3 and P1000-DANS; (b) H20, H20-N3 and H20-DANS.

O

O OH

N H

(a)

O N H

Ms

(b)

N3

N H

O H N O HN

HN

SO2 (c)

+

N N N

H N

SO2

N3 N N

Scheme 5. Synthesis of DANSmod. (a) MsCl, py, r. t., 24 h; (b) NaN3, DMF, 90 °C, 18 h; (c) DANSyne, CuSO4, (+) Na L-ascorbate, CH2Cl2/H2O (1:1), Ar, r. t., 3 h.

296

S. Medel et al. / European Polymer Journal 59 (2014) 290–301

2100 cm1. Therefore, we assumed that H20-DANS presents an average value of 12 dansyl chromophore units per molecule. The overall yield for the dansyl hyperbranched polymer was 84%. The analysis of the reactions through gel permeation chromatography indicated that the molecular weight distribution is maintained after the three synthetic steps (Fig. 1).

a reported procedure as described in Section 2. As occurred with P1000-N3, the 1H NMR spectrum and elemental analysis confirmed a partial functionalization of the hydroxyl groups, pointing out an average of twelve azide groups per molecule. The following click reaction occurred easily in a good yield (93%). The reaction was monitored by FTIR–ATR spectroscopy and the complete absence of azide groups was assured by the disappearance of the peak at

NH SO2

N

O S N O H

N

N

N N N

O

OH

N N N

O O O O

N

N N

DANSyne

O NH S

N

N

O O

N

O

N

O O O

O

N

N N

N N N

N N

O HN

N

N N

H N

HN

SO2

N

O

NH O S O

S

O

N N N

O NH S O

DANSmod

O HN S O

N N

N

P1000-DANS N

N SO2 HN N

O2 S N H

O2 S N H

N N

N

N

N

N N O

N

OH

O

O O

O O

O

O O

O

NH N N

O 2S

O

NH

O

O

O

O O O

O

OH

O

O

O

O

O

O

N

O O

HO

N

O

N

N

N N

N O 2S

NH

N N

H N

S O2

H N S O2

N

N

O

N

N

SO2

N N N

O

O

O O

N

HN

O

O

O

O

N

N O

O

O

N N N

N

O

O

N

N

O

O

O

SO2

HN

O

O

O HO

O 2S

O

O

O

O O

N

N

N

N

SO2

HN

N N

H20-DANS Scheme 6. Structures of the probes.

N

N N

H N

S O2 N

297

S. Medel et al. / European Polymer Journal 59 (2014) 290–301

A progressive shift to higher molecular weight of the starting HBPs is found when the successive functionalization steps take place, and the pattern of the molecular weight distribution of the final products is almost identical to that of their respective starting materials. The fluorescent low molecular weight model compound of P1000-DANS has been also synthesized as to have the same chemical branch structure of the hyperbranched P1000. The possible effect of the triazole ring as well as the benzamide moiety over the photochemistry of the dansyl group could then be analyzed. For the synthesis of the model compound of the fluorophore unit, N-(2-azidoethyl)benzamide was first prepared as described above and then subjected to the click conditions in the presence of DANSyne chromophore (Scheme 5). The structures of all the probes are summarized in Scheme 6.

3.2. Photophysical properties The absorption and fluorescence emission spectra of DANSyne, DANSmod, P1000-DANS and H20-DANS were registered in several solvents. All of them display the characteristic spectroscopic features of dansyl units with intense absorption bands in the near UV spectral region (kmax around 252 and 340 nm, pp* and CT transitions, respectively) and a strong fluorescence emission in the visible region (Table 2 and Fig. 2). The solvent polarity exerts little effect on the groundstate electronic transition of the chromophore, being all the spectra similar with very little shift in the peak maximum. In contrast, the emission spectra progressively shift to red with an increase in the solvent polarity, with the wavelength of maximum emission (kem) moving from around 450 to 528 nm when the solvent is changed from cyclohexane to methanol. This chromic behavior is associated with the stabilization of the polar emitting excited states by polar solvents [41]. Fluorescent emission of P1000-DANS and H20-DANS shows practically the same strong positive solvatochromic behavior than DANSyne and even higher than its corresponding model. The emission band is shifted

