Novel Reversible Mechanochromic Elastomer with ... - ACS Publications

Research Article www.acsami.org

Novel Reversible Mechanochromic Elastomer with High Sensitivity: Bond Scission and Bending-Induced Multicolor Switching Taisheng Wang,† Na Zhang,† Jingwen Dai,† Zili Li,† Wei Bai,*,‡ and Ruke Bai*,† †

CAS Key Laboratory of Soft Matter Chemistry, Department of Polymer Science and Engineering, University of Science and Technology of China, Hefei 230026, P. R. China ‡ Department of Chemistry, University of MassachusettsAmherst, 300 Massachusetts Avenue, Amherst, Massachusetts 01003, United States S Supporting Information *

ABSTRACT: Although the rational designed mechanochromic polymer (MCP) materials have evoked major interest and experienced significant progress recently, it is still a great challenge to develop a facile and effective strategy for preparation of reversible broad-spectrum MCPs with a combination of wide-range color switch ability and high sensitivity, which thus make it possible to mimic gorgeous color change as in nature. Herein, we designed and synthesized a novel rhodamine-based mechanochromic elastomer. Our results demonstrated that the elastomer exhibited very promising and unique properties. Three primary fluorescence colors were presented during continuous uniaxial extension and relaxing process, and reversible broad-spectrum fluorescence color change could be achieved consequently. The fluorescence quantum yield of the opened zwitterion of this new mechanophore was as high as 0.67. In addition, the elastomer showed very high sensitivity to stress with a detectable activation strain of ∼0.24, which was much smaller than those reported in the previous literature reports. Meantime, the easy-to-obtain material, facile preparation, and good mechanical property also made it suitable for potential practical applications. KEYWORDS: mechanochromic, broad-spectrum, bond scission, bond bending, rhodamine, high sensitivity

■

INTRODUCTION The ability of responding to mechanical strain and transducing mechanical stimulus into optical signal is one of the most attractive phenomenon developed by natural evolution.1,2 For example, cephalopod, such as cuttlefish, octopus, and squid, can take advantage of dazzling patterns of colors produced by contracting and releasing the muscles to improve the success rate in hunting and escaping.3,4 Inspired by their natural counterparts, the rational designed mechanochromic polymer (MCP) materials have also evoked major interest and experienced significant progress in recent years.5−11 The growing trend of MCP is believed to be mostly derived from its potential in developing sensing or displaying devices that can be used to measure or monitor the change of stress or strain, thus leading to various latent applications such as memory chip, security communication, and human motion monitoring.12−15 To fully take advantage of the MCP materials, the ability to display broad-spectrum optical signal, optical reversibility, and high quantum yield are highly demanded besides general requirements such as good mechanical properties and weathering resistance. Currently, the most common method to prepare MCP materials is physical doping, in which dyes are generally dispersed in the polymer matrixes.16−20 However, most of these materials possess poor reversibility, thus limiting their applications. On the other hand, the mechanochromic © 2017 American Chemical Society

polymers consisting of mechanophores have also been exploited. The conversion is often based on the transfer of mechanical forces along a polymer chain to a responsive unit, the mechanophore, of which the optical properties can be modulated by force-induced bond scission or isomerization of the mechanophores.5,6,8,21−31 Although very promising, these mechanochromic polymers as bioinspired materials still suffer from some drawbacks. For instance, up to now, most of the mechanochromic polymers reported in the literature only display two-color switching due to their specific mechanistic models, which make them hard to mimic the gorgeous color change as in nature for wider applications. The spiropyran motif’s merocyanine form has been widely employed in recent years as a mechanophore32−38 that exhibits secondary fluorescence color transition under force. However, besides the extremely low fluorescence quantum yield of the merocyanine form (ϕf < 0.02), which is also known as a crucial disadvantage,39,40 the color gamut produced in those cases was quite narrow. Therefore, it is still a challenging task to develop a facile and effective strategy for preparation of Received: January 5, 2017 Accepted: March 14, 2017 Published: March 14, 2017 11874

