Article pubs.acs.org/JPCB
Immobilization and Surface Functionalization of Gold Nanoparticles Monitored via Streaming Current/Potential Measurements Kyrylo Greben,* Pinggui Li, Dirk Mayer, Andreas Offenhaü sser, and Roger Wördenweber Peter Grünberg Institute (PGI-8), Forschungszentrum Jülich GmbH, Jülich 52425, Germany ABSTRACT: A streaming current/potential method is optimized and used for the analysis of the variation of the surface potential upon chemical modifications of a complex interface consisting of different organic molecules and gold nanoparticles (AuNPs). The surfaces of Si/SiO2 substrates modified with 3-aminopropyltriethoxysilane (APTES), AuNPs, and 11-amino-1-undecanethiol (aminothiols) are analyzed via pH and time dependent ζ potential measurements that reveal the stability and modification of the surface and identify crucial parameters for each individual preparation step. For instance, surface activation and especially molecular adsorbate layers tend not to be stable in time, whereas the substrate and the AuNPs provide a stable surface potential as long as impurities are avoided. It is shown that the streaming potential/current technique represents an ideal tool to analyze and monitor the complex surfaces and their modification.
■
neurons.17−20 However, also the modification of biosensor surfaces often involves electrostatic interactions between surface tethered capture molecules and proteins, polypeptides, oligonucleotides, or nanomaterials, which are transducing or enhancing the sensor signal.21−23 The balance of surface charges can have a strong influence on the conformation of receptor molecules, on the efficiency of target binding, and finally on the detectable signal. Consequently, it is essential to be able to analyze the surface charge of a potential substrate for bioelectronic applications under conditions that are identical or at least comparable to the conditions used during deposition or immobilization of biomaterial. For example, since the adhesion occurs in the electrolyte solution, the cells sense the solid surface including ions from the solution, which are screening the surface. Therefore, the cells are sensible to the ζ potential of a given substrate. In this work, we modified planar biocompatible surfaces among others with molecular layers and gold nanoparticles that potentially could promote cell adhesion, as shown in ref 16. The focus of this work, however, lies in the analysis of the surface potential of each individual step in the immobilization and functionalization process. There are a number of conditions and limitations for the characterization technique of the surface potential of substrates for the envisaged bioelectronic experiment and application. For instance, (i) the measurements should be performed in a biocompatible environment, i.e., aqueous solutions, (ii) surfaces can be
INTRODUCTION The optimization of the interface between cells and substrate plays a major role in nowadays biomaterial and bioelectronic research.1 For example, (i) cell adhesion to artificial substrates represents a key element of biotechnological applications, (ii) the controlled immobilization of neurons and guidance of neurite outgrowths on the surface of an electronic sensor device gain increasing interest for fundamental aspects of neurobiology such as neuritogenesis and neural development, and (iii) a welldefined and stable contact between cells and the substrate’s surface is of importance for an optimal signal transfer in bioelectronic devices.2 Generally, the growth of any material (ranging from inorganic to organic) on a substrate strongly depends on the properties of the substrate, especially the surface properties. The chemical composition and structure of the substrate’s surface play a major role for the properties and quality of epitaxial inorganic films, whereas in the case of adhesion of biological material different strategies can be considered (e.g., biological, chemical, topographical, or physical).3 Besides the chemical contrast, the surface topography can act as a guiding cue for cell attachment.4,5 Even sub-100 nm structures are capable of influencing cells.6 Furthermore, gold nanoparticles (AuNPs), immobilized with defined interparticle spacing on surfaces7 and functionalized with organic ligands,8 can serve as both electrode surface9 carriers of single peptide guidance factors.10−14 Additionally, it has been recognized that, in particular, positively charged domains of peptides assist the adhesion of neurons.15,16 This finding was supported by the observation that positive charges associated with surface-bound, synthetic molecules containing amino groups can also promote the adhesion and growth of © XXXX American Chemical Society
Received: March 18, 2015 Revised: April 22, 2015
A
DOI: 10.1021/acs.jpcb.5b02615 J. Phys. Chem. B XXXX, XXX, XXX−XXX
Article
The Journal of Physical Chemistry B
Figure 1. Schematic description of the sample preparation, i.e., (A) Si/SiO2 substrate, (B) substrate functionalized with APTES, (C) immobilization of citrate-stabilized AuNPs, (D) removal of free citrates and APTES ligands, and (E) decoration of AuNPs with aminothiols. For details, see text.
