This is the peer reviewed version of the following article: Saiz-Poseu J., Mancebo-Aracil J., Nador F., Busqué F., Ruiz-Molina D.. The Chemistry behind Catechol-Based Adhesion. Angewandte Chemie - International Edition, (2019). 58. : 696 - . 10.1002/anie.201801063, which has been published in final form at https://dx.doi.org/10.1002/anie.201801063. This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Use of Self-Archived Versions. REVIEW The Chemistry behind Catechol-Based Adhesion J. Saiz-Poseu,* J. Mancebo-Aracil, F. Nador, F. Busqué, D. Ruiz-Molina* REVIEW Abstract: The wet adhesion of some marine organisms to especially relevant in mfp-3 and mfp-5 (21%[9] and 27%[10] DOPA almost any kind of surface has aroused increasing interest over content, respectively, for Mytilus edulis), key components of a recent decades. Numerous fundamental studies have been plaque that contacts surfaces during adhesion. DOPA was thus performed to understand the scientific basis of this behaviour, suggested to play a key role on this process, which was where catechols have been found to play a key role. Several experimentally confirmed in subsequent works.[11,12] novel bioinspired adhesives and coatings with value-added Currently, a wide consensus among the scientific community performances have been developed by taking advantage of the considers catechol as the origin of the astonishing adhesion knowledge gained from these studies. Excellent reviews of this capacities of mussels and other marine wildlife. This discovery, issue have been published due to the large amount of interest besides its raising of inherent interest, also represents a starting and generated literature. Nevertheless, no detailed compilation point in the development of novel synthetic wet adhesives and exclusively focused on the complex inner workings of these coatings, a scientific and technical challenge. This discovery is materials exists. Thus, the aim of this work is to review recent especially relevant in areas such as biomedicine, where high investigations that elucidate the origin of the strong and versatile adhesion capacities in aqueous environments are at a adsorption capacities of the catechol moiety and the effects of premium.[13,14,15,16,17] The need to adhere different body parts extrinsic factors playing important roles in the overall adhesion through suturing and wound sealing is a representative process, such as pH, solvent and the presence of metal ions. In example.[ 18, 19 ,20 , 21 ] However, catechols also represent a quite other words, it is our aim to review in detail the chemistry behind easy, reliable and effective way to functionalize surfaces. The paradigm is the in situ polymerization of dopamine under mild the astonishing properties of natural and synthetic catechol- basic conditions, which is broadly used by several different based adhesive materials. groups worldwide. Such polymerization generates a strongly adherent coating (primer) of polydopamine, which can be later functionalized with the aim of transferring given properties (e.g., 1. Introduction hydrophobicity/hydrophilicity or bioactivity) to the substrate.[22] According to the prospects previously described, the number of Many marine organisms have developed natural mechanisms to publications reporting catechol-based adhesives and coatings firmly attach to underwater surfaces. Their objective is to prevent has grown over the years. Most of them have been compiled in their movement by currents and tides so they can perform basic excellent reviews focusing on the material characteristics and vital functions such as feeding and reproduction. Representative applications.[23] An updated version gathering novel examples is examples of organisms with these mechanisms are sandcastle thus outside the scope of the present work. In contrast, at the worms,[1] barnacles [2,3] and mussels,[4,5,6,7] which mostly base time of writing this manuscript, detailed reviews analysing and their adhesion on the secretion of specialized proteins. For correlating the actors affecting the chemistry behind the instance, mussels use natural holdfasts called byssus to attach adhesion of these materials were lacking; these materials onto rocks and almost any kind of surface they find underwater. usually involve far more than catechols and related factors such These anchoring elements are composed of several proteins as their organization[24] and structuring at the nanoscale.[25] With that are secreted during the adhesion process.[1,7] A detailed this aim, we have classified this revision to cover two main study of these mussel foot proteins (mfps) showed that their areas: I) a description and understanding of the different most representative feature is an atypically high concentration of interactions and bonds that catechols establish with surfaces in the catecholic amino acid 3,4-dihydroxy-L-phenylalanine the anchoring step, i.e., the adsorption phenomena and II) (DOPA), which is obtained by the post-translational enzymatic factors that may influence the effectiveness of the overall hydroxylation of tyrosine.[ 8 ] Such a high concentration is adhesion (i.e., adsorption and bulk cohesion) such as catechol content, the presence of other species in the material backbone, [*] Dr. J. Saiz-Poseu, Dr. D. Ruiz-Molina pH, redox activity and the presence of ions in the environment. Catalan Institute of Nanoscience and Nanotechnology (ICN2), CSIC Finally, concluding remarks will gather the different actors and BIST, modulating the effectiveness of these systems while highlighting Campus UAB, Bellaterra, 08193 Barcelona, Spain future approaches to improving catechol-based adhesive E-mail: javier.saiz@icn2.cat dani.ruiz@icn2.cat materials and coatings. Dr. J. Mancebo-Aracil, Dr. F. Nador Instituto de Química del Sur-INQUISUR (UNS-CONICET) Javier Saiz-Poseu graduated in chemistry Universidad Nacional del Sur Av. Alem 1253, 8000 Bahía Blanca, Buenos Aires (Argentina) (2006) and received his PhD (2011) from the Universidad Autónoma de Barcelona. Co- Dr. F. Busqué directed by Félix Busqué and Daniel Ruiz- Dpto. de Química (Unidad Química Orgánica) Molina, his work focused on the synthesis, Universidad Autónoma de Barcelona characterization and macro/nanoscale Edificio C - Facultad de Ciencias, 08193 Cerdanyola del Vallès, Barcelona (Spain) application of catechol-based materials. After expanding the scope of these studies as a postdoctoral fellow, he is currently developing his career at the ICN2. REVIEW balance for the interaction of pyrocatechol and water on α- Daniel Ruiz-Molina earned his PhD on cristobalite (001) and β-cristobalite (111).[ 26 ] These two dendrimers, and after a postdoctoral position hydroxylated silica surfaces were selected because of their at UC San Diego working on switchable similarity to underwater amorphous silica, where surface silicon molecular materials, he secured a permanent atoms rapidly react with water to form silanols. The main results position at the CSIC. More recently, he from this work were that I) pyrocatechol showed a higher affinity moved to the ICN2, where he is leading the (14.15 and 11.65 kcal/mol for the (001) and (111) surfaces, Nanosfun group. His main research areas are respectively) than water molecules (1.98 and 0.57 kcal/mol) for the fabrication of biomimetic hybrid colloids both surfaces, II) the underlying lattice noticeably affects the and surfaces, encapsulation, nanoscale adsorption process (pyrocatechol can establish four hydrogen coordination polymers and smart optical bonds with the (001) lattice but only three with the (111) surface), materials and III) independently of the surface, catechols prefer to stand upright, i.e., nearly perpendicular to the surface plane, rather than lay flat. Taking into account this geometry and an average energy of approximately 3.7 kcal/mol per bond, it is feasible to 2. The nature of the catechol-surface consider hydrogen bonding (typically between 2.4-6.2 interaction kcal/mol)[27] the most important interaction with amorphous wet silica surfaces. Interrelated calculations[ 28 ] and ab initio Although very simple in its structure, composed of only a molecular dynamic (MD) simulations performed by Ganz et al. [29] benzene ring bearing two neighbouring hydroxyl groups, also supported these results. The authors demonstrated that catechol effectively interacts with almost any kind of surface. pyrocatechol displaces preadsorbed water molecules from the The secret to such a high versatility is associated with an substrate by competitive hydrogen bonds and the help of exceptional surface adaptation ability: depending on the dispersion forces from the phenylene ring. [29,30] Finally, both DFT substrate nature, catechols can be adsorbed by means of and MD calculations noted the torsion capacity of the hydroxyl different chemical interactions (Figure 1), ranging from weak bonds as the origin of the enhanced versatility of catechols to dispersion forces to covalent bonds, which are discussed next. effectively establish hydrogen bonds with different underlying lattices since they can freely rotate with respect to the phenylene ring to find an optimal adsorption geometry.[26,29] In addition to theoretical approaches, experimental methods also confirmed the role of hydrogen bonding in the interaction of catechols with polar substrates. For instance, Israelachvili et al.[4] used an atomic force microscope (AFM) and a surface force apparatus (SFA) for studying the adhesive properties of two mfps (mfp-1 and mfp-3) from M. edulis on a poorly adherent mica substrate at the microscale. At a slightly acidic pH, AFM experiments confirmed the fast adhesion of mfp-1 to this aluminosilicate, whereas SFA evidenced two different roles: mfp-1 behaved as a protective coating, and mfp-3 was the real gluing agent (Figure 2). Although both proteins have similar DOPA content, the smaller size of the latter favours diffusion into the gaps of the substrate, allowing mfp-3 to form more binding sites because of its higher mobility and flexibility. This example showed how the presence of DOPA, though necessary, is not Figure 1. Schematic representation of the four main catechol-surface enough for ensuring effective gluing. Regarding the nature of the interactions described in this section: A) hydrogen bonding, B) -stacking, C) interactions between these proteins and the mica substrate, the coordination and E) an example of covalent bonding with surface amines (via authors concluded that the energy values and the reversibility of Michael-type addition). the process were indicative of non-covalent interactions, probably hydrogen bonding, though electrostatic forces could 2.1. Non-covalent interactions not be excluded. Waite et al.[31 ] determined the adhesion capacity of mfp-1 to 2.1.1. Hydrogen bonding amorphous titania and mica surfaces with an SFA. This protein adhered strongly and reversibly to both substrates at a slightly Catechol groups have a strong affinity for hydrophilic surfaces acidic pH in a hydrated environment, with the adhesion strength thanks to their capacity to establish hydrogen bonds. How these of TiO2 being roughly double that of mica. The reasons for this interactions can be effective in an aqueous environment with a difference are I) the cumulative contribution of coordination and large excess of competing water molecules is unclear. To hydrogen bonds in the case of TiO2[32,33,34] and II) the surface answer this question, researchers performed theoretical studies. roughness (0.8 nm vs. 0.2 nm for TiO2 and mica, respectively) Density functional theory (DFT) calculations analysed the energy REVIEW conditioning the number of effective bonding sites and thus the A) adhesion.