pubs.acs.org/Langmuir © 2009 American Chemical Society
Extraordinary Adhesion of Phenylboronic Acid Derivatives of Polyvinylamine to Wet Cellulose: A Colloidal Probe Microscopy Investigation Shannon M. Notley,*,† Wei Chen,‡ and Robert Pelton‡ † Department of Applied Mathematics, Research School of Physical Sciences and Engineering, Australian National University, Canberra 0200 ACT, Australia, and ‡Department of Chemical Engineering, McMaster University, Hamilton, Ontario, Canada L8S 4L7
Received January 15, 2009. Revised Manuscript Received March 1, 2009 Typically, the adhesion between cellulose surfaces under aqueous conditions is very poor. Often, adsorbed polymers such as polyvinylamine (PVAm) are used to increase the wet strength; however, this provides only a minimal increase in the adhesion energy. Here, the adhesion between cellulose surfaces with adsorbed layers of phenylboronic acid derivatized polyvinylamine has been studied using colloidal probe microscopy as a function of pH. The adhesion due to the phenylboronic acid (PBA) groups grafted on the polyvinylamine backbone is almost 30 times greater, providing a new, exciting class of polymers using covalent linkages to improve the strength of the joint between cellulose surfaces. The measured surface forces on approach provided key information on the molecular conformation of the polymers at the cellulose-solution interface. At low pH, the three polymers tested, PVAm, PVAm-Ph (with pendant phenol groups), and PVAm-PBA (with phenylboronic acid groups) all had a relatively flat conformation at the interface, which is in agreement with the predictions based upon theory for highly charged polyelectrolytes adsorbing to an oppositely charged interface. With increasing pH, the charge on the polymers is reduced, eventually resulting in a more expansive conformation at the interface at pH 10 and above with the development of a steric interaction force. The onset of this steric force correlates well with the observed significant increase in the pull-off force upon separation of the cellulose surfaces. Furthermore, a greater increase in the adhesion was observed for PVAm-PBA in agreement with previous studies using macroscopic cellulose surfaces. This is attributed to the formation of boronic acid esters between the polymer and the cis diol groups on the cellulose surface.
Introduction It is often difficult to form strong adhesive joints between wet, hydrophilic surfaces. Tissue adhesives, for example, must function in wet, salty environments. Our interest in wet adhesion arises from a need in the papermaking industry for water-soluble polymers which will promote adhesion between cellulose fibers when they first come into contact in water. Cationic water-soluble polymers spontaneously adsorb onto wet cellulose surfaces owing to the presence of anionic carboxyl groups on most cellulose surfaces. However, polymer adsorption does not guarantee strong adhesive joints. For example, we have shown that cationic polyacrylamides, poly(diallyldimethyl ammonium chloride),1 polypeptides,2 and proteins adsorb onto cellulose but do not give significant wet adhesion to cellulose. An early promising candidate was polyvinylamine (PVAm). We have shown that polyvinylamine does strengthen cellulose-cellulose joints without a heating/curing step; however, it was necessary to nearly dry the joint to achieve a strong cellulose-cellulose adhesion when the joint was subsequently immersed in water. We proposed that when most of the free water was removed, the primary amine groups reacted with hemiacetal groups in the regenerated cellulose films to give imine and aminal linkages.3 *To whom correspondence should be addressed. E-mail: shannon.
[email protected]. (1) Feng, X.; Pouw, K.; Leung, V.; Pelton, R. Biomacromolecules 2007, 8(7), 2161–2166. (2) Kurosu, K.; Pelton, R. J. Pulp Pap. Sci. 2004, 30(8), 228–232. (3) DiFlavio, J. L.; Bertoia, R.; Pelton, R.; Leduc, M. In The Mechanism of Polyvinylamine Wet-Strengthening, 13th Fundamental Research Symposium, Cambridge, U.K., 2005; Sampson, W. W., Ed.; FRC: Cambridge, U.K., 2005; pp 1293-1316.
