Quantitative Analysis of Thickness and pH-Actuation of Weak

Subscriber access provided by Kaohsiung Medical University

C: Surfaces, Interfaces, Porous Materials, and Catalysis

Quantitative Analysis of Thickness and pHActuation of Weak Polyelectrolyte Brushes Gustav Ferrand-Drake del Castillo, Gustav Emilsson, and Andreas B. Dahlin J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.8b09171 • Publication Date (Web): 08 Nov 2018 Downloaded from http://pubs.acs.org on November 10, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Quantitative Analysis of Thickness and pHActuation of Weak Polyelectrolyte Brushes Gustav Ferrand-Drake del Castillo, Gustav Emilsson and Andreas Dahlin.* AUTHOR ADDRESS Department of Chemistry and Chemical Engineering, Chalmers University of Technology, 41296 Gothenburg, Sweden

ACS Paragon Plus Environment

1

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 36

ABSTRACT: Polymer brushes are widely used as surface coatings for various inert, functional or responsive interfaces. If the polymer can alter its protonation state (a polyelectrolyte), the brush can switch between a collapsed and swelled state with pH, which enables applications such as nanoscale actuators. However, changes in brush height as the polymer alters its charge state are not straightforward to measure accurately. Here we show how surface plasmon resonance can be used to determine the thickness of PE brushes both in their charged and neutral states. We use different methods to measure the heights of brushes consisting of poly(acrylic acid) and the polybasic poly(2-(diethylamino)ethyl methacrylate), both prepared by atom transfer radical polymerization. We find polymers in solution that can act as refractive index probes which do not interact with the grafted polyelectrolytes, thus providing an “exclusion height” of the brush. Importantly, the angular reflection spectrum can be used to directly identify if a probe is indeed non-interacting. Furthermore, using different non-interacting probes results in small but significant changes (~10%) in the exclusion height as long as the probe is reasonably large (approximately >2 kg/mol). These differences cannot be attributed to probe charge. Data from multiple brushes shows that the relative height increase (at physiological ionic strength), i.e. the “collapse ratio” upon charging due to pH alterations, increases with the absolute brush height. In addition, we show that the plasmonic response to the pH switching of the polyelectrolyte brush is opposite to the response of hydrophilic polymer brushes collapsing at the lower critical solution temperature. This phenomenon is explained by an increase in refractometric constant upon charging. Our study shows that surface plasmon resonance is an excellent tool for characterizing polyelectrolyte brushes and provides useful insights on pH actuation not easily obtained by other methods.


2

Page 3 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


INTRODUCTION End-grafter polymers stretch out from the surface when the grafting density is sufficiently high and assume the so-called “brush” configuration.1 Polymer brushes have a long history related to understanding the physics of macromolecules and have also attracted a lot of attention in various applications.2 The most known example is probably the use of hydrophilic polymer brushes to create surfaces that are inert towards adsorption of proteins and cell attachment. Lately, there has been a focus on brushes containing charged (or zwitterionic) monomers, which introduces several additional energetic and entropic effects related to counterions.3 A brush that carries a net positive or negative charge determined by the solution pH is known as a weak polyelectrolyte (PE) brush.3, 4, 5

In contrast to their neutral counterparts, PE brushes are often very “sticky” towards other

macromolecules and suitable for immobilizing biomolecules such as enzymes.6 From a fundamental point of view, PE brushes are interesting since surface sensitive techniques can be employed to provide information about the polyelectrolytes themselves as well as their complex supramolecular interactions.7 In addition, with PE brushes it is possible to create various responsive interfaces because, as the local pH is altered, the morphology switches between a compact (neutral, more hydrophobic) and an extended (charged, more hydrophilic) state. This makes it possible to construct nanoscale actuators,8 for instance through local pH changes,9 and enables new application such as transducers for label-free sensors.10 However, there are relatively few quantitative studies of actuation of PE brushes by pH. Much existing literature has instead focused on quantifying swelling/contraction by changes in ionic strength,11, 12, 13, 14 which is arguably a fascinating topic but less connected to sensors and actuators. In general, it is far from straightforward to probe the thickness of soft and hydrated organic layers. Atomic force microscopy operated in indentation mode is typically noisy and the contact point is