more than 70 nm when the solvent is varied from an apolar (cyclohexane) to a polar one (methanol). The ability of the studied molecules to emit the absorbed light energy is quantitatively characterized by the fluorescence quantum yield (UF). The results obtained (Table 3) show that UF values for all compounds are similar to those described in literature [30] for the dansyl chromophore. It is also noticed that both HBP derivatives have a similar UF than the low molecular weight compounds which means that chromophores maintain their photophysical properties when they are bonded to the HBP behaving as independent entities. This is a very interesting finding if hyperbranched derivatives are compared with fluorescent dansyl dendrimers of comparative molecular weight and number of chromophores. In dendrimers a strong quenching between chromophores is found with increasing generation number, appearing the excimer band corresponding to dansyl units. We have not detected the excimer band even for the highly modified H20-DANS molecule, which bears as much as 12 dansyl groups onto a relatively small architecture. We attribute this finding to the irregular structure of the HBP skeleton, which minimizes interchain interactions present in the high ordered conformation of perfect dendrimers. The solvatochromic behavior of the models, P1000DANS and H20-DANS has been studied by means of ET30 [42], p* [43] and SPP [44] empirical scales, which are commonly accepted to better describe the dependence of the maximum absorption and/or emission wavelength with solvent polarity because of their adequate measurement of dispersive, inductive and electrostatic probe-solvent interactions. In Table 4 the values used for the solvent coefficients in the different scales are listed. Results for ET30 and SPP scales are similar, and the variation of maximum absorption and emission bands for compounds versus solvent polarity is shown in Fig. 3. Using these scales a very low dispersion of the values is found, except for the extremely non polar solvent cyclohexane, in which kem is always lower than the expected value. When using the p* scale, the correlation is also very good, except for the hydrogen-bonding solvents (EtOH and MeOH), which are bathochromically shifted in respect

Table 2 Absorption and emission maximum wavelengths for the fluorescent derivatives in solution (k in nm). Solvents are listed in increasing order of polarity. For each solvent kex = kabs. Solvent

Cyclohexane Diethyl ether Toluene Tetrahydrofuran Chloroform Ethyl acetate Ethanol Acetonitrile Methanol

Dk (nm)

DANSyne

DANSmod

P1000-DANS

H20-DANS

kabs

kem

kabs

kem

kabs

kem

kabs

kem

340 335 340 336 344 337 335 340 336

451 484 487 497 499 500 523 526 528

– 335 340 335 344 336 336 341 336

464 486 487 496 498 499 522 525 525

340 332 345 335 343 335 335 339 336

454 483 489 497 496 500 520 523 525

– – 334 335 342 335 337 340 337

451 492 497 504 507 507 523 527 528

9

77

9

61

13

71

8

77

298

S. Medel et al. / European Polymer Journal 59 (2014) 290–301

Normalized FI (a. u.)

1.0

(a) DANSmod

3.3. Sensitive fluorescent films

Cyclohexane Diethyl ether Toluene THF Chloroform Ethyl acetate Ethanol Acetonitrile Methanol

0.8 0.6 0.4 0.2 0.0 400

450

500

550

600

650

λem (nm)

Normalized FI (a. u.)

1.0

Cyclohexane Diethyl ether Toluene

(b) P1000-DANS

THF Chloroform Ethyl acetate Ethanol Acetonitrile Methanol

0.8 0.6 0.4 0.2 0.0 400

450

500

550

600

650

λem (nm)

Normalized FI (a. u.)

1.0

(c) H20-DANS

Cyclohexane Diethyl ether Toluene

THF

0.8

Chloroform Ethyl acetate Ethanol Acetonitrile Methanol

0.6 0.4 0.2 0.0 400

450

500

550

600

650

λem (nm) Fig. 2. Fluorescence emission spectra for DANSmod (a), P1000-DANS (b) and H20-DANS (c) in several solvents (kex = 340 nm).

to the non-hydrogen bonding solvents, as described [45]. Correlations in hydrogen-bonding and no-hydrogen bonding solvents present the same slope (Fig. 4). The sensitivity of the probes towards changes in polarity can be measured as the slope of the solvatochromic plots, showing the hyperbranched derivatives the same sensitivity than the low molecular weight compounds.

