In-Situ Deactivation of Catechol-Containing Adhesive using Electrochemistry

Abstract

Marine mussels secret catechol-containing adhesive proteins that enable these organisms to bind to various surfaces underwater. Synthetic mimics of these proteins have been created to function as adhesives and coatings for a wide range of applications. Here, we demonstrated the use of in-situ electrical field stimulation to deactivate the adhesive property of catechol-containing adhesive that is in direct contact with a surface. Johnson-Kendall-Roberts (JKR) contact mechanics test was performed using a titanium (Ti) sphere in the presence of a pH 7.5 aqueous buffer. The Ti sphere also served as a conductive electrode for applying electricity to the adhesive, while a platinum (Pt) wire served as the counter electrode. Work of adhesion (W_adh) decreased with increased levels of applied voltage and current, exposure time to the applied electricity, and salt concentration of the interfacial buffer. Application of 9 V for 1 min completely deactivated the adhesive. UV-Vis diffuse reflectance spectra and tracking of catechol oxidation byproduct, hydrogen peroxide, confirmed that catechol was oxidized as a result of applied electricity. Contact mechanics testing further confirmed that the Young’s modulus of the adhesive increased by nearly 4 folds at the interface as a result of oxidative crosslinking, even though the modulus of the bulk of the adhesive was unaffected by applied electricity. The accumulation of hydroxyl ions near the cathode increased the local solution pH, which promoted oxidation-induced crosslinking of catechol and subsequently decreased its adhesive property. Tuning adhesive properties through in-situ electrochemical oxidation provides on-demand control over the adhesive, which will potentially add another dimension in designing synthetic mimics of mussel adhesive proteins.

Graphical Abstract

Introduction:

A smart adhesive can be activated or deactivated in response to externally applied stimuli. Smart adhesives are of great interest for various applications in the biomedical field (e.g., painless removal of wound dressing^1-2 and prosthetics limb attachment with tissue^3-4), various industries (e.g., on-demand attachment or detachment of components in the automotive industry^5-7 and structural joints^8-9), and the manufacturing sectors. However, most of the existing smart adhesives are designed to adhere to dry surfaces and have limited adhesion to wet surfaces¹⁰. Therefore, a smart adhesive that can form strong adhesion to wet surfaces with tunable adhesive properties is in great demand.

To design a wet adhesive, researchers have taken an interest in underwater living creatures that can attach themselves firmly to wet surfaces, such as marine mussels^11-12, barnacles¹³, sandcastle worms¹⁴, and octopi¹⁵. Among these biomimetic strategies, our interest is in mussel-inspired wet adhesion. Marine mussels secret protein-based adhesives that contain a unique amino acid, 3,4-dihydroxyphenylalanine (DOPA). The catechol side chain of DOPA is responsible for strong adhesion to various underwater surfaces^16-18. Many catechol-functionalized adhesives and coatings have been developed in the last decade for different applications ranging from multifunctional coatings¹⁹ to skin wound closure²⁰, hemostatic agent²¹, and drug delivery application²². Recently, catechol-containing smart adhesives that are responsive to light irradiation²³, change in temperature²⁴, or enzyme²⁵ have been reported. Similarly, our previous work demonstrated that pH can be used to modulate the interfacial binding property of catechol-based smart adhesives, by controlling the oxidation state of catechol^26-28. The reduced form of catechol exhibits strong interfacial binding strength to inorganic substrates. The reported interfacial binding strength between the reduced form of catechol and a titanium (Ti) substrate averaged around 800 pN, which is 40% that of a covalent bond²⁹. Catechol forms monodentate or bidentate coordination bonds with Ti³⁰. This binding strength is drastically reduced to 180 pN as a result of pH-induced oxidation of catechol²⁹. However, using change in pH as an externally applied stimulus is limited by the slow diffusion of ions into the adhesive network and is highly impractical.

