(PEG) Hydrogel - American Chemical Society

Feb 9, 2017 - School of Mechanical and Aerospace Engineering, Nanyang ... Department of Diagnostic Radiology, Singapore General Hospital, Singapore ...
1 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/accounts

Water-Responsive Shape Recovery Induced Buckling in Biodegradable Photo-Cross-Linked Poly(ethylene glycol) (PEG) Hydrogel Published as part of the Accounts of Chemical Research special issue “Stimuli-Responsive Hydrogels”. Abhijit Vijay Salvekar,† Wei Min Huang,*,† Rui Xiao,*,‡ Yee Shan Wong,§ Subbu S. Venkatraman,§ Kiang Hiong Tay,∥ and Ze Xiang Shen⊥ †

School of Mechanical and Aerospace Engineering, Nanyang Technological University, 50 Nanyang Avenue, 639798 Singapore, Singapore ‡ Institute of Soft Matter Mechanics, College of Mechanics and Materials, Hohai University, Nanjing, Jiangsu 210098, PR China § School of Materials Science and Engineering, Nanyang Technological University, 50 Nanyang Avenue, 639798 Singapore, Singapore ∥ Department of Diagnostic Radiology, Singapore General Hospital, Singapore 169608, Singapore ⊥ School of Physical and Mathematical Sciences, Nanyang Technological University, 50 Nanyang Avenue, 639798 Singapore, Singapore

CONSPECTUS: The phenomenon of recovering the permanent shape from a severely deformed temporary shape, but only in the presence of the right stimulus, is known as the shape memory effect (SME). Materials with such an interesting effect are known as shape memory materials (SMMs). Typical stimuli to trigger shape recovery include temperature (heating or cooling), chemical (including water/moisture and pH value), and light. As a SMM is able not only to maintain the temporary shape but also to respond to the right stimulus when it is applied, via shape-shifting, a seamless integration of sensing and actuation functions is achieved within one single piece of material. Hydrogels are defined by their ability to absorb a large amount of water (from 10−20% up to thousands of times their dry weight), which results in significant swelling. On the other hand, dry hydrogels indeed belong to polymers, so they exhibit heatand chemoresponsive SMEs as most polymers do. While heat-responsive SMEs have been spotted in a handful of wet hydrogels, so far, most dry hydrogels evince the heat and water (moisture)-responsive SMEs. Since water is one of the major components in living biological systems, water-responsive SMMs hold great potential for various implantable applications, including wound healing, intravascular devices, soft tissue reconstruction, and controlled drug delivery. This provides motivation to combine water-activated SMEs and swelling in hydrogels together to enhance the performance. In many applications, such as vascular occlusion via minimally invasive surgery for liver cancer treatment, the operation time (for both start and finish) is required to be well controlled. Due to the gradual and slow manner of water absorption for wateractivated SMEs and swelling in hydrogels, even a combination of both effects encounters many difficulties to meet the timerequirements in real procedures of vascular occlusion. Recently, we have reported a bioabsorbable radiopaque water-responsive shape memory embolization plug for temporary vascular occlusion. The plug consists of a composite with a poly(DL-lactide-co-glycolide) (PLGA) core (loaded with radiopaque filler) and cross-linked poly(ethylene glycol) (PEG) hydrogel outer layer. The device can be activated by body fluid (or water) continued...

Received: October 28, 2016 Published: February 9, 2017 © 2017 American Chemical Society

141

DOI: 10.1021/acs.accounts.6b00539 Acc. Chem. Res. 2017, 50, 141−150

Article

Accounts of Chemical Research

after about 2 min of immersion in water. The whole occlusion process is completed within a few dozens of seconds. The underlying mechanism is water-responsive shape recovery induced buckling, which occurs in an expeditious manner within a short time period and does not require complete hydration of the whole hydrogel. In this paper, we experimentally and analytically investigate the water-activated shape recovery induced buckling in this biodegradable PEG hydrogel to understand the fundamentals in precisely controlling the buckling time. The molecular mechanism responsible for the water-induced SME in PEG hydrogel is also elucidated. The original diameter and amount of prestretching are identified as two influential parameters to tailor the buckling time between 1 and 4 min as confirmed by both experiments and simulation. The phenomenon reported here, chemically induced buckling via a combination of the SME and swelling, is generic, and the study reported here should be applicable to other water- and non-water-responsive gels.

■

INTRODUCTION Hydrogels are defined by their ability to absorb a large amount of water or moisture without dissolving, and the absorbed water can hardly be removed even under pressure.1,2 Various hydrogels have been intensively investigated in recent years due to their great potential in a range of applications, in particular those stimuli-responsive hydrogels for biomedical applications.3−6 Among others, typical stimuli for the activation of wet hydrogels include temperature, pH value, and light.3,7−9 Moreover, the large amount absorbed water results in swelling, that is, significant volume expansion. Such a large volume change in the process of wetting and drying may result in remarkable change in surface morphology or overall configuration. As shown in Figure 1, during swelling upon wetting in water, wrinkles appear on the surface of a piece of poly(ethylene glycol) (PEG) hydrogel. The surface becomes smooth again only after a prolonged period of wetting, when the hydrogel almost reaches

its maximum swelling. Upon keeping the piece of wrinkled hydrogel in air for a while, the water distribution becomes uniform within the hydrogel eliminating the wrinkles.10−12 In Figure 2, a long piece of same type of PEG as used in Figure 1 is dried with a string (core filament) embedded inside. Depending on the initial position of the string, different shapes or configurations, for example, S shape, coil, and straight line, result. Orientation and conformational changes of molecular chains upon cooling (crystallization) and heating (melting) are novel ways utilized in polymers to reversibly switch between two shapes.13,14 Likewise swelling of a hydrogel or the difference in swelling ratio of a hydrogel composite provides an easy yet effective approach for water induced reversible shape switching and folding/unfolding.15−17 The same approach is also applicable to other gels activated by non-water solvents. However, the reversible nature implies that the associated shape switching is always between two shapes, one

Figure 1. Swelling of PEG hydrogel in 37 °C water: (a) side view; (b) cross-sectional view. Yellow sphere (6 mm diameter) for benchmark.