[ 35 , 36 , 37 ] The same group also studied the role of catechols in the TiO2 and hydroxyapatite adsorption of a peptide whose isoelectric point and hydrophobic characteristics mimicked those of mfp-3.[25] The adsorption was followed by a crystal microbalance with dissipation (QCM-D), and the results demonstrated that the presence of the catecholic amino acid DOPA accelerates the kinetics of the peptide adsorption on the hydroxyapatite surface. Catecholic peptides exhibited longer binding lifetimes than the B) analogous tyrosine model thanks to bidentate hydrogen bonding capacity, which contributes a stronger interaction to displace water molecules from the surface.[25] The inability of the tyrosine- based peptide to remove surface moisture was demonstrated by attenuated total reflection infrared (ATR-IR) spectroscopy with a flow cell set-up. The broad and positive intensity of the νOH C) mode, assigned to interfacial water molecules at the (1) hydroxyapatite, remained with time. These molecules would lead to weak hydrogen bond-based adsorption. In contrast, the intensity of the 4000 and 2500 cm−1 spectral regions of the DOPA-containing peptide decreased with time, indicating the removal of liquid-like interfacial water molecules. The same (2) authors stressed the relevance to adhesion of other contributing forces such as cation-π interactions between the aromatic ring of DOPA/Tyr residues and Ca(II) from hydroxyapatite. The importance of hydrogen bonding was also evaluated studying the adhesion capacity of DOPA-containing mfp-3 at different pH values.[38] By means of an SFA, Waite et al. showed (3) that the adhesion was enhanced at an acidic pH value (3), where the prevalent state is the catechol form. In this case, the distance between substrate oxygen atoms (approximately 0.28 nm) perfectly matches the distance between the hydroxyl groups of the DOPA moiety (approximately 0.29 nm). This relation Figure 2. A) Mussels of three species adhered to mica. The enlarged image together with the previously mentioned torsion adaptability of the shows one of the mussels bearing the weight of a mica sheet plus three hydroxyl groups to the underlying surface lattice[26,39] definitely congeners by means of only three byssal threads. B) Schematic drawing of a favours the formation of bidentate hydrogen bonds. When the byssal thread attached to a substrate. C) Two mica surfaces bridged by mfp-3. pH is increased to 5 or oxidizing agents are added, the quinone (Left) Fringes of equal chromatic order (FECO) images during an mfp-3 form prevails over the catechol. At this point, the adhesion bridging experiment. (Right) Schematic drawings of corresponding molecular decreased by more than 60%, confirming the importance of processes occurring at the junction: (1) Two mica surfaces in flat adhesive hydrogen bonding in the adsorption mechanism. This fact was contact in air. (2) Same surfaces after an mfp solution was injected between also supported by Robertson et al.,[ 40 ] who compared the them. (3) A configuration of surfaces immediately after release from adhesive adsorption properties of different catecholic and non-catecholic contact. Reproduced with the permission from Ref. [4]. Copyright The National materials. A comparison of pyrocatechol and 3-hydroxytyramine Academy of Sciences of the USA. with a mono-hydroxylated phenol showed that the catechol- bearing molecules adsorbed more strongly (approximately three Hydrogen bonding is also relevant on metal surfaces.[41, 42 ] orders of magnitude more strongly). The authors also suggested Rinderspacher et al.[43] have studied the adsorption of catechol that the main mechanism is a cooperative binding through and other phenolic compounds on hydroxylated and non- divalent hydrogen bonds on the hydroxyapatite crystal edges hydroxylated alumina using MD simulations. In the absence of (occupancy previously reported for analogous experiments on water, a bilayer was formed on both alumina surfaces: the alumina),[ 41 ] though additional contributions from electrostatic closest layer to the surface establishes hydrogen bonds with the and aromatic interactions could not be discarded. In this sense, aromatic rings lying flat, whereas the second layer remains the authors noted a feasible close packing of the catechols upright with the hydroxyls pointing at the surface. When water within the monolayer favoured by an almost vertical position of was considered, catechols displaced water molecules from the the molecules. This orientation would allow a cooperative substrate. In the case of non-hydroxylated alumina, the less interaction between aromatic rings, thus enhancing the rate and hydrophilic surface favours hydrophobic interactions, whereas energy of the adsorption by pulling other catechol groups from hydroxylated alumina gives rise to the strongest adsorption, solution into the monolayer.[40] mostly by cooperative hydrogen bonding with the surface REVIEW hydroxyl groups. The authors also remarked that among all the phenol. They also concluded (supported by previous results)[58] studied phenolic compounds, catechol establishes the most that graphene-like surfaces act as amphoteric adsorbents, being hydrogen bonds with the alumina surface per unit area. able to interact with both -electron donors and acceptors. The Finally, hydrogen bonding has also been reported for organic stronger the donor/acceptor character of the molecular  system substrates bearing polar groups. For instance, coating clays with is, the stronger the - interaction with the surface is. Thus, polydopamine can improve their dispersibility into some catechols per se, having marked  electron donor character, are polymeric matrices, though certain contributions of covalent excellent candidates for -stacking anchoring elements. bonding should not be completely excluded.[ 44 , 45 , 46 ] Wilker et Taking advantage of this feature, He et al.[59] have successfully al.[47 ] also studied the basic interaction of a synthetic DOPA- functionalized CNTs using a layer-by-layer approach with based polymer on different interfaces by sum frequency polydopamine. Characterization with transmission electron generation (SFG) vibrational spectroscopy. An enhanced microscopy (TEM) and X-ray photoelectron spectroscopy (XPS) ordering of the catechol moieties at the interface of demonstrated that CNTs are almost fully coated with an 8 nm poly(allylamine)/air was found and attributed to hydrogen thin polydopamine layer that enhances their dispersibility in bonding between DOPA hydroxyl groups and the primary amine water and their biocompatibility. A similar approach was followed groups of the surface polymer. by Lee et al.[60] to achieve a good colloidal dispersion of CNTs in water, allowing for their manipulation and incorporation into 2.1.2. - interactions conductive nanocomposite foams. In contrast, catechol-based coatings have also been used to obtain CNTs with strong When a catecholic species approaches an organic surface hydrophobic character. For instance, Ruiz-Molina et al.[61] used lacking polar groups, weak adsorption forces come into play. For the -stacking anchoring capacities of a polydopamine-like example, Israelachvili et al.[ 48 ] have demonstrated using polymer, obtained by polymerization of an alkylcatechol with NH3, combined experimental and theoretical studies that van der to effectively coat CNTs. The presence of the alkyl chains Waals and hydrophobic interactions play an important role in the provided the resulting material with hydrophobicity (Figure 3). In adsorption of mussel-derived peptide adhesives on wet organic this case, hydrophobic interactions can be discarded as the surfaces. However, among the different weak interactions that anchoring driving force since the coating process is performed in can be established, -stacking with other  systems such as non-polar solvents.[61] In addition, related studies with other non- graphite,[ 49 ] carbon nanotubes (CNTs)[ 50 , 51 , 52 , 53 ] and catecholic aromatic systems have confirmed that - stacking graphene[ 54 , 55 ] deserves special consideration because of the per se is strong enough to achieve efficient anchoring to this potential technological applications of these materials. kind of substrate.[52,53] Ruiz-Molina et al.[49] have studied the interactions of three different alkylcatechols with a highly oriented pyrolytic graphite (HOPG) surface by scanning tunneling microscopy (STM) at the A) solid/liquid interface. The studied molecules were synthesized bearing heptadecyl chains with the aim of favouring their ordering and stabilization on the HOPG surface. STM showed that monoalkylated species containing a bulky tert-butyl group, which hinders the catechol head, were unable to form structures on the surface. However, monoalkylated species lacking head hindrance formed well-defined patterns of rows with the catechol rings lying flat parallel to the surface. Thus,  interactions play an essential role in the adsorption of these alkylcatechols on B) HOPG since the parallel orientation of the catechol and the surface  system is needed. Finally, the strength of this interaction was also demonstrated by MD-ABF simulations, where a single molecule was transferred from the graphite to the bulk solution and the last part to leave the surface was shown to be the catechol ring.[49a] Alternatively, Ma et al.[56] have studied the adsorption of different aromatic compounds (pyrocatechol among them) on aqueous dispersions of multi-walled CNTs and powdered activated carbon and found that - stacking prevailed over other non- Figure 3. A) TEM images of multi-walled carbon nanotubes (MWCNTs) covalent interactions, though the influence of electrostatics coated with a polymer obtained after the reaction of 4-heptadecylcatechol with cannot be ruled out because of the complex adsorption trends. NH3. The green arrows mark the MWCNT wall; the orange arrows point at the Similar results were obtained by Xing et al., [57] who also studied coating thickness. B) Different behaviour of blank and treated MWCNTs the adsorption of several phenolic compounds on CNTs in water. dispersed in water and ethyl acetate. Reproduced with the permission from The adsorption constants of the non-aromatic molecule Ref. [61]. Copyright Wiley. cyclohexanol indicated significantly less affinity than those of REVIEW SFG studies of a DOPA-containing adhesive on deuterated Finally, combined DFT and STM studies[65] showed that isolated polystyrene (d8PS)[47] demonstrated that catecholic materials catechols tend to adsorb as a bridging bidentate structure can also interact by  stacking with polymeric matrices rich in parallel to the Ti5c rows on the rutile surface, with the benzene aromatic rings. After the adhesive matrix was cross-linked on the ring perpendicular to the surface plane. This work also d8PS surface, the intensity of SFG carbonyl peak at demonstrated the ability of catechols to easily exchange protons approximately 1663 cm-1 remained constant, whereas the with surrounding oxygen atoms, allowing them to move across intensity of the quinonic C=C stretching band at approximately the surface. Beyond demonstrating the diffusion capacity of 1610 cm-1 increased. The authors stated that this result these molecules on rutile, these results also brought relevant indicated an enhanced ordering of the quinone rings as a result information about the above-mentioned interconversion between of - stacking with the underlying aromatic d8PS rings. monodentate and bidentate coordination structures. [34] Finally, aromatic systems also strongly interact with cations.[62] In another interesting study by Grätzel et al.,[66] the adsorption of Nevertheless, other than a pioneering work by Waite et al.