An alternative approach to tune the oxidation state of catechol is through electrochemical oxidation. Various studies have demonstrated the electrochemical oxidation of soluble catechol derivatives using cyclic voltammetry technique in a three-electrode electrochemical cell^31-33. Electrochemical-induced oxidation and curing of catechol-containing adhesive was only demonstrated very recently^34-35. Also, tuning the oxidation state of catechol by electrochemistry was used in controlling the movement of a hydrogel actuator³⁶. Additionally, direct electrochemical oxidation of catechol-functionalized coatings was used to promote cell attachment and peptide immobilization through oxidation-induced crosslinking^37-38. However, to our knowledge, there is no direct evidence that correlated the effect of in-situ electrochemical oxidation on the adhesive property of catechol.

Here, we tested the feasibility of deactivating the adhesive property of catechol by in-situ electrochemical oxidation, when the adhesive is in direct contact with a surface substrate (Scheme 1). Johnson-Kendall-Roberts (JKR) contact mechanics test was used to evaluate changes in the interfacial binding property of catechol-containing adhesive with a Ti sphere in the presence of an interfacial buffer solution (pH 7.5) (Figure 1). The Ti sphere also served as a conductive electrode for applying electricity to the adhesive, while a platinum (Pt) wire was utilized as the counter electrode. The effect of applied voltage and current levels, exposure time, and salt concentration of the interfacial buffer on the adhesive property of catechol was determined. Electrochemical oxidation of catechol was confirmed using UV-Vis diffuse reflectance spectroscopy and tracking the production of hydrogen peroxide (H₂O₂), a byproduct generated during catechol oxidation³⁹. Finally, contact mechanics testing and oscillatory rheometry were performed to determine the effect of applied electricity on the mechanical properties at the interface and in the bulk of the adhesive, respectively.

Scheme 1:

(a) The reduced form of catechol exhibits strong adhesion to metal surface in the absence of applied electrical field and (b) in-situ electrical application oxidizes catechol to its quinone form, which exhibits weak adhesion.

Figure 1:

(a) Photograph and (b) schematic representation of custom-built JKR contact mechanics setup used for adhesion testing and in-situ electrochemical oxidation.

Results:

The model adhesive used in the experiment consisted of 10 mol% dopamine methacrylamide (DMA, Figure S1) copolymerized with methoxyethyl methacrylate (MEA) and 3 mol% methylene bis-acrylamide (MBAA) as the crosslinker (Figure S2). DMA consists of a catechol moiety linked to a polymerizable monomer. DMA remained redox active and exhibited similar anodic and cathodic peaks associated with the formation of semiquinone and quinone that were also found in N-acetyldopamine based on cyclic voltammetry experiments (Figure S3). FTIR spectra confirmed the presence of catechol (−OH 3383 cm⁻¹ and benzene rings 1486, 1583, and 1609 cm⁻¹) in the adhesive (Figure S4).

In a typical JKR contact mechanics test, the Ti sphere was brought into contact with the adhesive wetted with 200 μL of pH 7.5 Tris buffer containing 0.1 M NaCl. The Ti sphere was compressed at 1 μm/s until a maximum preload of 10 mN was reached and then retracted at the same speed (Figure S5). In the absence of applied electricity, the virgin adhesive exhibited a maximum tensile load of nearly 70 mN despite a preload of only 10 mN (Figure 2a). This indicated that catechol contributed to strong interfacial binding in the presence of water. To determine the effect of in-situ applied electrical filed on adhesion, up to 9 V was applied to the adhesive using the Ti sphere as the cathode and a Pt wire as the anode while the Ti sphere was still in direct contact with the adhesive. The interfacial binding properties of the adhesive decreased significantly with increased levels of applied voltage. The work and strength of adhesion (W_adh and S_adh, respectively) decreased by 96% and nearly 100%, respectively, at the highest voltage tested (9 V) when compared to the virgin adhesive (Figure 2b-c). S_adh decreased near linearly with increased applied voltage in the range of 1 to 5V (Figure S6). Adhesives oxidized by incubation in a pH 9 solution also exhibited nearly zero adhesive properties (Figure S7), confirming that the reduced form of catechol is responsible for strong wet adhesion. As expected, the measured current increased with increased applied voltage (Figure 2d).

