Journal of The Electrochemical Society, 162 (4) D174-D179 (2015) 0013-4651/2015/162(4)/D174/6/$33.00 © The Electrochemical Society
Continuous Electrophoretic Deposition and Electrophoretic Mobility of Ligand-Free, Metal Nanoparticles in Liquid Flow Sven Koenen,a Ren´e Streubel,a Jurij Jakobi,a Kerstin Schwabe,b Joachim K. Krauss,b and Stephan Barcikowskia,z a University of Duisburg-Essen and Center for Nanointegration b Hannover Medical School, 30625 Hannover, Germany
Duisburg-Essen (CENIDE), Essen 45141, Germany
Direct current electrophoretic deposition (DC-EPD) of ligand-free metal nanoparticles in a flow-through reactor is studied by analyzing the educt colloid and the outflow of the flow through chamber while the concentration of the colloid and the strength of the electric field is varied. Metal nanoparticles synthesized by pulsed laser ablation in liquid (PLAL) are used to ensure that the colloidal nanoparticle surface is free of any ligands and that the colloid’s stability and movement in an electric field is solely influenced by electrostatic forces. Electrophoretic mobility and deposition kinetics of these ligand-free nanoparticles on plain surfaces are examined for different electric field strengths. Additionally, a continuous liquid flow DC-EPD process is presented and optimized regarding deposition rate, colloid stability, and liquid flow rate. The reported parameter window for high deposition rates of nanoparticles without a negative impact on the colloid, allows to define an efficient stationary EPD process suitable for high throughput applications. © 2015 The Electrochemical Society. [DOI: 10.1149/2.0811504jes] All rights reserved. Manuscript submitted December 12, 2014; revised manuscript received January 28, 2015. Published February 7, 2015.
Electrophoretic deposition (EPD) is a process in which charged colloidal nanoparticles are guided by an applied electric field in order to adsorb suitable particles on an adsorbant substrate. An often-used application is the deposition of nanoparticles to create nano- and microstructures.1 Electrophoretic deposition generally requires a system containing two electrodes.2 Charged particles move in liquid in the direction of the oppositely charged electrode. During EPD, electrophoresis of charged particles in solution is followed by adsorption or assembling of particles on the oppositely charged electrode.2–4 This electrode is also the desired product of the EPD-based nanostructuring, e.g. for use as a medical device like a nanostructured neural electrode.5 EPD is a relatively simple and cost efficient technique. Therefore, it is highly suitable for coatings of a broad range of materials e.g. metals or biomaterials.5–9 As the applied electric field is directed perpendicularly to the curved surface of a shaped target and the particles follow this direction (with some limitations), EPD is predestined for controlled and fast coating of 3-dimensional materials.10 Since first experiments in the field of electrophoresis were performed by Bose in 174011 the process of EPD was studied by several researchers (which is reviewed elsewhere).1 The first study concerning the rate of nanoparticle deposition on a substrate was investigated by Hamaker in 1940.12 Since then EPD has been established as a reliable and promising application technique.13–15 Despite these different studies on the field of EPD there is still a need for a better understanding of the process itself. As a consequence, a suitable model system for EPD is desirable. EPD of nanoparticles is often carried out with a variety of different surfactants or ligands like citrate, polymers or other additives, which may, at least partially, block the particle’s surface.7,16 The approach in this work is to develop an EPD system independent from the impact of such ligands. Therefore, ligand-free and purely electrostatically stabilized nanoparticles are required, which can be synthesized using pulsed laser ablation in liquids (PLAL). During PLAL a pulsed laser is focused on a metal target that is placed in a liquid. The laser pulse is adsorbed by the metal target and generates a plasma plume that contains the ablated material.17 The plasma plume starts to grow, induces a shockwave in the liquid medium and leads to a cavitation bubble. The cavitation bubble expands and eventually collapses while upon its collapse the ablated nanoparticles are released into the liquid.17 The whole process starting from the laser shot and resulting in generated nanoparticles takes about hundreds of microseconds.18 To generate nanoparticle concentrations in the range of 50–100 μg/mL the process is repeated for
E-mail: [email protected]