6898 DOI: 10.1021/la900256s
In a research note in 2006, we reported that a phenylboronic acid (PBA) derivative of polyvinylamine (PVAm-PBA; see Scheme 1) can promote instantaneous adhesion between wet cellulose films without a drying or heating step.4 This approach was inspired by the well-documented ability of phenylboronic acid groups to condense with carbohydrates to give covalent rings. We believe that this is the first time that the boronatecarbohydrate interaction has been exploited to promote adhesion. More recently, we extended this work to document the effects of polymer structure, molecular weight, pH, and ionic strength on adhesion to wet cellulose. 5 The main conclusions from this work were as follows: 1. The boronic acid groups are responsible for the instantaneous wet adhesion in cellulose/PVAmPBA/cellulose laminates. The advantages conferred by the boronic acid groups were lost in the presence of sorbitol which complexes with boronate. 2. The greatest wet adhesion measurements were observed when the pH of the PVAm-PBA solution was 9-10 during polymer adsorption onto cellulose. Under these conditions, the net charge density and the solubility of PVAm-PBA was low, leading to high concentrations of adsorbed PVAm-PBA in the cellulose/PVAM-PBA/cellulose joints. 3. The wet adhesion was rather insensitive to pH under which the laminates were soaked and then tested. We propose that the high local concentration of ammonium ions on the PVAm backbone extend (4) Chen, W.; Lu, C.; Pelton, R. Biomacromolecules 2006, 7(3), 701–702. (5) Chen, W.; Leung, V.; Kroener, H.; Pelton, R. Langmuir, submitted.
Published on Web 04/02/2009
Langmuir 2009, 25(12), 6898–6904
Notley et al.
Article Scheme 1
the ionization pH range of phenylboronic acid below pH 7, enabling borate to carbohydrate bonding under acid conditions. Our past results were based on macroscopic adhesion measurements of the forces required to delaminate pairs of wet, regenerated cellulose films. The laminates were formed by adsorbing a saturated monolayer of PVAm-PBA onto two cellulose films which were then pressed together at room temperature.5 These macroscopic adhesion measurements suffer two key deficiencies: (a) the measurements give no direct information about the adhesive joint formation process and (b) neither the cellulose surfaces nor the adhesive layers were well-defined at the nanometer distance scale. Herein, we address these issues by employing colloidal probe microscopy with well-defined cellulose surfaces. Colloidal probe microscopy provides a convenient method for determining the forces on both approach and separation of a micrometer sized sphere and flat surface.6,7 This allows a measure of the extent of the adsorbed polymer layer away from the interface8-14 as well as the pull-off force as a function of solution conditions.15 In particular, we show herein that colloidal probe microscopy provides strong evidence for specific interactions between the phenylboronic acid groups and the cellulose surfaces. In addition, we show that in water there exists strong repulsive forces between cellulose surfaces coated with PVAmPBA, which means surfaces must be forced together to form the adhesive joint.
Experimental Section Preparation of Polyvinylamine and Polyvinylamine Derivatives. Linear polyvinylamine with a molecular weight of 150 kDa (BASF) was derivatized to give pendant phenol (PVAm-Ph) or phenylboronic acid groups (PVAm-PBA). The structures are shown in Scheme 2, and the preparation and solution properties have been described previously.16 In PVAm-PBA, 15% of the primary amines were substituted with phenyl boronic moieties. Similarly, PVAm-Ph had 16% substitution with phenolic groups. The polymer solutions were 100 ppm in 10 mM NaCl at pH 3.5. (6) Butt, H.-J. Biophys. J. 1991, 60, 1438. (7) Ducker, W. A.; Senden, T. J.; Pashley, R. M. Nature (London) 1991, 353, 239. (8) Biggs, S.; Healy, T. J. Chem. Soc., Faraday Trans 1994, 90, 3415–3421. (9) Biggs, S. Langmuir 1995, 11, 156–162. (10) Eframova, N. V.; Bondurant, B.; O’Brien, D. F.; Leckband, D. E. Biochemistry 2000, 39, 3441–3451. (11) Notley, S. M.; Biggs, S.; Craig, V. S. J. Macromolecules 2003, 36(8), 2903– 2906. (12) Notley, S. M.; Biggs, S.; Craig, V. S. J.; Wagberg, L. Phys. Chem. Chem. Phys. 2004, 6, 2379–2386. (13) McLean, S. C.; Lioe, H.; Meagher, L.; Craig, V. S. J.; Gee, M. L. Langmuir 2005, 21, 2199–2208. (14) Notley, S. M. J. Phys. Chem. B 2008, 112, 12650–12655. (15) Kappl, M.; Butt, H. J. Part. Part. Syst. Charact. 2002, 19, 129–143. (16) Chen, W.; Pelton, R.; Leung, V. Macromolecules, submitted.