Figure 2:

(a) Typical JKR contact curves, (b) W_adh, (c) S_adh, and (d) recorded current for JKR contact mechanics testing performed with in-situ application of voltage for 1 min. Voltage was applied using a Ti sphere and a Pt wire as a cathode and an anode, respectively, in the presence of an interfacial buffer (pH 7.5). * p < 0.05 when compared to the virgin adhesive (0 V). ^$ p < 0.05 when compared to all other group.

The images of catechol-containing adhesive after exposure to applied voltage can be found in Figure S8 a-c. The surface of the adhesive surrounding the cathode (Ti sphere) appeared red in color, an indication of catechol oxidation⁴⁰. With increasing applied voltage, this localized color change increased in surface area. On the other hand, the surface of the adhesive remained white near the anode (Pt wire). Litmus paper was used to determine the change in solution pH near each electrode. The solution near the cathode became basic (green color; pH ~ 10.0 for 9 V), whereas the solution became acidic (orange color; pH ~ 3.0) near the anode. Samples that did not contain catechol remained white, even though the litmus paper indicated similar pH changes in the interfacial liquid after the application of voltage (Figure S8 d-f).

Elevated applied voltage likely resulted in electrolysis of water (Scheme 2), which produced gaseous oxygen (O₂ (g)) and proton ion (H⁺) near the anode (Pt) and gaseous hydrogen (H₂ (g)) and hydroxyl ion (OH⁻) near the cathode (Ti)⁴¹. Generation of O₂ (g) and H₂ (g) results in bubble formation near the electrodes, which can cause the measured current to fluctuate^42-43. Current fluctuation was clearly visible with an applied voltage of 9 V (Figure 2d). Accumulation of OH⁻ near the Ti sphere and H⁺ near the Pt wire changed the local solution pH^44-48. Increased level of local solution pH near the Ti sphere promoted catechol oxidation and subsequently decreased the measured adhesive property. The interfacial buffer contained 0.1 M NaCl. The oxidation of chloride ions to chlorine gas (Cl₂ (g)) occurred near the anode⁴⁹ and would have minimal impact on the interfacial binding between Ti (cathode) and catechol.

Scheme 2:

Schematic representation of the proposed mechanism of in-situ deactivation of catechol-containing adhesive. (a) The application of the electrical field initiates water electrolysis, which produced hydrogen (H₂ (g)) and hydroxyl ion (OH⁻) near the cathode (Ti) and oxygen (O₂ (g)) and proton ion (H⁺) near the anode (Pt). Accumulation of OH⁻ near the Ti sphere and H⁺ near the Pt wire change the local solution pH to basic and acidic, respectively. (b) Catechol demonstrates strong adhesion toward Ti surface when no electrical field was applied. (c) After the application of an electrical field, the solution near the cathode (Ti) becomes basic (green), which resulted in the oxidation of catechol to its quinone and poorly adhesive form. The solution near the anode (Pt) became acidic (orange).

To confirm catechol oxidation, UV-Vis diffuse reflectance spectra of catechol-containing adhesive was obtained (Figure 3a). Catechol-containing adhesives exhibited a broad peak (400–600 nm) after exposure to elevated levels of applied voltage (5 and 9 V). Samples oxidized by incubation in a pH 9 solution also exhibited a broad peak in the same range of wavelength. This peak can be attributed to the oxidation products of catechol such as quinone as well as the dimers formed from oxidative crosslinking of catechol⁵⁰. The adhesives exposed to 1 V did not exhibit a similar peak potentially due to the reduced level of localized oxidation. Adhesives treated with a pH 3 solution exhibited increased reflectance at a wavelength lower than 400 nm due to the presence of catechol (λ_max = 280 nm)⁵⁰.

Figure 3:

(a) UV-vis diffuse reflectance spectra of freeze-dried adhesives after exposure to applied electrical potential. For comparison purposes, samples were immersed in either pH 3 or 9 solutions for 24 h to determine the spectra for the adhesive in the reduced (pH 3) and oxidized (pH 9) forms. (b) H₂O₂ concentration in the interfacial Tris buffer during in-situ application of voltage. * p < 0.05 when compared to other adhesives tested at same exposure time.