several minutes. Further details on the mechanism of PLAL can be found in the following review articles.18–20 PLAL is an established process in the field of nanotechnology.18,21,22 By PLAL a variety of nanomaterials can be synthesized that can be used in EPD. This variety is simply achieved by changing the solution medium or the metal target such as Au,23 Ag,24 Pt,25 Ni,26 Ti,27 Fe.28 Furthermore, PLAL enables the generation of nanoparticles without reducing agents or stabilizers because of the high surface charge29 and the electrostatic stabilization of particles.30 The surface charge of the laser generated nanoparticles derives from the fact that the surface atoms are partly oxidized. Different workgroups have shown that some surface atoms of gold nanoparticles bear the oxidation states of +1 and +3.23,29,31 The resulting negative zeta potential of the gold nanoparticles in water is due to the adsorption of oxygen and carbon dioxide forming Au-O− and AuCO3 − compounds. Furthermore, capped Au-O-Au groups were also reported on the surfaces of laser-fabricated gold nanoparticles in aqueous media.23,31 Notably, bare metal nanoparticles perfectly agree with the properties of the Stokes’ “hard sphere” (without coating), one of the boundary condition in modeling particle motion.32 Hence, those model particles could help to create an EPD model system. Furthermore, additive-free and reactant-free nanoparticles may present advantages for the application of EPD to biomaterials processing. Here, to comply with biological applications, it is not necessary to clean the colloidal nanoparticles before EPD or to remove the surfactant after EPD.33 The use of the EPD technique to deposit laser generated nanoparticles on surfaces has been reported in literature before.5 He et al. created ZnO films from ZnO nanoparticles in water to characterize the composition and also the charge of ZnO generated in water.34 Strauß et al. also used nanoparticles synthesized by PLAL for the deposition on biomaterials like stents35 and to study the influence of ligands on electrophoretic mobility.47 However, to the authors knowledge the fundamentals of EPD with laser generated nanoparticles e.g. deposition rate and electrophoretic mobility have not been investigated before in liquid flow. EPD can be carried out in different ways since an alternating current (AC) and a direct current (DC) can be used to generate an electric field to deposit nanoparticles on the respective surface.9 In literature AC-EPD is viewed as the more promising method to generate homogenous coatings on substrates in contrast to DC-EPD.9 Nevertheless, DC-EPD has been previously used to generate coatings with ligand-free and ligand-capped nanoparticles. One reason why AC-EPD is used more frequently is that in most cases ligandcapped nanoparticles are used for EPD. The problem that arises using ligand-capped nanoparticles for DC-EPD is that unbound ligand in the liquid may form a barrier in front of the electrode that hinders the deposition.36 To ensure a deposition without negative effects due to
Journal of The Electrochemical Society, 162 (4) D174-D179 (2015)
Figure 2. (Color online) Scheme of the self-designed flow-through EPD chamber with 16 slots for electrodes and an example of a nanoparticle coated surface. Figure 1. (Color online) Schematic overview of the study design: a) synthesis of ligand-free nanoparticles by laser ablation in liquids, b) EPD of charged, bare metal nanoparticles including determination of electrophoretic velocity of nanoparticles and electrophoretic deposition kinetics.
the ligands AC-EPD can be used to remove the barrier in front of the electrode. Another possibility to avoid the barrier formation caused by ligands is to implement a cleaning process after nanoparticle generation to remove excessive ligands.37–39 However, nanoparticle cleaning is generally a very time consuming process and quantitative removal cannot be guaranteed for all ligands.40 As our study utilizes totally ligand-free nanoparticles DC-EPD may be used without cross effects and we may assume that the direct current has no negative effect on the nanoparticles. In this work, DC-EPD of ligand-free gold nanoparticles (AuNP) in a flow-through chamber was systematically investigated. After characterization of the electrophoretic mobility of the metal colloid, the effect of the DC-EPD process on the mobility of the colloid is characterized, followed by definition of the optimal process window and quantification of the deposition rate. Materials and Methods In this study, the nanoparticles are generated using pulsed laser ablation in liquids and are afterwards deposited by EPD (Figure 1) where the colloids are analyzed before and after DC-EPD process. Gold foils/sheets (Goodfellow / 99.99 %) were used to synthesize colloidal gold nanoparticles by PLAL. The metal target was placed inside a self-constructed laser ablation chamber41 filled with ultrapure water (18.2 M.cm @ 25◦ C). No additional substances like ligands or salts were used. An Nd:YAG ns-pulsed laser (Rofin PowerLine E 20) with a repetition rate of 10 kHz and a pulse energy of 0.7 mJ was focused on the target with an F-Theta-lens ablating the noble metal causing cavitation, nanoparticle crystallization and dispersion into the liquid. EPD was performed using the self-designed flow-through chamber shown in Figure 2 which allowed the insertion of 16 working and counter electrodes plates. Before deposition the electrodes were purged with 2-propanol and ultrasonicated for 5 minutes. Deionized water was used as liquid for EPD since the colloidal stability of ligandfree particles benefits from low salinities.29,30 To determine the size and the stability of the nanoparticles before and after EPD UV-Vis-spectroscopy (Thermo Scientific Evolution 201), zeta potential analysis (Malvern Zetasizer Nano ZS) and transmission electron microscopy (TEM) (Philips CM12) were used.