the IEPs of given surfaces, which authors consider most informative. In order to optimize the measurement, a number of different approaches have been carried out. Both electrolyte and titration solution are aqueous solutions based on potassium chloride (KCl) in the case of working electrolyte and hydrochloric acid (HCl) in the case of the titration electrolyte with concentrations of 1 and 50 mM, respectively. Different types of purified water (Milli-Q water, distilled water, and double distilled water (DDW)) are tested. Only water with a conductivity 10 μS/cm is added in Figure 2. For this experiment, the values of the ζ potential are considerably lower, especially after titration, and the resulting IEP of 3.3 is unrealistically small. (iii) However, even if the measurement procedure is executed carefully, in the standard way, impurities seem to develop during the titration steps that reduce the IEP, e.g., to a pH of 3.65 in the case of the example shown in Figure 2. Therefore, we modified the standard measurement procedure by introducing an additional purging step after each titration step. In the standard procedure, after addition of the titration electrolyte into the working electrolyte, a number of measuring cycles are performed that deliver one averaged value for the ζ potential. Then, the next titration step takes place. By repeating the measurement with additional exchange of the working electrolyte between the reservoir (where the working electrolyte is purged with the N2) and the measuring cell, it seems to be possible to establish stable experimental conditions and obtain reliable results. As a result, the ζ potential increases with each data point taken at a given titration step. Figure 2 shows two examples of such a measurement. In one of the measurements (solid circles), only two titration steps are chosen. The impact of the additional purging of the electrolyte is clearly visible. We believe that the titration solution contains additional ions, since it is not purged. By adding the titration solution to the electrolyte, these impurities (e.g., hydrocarbonates formed by the dissolution of CO2) will directly be pumped into the measuring cell and lead to a reduction of the ζ potential. By repeating the measurement, the titrated working electrolyte is additionally purged; thus, the impurities are removed and the ζ potential and IEP increase. The second measurement shown in Figure 2 (open squares) is performed with smaller titration steps. The effect of the additional purging is also visible; moreover, both data sets are on one line and the IEP is as expected at a pH of 4.0. Therefore, we conclude that the additional purging with inert gas at each titration step is essential for obtaining reliable and reproducible data. In order to investigate the modification of the surface potential during the process of immobilization and functionalization of AuNPs on silicon oxide, detailed streaming current measurements are performed as a function of pH and time for each preparation step. The streaming current data are given in Figures 3−5, representing the pH dependence of the ζ potential (Figure 3), the time dependence of the ζ potential (Figure 4), and the IEPs of the different process steps (Figure 5). In the following, we will discuss these data for each individual process step separately. Step A, Pure Substrate. The n-doped and SiO2 terminated substrates represent the starting point of all sample preparations. These surfaces are typical representatives of relatively simple interfaces, where a solid inorganic surface and a liquid electrolyte are brought into contact. Nevertheless, the Si surface represents a well studied surface; different electrokinetic methods36−42 reveal small differences indicating that the
shell and the amino groups of silanes. This makes the surface amphoteric. Step D. Free citrate ligands and APTES ligands are removed by an oxygen plasma treatment (2 min at 200 W power and 1.4 mbar O2 pressure), leaving the bare AuNPs on the (again activated) SiO2 surface. Step E. Finally, the AuNPs are decorated with a new ligand shell that is comprised of linear molecules with a thiol headgroup and an amine functional group with 11-amino-1undecanethiol (aminothiols) in order to obtain a positively charged surface that could, for instance, be used as an attractive interface for (guided) growth of neurons. After each of these process steps, streaming current measurements are performed to characterize the resulting change of the surface charge. Furthermore, SEM images provide information about the distribution (and potential change of the distribution due to migration and clustering) of the AuNPs before and after streaming current measurement.
■
RESULTS AND DISCUSSION Test measurements using inert polypropylene foil (an isotactic polymer, supplied by Anton Paar GmbH) are executed in order to verify the calibration by observing the expected behavior of inert samples and an IEP of ∼4.35 These measurements are described below in detail in order to sketch the measurement procedure and the precautions taken to optimize the analysis. Figure 2 shows the optimization of the measurement procedure using polypropylene foils, as a tested calibration
Figure 2. ζ potential of polypropylene foils as a function of the pH value of the electrolyte obtained for different measurement conditions starting with 1 mM KCl electrolyte and decreasing the pH value by titration with 50 mM HCl electrolyte solution. Water with higher conductivity (triangles) yields the smallest ζ potential and IEP; in all other cases, DDW with conductivity