DOI: 10.1021/acsami.7b00176 ACS Appl. Mater. Interfaces 2017, 9, 11874−11881

Research Article

ACS Applied Materials & Interfaces

of THR was determined to be ϕf = 0.67, which was almost 2 orders of magnitude higher than that of the widely employed merocyanine. To achieve high sensitivity and strength, we used TAR as the cross-linker to prepare the mechanochromic elastomers through a three-step photopolymerization process (Scheme S3). First, a cross-linked polymer film was obtained by photopolymerization of ethyl acrylate (EA) and TAR in toluene under irradiation with ultraviolet light. This film was designated as “single network” or as first network (SNx, where x represents the cross-linker concentration; for example, SN1 was cross-linked at 1.0 mol % of monomer). Then the rhodaminederived mechanophores that contained SNx were stretched to prompt the sensitivity of the mechanochromic elastomers by swelling with EA, photoinitiator, and a certain amount of butanediol diacrylate (BDA). The double networks (DNx) were obtained by following photopolymerization. The triple networks (TNx) with higher level of prestretch were then prepared by repeating the swelling and photopolymerization procedure. Mechanical properties of the elastomers were tested, and the tunable mechanical performance could be achieved to satisfy the specific requirements in a wide range of applications. The results in detail are presented in Table 1. By comparing the elastomers with different levels of networks, i.e., SN1, DN1, and TN1 (Figure S1), both the elastic modulus and the tensile strength of them were increased in the order of SN1, DN1, and TN1. The modulus and the tensile strength of TN1 were higher by a factor of up to 1.9 and 18.2, respectively, compared with that of SN1. These results indicated that the three-step polymerization process was a facile route to enhance the mechanical properties of the elastomer. Moreover, higher content of cross-linker resulted in lower breaking stress and elongation. For example, the TN1 elastomer exhibited a breaking stress of 27.4 MPa and a stretch ratio of 3.2, which were much larger than those of the TN3 elastomer. A 1% content of cross-linker resulted in the highest prestretch (λprestretch = 2.4) of the first networks in TN1 elastomer, which would greatly facilitate the force-induced isomerization process of the mechanophore when it was stretched (Table S1). Dynamic thermomechanical analysis (DMA) and differential scanning calorimetry (DSC) measurements were performed to test the thermal properties of the TNx elastomers. The glass transition temperature (Tg) of the TN1 elastomer was measured to be −18 °C, which was similar to that of polyethyl acrylate reported previously, demonstrating its elastic character at room temperature.46 Moreover, a slight increase of Tg was observed when the concentration of TAR was varied from 1% to 3% as the polymer chain mobility would be restricted in higher cross-linking density (Figures S4 and S5). Because the TN1 elastomer possessed the highest prestretch of the first networks, we qualitatively examined the mechanoresponsive properties of it by stretching it by hand. Notably, the elastomer showed a vibrant fluorescent color change from pale blue to red with high contrast during stretching (Figure 1a). The transformation also could be observed by the naked eye from colorless to dark red. This fluorescent and visible color change could be ascribed to the force-induced isomerization of the rhodamine mechanophore from closed form to opened zwitterion, which dramatically redshifted the absorption band and turned on the fluorescence emission. The dependence of fluorescence intensity on the stretch ratio is shown in Figure 1b. The initial TN1 elastomer showed an emission band at 420 nm with a weak blue color

reversible broad-spectrum MCP with a combination of widerange color-switch ability and high sensitivity. In this article, we work on using a rhodamine-based single mechanophore to develop a new broad-spectrum MCP elastomer with the ability to display three primary colors, which exhibit broad-spectrum color-switch ability and high quantum yield. Rhodamine derivates are widely used for chemical or biological sensors because they can be easily transformed from a twisted structure to a planar zwitterion that exhibits high quantum yield.41,42 The well-studied property and easy accessibility make it an ideal candidate for practical broadspectrum MCP materials. Nevertheless, the use of rhodamine as a mechanophore for creating applicable polymers is still very rare. Notably, in a recent literature report by Jia et al., it was found that the rhodol fluorophore, a hybrid structure of fluorescein and rhodamine, 43−45 could act as both a mechanophore and a photochromic compound when covalently linked in a polyurethane elastomeric matrix.25 However, only two primary fluorescence color switchings were observed for the elastomer under mechanical stimulus. In addition, it was subjected to a disadvantage of low sensitivity in that the huge pressure used for activating the mechanophore was destructive to the elastomer, and no color change was found when stretching. Our results demonstrated that the elastomers as mechanochromic material exhibited very promising and unique properties. First, they could exhibit three primary fluorescence colors during the continuous uniaxial extension and relaxing process, and hence reversible broad-spectrum fluorescence color-change ability could be achieved. The fluorescence quantum yield of the opened zwitterion of the new mechanophore was as high as 0.67. What’s more, high sensitivity to stress was achieved with a detectable activation strain of ∼0.24, which was much smaller than those reported in the previous literature reports.5,22 To the best of our knowledge, this is a first example of MCP elastomer that can enjoy the combined benefits of broad-spectrum fluorescent color-switch ability, reversibility, and high sensitivity. Meantime, the easy-to-obtain material, facile preparation, and good mechanical property also make it suitable for potential practical applications.