3


Page 4 of 36

identified once the probe has started to compress the brush.15, 16 Quartz crystal microbalance with dissipation monitoring (QCMD) is another option, but accurate determination of thickness becomes difficult due to complicated multiparameter models for viscoelastic layers. Even in optical techniques like ellipsometry, thickness and refractive index (RI) are strongly coupled parameters and not straightforward to obtain separately. In fact, polymer brushes are particularly complicated to model accurately since (under reasonably good solvent conditions) the monomer density profile vs distance from the surface is a parabolic function,17 which makes the film heterogeneous in the direction perpendicular to the surface. Lately, several new methods for thickness determination by surface plasmon resonance (SPR) have been introduced. One method is based on utilizing changes in RI with wavelength,18 but such dispersive effects are very small and/or unknown for most organic matter. One can also utilize the difference in field extension with wavelength,19 but this approach can only measure heights accurately when the film is highly hydrated.16 Another SPR method is based on injections of non-interacting probes which provide an “exclusion height”, i.e. the characteristic height below which the probe molecules are expelled from the brush. The non-interacting probes method is highly precise20, 21, 22 and even applicable to dense films in its improved form.16 However, measuring the exclusion heights of PE brushes has never been done. The reason is that the probe, which in existing studies has been bovine serum albumin (BSA), must not be attracted to the layer for which the height should be determined. Proteins do typically bind strongly to polyelectrolytes,23 often even when carrying the same net charge.24 In this work we present a quantitative investigation of pH actuation of PE brushes obtained by new SPR methodology. We measure the height increase when the brushes go from the neutral to the charged state by pH changes in a water environment at physiological ionic strength. To provide


4

Page 5 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


a complete picture, we study both negatively charged poly(acrylic acid)11, 14 (PAA) and positively charged poly(2-(diethylamino)ethyl methacrylate)12, 25 (PDEA) brushes. Both brushes are prepared by atom transfer radical polymerization26 (ATRP), which can be monitored in real-time by SPR. Importantly, we show that macromolecules can be found which do act as non-interacting probes, thereby enabling accurate measurements of PE brush exclusion heights. Furthermore, by proper analysis, the SPR spectrum can be used to directly check if a probe is non-interacting and any influence from probe type can be elucidated. The measured exclusion heights are compared with a “de Feijter approach”27 based on measuring the surface coverage in the dry state and the refractivity of the polymer. Finally, we explain why the pH switching of the brush provides an SPR response which is the opposite of that observed from brush collapse at the lower critical solution temperature (LCST).16 Besides providing further insights into pH switching of weak PE brushes, this study provides useful methodology for characterizing thin hydrated layers with emphasis on thickness control.

MATERIALS AND METHODS Chemicals. H2O2 (30%) and NH4OH (28-30%) were from ACROS, while H2SO4 (98%) and ethanol (99.5%) were from SOLVECO. The chemicals for polymerization were 2(diethylamino)ethyl methacrylate (DEA), tert-butyl acrylate (inhibitor was removed from DEA and tert-butyl acrylate using an alumina column, after which they were stored at -20 °C and warmed to room temperature immediately before use), sodium acrylate, methanol, dimethylformamide, anisole, dichloromethane, methanesulfonic acid, ligand N,N,N’,N’’pentamethyldiethylenetriamine (PMDTA), CuBr2 (catalyst) and reduction agent L-ascorbic acid. The initiator bis[2-(2-bromoisobutyryloxy)undecyl]disulfide (DTBU) was used for monolayer