Figure 2. Typical shapes of hydrogel composites after drying. (a) S shape (cross-placement of core filament); (b) coil (eccentric/tangential placement of core filament); (c) straight line (coaxial concentric placement of core filament). Top, illustration of the initial position of the core filament; bottom, configuration after drying. 142

DOI: 10.1021/acs.accounts.6b00539 Acc. Chem. Res. 2017, 50, 141−150

Article

Accounts of Chemical Research corresponding to the dry state and the other corresponding to the wet state. The shape memory effect (SME) refers to the capability of a piece of severely deformed material to recover its original shape but only in the presence of the right stimulus, such as heat,18−21 chemical (including water/moisture22,23), and light.24,25 Such an effect to maintain the temporary shape is highly in demand in many engineering applications, including a range of biomedical devices.18,19,26−31 In technical terms, the process to fix the temporary shape is called programming, while the process to return to the permanent shape is called shape recovery. A material characterized with the SME may be called a shape memory material. Alternatively, the phenomenon similar to above-mentioned reversible shape switching between two shapes in hydrogels, upon wetting and drying is termed the shape change effect (SCE).32,33 Since a typical early work published in ref 34, only a handful research work on wet hydrogels with the SME have been reported (e.g., refs 35 and 36). While dried hydrogels may be considered as normal polymers and most polymers are intrinsically heat- and chemoresponsive shape memory materials (SMMs),17 a hydrogel may be tailored to have the SME or SCE via controlling the amount of water content in it.10,11 A combination of water-induced SME (shape recovery) and SCE (swelling) in programmed dry hydrogel could essentially overcome the limitations (e.g., limited reconfigurability) with either of them applied, hence facilitating the reshaping of design in various ways.37,38 For example, a piece of helical spring made of hydrogel may be dried and then straightened at high temperatures when it is soft. Upon wetting in water, the hydrogel recovers its original helical shape due to swelling.10 The dimension change in cross-section from a piece of straight dry hydrogel wire to fully swollen helical spring is achieved by water-induced SME and SCE. Such a device is required for applications such as vascular occlusion via minimally invasive surgery for liver cancer treatment.39 However, even for a small piece of hydrogel to achieve full swelling (refer to Figure 1), it may take longer than the time practically allowed in the medical procedure of vascular occlusion (normally around 2 min). Also the whole swelling process progresses in a gradual manner. Previously we have reported a bioabsorbable radiopaque water-responsive shape memory embolization plug for temporary vascular occlusion.40 The plug consists of a composite with a poly(DL-lactide-co-glycolide) (PLGA) core (loaded with radiopaque filler) and cross-linked PEG hydrogel outer layer. The device can be activated by body fluid (or water) in about 2 min of water immersion. The whole occlusion process is completed within a few dozens of seconds. The underlying mechanism is water-induced buckling, which occurs within a short time period and does not require complete wetting over the whole hydrogel. In this paper, we experimentally and analytically investigate the water-activated shape recovery induced buckling in this biodegradable PEG hydrogel to understand the fundamentals in precisely controlling the buckling time.

All chemicals were used as received. The deionized (DI) water used here was purified with a Milli-Q system (Millipore, USA). Standard photo-cross-linking method as reported in refs 41 and 42 was used to produce PEG hydrogel from macromolecular poly(ethylene glycol) diacrylate (PEGDA) solution in DI water (Mn 10 kDa). PEGDA powder (7.5 w/v %) was dissolved in DI water at room temperature (about 23 °C). Free radical photoinitiator, Irgacure-2959, dissolved in 70% ethanol was added to the PEGDA solution in 2 wt % relative to PEGDA powder. The solution mixture was added into cylindrical transparent plastic tubes and then cross-linked for 7 min under ultraviolet light (λ = 365 nm; power = 12 W) (Vilber Lourmat, France) at a 1 cm distance. Different sizes of tubes were used to fabricate PEG hydrogel samples of different diameters. The cross-linked hydrogel samples in a cylindrical shape were then dried in 37 °C incubator for 48 h. Herein, such samples are considered as dry samples in this study. Table 1 presents the diameters of the fabricated PEG samples in both wet and dry states. Table 1. Diameters (in mm) of Fabricated PEG Samples wet state

dry state

3.85 4.45 5.95 7.20

1.50 1.80 2.45 2.75

Differential scanning calorimetry (DSC) test (TA Instruments DSC Q10, USA) was carried out on the dry PEG sample (∼10 mg) for a thermal cycle between 0 and 90 °C at a heating/ cooling rate of 10 °C/min. From the resultant DSC data as shown in Figure 3, the melting transition temperature (Tm) of the dry PEG was identified as about 60 °C.