[25] pyrocatechol on an anatase (101) single-crystal and the reporting cation- interactions as a feasible adsorption driving crystallographically equivalent rutile (110) was compared. The force, catechol anchoring to surfaces by these means has not authors deposited the catechol by evaporation and been commonly described. In contrast, this family of interactions characterized it by XPS and near-edge X-ray absorption fine has been consistently used to enhance the bulk cohesion of structure (NEXAFS) experiments. The authors crosschecked the catechol-based adhesives with cations coming from both the collected data with DFT-optimized geometries to conclude that polymer backbone itself and the medium (vide infra). pyrocatechol binds more strongly to the anatase surface than to the rutile surface. Additionally, spectra obtained for O 1s 2.2. Chemical bonding suggested that pyrocatechol predominantly adsorbs at high coverages on both surfaces with a bidentate geometry (Figure 2.2.1. Coordination 4); both pyrocatechol O atoms are chemically equivalent as they appear as only strong peak, at 531.9 eV for anatase and 531.3 The interaction of catechols with metallic materials occurs eV for rutile. This hypothesis was also supported by the fact that mainly through coordination bonds between the hydroxyl the hydroxyl peak (532.8 eV for anatase and 532.9 eV for rutile) oxygens and the metal atoms of the surface with a strength that is small and probably includes a significant contribution from strongly depends on the metal. Though several metallic and surface hydroxyl groups bonded to titanium atoms. Thus, no metal oxide surfaces have been the focus of research regarding significant amount of monodentate coordination structures this issue, titanium oxide is one of the most widely studied bearing unreacted hydroxyls were present. because of its relevance in biomedical, catalysis, dye-sensitized solar cell (DSSC) and electrochemical applications, among others.[63] A) B) Titanium oxide. TiO2 is found in different forms, such as anatase and rutile. Since the underlying lattices of both structures are different, each having the possibility to present additional different face planes, specific studies should be performed for each case.[26] For example, Bowler et al.[39] developed DFT calculations to investigate the adsorption of pyrocatechol on the (100) rutile plane. The results showed that the most stable conformation is the monodentate coordination mode, whereas bidentate coordination is only favoured at defect sites where an oxygen atom has been removed from the surface. These results C) differ from those obtained by Diebold et al. [34] on a (110) rutile surface. The authors reported that a bidentate configuration is favoured at low surface coverages, whereas full coverages favour monodentate coordination. However, this assumption does not explain the observed experimental data, which suggested a mixed structure of alternating monodentate and bidentate coordination. Since both geometries are very close in energy, monodentate and bidentate structures could easily interconvert via proton exchange with the underlying substrate. Figure 4. A) Pyrocatechol molecule (optimized geometry). Grey spheres are Thus, the adsorption of catechols through coordination would carbon atoms, red spheres are oxygen atoms, and white spheres are not only depend on the surface but also on its coverage. Time- hydrogen atoms. B) A pyrocatechol molecule adsorbed on an anatase TiO2 dependent DFT calculations[ 64 ] demonstrated that molecular (101) cluster. C) Schematic picture of the three structures catechol may adopt adsorption by hydrogen bonding represents a less favourable on a TiO2 surface. Reproduced with the permission from Ref. [66]. Copyright situation in these systems. American Chemical Society. REVIEW Although photoemission data are not enough to differentiate Therefore, though the electronic properties of the system can be bridging and chelating bidentate coordination modes, NEXAFS tuned by the functionalization of the catechol ring, the stability of simulations supported the bridging bidentate adsorption as the the coordination bond is not considerably affected. most realistic mode. Regarding geometry, DFT calculations Finally, with regard to the translation from discrete catechols to showed a 27 ± 6° tilting of the aromatic rings on anatase and a structures that are more complex and functional, Waite et al. [25] 23 ± 8° tilting on rutile. These data were in agreement with used QCM-D and flow ATR-IR to study the adsorption of an previous results by Diebold et al.[34], where angles of ± 15-30° to mfp3S-pep peptide on TiO2 (anatase). These peptides adsorbed the surface normal were found using angle-resolved ultraviolet as a mixture of bridging bidentate coordination structures where photoelectron spectroscopy (ARUPS). hydrogen bonding and electrostatic interactions all contribute. In Catechol derivatives with greater structural complexity than another work, Messersmith et al.[70] improved the resistance of pyrocatechol have also been studied. For instance, studies of TiO2 surfaces to the adhesion of proteins by attaching the adsorption of dopamine on anatase (101) by DFT methoxypolyethylene glycol (mPEG) strands tailored with calculations[33] and on a rutile (110) single-crystal with catecholic DOPA. The authors demonstrated by XPS, synchrotron XPS and NEFAXS[ 67 ] showed that the bidentate spectroscopic ellipsometry (ELM) and optical waveguide coordination is the most preferable adsorption mode on both lightmode spectroscopy (OWLS) that fast and irreversible substrates. Regarding the orientation, analysis of the NEXAFS coordination occurs through the DOPA moieties. Spencer et spectra revealed that dopamine adopts an upright orientation al.[ 71 ] also demonstrated the hydrophilization and with an angle of 78 ± 5°, twisted 11 ± 5° from the (001) direction hydrophobization of a TiO2 surface by anchoring nitrodopamine (Figure 5). This value is higher than that previously obtained for and perfluoro-alkyl-nitrodopamine, respectively. Static contact pyrocatechol, reinforcing Grätzel’s work.[ 68 ] Similar binding angle measurements clearly showed the effect of these species energies and peak distributions were found in the O 1s on the surface, leading to contact angles of 10º and 105º for photoemission spectra, suggesting that dopamine also these cases, respectively. Additionally, XPS analysis clearly establishes bidentate coordination with titanium atoms. The demonstrated that the catechol moieties coordinated the surface main difference between pyrocatechol and dopamine was titanium atoms. Two different, coexisting coordination structures observed by combining computer modelling with NEXAFS: while were found, according to the presence of two peaks in the N 1s dopamine molecules stand practically upright (± 5°), almost spectra for the NO2 group. This result was in agreement with parallel to the surface normal, pyrocatechol exhibits a different that found for full coverages of pyrocatechol on rutile: [34] a tilt angle by almost 20º. This difference was attributed by the bridging bidentate mode and a monodentate mode. authors[67] to a higher coverage on anatase (101), which led to an increase in the steric repulsion between adsorbate molecules, Water effects. Studies under ultrahigh vacuum conditions are thus forcing them to adopt a more vertical orientation. essential to establish the scientific basis and understand the concepts and fundamentals behind the interactions of catechols with substrates. Nevertheless, for most real cases, the presence of water at the metal-catechol interface must be considered as it may clearly influence their interaction.[ 72 ] Therefore, different theoretical and experimental studies reporting the adsorption of catechols on different surfaces in the presence of water have been reported. All these studies showed that the adsorption energy is considerably decreased by the presence of surface water.[32,73] This phenomenon can be explained considering the energy cost associated with the removal of chemisorbed water molecules from the substrate before catechol adsorption. In addition, the presence of surface solvent molecules can also modify the structure of the final coordination complex, as demonstrated by Metiu et al.[73] Ab initio DFT-MD and ground- Figure 5. A dopamine molecule adsorbed on a rutile TiO2 (110) cluster. state calculations showed that pyrocatechol coordinates in the Titanium is represented by light blue, oxygen by red atoms, carbon atoms by bridging bidentate mode on TiO2 under vacuum conditions, grey, and hydrogens by white. Reproduced with the permission from Ref. [67]. whereas the monodentate coordination mode becomes the Copyright American Chemical Society. preferred configuration in the presence of water. Janković et al.[ 69 ] modified nanocrystalline TiO2 nanoparticles Other metallic surfaces. Though less frequent, the interaction of with catechols bearing different donating and withdrawing catechols with other surfaces has also been reported. For groups. In all the cases, ATR-IR measurements showed that instance, Zhitomirsky et al.[74] studied the adsorption of caffeic both hydroxyls of the catechol ring dissociate and bind the acid on both rutile (110) and wurtzite, a zinc oxide commonly titanium surface in a bridging bidentate structure, as later used for sensors and photovoltaic and optical devices. DFT confirmed by using Job’s method. Interestingly, the dissociation calculations showed that in both cases, the most favourable constants of the complexes showed no significant differences. coordination is a bridging bidentate structure (as previously REVIEW reported for isolated pyrocatechol molecules on TiO2), though a modification of two types of glassy carbon electrodes under higher affinity was exhibited for rutile than wurtzite. In another basic conditions in the presence of catecholic species. The study, strong coordination to zinc oxide quantum dots was also nucleophilic groups generated on the electrodes were able to observed by ligand exchange experiments. FT-IR attack the in situ-generated quinones, most likely by a Michael- measurements demonstrated that catechol replaces almost the type addition. This reaction gave rise to their covalent linking, as 65% of the initially surface-attached acetate.[ 75 ] The authors confirmed by XPS and FT-IR measurements. claimed that this displacement is due to the electron Finally, considering the small number of fundamental studies delocalization caused when the two hydroxide groups originating regarding the covalent anchoring of catechols to surfaces, it is from the dissociation of catechol hydroxyls bind to Zn(II) surface worth to mention the ability of catechol-based materials to atoms. covalently attach other species. In this sense, several works can Wesselink et al.[41] demonstrated that the mechanisms of be found where polydopamine[22, 78 ] or polydopamine-like catechol and other phenolic compounds adsorbing on different materials[ 79 ] are functionalized by the covalent attachment of alumina forms such as gibbsite, boehmite and non-crystalline given species. Though not directly related to the surface alumina were similar. FT-IR measurements showed that the anchoring phenomena, such studies are examples of the catechol-surface bond is chemically affected similarly to the covalent approach potentiality. chelation of Al(III). From these results, the authors concluded that the dominant coordination to the alumina surfaces must be A) 1:1 bidentate complexation. The dominant crystal surfaces of aluminium oxides were unreactive towards catechols, and adsorption was attributed to Al-OH groups located at edge faces. Thus, non-crystalline oxides seem to be more reactive per unit area than the crystalline minerals boehmite and gibbsite. In later work,[42] the adsorption of pyrocatechol violet on boehmite in water was studied, with similar results. According to the observed stoichiometry, the catechol adsorbed on aluminium oxide coordinates as a bidentate complex rather than a B) monodentate complex, which is the same as the Al(III) complexes in solution. 