■

RESULTS AND DISCUSSION As an initial step, a novel triacrylic ester of rhodamine (TAR) was designed and synthesized conveniently from commercially available rhodamine 6G. To increase the quantum yield of the ring-open zwitterion, two electron-donating tertiary amino groups were incorparated into the phenyl rings and trihydroxy rhodamine derivative (THR) (Scheme 1) was obtained. Then, TAR as cross-linker for photopolymerization was prepared through the reaction of THR with acryloyl chloride (Scheme S1). The fluorescence quantum yield of the opened zwitterion Scheme 1. Chemical Structures of THR and TAR

11875


Research Article


Table 1. Mechanical Properties of Elastomers; Shown Are the Young’s Modulus E (in MPa), True Stress at Break (in MPa), and Stretch Ratio at Break (λbreak) 1% cross-linker TAR

2% cross-linker TAR

3% cross-linker TAR

sample

T (°C)

E

σbreak T

λbreak

E

σbreak T

λbreak

E

σbreak T

λbreak

SN DN TN

25 25 25

0.9 1.2 1.7

1.5 18.5 27.4

2.5 3.0 3.2

1.5 1.9 4.0

1.9 10.1 27.0

1.9 2.4 2.9

2.1 2.3 5.4

1.0 5.3 8.6

1.4 2.3 1.5

Figure 1. (a) Fluorescent (left) and optical (right) images of manually stretched dumbbell-shaped TN1 elastomer. (b) Normalized FL intensity of TN1 elastomer during continuous extension. (c) Normalized FL intensity of TN1 elastomer under tensional and relaxed states. (d) Optical images of TN1 elastomer after unloading at room temperature and on exposure of UV light (λex = 340 nm). (e) Chromaticity parameters of TN1 elastomer at different stretch ratios and in relaxed state.

state in 3 min under irradiation of 340 nm ultraviolet light (100 W) (Figure 1d). Although the photoinduced transformation of the spirolactam to the open form was reported,47,48 we found that the newly synthesized compound THR was fairly stable under UV irradiation, which was proved by NMR spectroscopy (Figure S13). Figure 1e shows the sequential changes in the chromaticity parameters of the TN1 elastomer. As the stretch ratio increased, the overall chromaticity coordinates of the elastomer shifted from the blue region (x = 0.201, y = 0.165) to the red region. Once the stress was removed, the chromaticity coordinates moved to yellow region (x = 0.337, y = 0.395) immediately. This was a rather interesting and rare phenomenon that force can induce three primary fluorescence color transitions in a single mechanophore. Because the three primary fluorescence colors were located in the blue, yellow, and red regions separately, the elastomer could produce rich fluorescence colors by rational tuning of the stretch ratio, which was useful for various of purposes that require broad-spectrum colors (see Movie S1 in Supporting Information). To understand the role of the mechanophores in the elastomers, we also prepared the control composite elastomers by simply doping THR into cross-linked polyethyl acrylate. No visible and fluorescence color changes were observed when the composite elastomer was stretched (Figure S14). These results indicated that the force-induced ring-opening reaction of the spirolactam

that was originated from the closed form of the mechanophore (Figure S7). A very weak broad band ranging from 500 to 625 nm was also observed for the initial elastomer as a result of slight isomerization of the mechanophore. When the stretch ratio continuously increased (Figure 1b), a new band at 600 nm emerged with a red emission. Similar changes were also observed in the absorption spectra. The pristine TN1 elastomer film was transparent without any absorption bands at >450 nm. However, a broad absorption band at 564 nm appeared when the film was stretched (Figure S8). It was very interesting that the fluorescence color of the film immediately turned from red to yellow when the stress was released. Actually, both absorption and fluorescence emission spectra were blue-shifted by approximately 28 and 50 nm, respectively, as soon as the stress was unloaded (Figures 1c and S9). The most important issue was that the fluorescence color change between yellow and red could be repeatedly switched back and forth by cycling stress loading and stress unloading, which could be spectroscopically quantified (Figures S10 and S11). After unloading the stress, the elastomer could go back to the colorless state at room temperature within 5 h automatically (Figures 1d and S12) or 2 min by heating it at 70 °C, and this stretch-induced color change could be repeated over many cycles. What was beyond our expectation was that the TN1 elastomer could also return to the colorless state from the red 11876