Langmuir 2009, 25(12), 6898–6904
Scheme 2. Molecular Structures of PVAm and Its Derivatives
Cellulose Surfaces. Cellulose thin film surfaces were prepared according to a previously described method.17,18 A total of 0.1 g of cellulose derived from dissolving grade pulp (Domsjo Fabriken, Sweden) was dissolved in 20 mL of a 50% w/w solution of n-methylmorpholine oxide in water (Sigma-Aldrich, Australia) under continuous stirring at 110 °C for approximately 1 h or until a clear solution was observed free of any undissolved cellulose material. This cellulose solution was then diluted with dimethyl sulfoxide (Sigma-Aldrich, Australia) to give a volume of 50 mL which decreased the viscosity sufficiently for subsequent spin-coating. This solution was then spin-coated at 1500 rpm for 45 s onto oxidized silicon wafers which were pretreated with a cationic anchoring layer of polyvinylamine (BASF Ludwigshafen, Germany). The cellulose surfaces were then regenerated in Milli-Q standard water before drying under nitrogen and storage in a desiccator. This produced supported cellulose thin films of approximately 40 nm thickness measured using ellipsometry (Beaglehole Instruments, New Zealand). Prior to use in force measurements, the cellulose surfaces were imaged using tapping mode atomic force microscopy (Veeco Inc.) to ensure that they were smooth and continuous. Force and Adhesion Measurements. The surface forces between the cellulose thin film and a cellulose sphere (Monogel, Helsingborg, Sweden) were measured using colloidal probe microscopy. Cellulose spheres in the size range of 10-14 μm in radius were used with their surface roughness previously characterized.19 The properties of these cellulose spheres, in terms of their surface roughness and chemistry including charge and crystallinity, have been discussed previously where they have been used in surface forces measurements.19-23 The spheres tend to have little surface charge (typically, the surface potential is about -5 mV) and are smooth over a significant area to allow high quality surface forces measurements. The only major difference between the cellulose spheres and the cellulose thin film lies in their crystallinity. The spheres are amorphous, while the film contains areas of amorphous cellulose as well as the cellulose II crystalline form. The cellulose spheres were attached to triangular shaped atomic force microscopy cantilevers (Veeco Inc.) with a small amount of epoxy according to the method of Ducker et al.7,24 The spring constant of the cantilevers was measured to be 0.35 N/m using the thermal noise method.25 A multimode scanning probe microscope (Veeco Inc.) was used for the measurement of the apparent surface forces between the cellulose surfaces in the absence and presence of the polymeric additives. A detailed description of the force-distance experiment in aqueous solutions using this colloidal probe method has :: (17) Falt, S.; Wagberg, L.; Vesterlind, E. L.; Larsson, P. T. Cellulose 2004, 11, 151–162. (18) Gunnars, S.; Wagberg, L.; Cohen-Stuart, M. A. Cellulose 2002, 9, 239–249. (19) Notley, S. M.; Pettersson, B.; Wagberg, L. J. Am. Chem. Soc. 2004, 126(43), 13930–13931. (20) Notley, S. M.; Wagberg, L. Biomacromolecules 2005, 6(3), 1586–1591. (21) Notley, S. M.; Eriksson, M.; Wagberg, L.; Beck, S.; Gray, D. G. Langmuir 2006, 22(7), 3154–3160. (22) Notley, S. M.; Norgren, M. Langmuir 2006, 22(26), 11199–11204. (23) Notley, S. M. Phys. Chem. Chem. Phys. 2008, 10, 1819–1825. (24) Ducker, W. A.; Senden, T. J.; Pashley, R. M. Langmuir 1992, 8, 1831–1836. (25) Hutter, J. L.; Bechhoefer, J. Rev. Sci. Instrum. 1993, 64, 1868–1873.