2.2.2. Covalent bonding Even though covalent bonding has not been as widely explored as the previously described families of interactions, there are examples of catechols covalently attaching to organic substrates. In a pioneering and conceptual study, Messersmith et al.[ 76 ] functionalized an AFM tip with a single DOPA residue. The Figure 6. A) F-D curves for the interaction of a DOPA-modified AFM tip with authors studied its interaction with metal oxide and organic an organic surface. The presence of DOPA was confirmed at neutral pH on Ti substrates, the latter material simulated with an amine-modified (top curve), showing the expected pull-off force of 800 pN. The same tip was silicon surface (Figure 6). At a basic pH of 9.7, the AFM strongly allowed to interact with an amine-presenting organic surface at pH 9.7, which attached to the surface after a certain number of pull-off cycles exhibited a pull-off force of 2.2 nN (middle curve). This value is consistent with with a force of 2.2 nN. No adhesion was then detected in the covalent bond rupture. Subsequent F-D curves (after 800 contact/pull-off subsequent pull-off events (Figure 6A), which could be evidence cycles) showed no interactions (bottom curve). B) Scheme of the covalent of a covalent linkage between the oxidized catechol (now in bond formation between DOPA and amines at the organic surface, possibly quinone form) and the amine groups of the substrate. The via a Michael addition-type of reaction. Reproduced with the permission from position of the nucleophilic attack of the amine on the quinone Ref. [76]. Copyright The National Academy of Sciences of the USA. ring was not specified, though based on previous works, the authors stated that a Michael-type addition could occur. In a later work, SFG spectroscopy was used by Wilker et al. [47] to 3. Beyond catechol-surface interactions study the structure of a DOPA-containing adhesive at different interfaces. On poly(allylamine) (PAA), oxidized DOPA moieties 3.1. Catechol content and molecular weight established covalent bonds with the substrate. However, as the infrared signals of imines can overlap with C=C and C=O In addition to promoting surface adsorption, catechol-catechol stretching and -NH2 bending bands, these results are far from interactions contribute to adhesive cohesion through their cross- being conclusive. linking. From a biological point of view, covalent cross-linking In a more practical approach, Zen et al. [77] reported the covalent raises many doubts since the presence of quinones (obtained attachment of single pyrocatechol molecules to an organic after oxidation of the corresponding catechol moiety) is substrate. The experiments consisted of the electrochemical mandatory. However, it is well known that mussels form a REVIEW reductive environment during adhesive plaque formation, Thanks to their chemical structures, the chains of these precisely to prevent DOPA from oxidizing.[80] In 1999, Waite and polymers with enough cohesion forces interact to a high degree. co-workers[81] had already established the first evidence for the However, the gradual addition of small amounts of NaIO4 minority formation of quinone-derived cross-linking in mussel progressively decreased the adhesion of the materials until their byssal plaques. They estimated that this reaction only occurs in total disappearance. The authors claimed that I) some of the 1 DOPA pair per 1800 amino acid residues (average for bulk catechols were oxidized to quinones, which significantly lowered byssal plaque material). However, covalent cross-linking is a the adhesion,[38] and II) the addition of cross-linkers to the commonly followed approach in synthetic catechol-based polymer considerably increased the stiffness.[96] materials. As a general trend, larger catechol content results in To the best of our knowledge, the lowest optimum catechol greater cohesion forces. However, there is a limit; an excess of content within a polymer to induce strong adhesion ranges from catechol may result in excessive cohesion forces, generating a 5-8 mol% for a polylactic acid-based polymer,[75,88,89] dopamine rigid-like material with fewer adhesive surface interactions. In methacrylamide-based polymer[90] and bis-phosphate ester marine mussels,[7,82] optimum DOPA amounts range from 3 to grafted oxetane-based polymer.[91] These polymers can 30% of total amino acid content,[83] being close to 10 mol% on establish hydrogen bonds or van der Waals forces within the average.[84,85] Based on this observation, several groups focused material, increasing cohesion and cross-linking. Hence, less on the synthesis of catechol-based adhesive materials with catechol is required to obtain higher overall adhesion. In contrast, optimized catechol contents.[ 86 , 87 , 88 , 89 , 90 , 91 ] Mussel-inspired one of the highest catechol contents (approximately 30-33 biomimetic polymers have been mainly prepared from mol%,) was used for polystyrene-based polymers,[86,100] where polyesters,[88,89] polyamides,[ 92 ] polyacrylates,[ 93 , 94 , 95 , 96 ] weaker interactions between polystyrene macromolecules occur. polyethylene glycols,[16, 97 , 98 , 99 ] polyoxetanes,[87,91] In between these extremes, several works report optimum polystyrenes[86, 100 ] polypeptides[ 101 , 102 , 103 ] and catechol contents close to 10-15 mol%.[85,87,107] polysaccharides.[104,105] The results defined a range of optimal This parameter is also strongly dependent on the polymer catechol content close to that observed in mfps but with molecular weight (Figure 8).[100] differences depending on the nature of the material used. However, despite this growing number of synthetic polymers mimicking mussel adhesive proteins, the number of detailed and A) systematic studies aimed at elucidating this issue is rather limited. This lack of studies and the wide variety of catechol- based polymers, techniques used for measuring adhesion, substrates, environments of operation and curing conditions, make any conclusive and realistic comparison under standard conditions difficult. For instance, the use of oxidative cross- linkers generally enhances bulk adhesion.[86-89,91,100,106] However, Stuart and Kamperman[90] described the opposite effect for dopamine methacrylamide (DMA) and 2-methoxyethyl methacrylate (MEA) (Figure 7). B) Figure 8. Effect of molecular weight on the adhesion of poly[(3,4- dihydroxystyrene)-co-styrene] before (A) and after (B) cross-linking with Figure 7. Dry adhesion of P(DMA0.05-co-MEA0.95) in the presence of an [N(C4H9)4](IO4). The number average molecular weight (Mn) is shown in red, oxidant (NaIO4). Each data point resulted from an average of three to five and the weight average molecular weight (Mw) is depicted in black. Adhesion measurements. Reproduced with the permission from Ref. [90]. Copyright testing was conducted in shear on aluminium substrates. Reproduced with the Royal Society of Chemistry. permission from Ref. [100]. Copyright American Chemical Society. REVIEW Overall, low molecular weight species have better mobility and therefore provide better wetting for enhanced surface interaction.[ 108 ] However, this improvement is countered by decreased cohesiveness. In contrast, higher molecular weights effect supplementary chain entanglements that contribute to increased cohesive bonding.[109] This concept was nicely shown by Wilker et al.,[100] who studied the effect of molecular weight in poly[(3,4-dihydroxystyrene)-co-styrene] while maintaining a 30 mol% catechol content with and without the addition of cross- linkers.[86] The authors concluded that the absence of cross- linkers generally results in higher adhesion for polymers with higher molecular weights. However, the addition of (Bu4N)IO4 as a cross-linking agent has the opposite effect; under these conditions, greater adhesion is exhibited by polymers with lower molecular weights. On the other hand, if a copolymer based on polystyrene was replaced by polylactic acid, much less catechol was required to achieve high bulk adhesion (approximately 7 mol%), despite having a lower molecular weight.[87] As the authors noted, a smaller amount of catechol moieties is needed thanks to the stronger interchain interactions between polylactic acid strands than between polystyrene chains. 3.2. Roles of other functional groups in the polymeric backbone Beyond catechol, the presence of other species such as cationic, anionic, non-ionic polar and non-polar functional groups are also relevant in the wet adhesion of mfps.[ 110 ] Accordingly, novel approaches to incorporate some of these functionalities in order to achieve synergism with catechol derivatives have been developed mainly in two specific areas: I) surface or structure conditioners and II) cross-linkers, most of which are analysed next in more detail. Surface and structure conditioners. The success of synthetic adhesives in wet environments is most often conditioned by a previous cleaning of the substrates, either by chemical treatments and/or partial drying. [101] An interesting alternative to avoid this preliminary step is the introduction of surface conditioners within the polymeric backbone to improve the close contact with the surface. The objective is to minimize the effect of detrimental agents such as certain solvents or salts. For instance, several works have demonstrated that the presence of cationic groups (such as iminium, ammonium or guanidinium groups) in the material backbone significantly enhances its underwater adhesion (Figure 9), [101,111,112,113,114,115,116,117] which is primarily the result of surface salt displacement from the Figure 9. The synergy of catechol and lysine (Lys) in siderophore adhesion. substrate because of the competition with the cationic groups A) Structure of the Tren scaffold. B to G) The R groups appended to Tren. H) included in the polymer backbone. The flexibility and length of The average adhesion energy required to separate two mica surfaces the cationic and the catecholic groups as well as the distance adsorbed with 1 nmol (20 mM, except where indicated at 200 mM) of the between them are crucial for the adhesion enhancement.[113,115] homologue in buffer at pH 3.3 after 10 minutes of contact. I) DT of the Finding the optimum charge is challenging as an excess can be siderophore monolayer between two mica surfaces at 10 mN/m compressive detrimental due to I) the prevalence of charge-charge repulsions load. The film thicknesses correspond to the adhesion energy displayed in (H). or II) competition between cationic and catechol moieties for A decreased film thickness (<12 Å) indicates that homologues B, C, D, and E negative surfaces.[111,117,118] (200 mM) adsorb, displace hydrated salt at the mica surface, and mediate adhesion between two mica surfaces. Reproduced with the permission from Ref. [113]. Copyright Science. REVIEW Balances of hydrophilic/hydrophobic forces are also relevant for Finally, thiol groups also improve adhesive strength by the favouring interactions with surfaces. For example, if the polymer oxidation of sulfhydryl groups to disulfide bonds or, more backbone is hydrophobic (such as that of polystyrene or specifically, through a click thiol-ene reaction.[ 128 , 129 ] polyoxetane), it tends to avoid hydrophilic surfaces, weakening Independently of the type of interaction, a significant increase in the binding of the catechol moieties.