5


Page 6 of 36

formation on gold. Monolayers of oligo(ethylene glycol) thiols was prepared using HS-C11-EG6OCH2-OH (ProChimia, Gdansk, Poland). Buffers used in this work was phosphate buffered saline (PBS) tablets (0.01 M phosphate, 0.13 M NaCl, pH 7.4). Bulk polymers used as probes for PE brush height determination were PAA as sodium salt (30 kg/mol in aqueous solution), PAA (2 kg/mol in powder form and 100 kg/mol in liquid form), 35 or 8 kg/mol poly(ethylene glycol) (PEG), and poly(4-styrenesulfonic acid) (PSS, 70 kg/mol in aqueous solution). Water used in this work was ASTM research grade Type 1 ultrafiltered water, referred to as milli-Q-water from here on. All chemicals were purchased from Sigma-Aldrich unless stated otherwise. Cleaning. Prior to surface functionalization SPR chips (50 nm Au on glass) were cleaned in piranha wash (H2SO4:H2O2, 3:1 v/v) for 10 min followed by rinse in milli-Q, followed by RCA1 wash (H2O:H2O2:NH4OH 5:1:1 v/v at 75 °C for 20 min), followed by another rinse in milli-Q and sonication in ethanol. Gold sensor crystals for complementary measurements using quartz crystal microbalance with dissipation monitoring (QCMD) were purchased from Biolin Scientific and treated similarly. SAM Formation. Clean SPR chips were immersed in 2 mM initiator solution in ethanol. The substrates were exposed for 12-18 h. After incubation substrates were sonicated and rinsed in ethanol for 1 min then dried with N2. Polymerization. Activator regenerating electron transfer ATRP was used to make PAA and PDEA polymer brushes in a manner similar to previously published ATRP protocols.12, 16, 25, 28 For PDEA polymerization a solvent mixture of methanol and water 4:1 (v/v) was used, while for PAA the solvent composition was instead 1:4 (v/v) to promote initiation of the reaction. The monomer and reaction medium were deoxygenated in N2 separately for 30 min in septa covered Schlenk flasks. Transfer of reaction mixture and monomer into reaction vessel (screw-top jar with rubber


6

Page 7 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


septa) with initiator-prepared samples was performed with needle and vacuum pump. The reaction was initiated by the addition of ascorbic acid using a needle. For PDEA the final concentrations of each component in the reaction medium were: [monomer] = 1.24 M, [CuBr2] = 0.6 mM, [PMDTA] = 6.2 mM and [ascorbic acid] = 6.2 mM. For PAA the final concentrations of each component were: [monomer] = 2.64 M, [CuBr2] = 2.6 mM, [PMDTA] = 26.5 mM, [ascorbic acid] = 26.5 mM. Reactions were quenched by immersing the samples in pure ethanol. An alternative method of making PAA was also used with tert-butyl acrylate as the starting monomer, which is converted to PAA by post-modification after polymerization by cleaving of the tert-butyl group.29, 30, 31 This alternative path differ primarily by only using organic solvents as the reaction medium consisting of dimethylformamide and anisole. Deoxygenation, transfer or liquids and ignition of reaction was performed in the same way as for the other ATRP reactions. The final concentrations of each component in the reaction medium were: [monomer] = 1.24 M, [CuBr2] = 0.6 mM, [PMDTA] = 6.2 mM and [ascorbic acid] = 6.2 mM. Poly(tert-butyl acrylate) (PTBA) was converted to PAA by 15 min exposure to 0.2 mM methane sulfonic acid in dichloromethane, followed by rinsing in dichloromethane and ethanol (Scheme 1). Throughout this study no differences were observed for PAA brushes synthesized via route II in Scheme 1 (as compared to route I).


7


Page 8 of 36

Scheme 1. Preparation of PE brushes on gold. Self-assembly of polymerization inhibitor (top) and polymerization reaction schemes for (I) PAA, (II) PTBA (afterwards converted to PAA) and (III) PDEA. SPR Measurements. Sensor surfaces with a 50 nm gold layer were used in all experiments. Measurements were performed on a SPR NaviTM 220A instrument (BioNavis). The total internal