The concentration of H₂O₂ in the interfacial Tris buffer solution was determined to further confirm the oxidation of catechol. H₂O₂ is a byproduct generated during catechol oxidation (Figure S9)³⁹. The measured concentration of H₂O₂ increased with increasing applied voltage and exposure time (Figure 3b), which is indicative of an increased level of catechol oxidation. The highest amount of H₂O₂ was measured for the adhesive exposed to 9 V for 2 min (45 μM). Without any applied voltage (0 V), no H₂O₂ was generated from the adhesive. When electrical potential was applied to the interfacial buffer in the absence of the adhesive, a small amount of H₂O₂ was detected and the concentration increased with increasing applied voltage (Figure S10). H₂O₂ could be generated due to the reduction of O₂ (g) or the combination of H₂ (g) and O₂ (g)^51-52. However, the amount of H₂O₂ generated by the electrolysis of water was significantly lower than the amount generated by electrochemical oxidation of catechol.

Contact mechanics tests and oscillatory rheometry experiments were performed to investigate the effect of electrochemical oxidation on the mechanical properties at the interface and within the bulk of the adhesive, respectively, after the application of electricity. The contact curves (Figure S11) were fitted with the Hertzian model (Equation S8) to determine the Young’s modulus of the adhesive at the interface (E_inf) ⁵³. E_inf increased with increased applied voltage (Table S1). At the highest voltage level (9 V for 1 min), E_inf reached 76 kPa, which is nearly 4 folds higher when compared to that of the virgin adhesive (E_inf = 20 kPa). On the contrary, the storage modulus (G′) remained around 4 – 5 kPa for all samples (Figure S12). This indicated that the overall stiffness of the adhesive did not change and the increase in stiffness occurred only at the interface. Cross-sectional views of the adhesive samples also confirmed that oxidation (red color) occurred only near the interface (Figure S13), with a maximum depth of around 0.2 mm from the surface of the adhesive treated with 9 V. Additionally, adhesive property measured on the opposite side of the oxidized surface was not affected by the electrochemical treatment (Figure S14). These results collectively indicated that oxidation occurred mainly at the interface where electrochemical oxidation took place.

Electrochemical oxidation-induced catechol crosslinking was further confirmed using a model linear copolymer, p(DMA-co-MEA) (Figure S15). Fourier-transform infrared spectroscopy-attenuated total reflection (FTIR-ATR) spectrum of p(DMA-co-MEA) exhibited three peaks at 813, 782, and 754 cm⁻¹ (Figure S16), which were attributed to the C-H bending of the aromatic ring⁵⁴. The intensity of these peaks reduced with increased level of applied voltage, potentially due to the oxidative crosslinking of catechol. Additionally, p(DMA-co-MEA) was initially soluble in both aqueous and various organic solvents (e.g., tetrahydrofuran, dimethyl sulfoxide, and chloroform) and became insoluble in these solvents after electrochemical treatment (Figure S17), potentially due to chemical crosslinking of catechol. ¹H NMR spectra of N-acetyldopamine also confirmed catechol crosslinking. After electrochemical treatment (9V for 1 min), the peaks associated with the aromatic protons became broad in shape as opposed to sharp peaks found in the unoxidized N-acetyldopamine (Figure S18). This result indicated that oxidative crosslinking and polymerization of the catechol had occurred⁵⁵. Oxidized catechol is highly reactive and can irreversibly crosslink to form oligomers consisted of up to 6 catechol residues⁵⁶.

The temperature of interfacial buffer increased by only 1.22 ± 0.1 °C during electrochemical treatment using the highest applied voltage of 9 V (Figure S19). The increase in temperature was relatively low, potentially due to the high specific heat of water. Although we have previously demonstrated that the rate of catechol oxidation increases at an elevated temperature³⁹, the small increase in temperature associated with electrochemical treatment likely had a minimal effect on catechol oxidation.