Figure 3. DSC curve of dry PEG hydrogel.

Gel fraction, which indicates the degree of cross-linking, was measured as 86.3%. Swelling ratio of dry PEG as a function of immersion time in 37 °C DI water is plotted in Figure 4. Refer to ref 40 for the details of these two experiments and the formulas used for calculation of gel fraction and swelling ratio. Dry PEG samples were heated in an oven at 70 °C (above its Tm) for 20 min and then stretched to the required programming strains. The applied constraint was removed only after the stretched samples were cooled back to the room temperature.

■

MATERIALS, SAMPLE PREPARATION, AND EXPERIMENTAL PROCEDURES Poly(ethylene glycol) diacrylate (PEGDA) (Mn = 10 000) and Irgacure 2959 (2-hydroxy-1-[4-(hydroxyethoxy)phenyl]-2-methyl-1-propanone) were purchased from Sigma-Aldrich, Singapore. 143

DOI: 10.1021/acs.accounts.6b00539 Acc. Chem. Res. 2017, 50, 141−150

Article

Accounts of Chemical Research

Figure 4. Swelling ratio of PEG hydrogel as a function of immersion time in 37 °C DI water.

A MTS Criterion C42 machine (MTS Systems, USA) with a 50 N load cell was used to investigate the mechanical properties of dry samples with gage length of 30 mm at 25 °C. Uniaxial tensile tests were carried out at a strain rate of 5.56 × 10−3/s. Typical stress vs strain results of dry stretched PEG samples with different programming strains are presented in Figure 5.

Figure 6. Cyclic tensile test of dry PEG hydrogel at 70 °C.

molecular arrangement or slippage of chains may cause permanent deformation in the observed residual strain.43 Samples with different original diameters were hydrated in 37 °C DI water for 15 min. After the immersion, all samples virtually became fully transparent and solvated (refer to Figure 1). These samples without any programming strains were used for a series of uniaxial tensile tests on MTS Criterion C42 machine with the same testing parameters as mentioned above. The dimensions of the samples after solvation were used in the calculation of stress and strain. Figure 7 presents typical stress vs strain relationships

Figure 5. Effects of programming strain on the stress vs strain relationship of dry PEG hydrogel.

Herein, the stress and strain used in the course of this study are meant for engineering stress and engineering strain. As we can see, although the yield stress is increased significantly with the increase of programming strain, the stress vs strain curves in the early loading stage are about the same. The Young’s modulus of the dry stretched PEG can be obtained by measuring the slope in the early loading part of the stress vs strain curve. Cyclic tensile tests were carried using MTS Criterion C43 machine (MTS Systems, USA) with a 500 N load cell and a heating chamber. A dry hydrogel sample with 2.45 mm diameter and 10.5 mm gage length was fixed between the two jaws inside the heating chamber set at 70 °C. Cyclic tensile test for three different strains points (300%, 500%, and 700%, in ascending order) was then carried out at a strain rate of 0.079/s. The corresponding stress vs strain curves reported in Figure 6 depict the viscoelastic behavior and hysteresis of PEG hydrogel above its Tm with higher recovery strains. Redistribution of

Figure 7. Typical stress vs strain relationships in uniaxial tension to fracture after wetting the PEG hydrogels in 37 °C DI water for 15 min for samples of different original diameters.

until fracture. As we can see, for a sample with a larger original diameter, the corresponding slope in the earlier region of stress vs strain curve is higher, which indicates less solvation or far from equilibrium swelling in the larger sized sample upon immersing in water for 15 min. Another important feature observed here is limited capability in stretching in all samples. Hence, an alternative programming approach, that is, drying of stretched wet hydrogel as reported in ref 11 is not applicable to this hydrogel to achieve a high amount of stretching. 144

DOI: 10.1021/acs.accounts.6b00539 Acc. Chem. Res. 2017, 50, 141−150

Article

Accounts of Chemical Research In order to find the recovery force or stress generated in preprogrammed samples upon immersing in 37 °C water, isostrain tests were carried out on the MTS Criterion C42 machine with MTS Bionix EnviroBath setup. The testing process is as follows: the prestretched samples with 30 mm gage length were fixed between two grippers inside the EnviroBath, and then 37 °C DI water was circulated; recovery force induced upon wetting the sample in water was recorded. Figure 8a shows the typical recovery force vs time curves of original 2.45 mm diameter PEG samples with different programming strains (gage length 30 mm).

the result of water induced shape recovery.10,11,22 Upon continued immersion, a sudden upsurge in recovery stress (indicated by the arrow) in all samples is observed. Given the instant nature of such upsurge in recovery stage, the only possible cause is buckling as previously reported.44,45 However, after the maximum stress is reached, the swelling effect becomes relatively more influential, so that the observed stress starts to gradually decrease. While the final stress (within the time frame of our experiments) increases with the increase in programming strain, a higher programming strain induces earlier buckling. On the other hand, if the applied programming strain is constant, while the original sample diameter is a variable, a similar trend is still observed in the stress vs time relationship as shown in Figure 9. However, the stress increases faster in the sample

Figure 9. Evolution of recovery stress upon wetting in 500% prestretched PEG samples with 30 mm gage length and different original diameters.

with a larger original diameter, while the upsurge in stress occurs earlier in the sample with a smaller original diameter. In a similar experimental setup with sample fixed in a top grip and free at the bottom end, upon wetting the preprogrammed samples in 37 °C DI water, shape recovery or change was recorded by a video camera. Figure 10 depicts the snapshots of one typical test,

Figure 8. Evolution of recovery force (a) and stress (b) upon wetting in original 2.45 mm diameter PEG samples with different programming strains.