8

Page 9 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


reflection (TIR) and SPR angles were recorded by continuous goniometric scanning on the three laser wavelengths 670, 785 and 980 nm and in two flow cell channels. The flow rate used was 20 µL/min in PBS or TRIS buffer. Brush thickness measurements were performed by dissolving 2030 g/L of each probe in the buffer used. The buffer pH was adjusted using 1 M HCl or 1 M NaOH to pH 4.0 or pH 8.0. The pH was tested using a Metrohm Porotrode. After dissolving acids and bases the pH was tested again and if required readjusted to pH 4.0 or 8.0. In all experiments the ionic strength due to added HCl or NaOH did not significantly change when buffers or samples are titrated to the desired pH (up to 5% of the background buffer salt). Buffer solutions were degassed prior to experiments. Spectral Analysis. The DTBU monolayer was assumed to have nDTBU = 1.45.32 The RI of the dry polymer brushes was set to nPAA=1.52233, 34 or nPDEA=1.517,13, 34 i.e. the values for the polymers as bulk materials. The Fresnel models,35 their implementation in MATLAB and their validity for modelling the angular SPR spectrum are described in previous reports.16, 22 A summary of RI and thickness values used for all layers of the bare and initiator-modified SPR sensor (for each laser wavelength) is given in Supporting Information. After obtaining the thickness of the dry layers from the Fresnel models, the mass coverage was determined by dividing with the density of the polymer, which was assumed to be approximately 1.22 g/cm3 for both PAA and PDEA.6

RESULTS AND DISCUSSION Brush Synthesis. In order to get a first approximation for the thickness of the PE brushes and to run the polymerization for an appropriate time, we monitored the ATRP reaction in real-time using SPR as described previously for poly(N-isopropylacrylamide).16 Figure 1A shows the evolution of the SPR reflectivity spectrum after initiating the reaction (at 0 min) by injecting


9


Page 10 of 36

ascorbic acid, reducing Cu ions so that the tertiary Br can be removed from the surface anchored initiator to start the ATRP. Since the TIR angle only shows minute changes (Figure 1A), the RI of the bulk liquid remains constant36 and hence it is confirmed that only surface reactions are monitored.16 The shift in resonance angle vs time is plotted in Figure 1B. Importantly, as the brush thickness is expected to be comparable to the evanescent field extension, this signal is not simply proportional to brush thickness.16 To obtain the accurate brush height we fitted Fresnel models and used a constant RI for the growing film set to n = 1.452 obtained by fitting (R2 = 0.9924) data in Figure 1B to an exponential function.16 This value is significantly higher than the RI of the water/methanol solvent (1.352) and closer to that of the dry polymer (1.517), which confirms that the polymerization medium is not a good solvent (as the volume fraction of polymer is over 50%). This is in part due to the fraction of water introduced to the methanol.25 Thus the growing film can be fairly accurately modelled as a single homogenous layer.16 Figure 1C shows the calculated thickness evolution with time and an excellent fit to the expected growth kinetics:37 𝑑(𝑡) = 𝛼 log(1 + 𝛽𝑡)

(1)

For the ATRP of PDEA in Figure 1C, we fitted α = 130 nm and β = 0.046 min-1 for PDEA. The parameters were comparable for PAA. Figure 1C also includes a few values for dry brush height for PDEA obtained by terminating the reaction and measuring the SPR spectrum in air, as we have done for other polymers in previous studies.16, 22, 38, 39 We note that the initial growth rate is ~6 nm/min and decreases slowly, i.e. the polymerization is fairly “living” but termination events start to become significant after ~10 min. One reason is the presence of water.26 The variation in dry thickness was around 10% when preparing brushes repeatedly by our protocol. After synthesis the PE brushes were chemically characterized by several standard surface analysis methods (Supporting Information).


10

Page 11 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1. Example of real-time polymerization SPR data (here for PDEA). (A) Reflectivity as a function of incident angle for some of the measured spectra throughout the polymerization. The TIR angle inset shows that the bulk solution refractive index stays stable throughout the polymerization. (B) Plasmon angle shift as function of polymerization time. (C) Thickness in the polymerization medium shown as function of time where dots show the corresponding Fresnel model results. The solid line shows the fit to Equation 1 where dashed lines are 95% confidence intervals. Some values for the dry thickness are also shown (cross marks).


11


Page 12 of 36

Response to pH and Determination of pKa. The switching behavior of the weak PE brushes was investigated using multiple techniques by injecting buffers of different pH and monitoring the response. For all buffers we maintained a physiological ionic strength and hence the PE brushes are much thicker than the Debye length (

Quantitative Analysis of Thickness and pH-Actuation of Weak

Recommend Documents