[119] In this way, the design the cross-linking necessarily encompasses significant catechol of copolymers containing catechol residues located between consumption and therefore a reduction of surface adhesion. This hydrophilic blocks (e.g., PEG and polyvinylpyrrolidone (PVP)) problem has been circumvented on mussels limiting DOPA ensures good wetting or adhesive contact (biologic scaffolds and oxidation during adhesive plaque formation by imposing an tissue/implant surfaces).[ 120 ,118,119] Conversely, the inclusion of acidic medium based on the thiol-rich protein mfp-6, which DOPA in a hydrophobic aromatic sequence prevents catechol restores DOPA by coupling the oxidation of thiols to moieties from being exposed to the surrounding aqueous dopaquinone reduction.[128] Mimicking this strategy, some environment, retarding their oxidation.[121] authors have established alternative approaches based on the use of synthetic adherers with a low pKa to favour cross-linking Cross-linkers. Strong or weak nucleophiles in the polymeric at an acidic pH and/or reduce the quinone back to catechol to chain, such as amines or thiols, can react with oxidized improve adhesion.[80] For example, thiourea derivatives, which catechols through Michael-type additions or Schiff base are excellent nucleophiles and exhibit convenient reducing reactions (Figure 10).[ 122 , 123 ] This process enhances the power under acidic conditions, have been used to improve the cohesion by means of covalent cross-linking. Furthermore, some adhesion properties of catechol-conjugated gelatin hydrogels. functional groups such as protonated amines can also establish Also in this context, electron-withdrawing groups (such as cation- interactions, which represent one of the strongest non- chlorine and nitro), which decrease the pKa of the catecholic covalent interactions in water.[62] Cation- interactions enhance hydroxyls, can lead to higher reactivity towards nucleophiles the adsorption of catechols to charged surfaces[ 124 ] and the found on tissue substrates, even at mildly acidic pH values. This cohesive properties of materials rich in aromatic and cationic change results in a significant increase in the interfacial binding functional groups.[125] These interactions can also complement strength of catechols under acidic conditions since they are the weaker adhesion of the quinone moieties in oxidizing minimally hampered by pH changes.[1,130] environments. This supplementation enables interfacial contributions that assist material-independent adhesion,[ 126 ] 3.3. pH and redox: the balance between adsorption and particularly between protonated nitrogen-based groups (cation cohesion forces donors) and polyindolic rings ( donor).[114,127] The redox transition of catechols to quinones upon a pH A) increase and/or the addition of an oxidant is fully established.[131] In the previous sections, the mechanism by which the catechol moiety is mainly responsible for the adhesiveness of catechol- based materials via either non-covalent interactions (Section 2.1) or chemical bonding (Section 2.2) has been shown. However, cohesion comes from the ability of the reactive quinone state to generate cross-linked structures, mainly in synthetic materials (Section 3.1). Within this context, acidic and neutral pH values are necessary to ensure the proper fraction of the catecholic form (Figure 11).[132,133,76] In contrast, basic pH values enable cross-linking, which is necessary to achieve well- B) adjusted cohesiveness. Achieving a proper pH balance is therefore crucial to ensuring optimized adhesion/cohesion ratios and has become the focus of research groups. Waite and co-workers[134] have used an SFA to determine the force-distance profiles and adhesion energies of mfp-3 on a TiO2 surface at three different pH values (pH 3.0, 5.5 and 7.5). The results showed that the strongest adhesion occurs at pH 3. An increase of the pH to 5.5 decreases the measured adhesion forces. However, a further increase to 7.5 recovers the original adhesion value. This behaviour was rationalized as a balance between two opposing simultaneous effects during the pH Figure 10. A) Catechol oxidative chemistry explored from mfp and B) a stable variations: I) DOPA oxidation, which diminishes the adhesion, functional multilayer film prepared by the integration of catechol oxidative and II) the binding of a single DOPA molecule to the TiO 2 chemistry into layer-by-layer assembly. PAH is poly(allylamine hydrochloride), surface, which changes from hydrogen bonding to coordinative and PAA is poly(acrylic acid). Reproduced with the permission from Ref. [123]. bonding, thus increasing the binding strength. The same authors Copyright American Chemical Society. also investigated the pH dependence of the adhesion of the REVIEW electrophoretically slow protein mfp-3 (a hydrophobic variant with a smaller percentage of DOPA than the fast variant). At pH A) 3, where DOPA is stable, both proteins showed similar behaviour, with adhesion to mica proportional to the molar percentage of DOPA. At pH 5.5 and 7.5, however, the adhesion of the mfp-3 slow was almost half of that found for the mfp-3 fast, suggesting that DOPA in mfp-3 slow is less prone to oxidation. Accordingly, cyclic voltammetry measurements proved that the oxidation potential of DOPA in mfp-3 slow is significantly higher than that in mfp-3 fast at pH 7.5. This effect is attributed to the high proportion of hydrophobic amino acid residues near DOPA.[121] In a related study,[110] synthetic copolyampholytes have been obtained and formulated as coacervates for adhesive deposition on surfaces. The synthetized copolyacrylates combine catechol units with amphiphilic and ionic functionalities present in mfp-3 slow. Some of these copolymers formed B) coacervates at pH 4 and showed strong adhesion to mica. Increasing the pH to 7 after coacervate deposition doubled the bonding strength without oxidative cross-linking. UV-visible spectra of the materials showed similar signals at different pH values; quinone absorption bands were not observed. The authors claimed that this is an example of how hydrophobic residues in the copolymer can provide stability against catechol oxidation. Reches et al.[133] also performed a systematic study of the pH effect on the rupture forces between DOPA and related amino acids and a chemically well-defined TiO2 surface using single- molecule force spectroscopy (SMFS). At neutral pH values, Figure 11. A) Histogram of the rupture force values between DOPA and the DOPA presented a single strong average adhesion value of 383 TiO2 surface (loading rate of 4.6 ± 0.7 nN/s in Tris buffer (50 mM, pH 7.2)). B) ± 21 pN, in concordance with previous reports on the strong Histograms of the forces required to rupture DOPA from the TiO2 surface at a binding of this species to inorganic surfaces.[76] In basic medium loading rate of 4.5 ± 0.6 nN/s in Tris buffer (50 mM, pH 9.8). The strong (pH 9.8), DOPA showed a bimodal force distribution with interactions are assigned to the non-oxidized DOPA (DOPA-enol) and TiO2, average adhesion force values of 108 ± 31 pN (75%) and 362 ± whereas the weak interactions are assigned to dopaquinone and TiO2. 21 pN (25%), which were attributed to dopaquinone and DOPA, Reproduced with the permission from Ref. [133]. Copyright Royal Society of respectively. The study also evaluated the influence of DOPA Chemistry. substituents in the adhesion. At neutral pH values, interaction forces for N-Boc-6-hydroxy-L-DOPA were slightly higher than Researchers have also developed different strategies to cross- those of DOPA. However, in the case of N-Boc-6-nitro-L-DOPA, link polymers at acidic pH values by introducing additional a bimodal force distribution was obtained with similar values and moieties[80a] or diminishing the pKa value of the catecholic distributions to those obtained for DOPA at basic pH (Figure 11). hydroxyls.[130] Other works have published an intelligent In a work by Lee and co-workers,[135] the effect of pH on the approach based on boronate-catechol complex formation and its intermolecular cross-linking rate of a 4-armed dopamine- pH-dependent reversibility. Waite and co-workers[ 136 ] followed terminated PEG was studied. Such hydrogels formulated at this strategy with a twofold objective: I) retard DOPA oxidation at slightly acidic pH values (between 5.7 and 6.7) experienced a neutral pH and II) maintain adhesive properties under oxidizing reduced curing rate after periodate oxidant addition, exhibiting conditions through the dissociation of the complex. In another poor mechanical and adhesive properties. At pH 8, an increase work, Lee and co-workers[137] described the preparation of pH- in the curing rate was observed. However, the resulting responsive adhesive hydrogels by copolymerization of dopamine materials again showed a limited mechanical and bioadhesive methacrylamide and 3-(acrylamido)phenylboronic acid. These performance. Finally, adhesive hydrogels formulated at pH 7.4 materials showed great adhesive properties at pH 3 that were were obtained at a faster curing rate with a good balance drastically reduced at pH 9. At acidic pH values, the catechol between efficient interfacial binding and mechanical properties. and borate moieties contributed to strong interfacial binding with The evolution of the chemical species resulting from addition of the substrate; at basic pH values, the formation of an internal the oxidant agent was tracked using UV-visible spectroscopy. boronate-catechol complex occurred (as confirmed by means of This technique revealed that the stability of the quinone FT-IR and rheometry measurements), reducing the adhesive intermediate generated by the oxidation of dopamine increases properties of the hydrogel. This process was demonstrated to be under acidic conditions. reversible for several pH changes. REVIEW In a more recent work, Sohn and co-workers[138] studied pH- 3.4. Presence of ions responsive silica nanoparticles functionalized with catechol- modified hyaluronic acid (HA-CA) (Figure 12). The competition The mfp adhesion process occurs in two stages:[140] I) catechol between adhesive and cohesive forces within the material led to anchors to the surface and II) a slow (1-4 hours) curing process the formation of three different structures: I) an unstable where extensive cross-linking within the protein occurs. The catechol-modified HA-CA shell at pH 5.5; II) monodisperse objective for the cross-linking is twofold: improving the spherical silica HA-CA-coated particles at pH 7.4; and III) an mechanical performance of the threads and the formation of amorphous HA-CA layer at pH 8.5. This behaviour was a adhesive plaques.[ 141 ] In addition to the covalent interactions consequence of the pH-dependent redox transition of catechol, described in the previous sections, cross-linking through affording strong adhesion of HA-CA to silica at pH 5.5 and transition-metal ions represents the most used mechanism.[142] structural cohesiveness at pH 7.4. The chemical composition of In fact, mussels concentrate metal ions such as iron, zinc, HA-CA was analysed by FT-IR. As expected, the C=O stretching copper, and manganese from seawater [143] into their adhesive at vibration at 1715 cm−1, attributed to the quinone state, was levels much higher than those found in open ocean waters.[144] mainly measured at pH 7.4. The influence of pH on the catechol For this concentration to happen, metal ions coordinate to the to quinone transition was also measured using XPS and showed dianionic catecholates of mfps to form mono-, bis-, or tris- the variation of representative C-C/C-H and C=O species when coordination complexes, depending on the valency of the metal the pH was increased. ion, catechol-to-metal ion molar ratios and pH.