Correlating the mass and mechanical property changes during the degradation of PEG-based adhesive

Abstract

Change in mechanical property of a degrading adhesive is critical to its performance. However, characterization of degradation behavior is often limited to tracking its mass loss. 4-armed PEG end modified with dopamine (PEG-DA) was used as a model bioadhesive to correlate its change in mass with change in mechanical property. Shear modulus (G) was calculated based on the mass and average molecular weight between crosslinks $\left({\overline{M}}_c\right)$ of PEG-DA, while the storage modulus (G’) was determined by oscillatory rheometry. G decreased slowly within the first week of degradation (10% reduction by week 2), while G’ decreased by 60% during the same period. This large discrepancy is due to the partially disconnected and elastically ineffective PEG polymer, which is trapped within the adhesive network. This resulted in minimal mass change and higher calculated G value during the earlier time points. Therefore, tracking mass loss profile alone is inadequate to completely describe the degradation behavior of an adhesive. Additionally, PEG-DA was coated onto magnetoelastic (ME) sensors, and the change in the resonance amplitude of the sensor corresponded well with dry mass loss of PEG-DA. ME sensing provide a non-destructive method to track the mass loss of the coated adhesive.

Graphical Abstract

1. Introduction

Tissue adhesives are widely used for wound closure, surgical repair, and reconstruction of injured or degenerated soft tissues ^1–3. Bioadhesives have advantages over sutures, which are often associated with chronic pain, prolonged surgical time, fluid leakage and compliance mismatch ^4–8. Unlike sutures that require minimal footprint for repair, bioadhesives require increased overlapped surface area between the adhesive and the substrates to achieve equivalent fixation strength ^9,10. As such, prolonged presence of the adhesive can act as a barrier to the union of wound edges. One of the key criteria for a tissue adhesive is its ability to degrade ^11,12. However, the adhesive needs to maintain its mechanical integrity for a period of time immediately after surgery to prevent premature rupture of the repaired tissue ¹³. Therefore, there is a need to track the change in the mechanical behavior of a biodegradable tissue adhesive over time.

Despite the need to characterize the effect of degradation on the mechanical property of tissue adhesives, researchers typically only determine the mass loss of an adhesive during degradation ^14–17. Some reported the effect of degradation on the swelling ratio ^18,19 or dimension ²⁰ of the adhesive, which could potentially yield additional information relating to the network architecture. However, mechanical properties of adhesives were typically determined immediately after curing but were not tracked over time ^14–17. To our knowledge, very few publications have correlated the mass loss of an adhesive with change in its mechanical behaviors during degradation. Most importantly, the change in mass and mechanical properties may not occur at the same rate ²¹. Therefore, it is insufficient to use mass loss as the sole method for characterizing the degradation behavior of an adhesive.

Both mass and mechanical property tracking are destructive tests, and it is cost-prohibitive to conduct comprehensive studies to evaluate multiple parameters during the degradation of a tissue adhesive. Magnetoelastic (ME) sensing technology can potentially be used to remotely track the degradation behavior of tissues adhesive. ME sensor vibrates when excited by a magnetic field, and the vibration amplitude and frequency can be remotely characterized using a coil antenna ²². By tracking the relative change in its resonance frequency and/or amplitude, ME sensor can be highly sensitive to minute changes in mass and elasticity on its surface ²³. ME sensors have been used to detect chemical and biological agents ^24,25, change in coating viscosity ^26,27, and biointerfacial binding events (e.g., cellular attachment, proliferation) both in culture and in vivo ^28,29.

In this paper, we characterized the degradation behavior of a model tissue adhesive by correlating the tracked changes in mass, mechanical property, and resonance frequency and amplitude of adhesive-coated ME sensors. A 4-armed poly(ethylene glycol) end modified with dopamine (PEG-DA, Figure 1) was prepared as the model tissue adhesive. Dopamine contains a catechol moiety that mimics the interfacial binding and curing chemistry of 3,4-dihydroxyphenylalanine (DOPA), found in large abundance in mussel adhesive proteins ^30,31. Various versions of catechol-modified PEG have previously demonstrated strong adhesion to wetted tissue surfaces ^32–34. PEG was chosen as a polymer backbone due to its well-known biocompatibility and various PEG-based bioadhesives (DuraSeal, Integra LifeSciences; CoSeal, Cohesion Technologies) are commercially available ^3,35. Dopamine is linked to PEG through a glutaric acid using a hydrolysable ester bond. PEG-DA will solidify through chemical crosslinking of terminal dopamine moieties and degrade through hydrolysis of the ester bond.