The emission band of dansyl group is very sensitive to changes in polarity and viscosity of its surrounding medium [40]. Rectangular films of 4  1.3 mm2 were easily prepared by photoinitiated polymerization of a 1:1 mixture of HDDMA and HEMA. The formulations included DANSyne, P1000-DANS and H20-DANS as dopants, having the samples a chromophore content of 105 M. These monomers were chosen because the resultant film is slightly swellable in acetonitrile, and, in addition, all compounds show good solubility in the solvent and the monomers. When the films are immersed in acetonitrile under stirring, a continuous extraction of the low molecular weight probe, DANSyne, can be appreciated measuring the fluorescence emission of the solvent (Fig. 5), whereas negligible extraction of the hyperbranched probes is found. Diffusion of liquids through crosslinked polymers is a complex process, where the solvent usually acts as a softening agent increasing the space between polymer chains and producing a plasticizing effect. This behavior favors the mobility of molecules incorporated to the polymer network due to the free volume increase. In the case of mobility-sensitive or rigidochromic fluorescent probes this process could be monitored by the decrease of the fluorescence emission band. The swelling uptake of the doped films in acetonitrile has been monitored by measuring the variation of the fluorescence emission with time. Results when using hyperbranched probes as dopants are shown in Fig. 6. As can be seen, the swelling process of the films is very rapid, reaching the equilibrium after 18 min. The position of the emission band is only shifted 3 nm (from 494 to 497 nm), so the decrease in fluorescence is mainly due to the increase in mobility inside the film, this is, the plastification effect of the solvent. The total solvent uptake measured after equilibrium has been 21 mg, which is as low as 14% w/w. Given the small dimensions of the sample and the velocity of the process, gravimetric analysis is impossible. However, when DANSyne is used as dopant, a continuous decrease of the fluorescence emission is found (Fig. 7), and the obtained two-slope plot arises from the fact that two processes are simultaneously taking place: the swelling of the network and the extraction of the dopant. After the swelling is completed (t = 18 min), only the extraction of the probe is responsible for the decrease in the fluorescence.

Table 3 Fluorescence quantum yields for all derivatives in different solvents. For each solvent kex = kabs. Solvent

Acetonitrile Ethanol Methanol Chloroform Tetrahydrofuran

U DANSyne

DANSmod

P1000-DANS

H20-DANS

0.21 0.19 0.18 0.39 0.27

0.25 0.28 0.21 0.40 0.33

0.21 0.21 0.16 0.36 0.35

0.24 – – 0.34 0.32

299

S. Medel et al. / European Polymer Journal 59 (2014) 290–301

Table 4 Values for the ET30, SPP and p* parameters of the solvents employed (1 – cyclohexane, 2 – diethyl ether, 3 – toluene, 4 – tetrahydrofuran, 5 – chloroform, 6 – ethyl acetate, 7 – ethanol, 8 – acetonitrile and 9 – methanol).

ET30a (kcal/mol) SPPb

p*c

3

4

5

6

7

8

9

34.5 0.694 0.27

33.9 0.66 0.54

37.4 0.838 0.58

39.1 0.786 0.58

38.1 0.795 0.55

51.9 0.853 0.54

45.6 0.895 0.75

55.4 0.857 0.60

Ref. [42]. Ref. [44]. Ref. [43].

550

550

(b) DANSmod 500

450

450

Emission Absorption

500

400

λabs (nm)

S = 2.61

λem (nm)

λabs (nm)

500

400

350

350

30

35

550

550

(a) DANSyne

40

45

50

450

450

Emission 400

400

Absorption

350

350

30

55

35

40

45

50

55

E T (30)

ET (30) 550

550

550

(c) P1000-DANS 500

(d) H20-DANS

550 500

450

Emission Absorption

400

λabs (nm)

450

λem (nm)

S = 2.39 500

λabs (nm)

500

S = 2.27

λem (nm)

c

2

30.9 0.557 0.00

500

S = 2.31

450

450

Emission Absorption

400

λem (nm)

a b

1

400

400 350

350

350 350 30

35

40

45

50

55

30

35

40

45

50

55

E T (30)

E T (30)

Fig. 3. Correlation of absorption and emission maximum wavelengths for the probes with ET30 values. Solvents are listed in Table 4.

60

500

6 3

λabs (nm)

1

450

4

5

8

em

DANSyne DANSmod P1000-DANS H20-DANS

400

abs 350

500 450 400

40

FI (a.u.)