Research Article


Figure 2. (a) Optimized molecular geometries of closed spirolactam (left) and opened zwitterion (right). (b) Mechanically triggered ring-open reaction and bond-bending-induced secondary fluorescent color transition. (c) Illustration of the torque M generated by the forces in three directions. The force was assumed to be equally distributed across all three junctions. (d) Fluorescence spectrum of TN1 elastomer saturated with CF3COOH.

occur in the presence of acid such as trifluoroacetic acid.47,51 Because the negative charge (CF3COO−) was adjacent to the positive charge in this situation, we speculated that the dipole moment was dramatically decreased, which would planarize the open state. Therefore, to further confirm the mechanism of the secondary color transition, the fluorescence spectrum of TN1 elastomer saturated with CF3COOH gas was studied. It was found that the red fluorescence appeared at 600 nm after the reaction (Figure 2d), indicating that the planar zwitterions were formed in this situation. This result was further confirmed by the simulation data. As shown in Figure S17, the bond angle of C1−C2−O was calculated to be 175° for the zwitterions with trifluoroacetic anion, which was consistent with our hypothesis. To determine whether the observed color change was dependent on linking positions or was the result of other factors, we also synthesized two control mechanophores with two cross-linking junctions (Figure 3a and Scheme S2) and then prepared the triple network elastomers following the same procedures for TN1. These two elastomers were referred to as C-3 and C-6. Both C-3 and C-6 exhibited a weak broad band at 550 nm before mechanical stretching (Figure 3b and c). This was because of a slight isomerization of the mechanophore, similar to that of TN1. Because the two cross-linking junctions of cross-linker 6 were located on the xanthenes, the force could not be transferred across the sensitive C−N spiro bond. Indeed, the fluorescent band at 550 nm almost remained the same with increasing stretch ratio (Figure 3c), indicating that the mechanophore was not activated in C-6 elastomer. Unlike the mechanophore in C-6 elastomer, cross-linker 3 was linked in a manner that the bond-scission reaction was not hindered consequently. We could see an obvious increase in fluorescence band at 550 nm with a red-shift of ∼8 nm when the C-3 elastomer was stretched continuously (Figure 3b). The red-shift in fluorescence was a result of bond bending induced by planarization of the mechanophore, which however was restricted in cross-linker 3 with two linking junctions. Compared with the TN1 elastomer, the value of red-shift for

would occur only when the mechanophore was incorporated in the polymer networks as a linking unit, thus endowing the elastomer with a mechanochromic property. To gain more insights into the mechanochromic property of the elastomer, theoretical calculations were performed on the closed and opened form of the mechanophore to optimize their molecular geometries (Figure 2a). To reduce computation time, the hydroxyls on the tertiary amino groups were ignored. To our surprise, the zwitterion of the mechanophore did not adopt a planar conformation. The angle of C1−C2−O was calculated to be 134°. Although the positive charge on the opened zwitterions could be delocalized over the π extended xanthene rings, the negative charge was localized at nitrogen. Therefore, such zwitterions have a large dipole moment (DM = q × d) because of the relative long distance between charges, which may lead to the formation of this bent geometry. The energy barrier for the C−N bond-scission reaction was simulated to be 96.2 kJ/mol (Figures S15 and S16), which was much lower than the breaking energy for the C−C or C−O bond.49,50 During the continuous uniaxial extension, the stress was transferred to the mechanophores through the cross-linked network of the elastomer. Once the C−N bond of the twisted spirolactam was broken, the torque (M = F3 × d) applied by the polymer chains would lead to the bending of the C1−C2 bond (Figure 2c), which planarized the conformation of the zwitterions. This would facilitate the delocalization of the π electrons in the phenyl ring to that of the xanthene, resulting in red-shifts in both the absorption and fluorescence spectra. When the stress was removed, the zwitterions immediately went back to the bent geometry, leading to an obvious blueshift in the fluorescence spectra (Figure 2b). The bending energy of the C−C bond is ∼30 kJ/mol, which is much lower than the breaking energy of the C−N bond in spirolactam. As soon as the C−N bond broke, the zwitterion planarized instantaneously. This might explain why the emission band at 600 nm appeared at the early stage of uniaxial stretching. As is well-known, the ring-opening reaction of the spirolactam could 11877