[145] In contrast to covalent cross-linking, metal-ligand coordination is reversible, A) thus giving elasticity and self-healing properties to the mfps.[146] Besides cross-linking, metal ions also facilitate the adhesion of catechols to different surfaces. The roles of the previously studied metal ions are described next. Iron. The cross-linking of mfps in the presence of Fe(III)[147] is quite strong. The DOPA-Fe(III) complex exhibits very high binding and stability constants, and the bond can be reversibly broken.[148] A Raman microscopy map of DOPA coordination to Fe(III) throughout the byssal plaque has evidenced a predominance of iron in the outer cuticle and central bulk of the plaque, where mfp-1 and mfp-2 are the prevalent proteins. In B) C) contrast, less iron was found near the interface between the plaque and the substrate, where mfp-3 and mfp-5 are predominant. There is a reason for this distribution: for mfp-1 and mfp-2, which form the cuticle and the central bulk of the plaque, respectively, the prevalence of Fe(III)-DOPA complexation is essential for their structural integrity and unique mechanical properties[ 149 , 150 ] as this complexation provides hardness, extensibility and self-healing properties.[ 151 ] No Figure 12. A) Schematic illustration of silica@HA-CA formation at different pH involvement in a specific physiological function is known.[ 152 ] values: I) an unstable catechol-modified HA-CA shell at pH 5.5, II) Additionally, fluorescence microscopy and elemental analysis monodisperse spherical silica HA-CA-coated particles at pH 7.4, and III) an have demonstrated that Fe(III) and Ca(II) coexist in mfp-1 in the amorphous HA-CA layer at pH 8.5. B) Bio-TEM image of silica@HA-CA cuticle of Mytilus galloprovincialis byssal threads, leading to the particles at pH 7.4 and C) Bio-TEM image of hollow silica@HA-CA particles at proposal of Ca(II) having similar functions to Fe(III). In fact, pH 7.4 after silica core removal. Reproduced with the permission from Ref. chelation and the removal of both ions from the cuticle resulted [138]. Copyright Elsevier Ltd. in 50% hardness reduction and a disruption of cuticle integrity.[153] In contrast, whereas the role of the DOPA-Fe(III) Finally, Wilker et al.[ 139 ] focused on the rich redox chemistry bond in mfp-1 and mfp-2 is well-established, its function in mfp-3 taking place in mussel wet adhesion to rationally design and and mfp-5 remains unclear, especially considering their role at develop new antifouling coatings for ships. The addition of the the byssus/surface interface.[154] usual antioxidants that are insoluble in water to a commercial In addition to being a coordinative cross-linker, Fe(III) also ship coating decreased mussel adhesion to aluminium plaques induces DOPA oxidation and subsequent aryl cross-linking in artificial seawater (pH 7.9). Interestingly, a correlation was during mfp curing.[155] Electron paramagnetic resonance (EPR) established between decreased adhesion and a low of mfp-1 and mfp-2 precipitated upon addition of Fe(III) at acidic electrochemical potential of the antioxidant. This result may pH values revealed the presence of radical species. These indicate that these species disrupt the curing of mfps by species were assigned as both a high-spin Fe(III) and an reducing quinone intermediates to the catechol state. organic radical, presumably DOPA-semiquinone, which is known to undergo aryl coupling.[131] These results are also supported by REVIEW theoretical calculations that demonstrated that the strength with A) which Fe(III) coordinates the catechol ligand is not particularly strong. Therefore, the strong cross-linking in the presence of Fe(III) can be ascribed to this ion’s additional ability as an oxidant to induce covalent coupling of the catecholic groups of mfps.[156] In addition to the fundamental studies previously described, several other revisions have highlighted the role of iron in the adhesive properties of mfp-inspired materials.[140] Such works include AFM studies,[157] binding of DOPA-containing mfps[158] and hydrolysed peptides,[ 159 ] spectroscopic and chemical characterization of the complexes formed during cross-linking[160] and metal-mediated cross-linking in mfps.[161] In a recent study, Cha et al.[162] confirmed DOPA-Fe(III) complexation in mfp-3F and mfp-5 and investigated its effect on both cohesion and B) surface adhesion (Figure 13). At low pH values (similar to those of the local acidic environments during the secretion of adhesive proteins), DOPA-Fe(III) complexation decreases. Under these conditions, SFA measurements showed strong surface adhesion and weak cohesion. Alternatively, higher pH values (similar to those found in seawater) favour catechol deprotonation and thus the formation of more iron-catechol bonds and a consequent remarkable increase in cohesion forces. All the fundamental knowledge regarding the role of iron in the adhesion properties of byssal threads [ 163 ] has fuelled the development of novel materials and applications, e.g., surgical adhesives and orthopaedic cements,[164] versatile metal-phenolic platforms used to engineer nanomaterials and biointerfaces, [165] dendritic polymers as universal multifunctional coatings [166] and novel hydrogel actuators inspired by the reversible mfp chemistry.[167] The reversibility of the DOPA-iron bond has also been used to fabricate self-healing hydrogels with stretching capacities ten times their original lengths.[ 168 ] This high extensibility can be completely recuperated in under 20 min, even after they’ve been cut apart. In addition, the hydrogels respond to multiple stimuli such as mechanical force, temperature, and certain chemicals because of the dynamic catechol-Fe(III) bond. In another example, polyoxetane copolymers with grafted catechol moieties showed strong bonding upon addition of FeCl 3 as a cross-linker. Polymers containing up to 15.5 mol% catechol content showed the strongest bonding to a variety of substrates, including sanded stainless and porcine skin.[87] Chelation with iron can even favour much higher bonding strength in seawater than under dry conditions. Wan et al.[119] reported a catechol-containing PVP backbone that upon addition Figure 13. A) Schematic illustration of mussel byssus showing the location of of FeCl3, increases in bond strength to a maximum of 1.63 MPa mfp-1-6 proteins and the relative content of DOPA, as indicated by the colour (average: 1.33 MPa). The authors argue that there are two main gradient. Mfp-3F and mfp-5 are located at the plaque substrate interface reasons for this outstanding adhesion: I) amides within PVP mediating surface adhesion. DOPA-Fe(III) complexes are found everywhere establish additional hydrogen bonds with the substrate, working on the byssus from interface to cuticle. B) Amino acid sequence information together with catechols to repel water molecules from the for DOPA-incorporated recombinant foot proteins 3F and 5 (drfp-3F and drfp- surface, and II) water and Fe(III) slowly diffuse into the polymer 5); ∼94% DOPA incorporation yield indicates that almost all tyrosine residues thanks to the hydrophilic character of the PVP backbone, thus were converted to DOPA. Reproduced with the permission from Ref. [162]. promoting a more homogeneous coordination cross-linking. Copyright American Chemical Society. REVIEW Finally, proper amounts of dopamine and metal ions endow an 4. Summary epoxy adhesive with better water resistance and excellent adhesion properties.[106] The authors found that though Several studies in the literature aimed to understand the dopamine itself increases the adhesion strength of a material adhesion properties of catechol-based materials. On grouping all and its resistance to water, complexation with Fe(III) (for a these examples, one should consider two main different factors, dopamine content less than 5 wt%) further improves its both crucial for optimal performance and requiring adequate performance. UV-visible and X-ray diffraction measurements balancing: I) adsorption of the catechol moiety on the surface confirmed that at high dopamine-Fe(III) ratios, bis-catecholate (valuable for both coatings and adhesives) and II) cohesion of complexes are formed, favouring cohesion and thus the overall the whole material by cross-linking (mainly for adhesives). adhesion. Adsorption. Catechol establishes numerous interactions (both Copper and manganese. QCM-D, dynamic light scattering (DLS) non-covalent forces and chemical bonding) whose prevalence and small-angle X-ray scattering (SAXS) have been successfully depends on the nature of the substrate to be attached. For used to study the intra- and intermolecular cross-linking of mfps instance, thanks to the presence of the two neighbouring with different metal ions.[ 169 ] Specifically, the viscoelastic hydroxyl groups, catechols can anchor to inorganic surfaces by properties of the M. edulis foot protein Mefp-1 adsorbed on means of hydrogen bonding (mainly inorganic substrates lacking modified hydrophobic gold surfaces after the addition of Cu(II), transition metals, such as silica and mica) or coordination (on Mn(II) and NaIO4 were studied. The reduction in viscoelasticity metal oxides, preferentially at defect sites where followed the order NaIO4 > Cu(II) > buffer control > Mn(II). While undercoordinated metal atoms favour the dissociation of Cu(II) participates in intermolecular cross-linking via metal hydroxyl groups and coordination to the metallic centre). These complexation and increases the abrasion resistance of the Mefp- interactions, even non-covalent ones, are strong enough to 1 layer, NaIO4 mainly resulted in the intramolecular formation of displace preadsorbed water molecules, allowing the catechols to di-DOPA cross-links, thus failing to induce larger intermolecular adsorb in wet environments. In addition, the torsion capacity of aggregation. In a related work, the comparative adsorption of the these hydroxyls seems to be responsible for an enhanced Mefp-1 cross-linked with NaIO4 and catechol oxidase on two capacity to adapt to the underlying surface lattice, which would different surfaces was studied via surface plasmon resonance improve the efficacy of the interaction. On the other hand, the (SPR) and QCM-D. A negatively charged polar SiO2 surface and catechol benzene ring allows for establishing different an electrically inert non-polar CH3-terminated thiolated gold interactions with organic surfaces, such as -stacking, which is surface were used. Upon cross-linking of the Mefp-1 formed on of relevance mostly when catechols approach substrates rich in the CH3 surface, the rigidity of the adlayer(s) increased aromatic systems (e.g., graphene, CNTs or polymers containing significantly. A similar increase in the rigidity was also observed aromatic groups). In these cases, though hydrophobic upon addition of Cu(II), suggesting that the high level of metal interactions can also mix and contribute to these supramolecular ions present in the byssus thread might be essential for the entities, interactions alone are effective, even in non-polar cohesive and adhesive contributions of this protein.