Figure 1.

The chemical structure of PEG-DA (A). Schematic showing the crosslinking of terminal dopamine in the presence of a network (B).

During the degradation of PEG-DA adhesive, both dry and wet mass of the adhesive were tracked. This information was further used to calculate the average molecular between crosslinks $\left({\overline{M}}_c\right)$, effective crosslinking density (v_e), and the shear modulus (G) of the adhesive. Oscillatory rheometer was used to track the changes in the mechanical properties of PEG-DA. Finally, PEG-DA was coated onto ME sensor strips and the change in the resonance frequency and amplitude was determined.

2. Experimental Section

2.1. Materials

PEG-DA was prepared using published protocols ^36,37 using 10 kDa 4-arm PEG from JenKem Technology USA (Plano, TX). ¹H NMR spectroscopy confirmed the structure of PEG-DA (Figure S1). Sodium periodate (NaIO₄) (ACS reagent > 99.8%) and Tris hydrochloride were purchased from Sigma-Aldrich and phosphate buffer saline (PBS) was purchased from Fisher Scientific. Dopamine hydrochloride was purchased from ACROS Organics. Metglas 2826MB (Fe₄₀Ni₃₈Mo₄B₁₈) ribbon was purchased from Metglas, Inc. Parylene C was purchased from Specialty Coating Systems.

2.2. Preparation of PEG-DA adhesive samples

Adhesive samples were prepared by mixing equal volume of two precursor solutions consisted of 300 mg ml⁻¹ of PEG-DA in 2X PBS and 22 mg ml⁻¹ of NaIO₄ in deionized (DI) water to form an adhesive hydrogel with a final PEG concentration of 150 mg ml^-1. The two precursor solutions were allowed to solidify in a mold formed from two glass plates sandwiched between a 1.5 mm-thick silicone rubber spacer for 24 h. The hydrogel sheet was further cut into disc shape using a biopsy punch with a diameter of 10 mm. Degradation of hydrogels was performed by incubating each disc sample in 10 ml of PBS buffer (pH 7.4) at 37 ˚C. The PBS buffer was replaced every week to maintain the pH value at 7.4. At a given time point, hydrogel samples were retrieved for testing.

2.3. Lap shear adhesion test

The adhesive strengths of the hydrogels were determined by the lap shear adhesion test with a modified protocol ³⁷. The bovine pericardium was cut into 2.5 cm × 2.5 cm strips and hydrated in PBS (pH = 7.4). PEG-DA adhesives were cured between two partially overlapping bovine pericardial with an overlapping area of 2.5 cm × 1 cm. The adhesive joint was compressed with a 100 g weight for 5 min and further conditioned in PBS (pH = 7.4) at 37 °C for overnight prior to the testing. The dimensions of the contact area of each adhesive joint were measured before testing. A commercial PEG-based sealant, DuraSeal (Integra LifeSciences, Inc.), was prepared the same way and tested for comparison. The adhesive joints were pulled to failure at a rate of 0.1 mm s⁻¹ until the tissues separate using an Electroforce® machine (Bose Electroforce Group, MN). The adhesive strength was determined by the max load divided by the initial contact area of the adhesive joint.

2.4. Determination of mass loss during degradation

Hydrogel discs were collected and dried in vacuum for at least 48 hours. The mass of the hydrogel in the swollen (M_s) and dried (M_d) states were determined with an analytical balance. The swelling ratios were determined by the following equation:

$$swelling\hspace{0.17em}ratio=\frac{M_s-M_d}{M_d}$$ (1)

The masses of hydrogels were further used to determine the average molecular weight between crosslinks $\left({\overline{M}}_c\right)$, using the following Flory-Rehner equation ³⁸:

$$-\left[\ln \left(1-v_{2m}\right)+v_{2m}+{{\chi }_1{v}_{2m}}^2\right]=\left(V_1/\overline{v}{M}_c\right)\left(1-2{\overline{M}}_c/M\right)\left({v_{2m}}^{1/3}-v_{2m}/2\right)$$ (2)

where M is the starting molecular weight of PEG-DA (10992 Da), V₁ is the molar volume of water (18.1 mol cm⁻³), $\overline{v}$ is the specific volume of the polymer (0.8904 cm³ g⁻¹) ³⁹, χ₁ is the polymer-solvent interaction parameter for PEG and water (0.462) ⁴⁰, and v_2m is the volume fraction of polymer in the swollen hydrogel. v_2m is determined by the following equation ³⁴:

$$v_{2m}=\frac{1}{{\rho }_p\left[\left(M_s/M_d\right)-1\right]+1}$$ (3)

where ρ_p is the density of PEG (1.123 g cm⁻³) ³⁹.