DOI: 10.1021/la900256s
6899
Article
Notley et al.
been published elsewhere,26 so only a brief description will be given here. The deflection of the free end of the cantilever is monitored as a function of surface separation as the fixed end is moved perpendicular to the plane of the flat substrate. The deflection of the cantilever spring is converted to a force through application of Hooke’s law with the experimentally determined spring constant. The optical sensitivity is measured from the “constant compliance” region of the force-distance curve where the linear motion of the substrate results in a linear deflection of the cantilever. The onset of this region is used to define the zero separation point between the sphere and flat substrate; however, in the case of adsorbed polymer layers, some error may be expected due to trapped or pinned polymer chains between the surfaces.27,28 The measured force is normalized by the spherical probe radius in order to convert to the potential energy of interaction through the Derjaguin approximation.29 The forces on approach between cellulose surfaces in the absence and presence of the polymeric additives were measured under aqueous solution conditions with the pH varied and a constant ionic strength maintained. Furthermore, the adhesion was quantified by measuring the pull-off force as the cellulose surfaces were separated. A minimum 100 force curves were measured for each polymer and each pH. Force curves presented here are representative for each condition with the adhesion values reported as an average for all force curves. Care was taken to maintain a constant maximum applied load in all experiments by using the trigger settings of the atomic force microscopy (AFM) equipment.
Figure 1. AFM tapping mode height image of the cellulose thin film supported on silica. Image is 1 μm 1 μm with z scales of 20 nm.
Results The adhesion between cellulose surfaces will be influenced by the true molecular contact area. This will be governed by the surface roughness of the cellulose surfaces. Ideally, the surface will have a minimal amount of roughness or, alternately, be reasonably deformable under the somewhat low loads experienced in the AFM force experiment. AFM tapping mode imaging was performed on both the cellulose thin films as well as the cellulose spheres prior to use in the force measurements. An example image is shown in Figure 1. The roughness of the cellulose thin film was significantly lower than the contacting area of the cellulose sphere. The surface roughness (root-mean-square) was determined from AFM images and found to be 2.5 nm for the film over a 1 μm2 image, which is in agreement with previous results using similar materials.19,21 Prior to injection of the polymers into the AFM liquid cell, the forces between the cellulose surfaces were measured as a function of pH. Previous studies have shown that there is a slight increase in the surface potential of the cellulose surfaces as the pH of the solution is increased to above 5. This has resulted in a change from a purely attractive van der Waals interaction at lower pH to a repulsion of electrostatic origin at pH greater than 5.19-21 Figure 2 shows the forces on approach between the cellulose flat surface and sphere as a function of pH. The data shown here agree with those presented in previous studies and demonstrate that there is a minimal charge on the surfaces under the solution conditions used for the adsorption of the polymers here in this study. At pH 3, the interaction can be fit using the nonretarded Hamaker equation with a Hamaker constant of 5 10-21 J. At pH 5 and 8, the forces may be fit to the Derjaguin-Landau-Verwey-Overbeek (26) (27) (28) 5691. (29) (30)
Figure 2. Surface forces on approach of a cellulose thin film and cellulose sphere as a function of pH: (b) pH 3.8; (0) pH 5, and () pH 8.
(DLVO) theory30,31 within the limits of constant charge and constant surface potential32 with fitted surface potentials of -13 and -24 mV, respectively, with a Debye length of 25 nm. The quality of the fits to DLVO theory for the interaction between these similarly prepared cellulose thin films and cellulose colloidal probe as a function of aqueous solutions conditions has been presented elsewhere.20,21,23 The adsorption of the PVAm and derivatives to the cellulose surfaces is expected to be dominated by the ionic interactions between the positively charged polymers at pH 3.8 and the slightly negatively charged cellulose. The polymers were adsorbed from a 100 ppm solution of NaCl with a concentration of 0.1 mM. Figure 3 shows the interaction between cellulose surfaces after the adsorption of the PVAm and derivatives at pH 3.8. The first point to note is the change in the interaction potential from attractive in the absence of the polymers to repulsive for all polymers. These force-distance curves were measured in the absence of free polymer in solution; that is, after adsorption, the liquid cell was flushed with pH adjusted electrolyte solution to remove any unadsorbed polymer. Thus, because
Senden, T. J. Curr. Opin. Colloid Interface Sci. 2001, 6, 95. Murat, M.; Grest, G. S. Macromolecules 1996, 29, 8282. Guffond, M. C.; Williams, D. R. M.; Sevick, E. M. Langmuir 1997, 13, Derjaguin, B. V. Kolloid Z. 1934, 69, 155. Deryagin, B.; Landau, L. Acta. Phys. Chem. 1941, 14, 633–662.