Instead of plotting recovery force against time, we may plot the corresponding engineering stress as a function of immersion time to reveal the fundamentals in a dimensionless manner. As shown in Figure 8b, the evolution of recovery stress in all samples with the same original dry diameter of 2.45 mm follows more or less the same increasing path in the early wetting stage. According to Figure 4, the gel should swell continuously upon wetting in water, so that the effect of swelling should result in compressive stress within the time frame of our experiments. Hence, the observed gradual increase in tensile stress should be

Figure 10. Sequence of water-responsive shape recovery induced buckling and swelling in 400% prestretched dry PEG wire (original diameter 2.45 mm).

for the 30 mm free length of sample with original diameter 2.45 mm and programming strain of 400%. The same tests were performed on samples with different programming strain and original diameter, and the corresponding buckling time was recorded. 145

DOI: 10.1021/acs.accounts.6b00539 Acc. Chem. Res. 2017, 50, 141−150

Article

Accounts of Chemical Research

Figure 11. Shape change of PEG sample (with blue mark for better visualization): (a1) after 600% prestretching; (a2) after wetting in 4 mm diameter tube; (a3) after drying; (b1) prestretched PEG in the catheter loader for delivery; (b2) 4 mm diameter tube is fully blocked. Yellow sphere (6 mm diameter) for benchmark.

■

FUNDAMENTALS, SIMULATION, AND COMPARISON The mechanism of the above water-responsive process is illustrated in Figure 13 and is explained as follows. In the

Another series of wetting tests on free-standing samples was carried out to simulate the real situation of water induced buckling within a confined space for the temporary vascular occlusion application.40 The original diameter of PEG sample was 2.45 mm, and it was 600% prestretched, so that its diameter was reduced to 0.87 mm and thus it could be fitted into a 4F catheter for delivery and deployment [refer to Figure 11a1,b1]. Upon wetting of a piece of prestretched PEG wire (about 5 cm long) in a 4 mm diameter tube by 37 °C DI water with a flow rate of 120 mL/min for 2 min, the tube was fully blocked [Figure 11b2] and the flow of water was fully stopped as well. Refer to ref 40 for the details of the testing process and Figure 12

Figure 13. Illustration of mechanism of water induced SME and swelling in PEG hydrogel. Figure 12. Images showing the process of buckling induced occlusion with time (PEG sample marked in red for illustration).

cross-linked PEG hydrogel, chemical covalent cross-links serve as the permanent netpoints responsible for holding the permanent shape. Upon wetting in water, the material swells continuously until the maximum is reached (Figure 13e). During the process of programming (e.g., uniaxial stretching) at 70 °C, which is above the Tm of PEG crystalline phase, amorphous molecular

for snapshots of the process of occlusion (another sample marked in red for better visualization). After being taken out of the tube, the PEG maintained its shape [Figure 11a2] and gradually shrank back to its original dimensions after drying [Figure 11a3]. 146

DOI: 10.1021/acs.accounts.6b00539 Acc. Chem. Res. 2017, 50, 141−150

Article

Accounts of Chemical Research chains get oriented along the direction of deformation with loss of entropy and gain of elastic energy (Figure 13b). The temporary deformed shape is fixed by PEG recrystallization as the temperature is lowered to room temperature (Figure 13c). PEG crystallites act as the reversible switch points. Upon immersion in water at 37 °C, water molecules diffuse into the system and as the PEG crystalline phase “dissolves”, the network becomes fully amorphous,46 actuating the shape recovery of PEG in conjunction with swelling (Figure 13d). After a prolonged period of immersion in water, the final shape of the hydrogel (Figure 13e) is essentially the same as that without programming. As we can see, the shape change upon wetting a piece of programmed hydrogel involves two simultaneous processes, namely shape recovery and swelling. Thus, the total absorbed solvent may be decomposed into two parts. One is for shape recovery as it causes the plasticization effect to reduce the Tg of a polymeric material (such as the polyurethane reported in ref 22) or dissolves the crystalline phase (as in this PEG), while the rest of the absorbed solvent largely contributes to swelling, which could be significant (for example, in wetting hydrogels with water) or not easily noticeable by naked eye (e.g., wetting the polyurethane with water as reported in ref 22). The actual ratio between these two parts is also dependent on the exact polymeric material and the applied solvent.10,11,45,47−49 The constitutive model developed in ref 50 was originally meant for chemoresponsive SME in polymers, in which swelling is limited. After integrating with the swelling effect, we extend this model to simulate the response of a piece of programmed hydrogel upon wetting in water. We consider that three samples (original diameter 2 mm) are programmed with three different uniaxial tensile strains of 30%, 100%, and 600%. The equilibrium solvent concentration is set as 0.85 and the swelling ratio (in volume) as 6.7 for the time of recovery process, which was within the initial half an hour. After shape recovery, pure swelling occurs. The swelling ratio can reach as high as 30. The diffusion of water into PEG hydrogel matrix is assumed to follow the Fick’s law with a diffusion coefficient of 0.0003 mm2/s. The solvent concentration on the boundary is slowy ramped to the equilibrium in 600 s. All other parameters used in current simulation are the same as those reported in ref 50. Readers may refer to ref 50 for their actual values and the detailed modeling process. Due to the nature of symmery, as shown in Figure 14, only one-quarter of the sample is investigated via the finite element method (FEM). Refer to the color bar at the bottom right corner of Figure 14 for the concentration of solvent (water) during wetting. As diffusion is intrinsically time-dependent, wetting starts from the outer surface of the sample, and gradually moves inward. Hence, a gradient of solvent concentration, from the maximum at the outer surface to the minimum at the center, results. Consequently, shape recovery and swelling occur simultaneously in the wetted part of the sample, while the inner dry part remains hard at the beginning. Since all samples have the same original diameter, as expected, water penetration in the sample programmed with a larger strain is faster due to its smaller diameter after programming. Since all samples are programmed via uniaxial stretching, water induced shape recovery shortens their length, while the nature of swelling means expansion. Simulation in Figure 14 indicates that the length change upon wetting in water follows different patterns according to the magnitude of programming strain. For a less stretched sample, expansion is dominant, while in a