The effective crosslinking density (v_e) was determined using the following equation ⁴¹:

$$v_e=\frac{\rho }{M_c}$$ (4)

where ρ is the density of PEG-DA. Since the mass of PEG accounts for greater than 91% of PEG-DA, the density of PEG was used for calculation.

Shear modulus (G) was calculated with the following equation ⁴²:

$$G=\left(1-\frac{2}{f_{\text{eff}}}\right)v_{e}RT{v}_{2m}$$ (5)

where f_eff is the effective functionality, R is the gas constant, and T is the temperature in Kelvin. f_eff was set as four as the branched PEG has 4 arms.

2.5. Oscillatory rheometer

Samples were subjected to amplitude sweep experiment (0.01–500 strain, frequency = 0.1 Hz) using an oscillatory rheometer (HR-2, TA Instruments, DE) to determine the storage (G’) and loss (G”) moduli, and loss tangent (tan δ). The samples were tested using a parallel plate set up with a gap distance that was 85% of the measured thickness of each sample. The thickness of the hydrogel discs was measured using a digital Vernier caliper before testing.

2.6. Track degradation using ME sensors

PEG-DA adhesive was coated onto ME sensor strips. First, Metglas 2826MB strips (length = 12.7 mm; width = 5 mm; thickness = 30 µm) were coated with 6–7 µm-thick Parylene C to prevent sensor corrosion in an aqueous environment using a Parylene deposition system (PDS 2010 Labcoater® 2, Specialty Coating Systems, Inc.) ²⁸. Parylene C-coated strips were rinsed with ethanol and dried in vacuum. Parylene C-coated strips were further treated with 10 mL of 10 mg mL⁻¹ dopamine in 10 mM Tris-HCl buffer (pH 8.5) to create a thin polydopamine primer layer ⁴³. Each side of the ME sensor strip was exposed to the dopamine solution for 30 minutes each and the coated strips were dried in vacuum. Finally, PEG-DA was coated onto the sensor strip using a spin coater (WS-650–23B Spin Coater, Laurell Technologies Corporation). 12 µl of ethanol was first added to cover the surface of the polydopamine-coated sensor strip. 6 µl each of the 2 precursor solutions were first mixed in a glass vial and immediately transferred to the surface of the sensor and spun at 1500 RPM for 4 minutes. The adhesive-coated sensors were immersed in DI water for 24 hours to remove impurity, dried and stored in vacuum. The masses of the sensors were determined before coating and after the application of each coating layer. Fourier-transform infrared spectroscopy (Perkin Elmer Spectrum One) was utilized to verify the existence of PEG-DA coating. PEG-DA-coated sensor was cut in half to visualize the morphology of the cross-section using a field emission scanning electron microscope (FESEM; Hitachi S-4700, Hitachi High Technologies America, Inc.).

Each PEG-DA-coated sensor (n = 7) was submerged in 2 mL of PBS buffer (pH 7.4) at 37°C. Every two days, the resonance amplitude and frequency of each sensor strip were determined using a custom ME resonant sensor detector ^27,43. At each time point, three frequency sweep experiments (frequency = 150–170 kHz with a resolution of 20 Hz) were performed for each sensor using a 2000 mV DC offset. During each frequency sweep operation, the sensor detector first generated an excitation sinusoidal magnetic field (100 cycles at the start frequency of 150 kHz) through the coil antenna to energize the sensor into vibration. The excitation field was then turned off to allow the sensor to generate an exponentially decaying sinusoidal magnetic field, a process that was termed as the “ring down” of the sensor. The ring-down signal was captured by the detector through the same coil antenna. The detector continued to increase the excitation field frequency while recording the ring-down duration at each frequency. The resonance frequency of the sensor was determined as the frequency with the longest ring-down duration, and the resonance amplitude was defined as the number of periods of the rind-down signal at resonance (referred to as the “count”).

2.7. Statistical analysis

Statistical analysis was performed using SigmaPlot® software (version: 12.0). Student’s t-test was performed for comparing means between two groups using a p-value of 0.05.