6900 DOI: 10.1021/la900256s
(31) Verwey, E. G. W.; Overbeek, J. T. G. Theory of the stability of lyophobic colloids; Elsevier: New York, 1948. (32) Chan, D. Y. C.; Horn, R. G. J. Chem. Phys. 1985, 83, 5311.
Langmuir 2009, 25(12), 6898–6904
Notley et al.
Figure 3. Interaction between cellulose surfaces with adsorbed PVAm derivatives at pH 3.8: (2), PVAm, (9), PVAm-Ph, and (]), PVAm-PBA.
the interaction has changed from attractive to repulsive, adsorption of polymer must have taken place. At this low pH, the PVAm and derivatives should have a relatively high positive charge which may interact strongly with the low negatively charged cellulose. Thus, a flat conformation of the polymers and low adsorbed amount at the solid-liquid interface may be expected. This is supported by the lack of any significant steric repulsive force in the observed forces on approach between the cellulose surfaces. Indeed, the surface forces on approach may be fit to DLVO theory under these solution conditions, with an example of the good agreement between the theory and experiment shown in Figure 4 for the adsorbed layers of PVAm-PBA on cellulose. Here, significant charge reversal due to the adsorption of the polymer must occur such that a surface potential of 39 mV is developed. The surface forces on approach for both the PVAm and PVAm-Ph may also be similarly fit within the limits of constant charge and constant potential with surface potentials of 38 and 36 mV, respectively; however, there is a slight difference in the fitted Debye lengths of 24.5 and 17 nm. This charge reversal mechanism upon adsorption of the PVAm and derivatives is supported by force-separation data collected at different pH values. As the solution pH is increased, the charge density on the polymers decreases. Figure 5 shows the force interaction curves as a function of pH for the cellulose surfaces with adsorbed PVAm-PBA. As the pH is increased from 3.8 to 7, the surface forces may be fit within the limits of constant charge and constant potential with progressively lower surface potentials. At pH 3.8, a surface potential of +39 mV is measured, while at pH 7 the surface potential has dropped to +30 mV. If the solution pH is further increased to 10, the data can no longer be fit using DLVO theory and now a steric repulsive interaction is observed. The steric repulsive force observed at high pH may be fit using various theories for the overlap of adsorbed polymer layers. These include the Alexander-de Gennes33,34 theory for neutral brush layers with a constant segment density profile (given by eq 1) and the Milner-Witten-Cates35 theory which assumes a parabolic segment density profile away from the interface (eq 2). The normalized force-distance curves may (33) Alexander, S. J. Phys. (Paris) 1977, 38, 983. (34) de Gennes, P.-G. Macromolecules 1981, 14, 1637–1644. (35) Milner, S. T.; Witten, T. A.; Cates, M. E. Macromolecules 1988, 21, 2610– 2619.
Langmuir 2009, 25(12), 6898–6904
Article
Figure 4. Fit to DLVO theory of the PVAm-PBA adsorbed layers on cellulose at pH 3.8 and with an added background ionic strength of 0.1 mM of NaCl. Data (b) are fit to the DLVO theory within the limits of constant charge and constant potential with a Hamaker constant of 5 10-21 J, ψ0 of +39 mV, and κ-1 of 21 nm.
