Effect of Mechanical Stress on Spiropyran-Merocyanine Reaction Kinetics. in a Thermoplastic Polymer

Transcription

1 Supporting Information for: Effect of Mechanical Stress on Spiropyran-Merocyanine Reaction Kinetics in a Thermoplastic Polymer Tae Ann Kim,, Brett A. Beiermann,, Scott R. White,, Nancy R. Sottos *,, Department of Materials Science and Engineering, and Aerospace Engineering, University of Illinois at Urbana Champaign, Urbana, Illinois 61801, United States Beckman Institute of Advanced Science and Technology, University of Illinois at Urbana Champaign, Urbana, Illinois 61801, United States 1. General experimental details Synthesis of spiropyran-linked polyurethane mg of dihyrdoxy spiropyran (SP) and 36 mg of 1,4- diazabicyclo[2.2.2]octane (DABCO: 99%, Sigma-Aldrich) were dissolved in 30 ml of tetrahydrofuran (THF: anhydrous, 99.8%, Sigma-Aldrich) mg of 4,4 -methylenebis(phenylisocyanate) (MDI: 98%, Sigma-Aldrich) dissolved in 10 ml of THF was added to the solution and stirred at 60 o C for 1 h. The reaction mixture was filtered through a 0.45 µm syringe filter and added into a solution of 9.27 g of poly(ethylene glycol) (PEG, MW=200, Sigma-Aldrich) and 4.5 mg of DABCO in 4.5 ml of THF. Excess THF was removed under vacuum by slowly increasing temperature from RT to 70 o C to avoid bumping. After complete removal of the solvent, 7.23 ml of hexamethylenediisocyante (HDI, 98%, Al-

2 drich) was added into the mixture as a chain extender under N2, poured into a Delrin mold, and degassed. The mold was kept in an N2-purged oven at 70 oc for 2 days. The final dog-bone shaped tensile specimens (Figure S1) were detached from the mold and kept under ambient conditions for 24 h before testing. Characterization. The molecular weight and polydispersity of SP-PU were measured by gel permeation chromatography (GPC, Waters) with a refracted index detector (Waters 2414) at a flow rate of 1 ml/min. THF was used as an eluent, and the calibration of molecular weights was done by polystyrene standards. Figure S1. Pristine dog-bone shaped tensile specimens after removal from the mold (left). Observed color change due to keeping the specimens in the dark (right). 2

3 2. Fluorescence measurement Combined mechanical and optical testing setup. Fluorescence images of specimens were acquired during uniaxial tensile loading. Tensile force was applied to the samples using a horizontally oriented uniaxial load frame (IMAC Motion Control Corp.), with both actuators translating simultaneously in opposite directions so that the center of a tensile sample remained in the field of view. The pulling speed was set to 0.1 s -1 until the desired deformation ratio (stretch) was reached. Applied force was measured by 50-lb capacity load cell (Honeywell Sensotec). The corresponding true stress was calculated from in situ optical measurement of the dimensions of the deformed samples. 1 A 532-nm laser diode was used for the excitation light source, with power set to 2.5 mw using polarization optics. The light was circularly polarized prior to incidence on the sample. During the measurement, exposure was constant for all samples. Fluorescence images were recorded on a color camera (AVT Stingray F504c) and all measurement conditions were controlled by NI LabVIEW. Image analysis for fluorescence. Captured fluorescence images were analyzed by an image analysis program (ImageJ). Fluorescence intensity was calculated by the red channel intensity of the color image, averaged over pixels in the gauge section of the sample. To normalize fluorescence intensity, the background intensity was first subtracted from the obtained value. Then, each value was divided by the maximum intensity value. At least 3 samples were measured for the experiments and the rate constants were averaged. 3

4 Figure S2. Engineering stress (The load divided by the initial cross-sectional area of the specimen) and thickness-corrected fluorescence intensity as a function of deformation ratio. Figure S3. Applied deformation ratio (λ), corresponding load and raw fluorescence intensity as a function of time. t 0 is defined when the deformation ratio reaches a constant value. 4

5 Fluorescence changes under dark conditions or during green (532 nm) laser exposure. To investigate the deactivation rate of MC under dark conditions, we measured the fluorescence intensity of a fully activated specimen by UV light (black dashed line in Fig. S4). The normalized fluorescence intensity did not change significantly after more than 9 hours, indicating that the transition from MC to SP in the dark is insignificant over the time scale of our experiments. Also, we measured the activation behavior of a fully bleached specimen under green laser (532nm) exposure (red dashed line in Fig. S4). No noticeable fluorescence increase was observed, indicating the activation of SP is significantly suppressed by the exposure of green light. Figure S4. Fluorescence intensity changes of activated samples under dark conditions (black square) and fully bleached specimen exposed to a green (532 nm) laser (red circle). The fluorescence imaging was started at t=0s. 5

6 Residual fluorescence intensity decay of specimens held at fixed deformation ratios after exposure to green (532 nm) laser light. Due to the thickness change at different deformation ratios (stretch), the fluorescence intensity was normalized by the maximum value for each experiment. Also, the background fluorescence intensity was not subtracted to reveal the change in the magnitude of fluorescence intensity at different deformation ratios (stretch). Figure S5. Normalized fluorescence intensity changes for different deformation ratios during continuous exposure to green (532 nm) laser light. Here, t 0 defines the time when the deformation ratio (λ) reaches a constant value. 6