3. Results

3.1. Lap shear adhesion test

The lap shear adhesion testing was performed using pericardium tissues (Figure S2). Lap shear strength of PEG-DA (2.2 ± 0.3 kPa) was not statistically different (p > 0.05) from that of a commercial PEG-based adhesive, DuraSeal (2.6 ± 0.9 kPa) (Figure 2). DuraSeal consists of a 4-armed PEG end-modified with N-hydroxysuccinimidyl-activated ester and tri-lysin peptide ³. Cured DuraSeal degrades through hydrolysis of terminal ester bonds similar to the degradation of PEG-DA. This result confirms that PEG-DA can function as bioadhesive and can be used as a model polymer for studying the degradation of PEG-based adhesives.

Figure 2.

Lap shear adhesion strength of PEG-DA and DuraSeal (n = 3).

3.2. Changes in the mass of the adhesive during degradation

PEG-DA adhesive was incubated at physiological conditions (pH 7.4, 37°C) and the mass of the adhesive at the swollen (M_s) and dried (M_d) states were recorded (Figure 3). M_s continued to increase as PEG-DA degraded, demonstrating a 2.5-fold increase by week 5. As the ester bond between PEG and glutaric acid hydrolyzes, the adhesive network swells and its water content increases with time. The swelling ratio of the adhesive increased from 10.6 ± 0.99 to 40.2 ± 3.92 during the first 5 weeks of degradation (Figure S3). However, M_d remained mostly constant for the first 3 weeks and did not exhibit an appreciable decrease until week 4 (a 25% reduction). The decrease in M_d is associated with the loss of PEG polymer. This result indicated that little or no loss of PEG occurred during the first 3 weeks of degradation. The adhesive fully degraded and completely dissolved in PBS by the 6^th week.

Figure 3.

Mass of the adhesive in the swollen (A) and dried (B) state during degradation (n = 3). The normalized mass is normalized to the mass of the adhesive at time = 0 week.

Both M_s and M_d were used to calculate the average molecular weight between crosslink $\left({\overline{M}}_c\right)$ using modified Flory-Rehner equation ³⁸. ${\overline{M}}_c$ is the average molecular weight of the polymer chain between two consecutive junctions or crosslinking points in a polymer network and is inversely proportional to the crosslinking density and mechanical properties of the network ^44,45. Prior to degradation, ${\overline{M}}_c$ averaged around 4,300 Da (Figure 4). This value is in agreement wih the architecture of PEG-DA network formed through dimer formation of dopamine, as PEG contains 4 chains of equal length (2,500 Da). ${\overline{M}}_c$ value increased over time as PEG-DA degraded. ${\overline{M}}_c$ value was used to further calculate the changes in the effective crosslinking density (v_e) and the shear modulus (G) of the adhesive network. v_e is inversely proportional to ${\overline{M}}_c$ and decreased with time (Figure 4B). v_e decreased to 80% of its original value by week 5. G was calculated to be around 6700 Pa before degradation and the average value of G decreased to 20% of its original value by week 5 (Figure 4C).

Figure 4.

Average molecular weight between crosslinks $\left({\overline{M}}_c\right)$ (A), effective crosslinking density (v_e) (B), and shear modulus (G) (C) of PEG-DA adhesives during degradation (n = 3). The data are normalized by the calculated values of the adhesive at time = 0 week.

3.3. Oscillatory rheometer data of PEG-DA adhesive

Oscillatory rheometry was performed to track the changes in the mechanical property of PEG-DA adhesive during degradation (Figure S4). Throughout the degradation process, the storage modulus (G’) of PEG-DA was significantly higher than those of loss modulus (G”) at any given time points, indicating the adhesive behaved as an elastic hydrogel ⁴⁶. Additionally, both G’ and G” values were independent of strain amplitude at a range between 0.1 and 10, which signifies the linear viscoelastic (LVE) range of PEG-DA. The data points at one strain amplitude within LVE range (strain = 0.75) were further plotted to visualize the changes in the shear moduli as PEG-DA degraded (Figure 5). G’ of the adhesive was around 5900 Pa at the beginning of the experiment, which is in agreement with the reported G’ values of PEG-dopamine based hydrogel prepared in a previous work ³⁷. This value is also in agreement with shear modulus (G = 6700 Pa) determined based on the mass and ${\overline{M}}_c$ of the adhesive (Figure 4C). However, G’ values determined by the rheometer decreased sharply to 2400 Pa, a 60% reduction in value, by week 2, which differs from the graduate decrease in G value calculated based on the mass of PEG-DA (a 4% reduction between weeks 1 and 2) (Figure 6). This indicates that modulus values derived from mass inaccurately predicted the mechanical property of the adhesive once it started to degrade. It is not until week 4 that the differences in the values of G’ and G were reduced.