Figure 5. Force-distance curves between cellulose surfaces with adsorbed PVAm-PBA as a function of pH: () pH 3.8, (b) pH 5, (]) pH 7, and (9) pH 10. Data for pH 3.8, 5, and 7 may be fit to DLVO theory within the limits of constant charge and potential with ψ0 of +39, +32, and +30 mV, respectively.
be fit to determine the grafting density Γ and the layer thickness L0. " -5=4 7=4 # F 16π D D 3=2 ¼ kB TL0 Γ 12 -7 -5 r 35 2L0 2L0 2 3 !2 !5 F 2L D 1 D 9 0 - 5 ¼ 4πP0 4 þ r 5 2L0 5 D 2L0
ð1Þ ð2Þ
with kB TN π2 l 4 P0 ¼ 2 12
!1=3 Γ5=3
where N is the number of polymer segments of length l. The steric repulsive force between the cellulose surfaces with adsorbed PVAm-PBA at pH 10 was fit to the Alexanderde Gennes theory and the Milner-Witten-Cates theory as shown in Figure 6. The layer thickness, L0, determined using these two theories agrees to within experimental error considering that the Milner-Witten-Cates theory was developed for two impenetrable brushes and not for a polymer brush confined by a DOI: 10.1021/la900256s
6901
Article
Figure 6. Interaction between two cellulose surfaces with adsorbed PVAm-PBA at pH 10. Solid line is the fit to the Alexander-de Gennes brush layer theory with a layer thickness of 16.5 nm and polymer chain spacing of 3.7 nm. Dashed line is the fit to the Milner-Witten-Cates theory with a layer thickness of 21 nm and prefactor of 0.37.
hard wall (a correction factor of L0 = 1.34L0* can be used to equate the two theories10). Both theories describe the interaction well at intermediate to high surface separations. It is interesting to note that neither theory is particularly satisfactory for short surface separations with an overestimation of the force predicted compared to that measured using the colloidal probe technique. This is most likely due to the splaying of polymer chains from the gap in between the surfaces effectively reducing the steric repulsive force.27,28 The forces on approach for all of the polymers showed a steric repulsive region at the high pH of 10 which is shown in Figure 7. The extent of the polymer layer is approximately the same for both the derivatized PVAm-Ph and PVAm-PBA; however, it is much reduced for the PVAm layers. The steric forces, in all cases, are substantially lower in magnitude than the forces of electrostatic origin for each of the polymers. This will be a key point in the discussion of adhesion presented below. The faster decaying repulsive forces as a function of surface separation should allow a more intimate contact between the two cellulose surfaces, giving rise to a stronger adhesion. The forces on separation of the cellulose sphere from the cellulose thin film were also measured as a function of pH and the type of adsorbed polymer. This allowed a measure of the relative adhesion in the system as determined from the maximum in the pull-off force. Care was taken to maintain a constant maximum applied load and unloading rate between experiments with different polymers. Some secondary adhesion events associated with the stretching of individual chains in a good solvent were also observed in between 2 and 5% of force curves for all polymers and at each pH studied (data not shown). These secondary events typically showed adhesive minima of at least 1 order of magnitude lower than any primary minima and at surface separations greater than 100 nm. Previously, the derivatized PVAm polymers were used in a study of the wet adhesion between never dried cellulose surfaces measured using peel testing.36 In that study, there was a significantly greater observed adhesion for the PVAm-PBA at low pH than for the PVAm and PVAm-Ph. Furthermore, as the pH was raised above the point where the charged boronate now existed, the adhesion between the cellulose surfaces with adsorbed layers of PVAm-PBA was greater still. Thus, there was a strong (36) Chen, W.; Lu, C.; Pelton, R. Biomacromolecules 2006, 7, 701–702.
6902 DOI: 10.1021/la900256s
Notley et al.
Figure 7. Interaction between two cellulose surfaces with adsorbed polymer layers at pH 10: (]) PVAm, (2) PVAm-Ph, and (0) PVAm-PBA.