[170] solvents. Beyond -stacking, covalent bonding is also an interaction between catechol and organic substrates. In these Others. Harrington and co-workers[171] have demonstrated that cases, catechols must be oxidized to the quinone form to favour different metal ions are equally sufficient as DOPA cross-linking the nucleophilic attack of groups such as amines or thiols. A agents. The authors performed the in vitro removal of native relative ease in undergoing redox processes would be a key DOPA-metal complexes using ethylenediaminetetraacetic acid. factor in favouring this anchoring strategy. After replacement of the metals with vanadium or aluminium, the Finally, it is important to remark that in most cases, the cuticle performance does not suffer significant changes, coexistence of different adsorption modes occurs. Thus, preserving most of the stiffness and hardness of the material exclusivity should not be generally expected in catechol-surface (above 80% in both cases). Wilker et al.[172] also studied the role interactions. of other metals in the cured glue of marine mussels. For this work, proteins containing DOPA were extracted from mussel Cohesion. Cohesion can be modulated by different factors such feet and mixed with potential cross-linkers, including metal ions as catechol content and the presence of specific functional (e.g., Na(I), Zn(II)), oxidizing transition metals (e.g., Fe(III) and groups in the material backbone. While the influence of the Cr2O72-), non-metallic oxidants (e.g., H2O2, IO4-), and oxidizing catechol content has been clearly identified, no specific enzymes (e.g., tyrosinase)). The compressibility and shear correlation between this content and adhesion has been properties of the resulting materials were investigated by means established due to the many relevant influential variables. of a penetration test, and the results showed optimal curing with Alternatively, the presence of strong or mild nucleophiles such oxidizing metal ions. as amines or thiols in the polymeric chain can enhance the Finally, Ni(II) amends the polymerization kinetics of degree of cross-linking and hence cohesiveness. polydopamine.[ 173 ] Spectroscopic and electron microscopy Factors external to the adhesive material itself, such as pH or studies revealed that chelation occurs, forming homogeneous the presence of metal ions, also modify the adsorption/cohesion Ni(II)-dopamine complexes and thus accelerating the assembly properties of the material. For instance, pH determines the redox of dopamine oligomers in polymerization. balance between the catecholic and quinonic forms and thus the REVIEW compromise between adsorption and cohesion capacities. As a general trend, a large amount of the catecholic form favours [ 5 ] L. Petrone, A. Kumar, C. N. Sutanto, N. J. Patil, S. Kannan, A. adsorption at the expense of cohesion and vice versa. Palaniappan, S. Amini, B. Zappone, C. Verma, A. Miserez, Nat. Commun. Nevertheless, this trend should not be extrapolated to organic 2015, 6, 8737. substrates, where interfacial quinones in the adhesive material [6] N. Farsad, E. D. Sone, J. Struc. Biol. 2012, 177, 613. play a crucial role as anchoring agents through covalent bonding. [7] B. P. Lee, P. B. Messersmith, J. N. Israelachvili, J. H. Waite, Annu Rev Mater Res. 2011, 41, 99. On the other hand, cohesive cross-linking relies on not only the [8] S. C. Daubner, T. Le, S. Wang, Arch Biochem Biophys. 2011, 508, 1. quinonic state but other factors since different metal ions can [9] V. V. Papov, T. V. Diamond, K. Biemann, J. H. Waite, J. Biol. Chem. 1995, interact with catechol moieties, enhancing the cohesiveness by 270, 20183. coordination. This strategy is observed in Nature, where [10] J. H. Waite, X. X. Qin, Biochemistry 2001, 40, 2887. mussels concentrate ionic species such as iron, zinc, copper, [11] J. H. Waite, J. Exp. Biol. 2017, 220, 517. and manganese from seawater into their adhesive proteins at [12] N. Bandara, H. Zeng, J. Wu, J. Adhes. Sci. Technol. 2013, 27, 2139. levels much higher than those found in oceans. Besides playing [13] J. Zhou, A. P. Defante, F. Lin, Y. Xu, J. Yu, Y. Gao, E. Childers, A. an active role as coordination cross-linkers, metal ions also lead Dhinojwala, M. L. Becker, Biomacromolecules 2015, 16, 266. [14] L. Han, Z. -M. Wang, X. Lu, L. Dong, C. -M. Xie, K. -F. Wang, X. -L. Chen, to the formation of semiquinones. These semiquinones react Y. -H. Ding, L. -T. Weng, Colloids Surf. B 2015, 126, 452. through a radical-radical mechanism, giving rise to covalently [15] C. E. Brubaker, P. B. Messersmith, Biomacromolecules 2011, 12, 4326. cross-linked species. However, an excessive degree of cross- [16] M. Brodie, L. Vollenweider, J. L. Murphy, F. Xu, A. Lyman, W. D. Lew, B. linking (either covalent or coordinative) can be detrimental P. Lee, Biomed. Mater. 2011, 6, 15014. because it may lead to an extremely stiff material and [17] D. G. Barrett, G. G. Bushnell, P. B. Messersmith, Adv. Healthcare Mater. consequent adhesion decrease. 2013, 2, 745. In summary, the roles of catechols and other agents (both [18] J. Guo, G. B. Kim, D. Shan, J. P. Kim, J. Hu, W. Wang, F. G. Hamad, G. Qian, E. B. Rizk, J. Yang, Biomaterials 2017, 112, 275. inherent and external to the material) have been determined in [19] M. Mehdizadeh, H. Weng, D. Gyawali, L. Tang, J. Yang, Biomaterials the optimal balancing of the adsorption/cohesion equilibrium and 2012, 33, 7972. thus in overall adhesion. This review established a [20] J. Guo, W. Wang, J. Hu, D. Xie, E. Gerhard, M. Nisic, D. Shan, G. Qian, comprehensive toolkit of such factors from a chemical point of S. Zheng, J. Yang, Biomaterials 2016, 85, 204. view and definitely provides readers a better understanding of [ 21 ] K. Huang, B. P. Lee, D. R. Ingram, P. B. Messersmith, these materials, which should therefore lead to an improvement Biomacromolecules 2002, 3, 397. of their properties. In other words, this review is aimed at being a [22] L. Haeshin, S. M. Dellatore, W. M. Miller, P. B. Messersmith, Science road map towards novel catechol-based adhesives. However, 2007, 318, 426. [23] a) P. K. Forooshani, B. P. Lee, J. Polym. Sci. A 2017, 55, 9; b) M. the application of the information in this review is not Krogsgaard, V. Nue, H. Birkedal, Chem. Eur. J. 2016, 22, 844; c) S. Moulay, straightforward and should be considered on a case-by-case Polymer Reviews 2014, 54, 436; d) J. Sedo, J. Saiz-Poseu, F. Busqué, D. basis mainly depending on the nature of the substrate and the Ruiz-Molina, Adv. Mater. 2013, 25, 653; e) E. Faure, C. Falentin-Daudre, C. conditions of the surrounding environment. Jerome, J. Lyskawa, D. Fournier, P. Woisel, C. Detrembleur, Prog. Polym. Sci. 2013, 38, 236; f) B. P. Lee, P. B. Messersmith, J. N. Israelachvili, J. H. Waite, Annu. Rev. Mater. Res. 2011, 41, 99; g) Q. Ye, F. Zhou, W. Liu, Chem. Soc. Acknowledgements Rev. 2011, 40, 4244. [24] M. Guardingo, E. Bellido, R. Miralles-Llumà, J. Faraudo, J. Sedó, S. Tatay, A. Verdaguer, F. Busqué, D. Ruiz-Molina, Small 2014, 10, 1594. This work was supported by project MAT2015-70615-R from the [25] W. Wei , L. Petrone , Y. Tan , H. Cai , J. N. Israelachvili , A. Miserez , J. Spanish Government and by FEDER funds. The ICN2 is funded H. Waite, Adv. Funct. Mater. 2016, 26, 3496. by the CERCA programme/Generalitat de Catalunya. The ICN2 [26] S. A. Mian, L. C. Saha, J. Jang, L. Wang, X. Gao, S. Nagase, J. Phys. Chem. C 2010, 114, 20793. is supported by the Severo Ochoa programme of the Spanish [27] A. J. Kinloch, J. Mater. Sci. 1980, 15, 2141. Ministry of Economy, Industry and Competitiveness (MINECO, [28] S. A. Mian, Y. Khan, U. Ahmad, M. A. Khan, G. Rahmanc, S. Ali, RCS grant no. SEV-2013-0295). We would like to thank J. Sedo for Adv. 2016, 6, 114313. fruitful discussions. [29] S. A. Mian, L. -M. Yang, L. C. Saha, E. Ahmed, M. Ajmal, E. Ganz, Langmuir 2014, 30, 6906. [30] S. A. Mian, X. Gao, S. Nagase, J. Jang, Theor. Chem. Acc. 2011, 130, Keywords: catechol • mussel • adhesion • polydopamine 333. [31] D. S. Hwang, M. J. Harrington, Q. Lu, A. Masic, H. Zeng, J. H. Waite, J. Mater. Chem. 2012, 22, 15530. [32] P. C. Redfern, P. Zapol, L. A. Curtiss, T. Rajh, M. C. Thurnauer, J. Phys. Chem. B 2003, 107, 11419. [1] C. J. Sun, A. Srivastava, J. R. Reifert, J. H. Waite, J. Adhes. 2009, 85, 126. [ 33] M. Vega-Arroyoa, P. R. LeBretona, T. Rajhb,P. Zapol, L. A. Curtiss, [2] J. -L. Jonker, L. Morrison, E. P. Lynch, I. Grunwald, J. von Byern, A. M. Chem. Phys. Lett. 2005, 406, 306. Power, Interface Focus 2005, 5, 20140062. [34] S. -C. Li, J. -G. Wang, P. Jacobson, X. -Q. Gong, A. Selloni, U. Diebold, J. [3] K. Kamino, Biochem. J. 2001, 356, 503. Am. Chem. Soc. 2009, 131, 980. [4] Q. Lin, D. Gourdon, C. Sun, N. Holten-Andersen, T. H. Anderson, J. H. [35] M. Benz, K. J. Rosenberg, E. J. Kramer, J. N. Israelachvili, J. Phys. Waite, J. N. Israelachvili, Proc. Natl. Acad. Sci. USA 2007, 104, 3782. Chem. B 2006, 110, 11884. REVIEW [36] B. Zappone, K. J. Rosenberg, J. N. Israelachvili, Tribol. Lett. 2007, 26, [73] H. H. Kristoffersen, J. -E. Shea, H. Metiu, J. Phys. Chem. Lett. 2015, 12, 191. 2277. [37] F. C. Teng, H. B. Zeng, Q. X. Liu, J. Phys. Chem. C 2011, 115, 17485. [74] T. Zhang, P. Wojtal, O. Rubela, I. Zhitomirsky, RSC Adv. 2015, 5, 106877. [38] J. Yu, W. Wei, E. Danner, J. N. Israelachvili, J. H. Waite, Adv. Mater. [75] W. Lin, J. Walter, A. Burger, H. Maid, A. Hirsch, W. Peukert, D. Segets, 2011, 23, 2362. Chem. Mater. 2015, 27, 358. [39] U. Terranova, D. R. Bowler, J. Phys. Chem. C 2010, 114, 6491. [76] H. Lee, N. F. Scherer, P. B. Messersmith, Proc. Natl. Acad. Sci. U.S.A. [40] W. M. Chirdon, W. J. O’Brien, R. E. Robertson, J. Biomed. Mater. Res. B 2006, 103, 12999. 2003, 66B, 532. [77] A. S. Kumar, S. Sornambikai, P. Gayathri, J. -M. Zen, J. Electroanal. [41] M. B. McBride, L. G. Wesselink, Environ. Sci. Technol. 1988, 22, 703. Chem. 2010, 641, 131. [42] S. L. Simpson, K. J. Powell, S. Sjöberg, J. Colloid Interface Sci. 2000, [78] a) M. Oroujeni, B. Kaboudin, W. Xia, P. Jönsson, D. A. Ossipov, Prog. 229, 568. Org. Coat .2018, 114, 154; b) C. -Y. Liu, C. -J. Huang, Langmuir 2016, 32, [43] I. -C. Yeh, J. L. Lenhart, B. C. Rinderspacher, J. Phys. Chem. C 2015, 5019; c) Y. Liu, K. Ai, L. Lu, Chem. Rev. 2014, 114, 5057; d) J. Cui, Y. Yan, G. 119, 7721. K. Such, K. Liang, C. J. Ochs, A. Postma, F. Caruso, Biomacromolecules [44] L. Yang, S. L. Phua, J. K. H. Teo, C. L. Toh, S. K. Lau, J. Ma, X. Lu, ACS 2012, 13, 2225. Appl. Mater. Interfaces 2011, 3, 3026. [79] S. Suárez-García, J. Sedó, J. Saiz-Poseu, D. Ruiz-Molina, Biomimetics [45] S. L. Phua, L. Yang, C. L. Toh, S. huang, Z. Tsakadze, S. K. Lau, Y.-W. 2017, 2, 22. Mai, X. Lu, ACS Appl. Mater. Interfaces 2012, 4, 4571. [80] a) Y. J. Xu, K. Wei, P. Zhao, Q. Feng, C. K. K. Choi, L. Bian, Biomater. [46] S. L. Phua, L. Yang, S. Huang, G. Ding, R. Zhou, J. H. Lew, S. K. Lau, X. Sci. 2016, 4, 1726; b) D. R. Miller, J. E. Spahn, J. H. Waite, J. R. Soc. Yuan, X. Lu, Eur. Polym. J. 2014, 57, 11. Interface 2015, 12: 20150614. [47] C. Leng, Y. Liu, C. Jenkins, H. Meredith, J. J. Wilker, Z. Chen, Langmuir [81] L. M. McDowell, L. A. Burzio, J. H. Waite, J. Schaefer, J. Biol. Chem. 2013, 29, 6659. 1999, 274, 20293. [48] Z. A. Levine, M. V. Rapp, W. Wei, R. G. Mullen, C. Wu, G. H. Zerze, J. [82] L. Li, H. Zeng, Biotribology 2016, 5, 44. Mittal, J. H. Waite, J. N. Israelachvili, J. -E. Shea, Proc. Natl. Acad. Sci. U.S.A [83] J. H. Waite, Integr. Comp. Biol. 2002, 42, 1172. 2016, 113, 4332. [84] L. M. Rzepecki, K. M. Hansen, H. J. Waite, Biol. Bull. 1992, 183, 123. [49] a) J. Saiz-Poseu, J. Faraudo, A. Figueras, R. Alibes, F. Busqué, D. Ruiz- [85] C. R. Matos-Pérez, J. J. Wilker, Macromolecules 2012, 45, 6634. Molina, Chem. Eur. J. 2012, 18, 3056; b) J. Saiz-Poseu, I. Alcon, R. Alibes, F. [86] C. R. Matos-Pérez, J. D. White, J. J. Wilker, J. Am. Chem. Soc. 2012, Busqué, J. Faraudo, D. Ruiz-Molina, CrystEngComm 2012, 14, 264. 