Figure 14. FEM simulation. Refer to color bar (bottom/right corner) for solvent concentration.

highly stretched piece, contraction occurs continuously. Same as shown in Figure 15, in which a piece of tough hydrogel was prestretched and then wetted in room temperature water, contraction-then-expansion was observed if a proper programming strain was applied. The above model is used to illustrate the shape change caused by both shape recovery and swelling effect. In order to simulate the experimentally observed buckling effect, a fully coupled chemical−mechanical model should be required.51 To simulate instability, geometry imperfections or mechanical perturbation also needs to be introduced.52,53 This will be a subject of future work. In the case of shape recovery from severe deformation, the SMM may become mechanically unstable, and thus buckling may be observed.12 Figure 16I reveals the underlying mechanism in buckling of a piece of programmed (via uniaxial stretching) poly(methyl methacrylate) (PMMA) upon immersing in ethanol for chemoresponsive shape recovery. Although there is not any external force, buckling is induced by the internal compressive stress in the wetted part of PMMA, since the sample has been prestretched. If the Young’s modulus of the dry sample is much higher than that of the wet one, according to ref 44, for buckling to happen, the critical value (rc) of radius r of the hard dry core may be estimated by (refer to Figure 16II) ⎡ ⎤1/2 π2 2 2L⎢ 2 2 ⎥ σ + 2 Er R σ − σ r = π Er ⎢⎣ ⎥⎦ 4L c

(1)

where R is the radius of the sample after prestretching, L is the length of the sample, Er is the Young’s modulus of the hard core, and σ is the compressive stress acting on the softened part. We assume σ = εER and L = L0(1 + ε), where ER is the modulus of the outer soft layer, ε is the programming strain, and L0 is the original length of the sample before programming. We may further assume that the volume is conserved when a piece of sample is stretched in its dry state. Thus, the radius after stretching can be approximated as R = R0(1 + ε)−1/2, where R0 is the original diameter of the dry sample. 147

DOI: 10.1021/acs.accounts.6b00539 Acc. Chem. Res. 2017, 50, 141−150

Article

Accounts of Chemical Research

Figure 15. Snapshot of a piece of prestretched tough hydrogel sample during wetting and then drying. Reproduced from ref 11, Copyright 2014, with permission from Elsevier.

Figure 16. Illustration of underlying mechanism: (I) (a) amorphous PMMA before stretching; (b) high temperature stretching of PMMA; (c) cooling and unloading; (d) penetration of ethanol and outer layer softening; (e) critical penetration depth is reached and sample is buckled; and (f) recovered shape after complete softening; (II) cross-section of panel I.d. Reprinted from Appl. Phys. Lett. 2011, 99, 131911 (ref 44) with permission. Copyright 2011 American Institute of Physics.

Instead of using the case II theory for diffusion of ethanol in PMMA in ref 44 for PEG hydrogel, the diffusion of water should follow the Fick’s law, which gives R − rc = vptc2, where vp is a constant to characterize the diffusion rate and tc is the critical bulking time. Thus, eq 1 can be rewritten as vptc 2 = R 0(1 + ε)−1/2 −

2 2 L0(1 + ε) π

⎡ ⎤1/2 2 2 π 2R 0 2E R ε ER ε ⎥ ⎢ ER ε + − ⎢⎣ Er 2 Er ⎥⎦ 4L0 2(1 + ε)3 Er

(2)

A close examination of eq 2 reveals that the critical bulking time is mainly determined by not only the original diameter of the sample but also the programming strain. Following parameters are used in eq 2 for numerical analysis of the buckling time of PEG hydrogel upon wetting in water. ER = 0.0216 MPa and Er = 696 MPa, which are calculated from the stress versus strain curves of fully swollen and dry specimens as shown in Figures 7 and 5, respectively; vp is set as 2.6 × 10−6 mm/s2, which is obtained through data-fitting of the critical buckling time in samples programmed with 500% strain and original diameter of 2.45 mm. In Figure 17, the calculated buckling time as a function of programming strain in original 2.45 mm diameter PEG hydrogel is plotted and compared with the experimental results of both one-end free (e.g., by observing the shape change in Figure 10) and constrained samples (by monitoring the recovery stress in Figure 8). Good agreement is observed in all of them. With the

Figure 17. Buckling time as a function of programming strain (original diameter 2.45 mm).