Research Article


Figure 3. (a) Chemical structures of control cross-linkers 3 and 6. Normalized FL intensity of C-3 (b) and C-6 (c) elastomer during continuous extension. The inset shows the images of the elastomer before and after mechanical stretching (the stretch ratio was ∼2.0). Figure 4. Stress and fluorescence intensity ratio of I600/I420 as a function of stretch ratio for DN1 (a) and TN1 (b) elastomer. The TN1 elastomer for finger-motion detection: stretch the finger (c), bend the finger (d), and stretch the finger again (e). (f) A red apple was written on a thin film of the TN1 elastomer, and it faded quickly in 3 min on exposure of UV light.

the C-3 elastomer was much smaller, which almost could not be detected by the naked eye (see Movie S2 in Supporting Information). We plot the fluorescence intensity ratio I600/I420 versus λPrestretch to examine the influence of the prestretching level of the first network on the mechanochromic behavior of the elastomer. Because the fluorescence signal output is ratiometric, this self-referencing two-band comparative method possesses a great advantage over the other “turn-on” MCP materials because of minimizing or eliminating external interference. The onset of detectable activation was defined as λ′. As seen in Figure 4b, in comparison with DN1, TN1 film showed a smaller value of λ′ (1.24) and a greater activation under uniaxial deformation. Compared with most of the mechanochromic polymers reported, which required high levels of strain to activate, often at the level of 5.0,5,29 the detectable activation strain for TN1 was much smaller. However, in the case of the SN1 film, no activation in fluorescence was found until it was broken into two pieces. These results suggested that high-level prestretching of the first network was crucial for producing high sensitivities in mechanochromic elastomers. In the basic action of making a fist, the skin on the finger joints repeatedly stretches and contracts by as much as 55%,52 which already exceeds the activation strain of TN1 elastomer (24%). A dumbbell-shaped TN1 elastomer was tightly attached to the middle of the finger by scotch tape. As the finger bent, the fluorescence color in the middle of the elastomer sensor was changed from blue to red, while it turned to yellow with finger straightening (Figure 4c, d, and e). When constantly bending and straightening the finger, the fluorescence color of

the elastomer could change from red to yellow repeatedly. Therefore, it is promising to use the elastomer as a sensor for detection of changes related to stretching and contracting or as an inviting candidate for biomimic mechanochromic materials. In addition, the TN1 elastomer film coated onto the white surface can be used as a mechanochemical writing/drawing tablet. Figure 4f shows an example made from TN1. A red apple was easy to draw on this thin film. When the colored film was irradiated with UV light (UV-340, 100 W), the red color faded very quickly in 3 min without any trace. The fact that such colored patterns can be recorded by gently drawing the surface of film several times (Figure S18) indicates the easy writing and photoerasing nature of the TN1 film.

■

CONCLUSION In summary, we have designed and synthesized a novel rhodamine-based mechanochromic elastomer that displays interesting and unique features. First, three-primary-color fluorescence emission stimulated by stress has been successfully achieved with single-mechanophore-containing elastomer via combination of mechanochemistry and bond-bending-induced comformation change. Meanwhile, the fluorescence change in the three primary colors was fast and reversible, which made this MCP material suitable for a strain sensor or a drawing 11878


Research Article

ACS Applied Materials & Interfaces board. Moreover, it not only possessed very high fluorescence quantum yield but also showed very high sensitivity to stress because of its distinctive triple network structure. Because of these unique properties as well as the facile and convenient preparation of the mechanophore TAR, this elastomer will be a promising candidate for biomimic mechanochromic materials and other practical applications. The work may provide a new strategy for the development of broad-spectrum MCP materials.