7 9 2

550

λem (nm)

550

DANSyne P1000-DANS H20-DANS

20

350

300

300 0.0

0.2

0.4

0.6

0.8

π* Fig. 4. Correlation of emission and absorption maximum wavelengths for the probes with p* values. Solvents listed in Table 4.

0 0

20

40

60

80

100

time (h) Fig. 5. Fluorescence variation of the solvent during the immersion of films in acetonitrile.

300

S. Medel et al. / European Polymer Journal 59 (2014) 290–301

P1000-DANS H20-DANS

FI (a.u.)

600

550

0

10

20

30

40

immersion time (min) Fig. 6. Fluorescence variation of P1000-DANS and H20-DANS during the immersion of films in acetonitrile.

DANSyne

FI (a.u.)

700

600

When the dansyl group is included in the HBP structure, the macromolecular architecture does not produce quenching or deactivation processes in the fluorophore moiety, even for the highest modified H20-DANS molecule, because the irregular structure of the HBP skeleton minimizes interchain interactions, in contrast to what is described for highly ordered dendrimers. The fluorescence emission of these compounds is highly sensitive to the polarity of their microenvironment, and no loss of sensitivity is obtained with the high molecular weight compounds. Sensitive films have been prepared using the probes as dopants. The fluorescence emission of the probes is also highly sensitive to the rigidity of their microenvironment, and the diffusion coefficient can be easily estimated by means of their fluorescence emission. The availability of reactants, the simplicity of synthesis and purification of the sensors, and the absence of extraction from a polymer matrix make these compounds adequate for large scale preparation and for a practical use as sensors in polymeric media. Due to the excellent photophysical properties observed, these polymeric probes are being used in ongoing work to assess their applicability as sensors in the study of different processes in polymeric media.

500

Acknowledgments 400 0

10

20

30

40

50

60

immersion time (min) Fig. 7. Fluorescence variation of DANSyne during the immersion of the film in acetonitrile.

Table 5 Diffusion coefficient values obtained via fluorescence measurements. Probe

D (cm2/s)

DANSyne P1000-DANS H20-DANS

1.1  106 7.5  107 3.5  107

Assuming a Fickian mechanism for the swelling, we have estimated the diffusion coefficient of acetonitrile into this acrylic network [37] (Table 5). As can be seen, the values obtained for the hyperbranched probes are very concordant, being the value obtained with the monomeric probe slightly higher, because the release of the probe contributes to the decrease of the fluorescence measurement. 4. Conclusions A new series of fluorescent hyperbranched polymers have been successfully prepared by click chemistry by introducing fluorophores into the side-chain of hyperbranched polymers. The synthetic procedure allows obtaining the polymers in good yields.

Financial support provided by the Ministerio de Economía y Competitividad (MINECO) through the Projects MAT 2009-09671 and MAT 2012-31709 is acknowledged. S.M. thanks MINECO for a FPI fellowship. References [1] Chen M, Yin M. Prog Polym Sci 2014;39:365–95. [2] Schäferling M. Angew Chem Int Ed 2012;51:3532–54. [3] The Handbook. A guide to fluorescent probes and labeling technologies. USA: Invitrogen Corp.; 2005. [4] Wang X, He Y, Wu J, Gao C, Xu Y. Biomacromolecules 2010;11:245–51. [5] Tanaka SI, Miyazaki J, Jin TDKT, Inouye Y. Angew Chem Int Ed 2011;50:431–5. [6] Nguyen TTT, Türp D, Wang D, Nölscher B, Laquai F, Müllen K. J Am Chem Soc 2011;133:11194–204. [7] Kainz QM, Schätz A, Zöpfl A, Stark WJ, Reiser O. Chem Mater 2011;23:3606–13. [8] Staneva D, Bosch P, Grabchev I. J Mol Struct 2012;1015:1–5. [9] Michael Pittelkow CBN, Broeren Maarten AC, van Dongen Joost LJ, van Genderen Marcel HP, Meijer EW, Christensen Jørn B. Chem – Eur J 2005;11:5126–35. [10] Ceroni P, Bergamini G, Marchioni F, Balzani V. Prog Polym Sci 2005;30:453–73. [11] Kim YH. J Polym Sci, Part A: Polym Chem 1998;36:1685–98. [12] Bruchmann B, Voigt B. Hyperbranched polymers: synthesis, properties and applications. John wiley &Sons; 2011 [chapter 16]. [13] Kim JS, Lee SY, Yoon J, Vicens J. Chem Commun 2009:4791–802. [14] Wang H, Qiu T, Song N, Li X. Polym Int 2010;59:1384–9. [15] Feng G, Liang J, Liu B. Macromol Rapid Commun 2013;34:705–15. [16] Li H, Wu H, Zhao E, Li J, Sun JZ, Qin A, et al. Macromolecules 2013;46:3907–14. [17] Bao B, Tao N, Ma M, Zhang L, Yuwen L, Fan Q, et al. Soft Mater 2014;12:73–8. [18] Jiang T, Jiang G, Wang X, Dong Y, Wei Z, Li X, et al. Des Monomers Polym 2014;17:576–81. [19] Kolb HC, Finn MG, Sharpless KB. Angew Chem Int Ed 2001;40:2004–21. [20] Lahann J, editor. Click chemistry for biotechnology and materials science. John Wiley & Sons: UK; 2009.