increase of programming strain from 300% to 700%, the buckling time is reduced from about 275 s to less than 100 s. Using the same set of parameters, simulation of buckling time as a function of original diameter is compared with the experimental data of both one-end free and constrained samples in Figure 18, in which a constant programming strain of 500% is applied. It is confirmed that same as that in the buckling time vs programming strain relationship, given a fixed programming 148

DOI: 10.1021/acs.accounts.6b00539 Acc. Chem. Res. 2017, 50, 141−150

Article

Accounts of Chemical Research

Polymer Engineering from India. His research interest includes shape memory polymers for biomedical applications, structure−property relationships, rheology, and processing of polymers. Wei Min Huang is an Associate Professor at the School of Mechanical and Aerospace Engineering, NTU, Singapore. He obtained his Ph.D. from Cambridge University, U.K. His research is about shape memory materials and technology. Rui Xiao is an Associate Professor of Engineering Mechanics at Hohai University. He obtained his Ph.D. from Johns Hopkins University in 2015. His research mainly focuses on constitutive modeling shapememory polymers and nonequilibrium behaviors of glassy amorphous polymers. Yee Shan Wong obtained her Ph.D. from the School of Materials Science and Engineering, NTU, Singapore, and did postdoctoral work in NTU with Prof. Subbu Venkatraman. She is experienced in the uses of biodegradable polymers in general and with shape memory of these polymers in particular, for biomedical devices and nanomedicine. Figure 18. Buckling time as a function of original diameter at a fixed programming strain of 500%.

Subbu S. Venkatraman graduated from Carnegie-Mellon University with a Ph.D. in Chemistry. Following industrial R&D stints at three California companies, he joined NTU in June 2000 and set up the Biomaterials effort at NTU and strengthened it over the years to focus on drug delivery, tissue engineering scaffolds, and medical device research.

strain, the buckling time observed in one-end free test is very close to that spotted by the start of sharp stress increase in the constrained test. With the increase of the original diameter from 1.5 to 2.75 mm, the buckling time is extended from about 60 s to around 160 s. At this point, we may conclude that buckling time in water induced shape recovery of prestretched PEG hydrogel can be well controlled by tailoring the original diameter, programming strain, or both. The actual buckling time can be estimated based on one set of parameters. Buckling does not require full wetting of the whole hydrogel, so it occurs within a much shorter period of wetting time.

Kiang Hiong Tay graduated from the National University of Singapore with MBBS, obtained his radiology specialist degree from the Royal College of Radiologists (U.K.), and did clinical as well as research Fellowships in Vascular and Interventional Radiology at the University of British Columbia (Vancouver, Canada) with Prof. Lindsay Machan. He has published on a variety of vascular and oncologic interventions including transarterial locoregional therapies for liver cancers. Ze Xiang Shen is a Professor in the School of Physical and Mathematical Sciences and the School of Materials Science and Engineering, NTU. He obtained his Ph.D. degree in physics from King’s College, University of London. His main research area is on the properties of nanomaterials.

■

CONCLUSIONS Our experiments and simulation prove that buckling induced by water-responsive SME and swelling in hydrogel can not only overcome the limitation with either of them being applied but also achieve time-controlled activation. Original diameter and programming strain are identified as two key parameters for tailoring the actual buckling (actuation) time within a wide range. Since buckling does not require full wetting of the whole hydrogel, it occurs within a much shorter period of wetting time. Although that demonstrated here is targeted for a particular application of vascular occlusion via minimally invasive surgery for liver cancer treatment using a biodegradable PEG hydrogel, the phenomenon reported here, that is, chemically induced buckling via a combination of the SME and swelling, is generic, and it can be extended to various other hydrogel materials and applications thereof.

■

■

ACKNOWLEDGMENTS This project is partially supported by AcRF Tier 1 (RG172/15) and A*STAR (Grant No. 132 148 0011). R. Xiao acknowledges the funding support from National Natural Science Foundation of China (Grant No. 11502068).

■

REFERENCES

(1) Raju, M. P.; Raju, K. M. Synthesis and water absorbency of superabsorbent copolymers. Int. J. Polym. Anal. Charact. 2003, 8, 245− 253. (2) Ahn, S.-k.; Kasi, R. M.; Kim, S.-C.; Sharma, N.; Zhou, Y. Stimuliresponsive polymer gels. Soft Matter 2008, 4, 1151−1157. (3) Qiu, Y.; Park, K. Environment-sensitive hydrogels for drug delivery. Adv. Drug Delivery Rev. 2001, 53, 321−339. (4) Gong, J. P.; Katsuyama, Y.; Kurokawa, T.; Osada, Y. Doublenetwork hydrogels with extremely high mechanical strength. Adv. Mater. 2003, 15, 1155−1158. (5) Sun, J.-Y.; Zhao, X.; Illeperuma, W. R.; Chaudhuri, O.; Oh, K. H.; Mooney, D. J.; Vlassak, J. J.; Suo, Z. Highly stretchable and tough hydrogels. Nature 2012, 489, 133−136. (6) Hoffman, A. S. Hydrogels for biomedical applications. Adv. Drug Delivery Rev. 2012, 64, 18−23. (7) Cai, S.; Suo, Z. Mechanics and chemical thermodynamics of phase transition in temperature-sensitive hydrogels. J. Mech. Phys. Solids 2011, 59, 2259−2278.