■

11.6. ESI-MS (m/z): calculated for C32H40N3O5 (M + H+), 546.29; found (M + H+), 546.36. Synthesis of Cross-linker TAR. THR (1.5 g, 2.7 mmol) and triethylamine (3.4 mL, 24.7 mmol) were dissolved in 60 mL of CH2Cl2 in a 100 mL bottom flask. A solution of acryloyl chloride (0.87 mL, 10.7 mmol) in 10 mL of CH2Cl2 was added dropwise into the flask at 0 °C. The mixture was allowed to warm to room temperature and stirred overnight. The solution was washed with diluted hydrochloric acid. After being dried with Na2SO4, the organic solvent was evaporated. The pure product was obtained by chromatography on silica gel using CH2Cl2 as eluent. Yield is 96%. 1H NMR (300 MHz, CDCl3): δ 7.99 (m, 1H), 7.52 (m, 2H), 7.06 (m, 1H), 6.91 (s, 2H), 6.42 (s, 2H), 6.40−5.73 (m, 9H), 4.25 (t, 4H), 3.79 (t, 2H), 3.53 (t, 2H), 3.34 (t, 4H), 3.13 (q, 4H), 2.04 (s, 6H), 1.08 (t, 6H). 13C NMR (100 MHz, CDCl3): δ 168.7, 166.1, 165.5, 153.3, 151.0, 150.1, 130.8, 130.7, 130.1, 129.7, 128.3, 128.1, 123.9, 113.2, 109.9, 64.4, 62.2, 61.2, 53.4, 50.8, 47.9, 38.9, 17.8, 12.3. ESI-MS (m/z): calculated for C41H46N3O8 (M + H+), 708.32; found (M + H+), 708.38. Synthesis of Compound 2. Compound 1 (5.0 g, 10.9 mmol) and K2CO3 (12.1 g, 87.4 mmol) were added into 40 mL of ethylene chlorohydrin. The reaction mixture was heated at 100 °C for 6 h. Excessive ethylene chlorohydrin was evaporated under vacuum. The residue was poured into water and extracted with CH2Cl2 three times. After being dried with Na2SO4, the organic solvent was evaporated. The pure product was obtained by column chromatography on silica gel (CH2Cl2/ethanol = 50:1). Yield is 51%. 1H NMR (300 MHz, CDCl3): δ 7.95 (m, 1H), 7.49 (m, 2H), 7.04 (m, 1H), 6.91 (s, 1H), 6.47 (s, 1H), 6.33 (s, 1H), 6.29 (s, 1H), 4.08 (t, 1H), 3.61 (br, 2H), 3.45 (m, 2H), 3.25 (m, 6H), 3.03 (q, 2H), 2.09 (s, 3H), 1.93 (s, 3H), 1.33 (t, 3H), 1.03 (t, 3H). 13C NMR (100 MHz, CDCl3): δ 170.1, 153.5, 151.5, 150.6, 150.4, 147.7, 132.9, 130.2, 129.9, 129.5, 128.4, 128.0, 123.8, 123.1, 118.5, 113.7, 110.4, 104.8, 96.5, 65.7, 62.3, 59.1, 53.8, 48.3, 44.7, 38.3, 17.9, 16.7, 14.7, 11.7. ESI-MS (m/z): calculated for C30H36N3O4 (M + H+), 502.26; found (M + H+), 502.36. Synthesis of Control Cross-linker 3. Acrylic acid (0.41 mL, 6.0 mmol), 2-chloro-1-methylpyridinium iodide (1.52 g, 6.0 mmol), and triethylamine (1.66 mL, 12.0 mmol) were dissolved into 50 mL of CH2Cl2 and stirred at room temperature for 30 min. Compound 2 (1.0 g, 2.0 mmol) was then added into the solution and stirred overnight. The solution was washed with diluted hydrochloric acid three times. After being dried with Na2SO4, the organic solvent was evaporated. The pure product was obtained by column chromatography on silica gel (CH2Cl2/ethanol = 50:1). Yield is 60%. 1H NMR (300 MHz, CDCl3): δ 7.96 (m, 1H), 7.46 (m, 2H), 7.03 (m, 1H), 6.89 (s, 1H), 6.20−6.45 (m, 5H), 5.70−6.15 (m, 4H), 4.22 (t, 2H), 3.78 (t, 2H), 3.48 (t, 2H), 3.30 (t, 2H), 3.22 (q, 2H), 3.08 (q, 2H), 2.01 (s, 3H), 1.89 (s, 3H), 1.32 (t, 3H), 1.04 (t, 3H). 13C NMR (100 MHz, CDCl3): δ 168.6, 166.0, 165.5, 153.6, 151.4, 150.8, 150.3, 147.6, 132.7, 130.8, 130.5, 130.4, 129.8, 129.6, 128.4, 128.3, 128.2, 123.8, 123.0, 118.3, 113.6, 109.9, 105.3, 96.5, 64.7, 62.2, 61.4, 53.4, 50.8, 47.9, 38.8, 38.3, 17.7, 16.7, 14.7, 12.3. ESI-MS (m/z): calculated for C36H40N3O6 (M + H+), 610.28; found (M + H+), 610.32. Synthesis of Compound 4. Rhodamine 6G (10 g, 20.8 mmol) was dissolved in 160 mL of acetonitrile. To this deep red solution was added n-butylamine (6.1 mL, 62.6 mmol). The reaction mixture lost color gradually. After being refluxed for 2 h, the reaction mixture was cooled to room temperature and poured into 150 mL of methanol. The solid was filtered and washed thoroughly with water and dried under vacuum to give 8.8 g of off-white product. Yield is 90%. 1H NMR (300 MHz, CDCl3): δ 7.92 (m, 1H), 7.42 (m, 2H), 7.01 (m, 1H), 6.35 (s, 2H), 6.24 (s, 2H), 3.50 (br, 2H), 3.23 (t, 4H), 3.11 (t, 2H), 1.90 (s, 6H), 1.32 (t, 6H), 1.08 (br, 4H), 0.67 (t, 3H). 13C NMR (100 MHz, CDCl3): δ 168.1, 153.8, 151.7, 147.3, 132.2, 131.4, 128.7, 127.9, 123.7, 122.7, 117.8, 106.5, 96.5, 64.9, 40.1, 38.4, 30.3, 20.3, 16.7, 14.8, 13.6. ESI-MS (m/z): calculated for C30H36N3O2 (M + H+), 470.27; found (M + H+), 470.40. Synthesis of Compound 5. Compound 4 (5.0 g, 10.6 mmol) and K2CO3 (11.8 g, 85.2 mmol) were added into 40 mL of ethylene chlorohydrin. The reaction mixture was heated at 100 °C for 24 h. Excessive ethylene chlorohydrin was evaporated under vacuum. The