Three supplementary videos captured the in-situ deactivation of the adhesive as a result of applied electrical potential (Supplementary Videos 1-3). A 1-g weight consisted of the Ti sphere and the indenter shaft was attached to the adhesive in the presence of interfacial buffer, in a setup that resembles the inverted JKR testing condition. A green light-emitting diode (LED) was attached in series with the electrodes and the source meter, which served as an indicator when the current was passing through the adhesive joint. The LED light turned off when the weight was detached. When a voltage of 9 V was applied, the weight detached within 7 sec. The detachment of the weight occurred due to a decrease in adhesive strength to the extent that the adhesive could no longer support the weight. When a lower voltage was applied, a significantly longer amount of time was required to deactivate the adhesive (23 sec and 2 min 23 sec for 5 and 1 V, respectively). These videos corroborate with JKR contact mechanics test results where higher applied voltage resulted in a larger decrease in adhesive property (Figure 2).

JKR contact mechanics test was also performed using the adhesive-contacting Ti sphere as the anode and the Pt wire as the cathode. Although both W_adh and S_adh values decreased with increasing applied voltage (Figure 4 and Figure S20), the extent of decrease in these values was significantly lower (~50% reduction at 9 V) when compared to adhesion tests performed using Ti sphere as the cathode (>96% reduction at 9 V; Figure 2). The anodic Ti oxidized catechol, which resulted in decreased adhesion. However, the anodic reaction associated with electrolysis of water also generated H⁺ and acidified the solution surrounding the anode⁴¹, which counteracted catechol oxidation. Cl₂ can also be generated near the anode from chloride ion, which can be further transformed into hypochlorite, chlorite, chlorate, or perchlorate ions.⁵⁷ These ions can potentially reduce the solution pH and counteract catechol oxidation. The photographs of the adhesive confirmed pH-induced oxidation near the cathodic Pt wire (red color) while the surface of the adhesive near the anodic Ti sphere only exhibited slight discoloration (Figure S8 g-i). UV-Vis diffuse reflectance spectra also did not exhibit similar peaks (400 – 600 nm) associated with pH-oxidized catechol (Figure S21). Given that using a cathodic Ti sphere as a contacting surface more effectively oxidized catechol and deactivated its adhesive property, subsequent experiments were performed using Ti as the cathode.

Figure 4:

(a) W_adh and (b) S_adh of catechol-containing adhesive tested with in-situ application of voltage for 1 min, using Ti sphere as the anode and Pt wire as the cathode, in the presence of an interfacial buffer (pH 7.5). * p < 0.05 when compared to the virgin adhesive (0 V).

In addition to varying the level of the applied electrical potential, the effect of other factors such as applied current level, exposure time, salt concentration of the interfacial buffer, and catechol concentration on adhesion was also investigated. Up to 10 mA of current was applied to DMA-containing adhesive. Both W_adh and S_adh decreased with increased level of electrical current application (Figure S22). As expected, the measured voltage across two electrodes increased with increased applied current (Figure S22d). When the adhesive was exposed to increased length of exposure to 5 V of electrical potential, the interfacial binding properties of the adhesive also decreased with increasing length of exposure to the electrical field (Figure S23). At the highest exposure time of 2 min, W_adh and S_adh values were reduced by 86% and 64%, respectively, when compared to the virgin adhesive (Figure S23). Similarly, increasing NaCl concentration in the interfacial buffer from 0.1 M to 1 M also resulted in further decreased adhesion as a result of increased conductivity in the adhesive joint as indicated by an increase in the measured current (Figure S24). Regardless of the treatment, the decrease in adhesion property was accompanied by an increased extent of catechol oxidation as indicated by an increased area of localized color change (red color) on the sample surface (Figures S22 e-g, S23 d-f, and S24 e-g). Finally, adhesives were prepared with varying DMA concentration (5 to 15 mol%). For the virgin adhesive, both W_adh and S_adh increased with increasing catechol content, indicating that catechol is responsible for strong interfacial binding in the presence of an aqueous solution (Figure S25). After electrochemical treatment (5 V for 1 min), both W_adh and S_adh decreased by 55-65% and 60-85%, respectively, when compared to the virgin adhesive.