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. Notes

The authors declare no competing financial interest. Biographies Abhijit Vijay Salvekar is a Ph.D. candidate at Nanyang Technological University (NTU), Singapore. He obtained M.Tech and B.Tech in 149

DOI: 10.1021/acs.accounts.6b00539 Acc. Chem. Res. 2017, 50, 141−150

Article

Accounts of Chemical Research (8) Marcombe, R.; Cai, S.; Hong, W.; Zhao, X.; Lapusta, Y.; Suo, Z. A theory of constrained swelling of a pH-sensitive hydrogel. Soft Matter 2010, 6, 784−793. (9) Satoh, T.; Sumaru, K.; Takagi, T.; Kanamori, T. Fast-reversible light-driven hydrogels consisting of spirobenzopyran-functionalized poly (N-isopropylacrylamide). Soft Matter 2011, 7, 8030−8034. (10) Zhang, J. L.; Huang, W. M.; Lu, H. B.; Sun, L. Thermo-/chemoresponsive shape memory/change effect in a hydrogel and its composites. Mater. Eng. 2014, 53, 1077−1088. (11) Zhang, J. L.; Huang, W. M.; Gao, G.; Fu, J.; Zhou, Y.; Salvekar, A. V.; Venkatraman, S. S.; Wong, Y. S.; Tay, K. H.; Birch, W. R. Shape memory/change effect in a double network nanocomposite tough hydrogel. Eur. Polym. J. 2014, 58, 41−51. (12) Huang, W. M.; Lu, H. B.; Zhao, Y.; Ding, Z.; Wang, C. C.; Zhang, J. L.; Sun, L.; Fu, J.; Gao, X. Y. Instability/collapse of polymeric materials and their structures in stimulus-induced shape/surface morphology switching. Mater. Eng. 2014, 59, 176−192. (13) Bothe, M.; Pretsch, T. Two-Way Shape Changes of a ShapeMemory Poly(ester urethane). Macromol. Chem. Phys. 2012, 213, 2378−2385. (14) Bothe, M.; Pretsch, T. Bidirectional actuation of a thermoplastic polyurethane elastomer. J. Mater. Chem. A 2013, 1, 14491−14497. (15) Shim, T. S.; Kim, S. H.; Heo, C. J.; Jeon, H. C.; Yang, S. M. Controlled origami folding of hydrogel bilayers with sustained reversibility for robust microcarriers. Angew. Chem., Int. Ed. 2012, 51, 1420−1423. (16) Zhang, Y.; Ionov, L. Reversibly cross-linkable thermoresponsive self-folding hydrogel films. Langmuir 2015, 31, 4552−4557. (17) Huang, W. M.; Zhao, Y.; Wang, C. C.; Ding, Z.; Purnawali, H.; Tang, C.; Zhang, J. L. Thermo/chemo-responsive shape memory effect in polymers: a sketch of working mechanisms, fundamentals and optimization. J. Polym. Res. 2012, 19, 9952. (18) Funakubo, H.: Shape Memory Alloys; Gordon and Breach Science Publishers: New York, 1987. (19) Lendlein, A.; Langer, R. Biodegradable, elastic shape-memory polymers for potential biomedical applications. Science 2002, 296, 1673−1676. (20) Xie, T. Tunable polymer multi-shape memory effect. Nature 2010, 464, 267−270. (21) Ecker, M.; Pretsch, T. Multifunctional poly(ester urethane) laminates with encoded information. RSC Adv. 2014, 4, 286−292. (22) Huang, W. M.; Yang, B.; An, L.; Li, C.; Chan, Y. S. Water-driven programmable polyurethane shape memory polymer: Demonstration and mechanism. Appl. Phys. Lett. 2005, 86, 114105. (23) Wang, C. C.; Huang, W. M.; Ding, Z.; Zhao, Y.; Purnawali, H. Cooling-/water-responsive shape memory hybrids. Compos. Sci. Technol. 2012, 72, 1178−1182. (24) Otsuka, K.; Wayman, C. M.: Shape memory materials; Cambridge University Press: Cambridge, 1998. (25) Huang, W. M.; Ding, Z.; Wang, C. C.; Wei, J.; Zhao, Y.; Purnawali, H. Shape memory materials. Mater. Today 2010, 13, 54−61. (26) Huang, W. M.; Song, C. L.; Fu, Y. Q.; Wang, C. C.; Zhao, Y.; Purnawali, H.; Lu, H. B.; Tang, C.; Ding, Z.; Zhang, J. L. Shaping tissue with shape memory materials. Adv. Drug Delivery Rev. 2013, 65, 515− 535. (27) Lendlein, A.; Behl, M. Shape-memory polymers for biomedical applications. Adv. Sci. Technol. 2008, 54, 96−102. (28) Feng, Y. K.; Zhang, S. F.; Wang, H. Y.; Zhao, H. Y.; Lu, J.; Guo, J. T.; Behl, M.; Lendlein, A. Drug release from biodegradable polyesterurethanes with shape-memory effect. J. Controlled Release 2011, 152, E20− E21. (29) Yoneyama, T.; Miyazaki, S.: Shape memory alloys for biomedical applications; Woodhead Publishing Limited: Cambridge, UK, 2009. (30) Metcalfe, A.; Desfaits, A. C.; Salazkin, I.; Yahia, L.; Sokolowski, W. M.; Raymond, J. Cold hibernated elastic memory foams for endovascular interventions. Biomaterials 2003, 24, 491−497. (31) Small, W.; Buckley, P. R.; Wilson, T. S.; Benett, W. J.; Hartman, J.; Saloner, D.; Maitland, D. J. Shape memory polymer stent with