Figure 5.

Storage (A) and loss (B) modulus, and loss tangent (tan δ) (C) of PEG-DA during degradation (n = 3). The values were determined at a strain of 0.75 and a frequency of 0.1 Hz.

Figure 6.

Normalized shear moduli for G calculated from the mass of the adhesive (□) and G’ determined by oscillatory rheometer (●).

The large difference between the values of G’ and G at weeks 2 and 3 (Figure 6) is likely due to the retention of elastically ineffective PEG polymer within the adhesive network. The PEG network is highly water-swollen and the hydrolysis of ester bonds occurred throughout the bulk of the network. Breaking ester bonds led to decreased crosslinking density and mechanical property. However, this resulted in dangling PEG chains that did not contribute to the elastic response during deformation. Additionally, it requires breaking all 4 terminal ester bonds and an increased mesh size for the 4-armed PEG polymer to diffuse out of network. The dry mass of the network did not change over the first 3 weeks of degradation (Figure 3B). These PEG molecules did not contribute to the elasticity of the polymer network and maintained the dry mass of the polymer. In the 4^th week, the mesh size of the polymer network become large enough for the PEG to diffuse out of the network, which resulted in the dry mass loss of the polymer. At this time point, the differences between the calculated G and G’ became small.

G” values for PEG-DA initially increased due to increased water content as the adhesive network swell (Figure 5B). This indicated an increased viscous property within the adhesive network. However, as the adhesive continued to degrade, there was a general decrease in the coverall modulus of the adhesive and G” decreased with time. The loss tangent (tan δ) is defined as the ratio between G” and G’, which gives the proportion of the dissipated energy and the stored energy ⁴⁷. During the degradation, tan δ value continues to increase, demonstrating a 9-fold increase by week 5 (Figure 5C). Both the reduction in the crosslinking density and infiltration of molecular water contributed to the increase in viscous property of the adhesive.

3.4. Tracking degradation using ME Sensors

PEG-DA was coated onto ME sensor strips. Polydopamine was used as a reliable surface primer to attached PEG-DA to the surface of the ME sensor ^48,49. The catechol and amine moieties present on the polydopamine surface resulted in covalent crosslinking with the PEG-DA during the oxidant-induced curing process of the adhesive ⁵⁰. FTIR spectra of the coated-strips exhibited peaks associated with PEG ether bonds (1103 cm⁻¹, -C-O-C-), carbonyl group (1727 cm⁻¹, ester linkage), and alkyl group (2878 cm⁻¹, -CH₂-) found in PEG-DA (Figure S5) ⁴³. Based on FE-SEM image, the thickness of the dried adhesive coating was around 32 µm (Figure S6). The dry mass of the PEG-DA coating was determined to be 0.0004 ± 0.0001 g. The sensor detector tracked the change in the resonance frequency and amplitude to determine their correlations with the degradation of PEG-DA. In the past, changes in resonance frequency have been used to determine changes in the mass of coating on the sensor ⁵¹. However, in the current study, change in the resonance frequency was minimal compared to the frequency measurement resolution in this experiment (Figure S7). This is potentially due to a very small amount of PEG-DA coating. On the other hand, the normalized resonance amplitude did not change over the first 3 weeks but started to decrease sharply in week 4 (Figure 7). The resonance amplitude was about 60% of the initial value after week 6 when the adhesive fully degraded. As the coated adhesive degraded over time, the adhesive swell and increased viscous damping, which reduced the resonance amplitude of the sensor vibration.

Figure 7.

Resonance amplitude of PEG-DA-coated ME sensors during degradation (resolution = 20 Hz, 2000 mV DC offset; n = 7).

The resonance amplitude results from the ME sensors resembled the changes in the dry mass of the bulk PEG-DA adhesive (Figure S8). Before the 4^th week, the dry mass of the adhesive remained constant because most of PEG polymers were either connected to or trapped within the polymer network. The water content in the network was low and the impedance between the water and the polymer chains was small. As such, the energy of vibration can pass through the network directly. Thus, the viscous damping from the coated adhesive was not significant during this period and the resonance amplitude did not change significantly. After the 4^th week, there was a large decrease in the dry mass of the adhesive, resulting in the release of PEG polymer from the adhesive network, which was then replaced by water molecules. It is likely that the replacement of PEG polymer by water molecule increased the mechanical damping on the sensor, which caused a reduction in the amplitude of the vibration ⁵². From our study, ME sensing provide a non-destructive method to track the mass loss of the coated adhesive.