To evaluate whether the electrochemically oxidized catechol can be converted back to its adhesive and reduced form, DMA-containing adhesive was incubated in a pH 3 solution for 24 h after electrochemical treatment. Based on JKR contact mechanics test (Table S2), pH 3 treated adhesive did not demonstrate a recovery in its adhesive property. On the contrary, the measured adhesion values decreased further. This is potentially due to the continued oxidation of catechol by the basic solution that was trapped within the adhesive network. Similar observation has been previously reported in pH-induced oxidation of catechol-containing adhesive²⁶. UV-vis diffuse reflectance spectra (Figure 3a), FTIR-ATR of p(DMA-co-MEA) (Figure S16), ¹H NMR of N-acetayldopamine (Figure S18) and increase in E_inf (Table S1) collectively suggested that irreversible oxidative crosslinking had occurred, which potentially limited the reversibility of catechol.

Discussion:

Taken together, this work demonstrated for the first time that it is feasible to directly deactivate catechol-containing adhesive by in-situ electrochemical oxidation. Electrolysis of water increased the solution pH near the cathode, which resulted in catechol oxidation (Scheme 2). In an oxygenated aqueous solution, catechol readily oxidizes in a basic solution (pKa = 9.2 and 14 for catechol)⁵⁸. Catechol oxidation occurred even when the adhesive was bound to a surface substrate. While it is possible to deactivate catechol-containing adhesive using electrical potential as low as 1 V, elevated applied voltage was required to rapidly deactivate the adhesive. This is potentially due to the poor conductivity of the model adhesive network (0.57 S/m, Table S3). This requirement may be problematic due to the generation of side reaction products associated with electrolysis of water^{41, 57, 59}. It may be necessary to increase the conductivity of the adhesive network to reduce the need for elevated electrical potential.

A custom-build JKR contact mechanics test setup was utilized to directly apply electricity to the adhesive joint. This setup enabled us to probe the interfacial interaction between the catechol-containing adhesive and Ti, while minimizing the contribution of the bulk property of the adhesive on the measured values. This differs from the oft-used lap shear adhesion testing or peel test, where the bulk mechanical property plays a significant role in the measured adhesion values⁶⁰. Additionally, we were able to characterize interfacial binding properties in the presence of water using a buffer solution of desired pH.

Elevated applied voltage likely resulted in irreversible crosslinking and the electrochemically oxidized catechol was unable to revert to its reduced and adhesive state. Catechol and the oxidized quinone can generate an aryloxy radical which leads to phenol coupling and dimer formation (Figure S27) ^61-62. A temporary protecting group such as a boronic acid, may be required to prevent catechol oxidation and crosslinking. Boronic acid forms pH dependent reversible complex with catechol and has been utilized in the synthesis of catechol-containing polymers⁶³. Additionally, smart adhesives consisting of boronic acid and catechol demonstrated multiple cycles of reversible adhesion, due to the ability of boronic acid to protect catechol from irreversible oxidation at a basic pH^26-28.

Nevertheless, the ability to tune the adhesive property using applied electricity as the external trigger is highly attractive and convenient. An electro-responsive adhesive can potentially be integrated with electronic devices, which will offer users unprecedented control over the interfacial binding properties of the adhesive. In the Supplementary Videos, we demonstrated that it is feasible to deactivate adhesion with a simple “push of a button”. Additionally, there are very limited examples of smart adhesive that can bind to wet surfaces. Catechol is widely used to impart adhesive property to polymers with strong, wet adhesive property^11-12. Although we had demonstrated in-situ deactivation using a model catechol-containing adhesive with an inert polyMEA backbone, it is potentially feasible to adopt the findings reported here to other catechol-based adhesives with a wide range of compositions and architectures.

Conclusion:

We developed a contact mechanics test setup to evaluate the effect of direct application of an electrical field on the adhesive property of catechol. Adhesive property of catechol-containing adhesive decreased with increased level of applied voltage and current, exposure time, and salt concentration of the interfacial buffer. From UV-Vis diffuse reflectance spectra and quantification of H₂O₂ generated at the interface, it was determined that in-situ electrochemical oxidation of catechol contributed to the decreased adhesive property. Electrochemical oxidation promoted catechol crosslinking and increased the Young’s modulus of adhesive near the interface of the adhesive. Catechol oxidation was confined to the interface and the stiffness within the bulk of the adhesive remained unchanged. Irreversible oxidative crosslinking of catechol limited the reversibility of inactivated catechol.