7 3. Stress relaxation of SP-PU Specimens were loaded under displacement control to a constant deformation ratio (stretch). The corresponding true stress was calculated from in situ optical measurement of the dimensions of the deformed samples. 1 The true stress relaxed over the first 25 min of the experiment and the relaxation response was fitted to a generalized Maxwell model (OriginPro (32-bit) b215). Three Maxwell elements (Table S1) were sufficient to fit the stress data with a coefficient of determination of Figure S6. Stress relaxation behavior of SP-PU under different deformation ratios (λ). Here, t 0 denotes the time when the λ reaches a desired value. 7

8 Table S1. Summary of the generalized Maxwell model parameters for SP-PU = σ n exp t σ t n τ n +σ 0 <Generalized Maxwell models> σ 0 σ 1 σ 2 τ 2 τ 3 τ 3 τ (MPa) (MPa) 1 (min) (MPa) (min) (MPa) (min) λ= λ= λ= λ= λ= λ= λ=

9 4. Calculation of rate constants and activation energies for both reactions SP k f MC (1) k r Rate constants. The total concentration of SP in a fully bleached state, [SP] 0, or total concentration of MC in a fully activated state, [MC] 0 is the sum of the current concentration of spiropyran, [SP(t)], and merocyanine, [MC(t)]. [ SP] 0 = [ MC] 0 = SP( t) + MC t (2) Under dark conditions, the reverse reaction is negligible and we only consider the forward reaction. For the forward reaction, the differential form of the rate law is, d MC t dt = k f SP( t) = k f SP ([ ] 0 MC( t) ) (3) After rearranging and integrating the equation from (0 to t, and 0 to [MC(t)]), we calculate the forward rate constant from, MC t = SP ( ) [ ] 0 1 exp k f t (4) During exposure to green (532 nm) laser light, the transition from SP to MC is insignificant. Hence, we assume the following form of the differential form of the rate law, d!" SP t # $ dt = k r MC( t)!" # $ = d SP ([ ] 0!" MC( t) # $ ) dt (5) After rearranging and integrating the equation from (0 to t, and [MC] 0 to [MC(t)]), we find the final formula of the reverse rate constant:!" MC( t) # $ = [ MC] 0 exp( k r t) (6) 9

10 Assuming that fluorescence intensity (I fl ) divided by the maximum fluorescence intensity (I fl,max ) is proportional to the relative concentration of MC, the concentration terms in Eqs. (2)-(6), are replaced by the fluorescence intensity.!" MC t # $ = MC t!" SP t # $0 MC!" # $ = I fl ( t) [ ] 0 = I fl ( t) [ ] 0!" MC t # $ MC ( ) (7) I fl,max = 1 exp k f t I fl,max = exp k r t (8) After normalizing the fluorescence intensity (PL) for each stress value, the data are fitted to Eqs. (7) and (8) using OriginPro (32-bit) b215. Figure S7. Representative change in normalized fluorescence intensity (SP to MC) under different average stress values (black square) and corresponding fit to Eq. (7) (red line). Here, t 0 corresponds to the time when the fluorescence imaging was started for the forward reaction kinetics. 10

11 Figure S8. Representative change in normalized fluorescence(mc to SP) under different stress values and corresponding fit to Eq. (8) (red line). Here, t 0 corresponds to the time when the fluorescence imaging was started for the reverse reaction kinetics. Activation energies. Both forward activation energy (ΔE a,f ) and reverse activation energy (ΔE a,r ) are assumed to be a linear combination of the inherent thermal activation energy, ΔE 0, the effect of the applied mechanical load, ΔE mech, and a photochemical term, ΔE λ,f : ΔE a, f = ΔE 0, f + ΔE λ, f + ΔE mech, f (9) ΔE a,r = ΔE 0,r + ΔE λ,r + ΔE mech,r (10) According to the Eyring-Polanyi equation, a reaction constant can be expressed as a function of activation energy: k f k T b h exp ΔE a, f = k T b k b T h exp ΔE 0, f + ΔE λ, f + ΔE mech, f (11) k b T 11

12 k r k bt h exp ΔE a,r = k bt k b T h exp ΔE 0,r + ΔE λ,r + ΔE mech,r (12) k b T The dependence of activation energy on an applied stress is assumed linear. By rearranging the equation and using the Taylor series approximation of exponential term, we arrive at Equation (5) in the main article. k f = k T b h exp ΔE 0, f + ΔE λ, f exp ΔE mech, f = k σ =0, f exp ΔE mech, f k σ =0, f 1 ΔE mech, f = k σ =0, f 1 σ v * f k b T k b T (13) k r = k bt h exp ΔE 0,r + ΔE λ,r exp ΔE mech,r = k σ =0,r exp ΔE mech,r k σ =0,r 1 ΔE mech,r = k σ =0, f 1 σ v * r (14) k b T All the measurements were done at room temperature. k σ=0,f and k σ=0,r were measured as s -1, s -1, respectively. Using these values, ΔE a,f and ΔE a,r were calculated at each stress level. Activation volumes were also derived from the obtained activation energy values and averaged. Reference (1) Beiermann, B. A.; Kramer, S. L. B.; May, P. A.; Moore, J. S.; White, S. R.; Sottos, N. R. Adv. Funct. Mater. 2014, 24 (11),