Figure 8. Typical force curves on separation for the cellulose surfaces functionalized with adsorbed layers of PVAm, PVAmPh, and PVAm-PBA.
dependency between the delamination force and the pH of the solution. Here, in this study, the pull-off force was measured for the PVAm, PVAm-Ph, and PVAm-PBA as a function of pH. A minimum of 100 force curves were used to statistically determine the pull-off force. Figure 8 shows typical force-distance curves on separation of the cellulose surfaces with the adsorbed polyvinylamine and derivatives at pH 10. As can be seen in Figure 8, the maximum pull-off force, Fmax as shown, was determined for each of the greater than 100 curves for each polymer and pH condition. The distribution of the magnitude of the pull-off force is shown in Figure 9 for PVAm and PVAm-PBA at pH 10. The pull-off force for each force-distance curve was determined, and a histogram prepared to demonstrate the adhesion between asperity contacts on both surfaces approximates a Gaussian distribution. At this pH, there is a 1 order of magnitude difference between the adhesion values for the PVAm and the phenylboronic acid derivative. This is in qualitative agreement with the previous study which used peel testing to determine the adhesion between cellulose coated with these polymers.36 The average pull-off force for each pH and the three polymers was determined along with the standard deviation. These values are collated in Figure 10. At pH less than 10, no primary adhesion, as measured as the pull-off force, was observed for the PVAm or PVAm-Ph. Occasionally, as mentioned above, some secondary adhesion events were observed due to the stretching of individual polymer chains; however, this contributes only minimally to the overall adhesion. At pH 10, a small pull-off force of 0.26 nN was measured for PVAm whereas the pull-off force for PVAm-Ph was Langmuir 2009, 25(12), 6898–6904
Notley et al.
Article
Figure 9. Distribution of pull-off forces between cellulose with adsorbed layers of PVAm (left) and PVAm-PBA (right).
Figure 10. Pull-off force between cellulose surfaces with adsorbed polymer layers as a function of pH: (O) PVAm, (2) PVAm-Ph, and (0) PVAm-PBA.
more than double at 0.55 nN. However, the adhesion between cellulose surfaces with adsorbed PVAm-PBA was significantly greater at all pH than that for either other polymer. Furthermore, there was a general trend of increasing pull-off force with increasing pH in agreement with the previous study.36
Discussion The development of adhesion between surfaces with adsorbed polymer layers can be thought of as a number of competing mechanisms. Strong adhesion may occur if the two surfaces are brought into very close contact such that the attractive forces due to the dispersion interactions may dominate. Any adsorbed polymer or polyelectrolyte layer will change the effective surface forces interaction in solution.37-39 Depending on the adsorbed layer conformation, adhesion will be either promoted or inhibited. For example, a highly extended polymer layer may bridge two surfaces, causing a weak attractive force and hence adhesion. Alternately, a high surface coverage and short extension of the polymer layer into solution may stabilize suspension against (37) Napper, D. H. Polymeric Stabilization of Colloidal Dispersions; Academic Press: New York, 1983. (38) Fleer, G. J.; Cohen Stuart, M. A.; Scheutjens, J. M. H. M.; Cosgrove, T.; Vincent, B. Polymers at interfaces; Chapman and Hall: London, UK, 1993. (39) Netz, R. R.; Andelman, D. Phys. Rep. 2003, 380, 1–95.
Langmuir 2009, 25(12), 6898–6904
aggregation. Furthermore, if polyelectrolytes are used, an added electrosteric force may prevent the close contact of the surfaces, thus reducing the adhesive interaction. These mechanisms assume that the particles or surfaces are coming together through either Brownian motion or hydrodynamic forces; however, in some instances, the surfaces will be forced together under an applied load or through capillary forces on drying. Here, the extension of the polymer away from the interface will still be highly influential. Extended polymer layers will be able to contribute to the adhesion through intermingling with polymer chains located on the adjacent surface, providing for an effectively welded joint. Upon separation, the interdigitated chains must be disentangled, a process requiring a finite time, with rapid separation resulting in chain rupture contributing to the measured adhesion energy.40 Real surfaces have a degree of roughness which will ultimately reduce the molecular contact area in an adhesive joint. Polymer layers which extend beyond the scale of this inherent roughness will hence provide the best opportunity for improving adhesion through contacts with other polymer segments or the other surface to which it was not originally adsorbed. The data in this current study support this idea. At low pH, all of the polymers under investigation essentially lie in a flat conformation at the interface. The best evidence of this fact is that the surface forces on approach show no steric interaction and can be well fit to the DLVO theory. Indeed, upon adsorption, there is a change from an attractive force to repulsive force which implies a charge reversal from slightly negative to positive. The result of this charge reversal is twofold: first, there is a relatively strong electrostatic repulsion which inhibits the close approach of the two surfaces, preventing the attractive dispersion forces from dominating the interaction; second, the polymer chains do not extend into solution to any great degree, preventing the development of adhesion through interdigitation of polymer segments. As the pH of the solution is increased, steric forces develop, which correlates well with the observed pull-off forces. At pH 10, all polymers show a steric interaction on approach as well as a pull-off force on separation. Interestingly, the extent of the steric force is greater for the derivatized PVAm-Ph and PVAmPBA than for the simpler homopolymer. This is perhaps due to the relatively fewer charged groups at the higher pH for the substituted PVAm polymers as the pKa of the amine groups is (40) Chen, Y. L.; Helm, C. A.; Israelachvili, J. N. J. Phys. Chem. 1991, 95, 10736.