EXPERIMENTAL SECTION

Materials. Ethyl acrylate (EA) and butanediol diacrylate (BDA) were purchased and filtrated through a column of basic alumina to remove the inhibitor. All other reagents were purchased and used without further purification. The UV-polymerizations were initiated by 2-hydroxyethyl-2-methylpropiophenone (HMP) under a UV lamp (UV-340, 100 W) with wavelength ranging from 295 to 365 nm. Measurements. 1H NMR was recorded on 300 MHz (Bruker ARX300), and 13C NMR spectra were recorded on Bruker 100 MHz spectrometer at room temperature with CDCl3 as the solvent and tetramethylsilane (TMS) as the internal standard. Electrospray ionization (ESI) mass spectra were obtained on a Finnigan LCQ Advantage ion trap mass spectrometer (ThermoFisher Corporation). Differential scanning calorimetry (DSC) measurements were carried out on TA Instruments DSC Q2000. Fluorescence spectra measurements were performed on a Shimadzu RF-5301PC spectrofluorophotometer. Absorption spectra were determined on a Pgeneral UV−vis TU-1901 spectrophotometer. The DMA measurements were conducted in the three points mode with DMA Q800. Uniaxial tensile tests were performed on TMA Q400. Specimens were cut in a dumbbell shape using a normalized cutter with a central part of 15 mm in length and 2 mm in width. The photoluminescence quantum yields of the rhodamine derivatives in an ethanol solution were determined relative to a quinine sulfate solution in a 1 N H2SO4 at room temperature, assuming a quantum yield of 0.546 when excited at 365 nm. All density functional theory (DFT) calculations were performed using the ORCA version 2.8 program package. 53 Geometry optimizations were performed using B3LYP density functional, while the final energy calculations were also performed with the B3LYP density functional. Synthesis of Compound 1. Rhodamine 6G (10 g, 20.8 mmol) was dissolved in 160 mL of acetonitrile. To this deep red solution was added monoethanolamine (3.7 mL, 62.6 mmol). The reaction mixture became gradually heterogeneous and lost color. After being refluxed for 2 h, the reaction mixture was cooled to room temperature. The solid was filtered and washed thoroughly with water and dried under vacuum to give 9.1 g of off-white product. Yield is 95%. 1H NMR (300 MHz, CDCl3): δ 7.94 (m, 1H), 7.47 (m, 2H), 7.05 (m, 1H), 6.35 (s, 2H), 6.28 (s, 2H), 4.20 (br, 1H), 3.54 (br, 2H), 3.45 (t, 2H), 3.26 (m, 6H), 1.92 (s, 6H), 1.35 (t, 6H). 13C NMR (100 MHz, CDCl3): δ 170.1, 153.9, 151.7, 147.5, 132.8, 130.4, 128.2, 128.0, 123.8, 122.9, 118.1, 105.2, 96.6, 66.0, 62.6, 44.7, 38.4, 16.7, 14.7. ESI-MS (m/z): calculated for C28H32O3N3 (M + H+), 458.24; found (M + H+), 458.32. Synthesis of THR. Compound 1 (4 g, 8.7 mmol) and K2CO3 (9.6 g, 69.8 mmol) were added into 40 mL of ethylene chlorohydrin. The reaction mixture was heated at 100 °C for 24 h. Excessive ethylene chlorohydrin was evaporated under vacuum. The residue was poured into water and extracted with CH2Cl2 three times. After being dried with Na2SO4, the organic solvent was evaporated. The pure product was obtained by column chromatography on silica gel (CH2Cl2/ ethanol = 35:1) to give 70% THR. 1H NMR (300 MHz, CDCl3): δ 7.99 (m, 1H), 7.54 (m, 2H), 7.07 (m, 1H), 6.92 (s, 2H), 6.50 (s, 2H), 3.95 (br, 1H), 3.65 (t, 4H), 3.46 (t, 2H), 3.31 (t, 2H), 3.24 (t, 4H), 3.05 (q, 4H), 2.25 (br, 2H), 2.11 (s, 6H), 1.06 (t, 6H). 13C NMR (100 MHz, CDCl3): δ 170.2, 153.1, 150.8, 150.2, 133.1, 130.3, 130.0, 129.5, 128.7, 123.8, 123.3, 113.4, 110.4, 65.4, 62.1, 59.1, 53.7, 48.2, 44.9, 17.9, 11879