4. Discussions

The curing of PEG-DA was accomplished by chemical crosslinking between terminal dopamine molecules, which are linked to the PEG chains with hydrolysable ester bonds. During the degradation process, ester bonds are broken and PEG chains progressively disconnected from the adhesive network (Figure 8). This resulted in decreased crosslinking density and stiffness of the polymer network. On the other hand, ${\overline{M}}_c$ value and water content of the adhesive network also increased over time, which resulted in an increased swollen mass and viscous property of the adhesive. In the first several weeks of degradation, branched PEG polymer remained connected to the adhesive network even though it becomes partially detached. As such, the dried mass of the adhesive remained constant. However, the dangling PEG chains do not contribute to the elasticity of the network, resulting in reduced mechanical properties. This contributed to the large discrepancy in the modulus measured by oscillatory rheometer and those calculated based on the mass of the adhesive during the first few weeks of degradation (Figure 6). When PEG is completely disconnected from the polymer network, the freed PEG polymer diffuses out of the adhesive network, which resulted in noticeable dry mass loss after week four. The vacancy left by detached PEG was occupied by water molecules, which contributed to the increased mechanical damping measured using ME sensor. Therefore, the resonance amplitude can be utilized for quantifying the loss of dry mass since their values change at almost the same time period.

Figure 8.

Schematic representation of how various parameters changed with the degradation of PEG-DA adhesive. Storage modulus (G’) and effective crosslinking density (v_e) of the adhesive increased with time while average molecular weight between crosslinks $\left({\overline{M}}_c\right)$, swollen mass (M_s), and loss tangent (tan δ) of the adhesive decreased. When up to 3 terminal ester bonds were hydrolyzed, the dry mass (M_d) of the adhesive and resonance amplitude (RA) of adhesive-coated ME sensors remained constant as the PEG polymer remained within the adhesive network. Both M_d and RA decreased when PEG is fully disconnected and diffused out of the adhesive network.

Taken together, our results indicated that tracking mass alone is insufficient in characterizing the degradation behavior of a tissue adhesive. The change in mass and mechanical property occurred at a different rate. Most notably, loss in mechanical property can take place without noticeable mass loss. This is especially important given that the bulk mechanical property of an adhesive directly affects its adhesive performance ⁵³. Other polyester-based adhesives have reported a similar observation, where the adhesive exhibited a large decrease in adhesive strength during the early time points of degradation, eventhough mass loss occurred at a much slower rate ²¹. Since mechanical strength is critical to the performance of tissue adhesives during degradation, it is necessary to characterize both mass and mechanical changes during degradation.

Although we have used PEG-DA as a model adhesive to study the degradation behavior using different approaches, various 4-armed PEG-based sealants are currently available in the market (DuraSeal, Integra LifeSciences; CoSeal, Cohesion Technologies).^3,35 Additionally, branch PEG and related polymers are often used as the backbone for designing research-grade tissue adhesives ^54–56. Specifically, these PEG-based adhesives solidify using terminal reactive functional groups linked to PEG using an ester bond and are expected to degrade in a similar manner as PEG-DA. Additionally, branched PEG polymers are widely used to formulate hydrogels for various biomedical applications such as tissue engineering ⁵⁷, carriers for drug delivery ^58,59, and tool for studying cell behavior ⁶⁰. Results reported here can potentially be useful for correlating mechanical properties and mass loss during the degradation of the PEG-based hydrogels.

5. Conclusions

The degradation behavior of PEG-DA was characterized by mass tracking, oscillatory rheology, and ME sensor. These results indicated that changes in mechanical property and mass loss occurred at a different rate. This discrepancy is due to the disconnected PEG chains that became elastically ineffective but remained in the adhesive network until PEG polymer can diffuse out of the network 4 weeks after the start of degradation. Additionally, resonance amplitude of ME sensor correlated well with changes in the dry mass of the adhesive, indicating this technology could become an attractive method to conveniently and cost-effectively characterize degradable polymers. Furthermore, the results in this work showed that mass tracking alone could not adequately characterize the degradation behavior of PEG-DA.