134, 9498. [50] X. Wang, Y. Liu, S. Tao, B. Xing, Carbon 2010, 48, 3721. [87] M. Jia, A. Li, Y. Mua, W. Jiang, X. Wan, Polymer 2014, 55, 1160. [51] C. Wang, S. Li, R. Zhang, Z. Lin, Nanoscale 2012, 4, 1146. [88] C. L. Jenkins, H. M. Siebert, J. J. Wilker, Macromolecules 2017, 50, 561. [52] R. J. Chen, Y. Zhang, D. Wang, H. Dai, J. Am. Chem. Soc. 2001, 123, [89] M. Ye, R. Jiang, J. Zhao, J. Zhang, X. Yuan, X. Yuan, Mater Sci: Mater 3838. Med 2015, 26, 273. [53] J. Chen, H. Liu, W. A. Weimer, M. D. Halls, D. H. Waldeck, G. C. Walker, [90] J. Yang, J. Keijsersa, M. van Heek, A. Stuiver, M. A. Cohen Stuart, M. J. Am. Chem. Soc. 2002, 124, 9034. Kamperman, Polym. Chem. 2015, 6, 3121. [54] J. Björk, F. Hanke, C. -A. Palma, P. Samori, M. Cecchini, M. Persson, J. [91] A. Li, M. Jia, Y. Mu, W. Jiang, X. Wan, Macromol. Chem. Phys. 2014, Phys. Chem. Lett. 2010, 1, 3407. 216, 450. [55] H. Hu, M. Chang, M. Zhang, D. Chen, J. Mater. Sci. 2017, 52, 8650. [92] L. Li, Y. Li, X. F. Luo, J. P. Deng, W. T. Yang, React. Funct. Polym. 2010, [56] F. -f. Liu, J. -l. Fan, S. -g. Wang, G. -h Ma, Chem. Eng. J. 2013, 219, 450. 70, 938. [57] D. Lin, B. Xing, Environ. Sci. Technol. 2008, 42, 7254. [93] K. Sarbjit, G. M. Weerasekare, R. J. Stewart, ACS Appl. Mater. Interfaces [58] W. Chen, L. Duan, D. Q. Zhu, Environ. Sci. Technol. 2007, 41, 8295. 2011, 3, 941. [59] X. Sun, H. Shao, K. Xiang, Y. Yan, X. Yu, D. Li, W. Wu, L. Zhou, K. -Fai [94] P. Glass, H. Y. Chung, N. R. Washburn, M. Sitti, Langmuir 2009, 25, So, Y. Ren, S. Ramakrishna, A. Li, L. He, Carbon 2017, 113, 176. 6607. [60] H. Kim, K. H. Ahn, S. J. Lee, Polymer 2017, 110, 187. [95] C. L. Gao, G. Z. Li, H. Xue, W. Yang, F. B. Zhang, S. Y. Jiang, [61] J. Saiz-Poseu, J. Sedó, B. García, C. Benaiges, T. Parella, R. Alibés, J. Biomaterials 2010, 31, 1486. Hernando, F. Busqué, D. Ruiz-Molina, Adv. Mater. 2013, 25, 2066. [96] H. Y. Chung, P. Glass, J. M. Pothen, M. Sitti, N. R. Washburn, [62] J. C. Ma, D. A. Dougherty, Chem. Rev. 1997, 97, 1303. Biomacromolecules 2011, 12, 342. [63] U. Diebold, Surf. Sci. Rep. 2003, 48, 53. [97] B. Malisova, S. Tosatti, M. Textor, K. Gademann, S. Zurcher, Langmuir [64] R. Sánchez-de-Armas, M. A. San-Miguel, J. Oviedo, A. Márquez, J. F. 2010, 26, 4018. Sanz, Phys. Chem. Chem. Phys. 2011, 13, 1506. [98] S. Yuan, D. Wan, B. Liang, S. O. Pehkonen, Y. P. Ting, K. G. Neoh, E. T. [65] S. -C. Li, L. -N. Chu, X. -Q. G, U. Diebold, Science 2010, 328, 882. Kang, Langmuir 2011, 27, 2761. [66] K. L. Syres, A. G. Thomas, W. R. Flavell, B. F. Spencer, F. Bondino, M. [99] K. H. Bae, Y. B. Kim, Y. Lee, J. Hwang, H. Park, T. G. Park, Bioconjugate Malvestuto, A. Preobrajenski, M. Grätzel, J. Phys. Chem. C 2012, 116, 23515. Chem. 2010, 21, 505. [67] M. J. Jackman, K. L. Syres, D. J. H. Cant, S. J. O. Hardman, A. G. [100] C. L. Jenkins, H. J. Meredith, J. J. Wilker, ACS Appl. Mater. Interfaces Thomas, Langmuir 2014, 30, 8761. 2013, 5, 5091. [68] K. Syres, A. Thomas, F. Bondino, M. Malvestuto, M. Grätzel, Langmuir [101] M. Yu, T. J. Deming, Macromolecules 1998, 31, 4739. 2010, 26, 14548. [102] J. Wang, C. Liu, X. Lu, M. Yin, Biomaterials 2007, 28, 3456. [69] T. D. Savić, M. I. Čomor, J. M. Nedeljković, D. Ž. Veljković, S. D. Zarić, V. [103] S. Saxer, C. Portmann, S. Tosatti, K. Gademann, S. Zurcher, M. Textor, M. Rakićc, I. A. Janković, Phys. Chem. Chem. Phys. 2014, 16, 20796. Macromolecules 2010, 43, 1050. [70] J. L. Dalsin, L. Lin, S. Tosatti, J. Vörös, M. Textor, P. B. Messersmith, [104] J. V. Alegre-Requena, M. Häring, R. P. Herrera, D. Díaz-Díaz, New J. Langmuir 2005, 21, 640. Chem. 2016, 40, 8493. [71] M. Rodenstein, S. Zürcher, S. G. P. Tosatti N. D. Spencer, Langmuir [105] J. Xu, S. Strandman, J. X. X. Zhu, J. Barralet, M. Cerruti, Biomaterials 2010, 26, 16211. 2015, 37, 395. [72] A. Vittadini, A. Selloni , F. P. Rotzinger , M. Grätzel, J. Phys. Chem. B [106] F. Dai, F. Chen, T. Wang, S. Feng, C. Hu, X. Wang, Z. Zheng, J. Mater. 2000, 104, 1300. Sci. 2016, 51, 4320. REVIEW [107] H. Wu, V. Sariola, J. Zhao, H. Ding, M. Sitti, C. J. Bettinger, Polym. Int. [139] C. A. Del Grosso, T. W. McCarthy, C. L. Clark, J. L. Cloud, J. J. Wilker, 2016, 65, 1355. Chem. Mater. 2016, 28, 6791. [108] M. E. R. Shanahan, A. Carre, Langmuir 1995, 11, 1396. [140] J. Monahan, J. J. Wilker, Chem. Commun. 2003, 1672. [109] L. E. Nielsen, R. F. Landel, Mechanical Properties of Polymers and [141] J. J. Wilker, Angewandte Chem. Int. Ed. 2010, 49, 8076. Composites 1994. [142] S. W. Taylor, J. H. Waite, M. M. Ross, J. Shabanowitz, D. F. Hunt, J. [110] S. Seo, S. Das, P. J. Zalicki, R. Mirshafian, C. D. Eisenbach, J. N. Am. Chem. Soc. 1994, 116, 10803. Israelachvili, J. H. Waite, B. K. Ahn, J. Am. Chem. Soc. 2015, 137, 9214. [143] a) S. G. George, B. J. S. Pirie, T. L. Coombs, J. Exp. Mar. Biol. 1976, 23, [111] J. D. White, J. J. Wilker, Macromolecules 2011, 44, 5085. 71; b) T. L. Coombs, P. J. Keller, Aquat. Toxicol. 1981, 1, 291. [112] D. Lu, Y. Zhang, T. Li, Y. Li, H. Wang, Z. Shen, Q. Wei, Z. Lei, Polym. [144] J. R. Donat, K. W. Bruland In Trace Elements in Natural Waters; Salbu, Chem. 2016, 7, 1963. B., Steinnes, E., Eds.; CRC Press: Ann Arbor, MI, 1995; 247. [113] G. P. Maier, M. V. Rapp, J. H. Waite, J. N. Israelachvili, A. Butler, [ 145 ] D. E. Fullenkamp, D. G. Barrett, D. R. Miller, J. W. Kurutz, P. B. Science 2015, 349, 628. Messersmith, RSC Adv. 2014, 4, 25127. [114] C. Lim, J. Huang, S. Kim, H. Lee, H. Zeng, D. S. Hwang, Angew. Chem. [146] N. Holten-Andersen, M. J. Harrington, H. Birkedal, B. P. Lee, P. B. Int. Ed. 2016, 55, 3342. Messersmith, K. Y. C. Lee, J. H. Waite, Proc. Natl. Acad. Sci. 2011, 108, [115] Y. Li, T. Wang, L. Xia, L. Wang, M. Qin, Y. Li, W. Wang, Y. Cao, Journal 2651. of Materials Chemistry B 2017, 5, 4416. [147] H. Zeng, D. S. Hwang, J. N. Israelachvili, J. H. Waite, Proc. Natl. Acad. [116] R. Wang, J. Li, W. Chen, T. Xu, S. Yun, Z. Xu, Z. Xu, T. Sato, B. Chi, H. Sci. 2010, 107, 12850. Xu, Adv. Funct. Mater. 2017, 27, 1604894. [148] a) D. J. Rubin, A. Miserez, J. H. Waite, Advances in Insect Physiology, [117] M. V. Rapp, G. P. Maier, H. A. Dobbs, N. J. Higdon J. H. Waite, A. Vol. 38; b) C. Jérôme, J. S. Stephen Eds.; Academic Press: London, 2010; Butler J. N. Israelachvili, J. Am. Chem. Soc. 2016, 138, 9013. Chapter 3, pp 75-133; c) E. Degtyar, M. J. Harrington, Y. Politi, P. Fratzl, [118] B. P. Lee, C. -Y. Chao, F. N. Nunalee, E. Motan, K. R. Shull, P. B. Angew. Chem., Int. Ed. 2014, 53, 12026. Messersmith, Macromolecules 2006, 39, 1740. [149] M. J. Harrington, A. Masic, N. Holten-Andersen, J. H. Waite, P. Fratzl, [119] A. Li, Y. Mu, W. Jiang, X. Wan, Chem. Commun. 2015, 51, 9117. Science, 2010, 328, 216. [120] M. Brodie, L. Vollenweider, J. L. Murphy, F. Xu, A. Lyman, W. D. Lew, B. [150] Fan Zhang, Majid Sababi, Tore Brinck, Dan Persson, Jinshan Pan,, Per P. Lee, Biomed. Mater. 2011, 6, O015014. M. Claesson, J. Colloid Int. Sci. 2013, 404, 62. [121] W. Wei, J. Yu, C. Broomell, J. N. Israelachvili, J. H. Waite, J. Am. Chem. [151] N. Holten-Andersen, G. E. Fantner, S. Hohlbauch, J. H. Waite, F. W. Soc. 2013, 135, 377. Zok, Nat. Mater. 2007, 6, 669. [122] J. Yang, M. A. Cohen Stuart, M. Kamperman, Chem. Soc. Rev. 2014, [152] a) B. J. Kim, S. Kim, D. X. Oh, A. Masic, H. J. Cha, D. S. Hwang, J. 43, 8271. Mater. Chem. B 2015, 3, 112; b) B. J. Kim, H. Cheong, B. H. Hwang, H. J. [123] J. Wu, L. Zhang, Y. Wang Y. Long, H. Gao, X. Zhang, N. Zhao, Y. Cai, Cha, Angew. Chem., Int. Ed. 2015, 54, 7318; c) B. J. Kim, D. X. Oh, S. Kim, J. J. Xu, Langmuir 2011, 27, 13684. H. Seo, D. S. Hwang, A. Masic, D. K. Han, H. J. Cha Biomacromolecules [124] K. V. Pillai, S. Renneckar, Biomacromolecules 2009, 10, 798. 2014, 15, 1579; d) Q. Li, D. G. Barrett, P. B. Messersmith, N. Holten- [125] a) M. A. Gebbie, W. Wei, A. M. Schrader, T. R. Cristiani, H. A. Dobbs, M. Andersen, ACS Nano 2016, 10, 1317. Idso, B. F. Chmelka, J. H. Waite, J. N. Israelachvili, Nature Chem. 2017, 9, [153] N. Holten-Andersen, T. E. Mates, M. S. Toprak, G. D. Stucky, F. W. Zok, 473; b) S. Das, N. R. Martinez-Rodriguez, W. Wei, J. H. Waite, J. N. J. H. Waite, Langmuir 2009, 25, 3323. Israelachvili, Adv. Funct. Mater 2015, 25, 5840. [ 154 ] a) S. Das, D. R. Miller, Y. Kaufman, N. R. Martinez-Rodriguez, A. [126] a) S. Kim, A. Faghihnejad, Y. Lee, Y. Jho, H. Zeng, D. S. Hwang, J. Pallaoro, M. J. Harrington, M. Gylys, J. N. Israelachvili, J. H. Waite, Mater. Chem. B 2015, 3, 738; b) W. Wei, J. Yu, M. A. Gebbie, Y. Tan, N. R. Biomacromolecules 2015, 16, 1002; b) S. Das, N. R. M. Rodriguez, W. Wei, J. Martinez-Rodriguez, J. N. Israelachvili, J. H. Waite, Langmuir 2015, 31, H. Waite, J. N. Israelachvili, Adv. Funct. Mater. 2015, 25, 5840. 1105; c) D. S. Hwang, H. Zeng, Q. Lu, J. Israelachvili, J. H. Waite, Soft Matter. [155] a) L. A. Burzio, J. H. Waite Biochemistry 2000, 39, 11147; b) Y. Miaoer, 2012, 8, 5640. J. W. Hwang, T. J. Deming, J. Am. Chem. Soc. 1999, 121, 5825. [127] S. Kim, A. Faghihnejad, Y. Lee, Y. S. Jho, H. Zeng, D. S. Hwang, J. [156] M. A. Matin, R. K. Chitumalla, M. Lim, X. Gao, J. Jang, J. Phys. Chem. B Mater. Chem. B 2015, 3, 738. 2015, 119, 5496. [128] J. Yu, W. Wei, E. Danner, R. K. Ashley, J. N. Israelachvili, J. H. Waite , [157] B. P. Frank, G. Belfort, Biotechnol. Prog. 2002, 18, 580. Nat. Chem. Biol. 2011, 7, 588. [158] S. W. Taylor, D. B. Chase, M. H. Emptage, M. J. Nelson, J. H. Waite, [129] B. J. Sparks, E. F. T. Hoff, L. P. Hayes, D. L. Patton, Chem. Mater. 2012, Inorg. Chem. 1996, 35, 7572. 24, 3633. [159] S. W. Taylor, G. W. Luther, J. H. Waite, Inorg. Chem. 1994, 33, 5819. [130] M. Cencer, M. Murley, Y. Liu, B. P. Lee, Biomacromolecules 2015, 16, [160] H. Ceylan, M. Urel, T. S. Erkal, A. B. Tekinay, A. Dana, M. O. Guler, 404. Adv. Funct. Mater. 2013, 23, 2081. [131] J. Yang, M. A. C. Stuart, M. Kamperman, Chem. Soc. Rev. 2014, 43, [161] M. J. Sever, J. T. Weisser, J. Monahan, S. Srinivasan, J. J. Wilker, 8271. Angew. Chem., Int. Ed. 2004, 43, 448. [132] P. Z. Araujo, P. J. Morando, M. A. Blesa, Langmuir 2005, 21, 3470. [162] B. Yang, C. Lim, D. S. Hwang, H. J. Cha, Chem. Mater. 2016, 28, 7982. [133] P. Das, M. Reches, Nanoscale 2016, 8, 15309. [163] G. P. Maier, A. Butler, J. Biol. Inorg. Chem. 2017, 22, 739. [134] J. Yu, W. Wei, M. S. Menyo, A. Masic, J. H. Waite, J. N. Israelachvili, [164] J. J Wilker, Current Opinion in Chemical Biology 2010, 14, 276. Biomacromolecules 2013, 14, 1072. [165] H. Ejima, J. J. Richardson, F. Caruso, Nano Today 2017, 12, 136. [135] M. Cencer, Y. Liu, A. Winter, M. Murley, H. Meng, B. P. Lee, [166] Q. Wei, K. Achazi, H. Liebe, A. Schulz, P.-L. M. Noeske, I. Grunwald, R. Biomacromolecules 2014, 15, 2861. Haag, Angew. Chem. Int. Ed. 2014, 53, 11650. [136] Y. Kan, E. W. Danner, J. N. Israelachvili, Y. Chen, J. H. Waite, PLoS [167] B. P. Lee, S.i Konst, Adv. Mater. 2014, 26, 3415. One 2014, 9 (10), e108869. [168] S. Hou, P.r X. Ma, Chem. Mater. 2015, 27, 7627. [137] A. R. Narkar, B. Barker, M. Clisch, J. Jiang, B. P. Lee Chem. Mater. [169] J. Hedlund, M. Andersson, C. Fant, R. Bitton, H. Bianco-Peled, H. 2016, 28, 5432. Elwing, M. Berglin, Biomacromolecules 2009, 10, 845. [138] J. Lee, K. C. Yoo, J. Ko, B. Yoo, J. Shin, S. J. Lee, D. Sohn, [170] C. Fant, K. Sott, H. Elwing, F. Hook, Biofouling 2000, 16, 119. Carbohydrate Polymers 2017, 164, 309. REVIEW [ 171 ] C. N. Schmitt, A. Winter, L. Bertinetti, A. Masic, P. Strauch, M. J. Harrington, J. R. Soc. Interface. 2015, 12(110):0466. [172] J. Monahan, J. J. Wilker, Langmuir 2004, 20, 3724. [173] L. Yang, J. Kong, D. Zhou, J. M. Ang, S. L. Phua, W. A. Yee, H. Liu, Y. Huang, X. Lu, Chem. Eur. J. 2014, 20, 7776. REVIEW REVIEW A suitable chemical toolbox to J. Saiz-Poseu,* J. Mancebo-Aracil, F. understand the different parameters, Nador, F. Busqué, D. Ruiz-Molina* both inherent to the polymer backbone and external (such as pH or the Page No. – Page No. presence of metal ions), influencing The Chemistry behind Catechol- the adhesion and cohesion of Based Adhesion catechol-based coatings and adhesives is given.