Research Article

ACS Applied Materials & Interfaces residue was poured into water and extracted with CH2Cl2 three times. After being dried with Na2SO4, the organic solvent was evaporated. The pure product was obtained by column chromatography on silica gel (CH2Cl2/ethanol = 70:1). Yield is 65%. 1H NMR (300 MHz, CDCl3): δ 7.96 (m, 1H), 7.50 (m, 2H), 7.05(m, 1H), 6.91 (s, 2H), 6.43 (s, 2H), 3.61 (br, 4H), 3.21 (t, 4H), 3.11 (t, 2H), 3.03 (q, 4H), 2.08 (s, 6H), 1.04 (m, 10H), 0.63 (t, 3H). 13C NMR (100 MHz, CDCl3): δ 168.1, 152.7, 150.5, 150.2, 132.5, 131.3, 130.2, 130.0, 128.5, 123.7, 123.0, 114.5, 110.2, 64.4, 59.1, 53.9, 48.3, 40.3, 30.4, 20.2, 17.8, 13.5, 11.6. ESI-MS (m/z): calculated for C34H44N3O4 (M + H+), 558.33; found (M + H+), 558.36. Synthesis of Control Cross-linker 6. Compound 5 (1.0 g, 1.8 mmol) and triethylamine (1.5 mL, 10.8 mmol) were dissolved in 50 mL of CH2Cl2 in a 100 mL bottom flask. A solution of acryloyl chloride (0.44 mL, 5.4 mmol) in 5 mL of CH2Cl2 was added dropwise into the flask at 0 °C. The mixture was allowed to warm to room temperature and stirred overnight. The solution was washed with diluted hydrochloric acid. After being dried with Na2SO4, the organic solvent was evaporated. The pure product was obtained by chromatography on silica gel using CH2Cl2 as eluent. Yield is 97%. 1 H NMR (300 MHz, CDCl3): δ 7.95 (m, 1H), 7.48 (m, 2H), 7.05 (m, 1H), 6.90 (s, 2H), 6.40 (s, 2H), 6.32 (d, 2H), 6.08 (q, 2H), 5.81(d, 2H), 4.21 (t, 4H), 3.30 (t, 4H), 3.09 (m, 6H), 2.04 (s, 6H), 1.03 (m, 10H), 0.62 (t, 3H). 13C NMR (100 MHz, CDCl3): δ 168.1, 166.1, 152.9, 150.7, 150.2, 132.4, 131.3, 130.8, 130.0, 128.3, 123.8, 123.0, 113.9, 109.8, 64.5, 62.1, 58.4, 50.9, 48.0, 40.3, 30.4, 20.2, 18.4, 17.6, 13.5, 12.2. ESI-MS (m/z): calculated for C40H48N3O6 (M + H+), 666.35; found (M + H+), 666.38. Preparation of the First Networks SNx. First networks were prepared by UV polymerization of a solution of EA (1 equiv), TAR cross-linker (1−3% mol), and photoinitiator HMP (1% mol). A 50% solution of the reactants in toluene was poured in a 0.36 mm thick glass mold and exposed to UV light. The polymerization was then left to proceed for 1 h. Any unreacted species were extracted with a mixture of toluene and petroleum ether for 2 days. Swollen first networks were then dried under vacuum at 80 °C. At this point we quantified the fraction of unreacted species to be