DOI: 10.1021/la900256s
6903
Article
shifted. The pull-off force scales with the steric interaction, indicating the importance of chain intermingling at the interface between the cellulose surfaces. However, the pull-off force for PVAm-PBA is some orders of magnitude greater than that for the other polymers, and indeed, a pull-off force is observed for the PVAm-PBA at all pH values which is not present for the other two polymers. This implies that another mechanism is perhaps of greater importance. It is well-established that boronic acid groups can form ester linkages with cis-diols which are present on most polysaccharides including in the macromolecular structure of cellulose,41,42 and this was the suggested mechanism behind the observed increase in wet adhesion in an earlier study.36 Previously, this esterification reaction has been exploited in the assembly of layer-by-layer grown polymeric multilayer structures.43,44 The PBA group binds relatively weakly to the cis-diol at lower pH; however, it is still strong enough to impart a pull-off force significantly greater than that of the unsubstituted PVAm. There is a small increase in the binding affinity as a function of increasing pH which correlates well with the observed increase in adhesion in the pH range of 3.8-7. Furthermore, at pH greater than the pKa of the PBA (between 8 and 9, depending on the presence of adjacent amine groups42), a charged boronate is formed which is considered to be far more reactive than the uncharged boronate. Thus, the adhesion in the pH region greater than 7 increases significantly (41) Niwa, M.; Sawada, T.; Higashi, N. Langmuir 1998, 14, 3916–3920. (42) Yan, J.; Springsteen, G.; Deeter, S.; Wang, B. Tetrahedron 2004, 60, 11205– 11209. (43) Recksiedler, C. L.; Deore, B. A.; Freund, M. S. Langmuir 2006, 22, 2811– 2815. (44) Zhang, D.; Tanaka, H.; Pelton, R. Langmuir 2007, 23, 8806–8809.
6904 DOI: 10.1021/la900256s
Notley et al.
because of the increased binding affinity but also because of the formation of a steric layer. This type of adhesion developed through the spontaneous chemical reaction of the boronate with the diol in the absence of any external stimulus such as heat is significant and represents an interesting new class of polymeric adhesives.
1.
2.
3.
Conclusions The influence of polymer structure and pH on adhesion closely agree with our previous macroscopic peeling measurements using very different cellulose surfaces. There exists a strong repulsion between aqueous cellulose surfaces saturated with PVAm-PBA, meaning the surfaces must be forced together to give an adhesive joint in water. At low pH, the replusion has an electrostatic character, whereas it is more typical of steric repulsion at high pH where the PVAm-PBA is zwitterionic. The specific, pH dependent bonding of phenylboronic acid to cellulose is responsible for the adhesion in water. We propose that the amine-rich environment in PVAm-PBA promotes the boronate ionization.
Acknowledgment. S.M.N. would like to thank Tim Senden and Vince Craig, ANU, for valuable discussions and continued guidance. Also S.M.N. would like to acknowledge the CRC Functional Communication Surfaces (CRC SmartPrint) for funding.
Langmuir 2009, 25(12), 6898–6904