S. Medel et al. / European Polymer Journal 59 (2014) 290–301 [21] Rostovtsev VV, Green LG, Fokin VV, Sharpless KB. Angew Chem Int Ed 2002;114:2708–11. [22] Soto-Cantu E, Lokitz BS, Hinestrosa JP, Deodhar C, Messman JM, Ankner JF, et al. Langmuir: ACS J Surf Colloids 2011;27:5986–96. [23] Sahoo NG, Rana S, Cho JW, Li L, Chan SH. Prog Polym Sci 2010;35:837–67. [24] Cai T, Li M, Neohab KG, Kang ET. J Mater Chem B 2013;1: 1304–15. [25] Qin A, Lam JWY, Tang BZ. Chem Soc Rev 2010;39:2522–44. [26] Qin A, Lam JWY, Tang BZ. Macromolecules 2010;43:8693–702. [27] Sumerlin BS, Vogt AP. Macromolecules 2010;43:1–13. [28] Binder WH, Sachsenhofer R. Macromol Rapid Commun 2007;28:15–54. [29] Roy RK, Ramakrishnan S. J Polym Sci, Part A: Polym Chem 2011;49:1735–44. [30] Vögtle F, Gestermann S, Kauffmann C, Ceroni P, Vicinelli V, Cola LD, et al. J Am Chem Soc 1999;121:12161–6. [31] Bolletta F, Fabbri D, Lombardo M, Prodi L, Trombini C, Zaccheroni N. Organometallics 1996;15:2415–7. [32] Deiters A, Cropp TA, Mukherji M, Chin JW, Anderson JC, Schultz PG. J Am Chem Soc 2003;125:11782–3.

301

[33] Rubinshtein M, James CR, Young JL, Ma YJ, Kobayashi Y, Gianneschi NC, et al. Org Lett 2010;12:3560–3. [34] Mroczkiewicz M, Ostaszewski R. Tetrahedron 2009;65:4025–34. [35] Yu T, Bai JZ, Guan Z. Angew Chem Int Ed 2009;121:1117–21. [36] Katritzky AR, Song Y, Sakhula R, Gyanda R, Meher NK, Wang L, et al. J Polym Sci, Part A: Polym Chem 2009;47:3748–56. [37] Bosch P, Fernández A, Salvador EF, Corrales T, Catalina F, Peinado C. Polymer 2005;46:12200–9. [38] Froehling P. J Polym Sci, Part A: Polym Chem 2004;42:3110–5. [39] (a) http://www.polymerlibrary.com.(b) (accessed 10.13). [40] Lakowicz JR. Principles of fluorescence spectroscopy. New York: Springer; 2006. [41] Katan C, Terenziani F, Mongin O, Werts MH, Porrès L, Pons T, et al. J Phys Chem A 2005;109:3024–37. [42] Reichardt C. Chem Rev 1994;94:2319–58. [43] Kamlet MJ, Abboud JL, Taft RW. J Am Chem Soc 1977;99:6027–38. [44] Catalán J, López V, Pérez P, Martin-Villamil R, Rodríguez JG. Liebigs Ann 1995;2:241–52. [45] Kamlet MJ, Abboud JML, Abraham MH, Taft RW. J Org Chem 1983;48:2877–87.