expandable foam: A new concept for endovascular embolization of fusiform aneurysms. IEEE Trans. Biomed. Eng. 2007, 54, 1157−1160. (32) Lendlein, A.; Kelch, S. Shape-Memory Polymers. Angew. Chem., Int. Ed. 2002, 41, 2034−2057. (33) Sun, L.; Huang, W. M.; Ding, Z.; Zhao, Y.; Wang, C. C.; Purnawali, H.; Tang, C. Stimulus-responsive shape memory materials: A review. Mater. Eng. 2012, 33, 577−640. (34) Osada, Y.; Matsuda, A. Shape memory in hydrogels. Nature 1995, 376, 219. (35) Hao, J. K.; Weiss, R. A. Mechanically Tough, Thermally Activated Shape Memory Hydrogels. ACS Macro Lett. 2013, 2, 86−89. (36) Skrzeszewska, P. J.; Jong, L. N.; de Wolf, F. A.; Cohen Stuart, M. A.; van der Gucht, J. Shape-memory effects in biopolymer networks with collagen-like transient nodes. Biomacromolecules 2011, 12, 2285−2292. (37) Yue, J. J.; Morgenstern, R.; Morgenstern, C.; Lauryssen, C. Shape Memory Hydrogels−A Novel Material for Treating Age-related Degenerative Conditions of the Spine. Eur. Musculoskeletal Rev. 2011, 6, 184−188. (38) Lauryssen, C.; Yue, J. J.; Jaramillo-de la Torre, J. J.; Chen, A.; Prewett, A. Novel application of biomedical hydrogels for treating degenerative conditions of the spine. Eur. Musculoskeletal Rev. 2010, 5, 36−38. (39) Mauro, M. A.; Murphy, K. P.; Thomson, K. R.; Venbrux, A. C.; Morgan, R. A.: Image-guided interventions: expert radiology series; Elsevier Health Sciences: Philadelphia, PA, 2013. (40) Wong, Y. S.; Salvekar, A. V.; Zhuang, K. D.; Liu, H.; Birch, W. R.; Tay, K. H.; Huang, W. M.; Venkatraman, S. S. Bioabsorbable radiopaque water-responsive shape memory embolization plug for temporary vascular occlusion. Biomaterials 2016, 102, 98−106. (41) Mironi-Harpaz, I.; Wang, D. Y.; Venkatraman, S.; Seliktar, D. Photopolymerization of cell-encapsulating hydrogels: Crosslinking efficiency versus cytotoxicity. Acta Biomater. 2012, 8, 1838−1848. (42) Hagel, V.; Haraszti, T.; Boehm, H. Diffusion and interaction in PEG-DA hydrogels. Biointerphases 2013, 8, 36. (43) Cantournet, S.; Desmorat, R.; Besson, J. Mullins effect and cyclic stress softening of filled elastomers by internal sliding and friction thermodynamics model. Int. J. Solids Struct. 2009, 46, 2255−2264. (44) Zhao, Y.; Wang, C. C.; Huang, W. M.; Purnawali, H. Buckling of poly(methyl methacrylate) in stimulus-responsive shape recovery. Appl. Phys. Lett. 2011, 99, 131911. (45) Wang, C. C.; Zhao, Y.; Purnawali, H.; Huang, W. M.; Sun, L. Chemically induced morphing in polyurethane shape memory polymer micro fibers/springs. React. Funct. Polym. 2012, 72, 757−764. (46) Gu, X.; Mather, P. T. Water-triggered shape memory of multiblock thermoplastic polyurethanes (TPUs). RSC Adv. 2013, 3, 15783−15791. (47) Yang, B.; Huang, W. M.; Li, C.; Lee, C. M.; Li, L. On the enects of moisture in a polyurethane shape memory polymer. Smart Mater. Struct. 2004, 13, 191−195. (48) Yang, B.; Huang, W. M.; Li, C.; Chor, J. H. Effects of moisture on the glass transition temperature of polyurethane shape memory polymer filled with nano-carbon powder. Eur. Polym. J. 2005, 41, 1123−1128. (49) Lu, H.; Huang, W. M.; Wu, X. L.; Ge, Y. C.; Zhang, F.; Zhao, Y.; Geng, J. Heating/ethanol-response of poly methyl methacrylate (PMMA) with gradient pre-deformation and potential temperature sensor and anti-counterfeit applications. Smart Mater. Struct. 2014, 23, 067002. (50) Xiao, R.; Nguyen, T. D. Modeling the solvent-induced shapememory behavior of glassy polymers. Soft Matter 2013, 9, 9455−9464. (51) Chester, S. A.; Di Leo, C. V.; Anand, L. A finite element implementation of a coupled diffusion-deformation theory for elastomeric gels. Int. J. Solids Struct. 2015, 52, 1−18. (52) Cai, S.; Chen, D.; Suo, Z.; Hayward, R. C. Creasing instability of elastomer films. Soft Matter 2012, 8, 1301−1304. (53) Kang, M. K.; Huang, R. Swell-induced surface instability of confined hydrogel layers on substrates. J. Mech. Phys. Solids 2010, 58, 1582−1598.

150

DOI: 10.1021/acs.accounts.6b00539 Acc. Chem. Res. 2017, 50, 141−150