For the characterization of the electrophoretic mobility of the ligand-free metal nanoparticles a miniature EPD set-up based on a sandwich of microscope slides with the electrodes and the colloid in between was built and placed under a dark field microscope (Leitz Orthoplan) coupled to a CCD-camera. Nanoparticle tracking video analysis42 allowed direct observation of Brownian motion and electrophoretic mobility. The limitation of this set-up is that one pixel represents 0.23 μm with a frame rate of 30 images per second. In order to accurately differentiate random movement by Brownian motion from directed movement due to applied electric fields, the nanoparticle has to travel at least one pixel per image. It should be noted that the nanoparticles do not necessarily show unidirectional movement in liquid and hence it has to be considered that the nanoparticle can travel further while still staying in the distance of one pixel. In that regard 0.23 μm × √ 2 is more accurate for the calculation which results in a distance of 0.325 μm per image and 9.75 μm/s. Below this boundary the movement cannot be exclusively associated with directed mobility caused by the applied electric field. The Brownian motion has to be considered when the distance goes below this threshold since the nanoparticles remain in the same pixel or can partially move in the opposite direction. For this study, gold nanoparticles (AuNP) were chosen due to their high scattering intensity, which leads to a high sensitivity in the detection of smaller nanoparticles in the dark field microscope. The movement of the 36.2 nm (hydrodynamic diameter) particles was analyzed for different electric field strengths. The measurement was repeated 20 times at every electric field strength. An external electric field may affect the nanoparticles’ stability against aggregation and sedimentation. To determine if the electric field has an effect on the size distribution of the colloid, the nanoparticles that remained in the liquid outflow after deposition were compared with the educt colloid using TEM-images. In order to select the best possible parameters for the electrophoretic deposition process (here defined by the highest deposited mass per time and the lowest nanoparticle agglomeration), electric field strength and liquid flow rate were varied and measurements were repeated three times for any given parameter set. UV-Vis-spectroscopy (Thermo Scientific Evolution 201) was used to further analyze differences in concentration and aggregation state of the colloids. Since the isosbestic point of gold (wavelength of 380 nm) is proportional to the concentration of gold, the deposited mass can be calculated directly from the UV-Vis-spectra using a suitable calibration curve.43
Journal of The Electrochemical Society, 162 (4) D174-D179 (2015) case the electric field strength exceeds 4 V/cm the electrophoretic net velocity increases. In this regime the directed electric forces seem to overshadow the Brownian motion. The electrophoretic velocity of the nanoparticles. Similar results were observed by Stotz et al.45 with ligand-capped nanoparticles, showing that a deformation of the nanoparticle’s double layer leads to a different electrophoretic mobility. The measured electrophoretic velocity fits well with the colloidal velocity simply based on nanoparticle’s zeta potential and the Smoluchowski equation,46 as shown in Figure 3. Therefore, the experiment shows that our results agree with the theory of the Stokes’ hard spheres model32 and can be used as a reference material to characterize the influence of ligands on the electrophoretic velocity in future work.47 However, subsequent to the electroosmosis to the electrode surface the nanoparticles also have to overcome the Coulomb repulsion of the surface to deposit on the electrodes according to DLVO-theory.6,29 At first glance this seems counterintuitive as the partly oxidized gold particles and anode surface are oppositely-charged and should be subject to Coulomb attraction. However, the electrokinetic charge of the nanoparticles in solution (−34.5 mV) is dominated by the electrical double layer, which yields an overall negative charge while the oxidized particle cores are positively charged. When the particle is in close proximity of the positively-charged electrode surface, the double layer can be deformed and Coulomb repulsion with the metal core of the particle has to be considered. In case the energy of the electric field overcomes the Coulomb repulsion, the nanoparticles will deposit and stick on the surface due to the attractive van der Waals forces. However, it needs to be considered that different parameters like friction and surface charge of the nanoparticles and also the viscosity of the liquid are important as well while evaluating the energetic contributions that influence this system. The reader should keep in mind that these energetic calculations are only valid when the electrophoretic mobility significantly exceeds the Brownian motion, as it was specified in the Materials and Methods section.
Figure 3. a) (Color online) Electrophoretic velocity of ligand-free gold nanoparticles and comparison of the measured velocity with the velocity calculated by the nanoparticle’s zeta potential (deviation of the calculated velocity derives from the measured zeta potential) b) size of the nanoparticles remaining in the colloid before and after electrophoretic deposition measured by SEM c) size of the nanoparticles in the colloid before electrophoretic deposition measured by dark field microscopy d) zeta potential of the nanoparticles used in the study.
The primary particle index (PPI) is given by the extinction at 380 nm divided by the extinction at 800 nm, and is proportional to the stability of a gold colloid against agglomeration.29,30 A vital aspect of EPD is precise knowledge of the deposition kinetics and the deposition rates of the particles during the process. Therefore, a “batch cell”-cuvette was used to perform online UV-Vis-measurements at the mass-proportional interband wavelength (380 nm) of gold. Results and Discussion Electrophoretic mobility in DC electric fields.— For the determination of the electrophoretic mobility of ligand-free AuNP the electric field was varied via the applied current and particles velocities were quantified by tracking them via dark field online imaging. In order to evaluate the different energetic contributions in the observed system we compared the thermal energy to the energy that affects a single particle in the electric field. Our calculations revealed that the thermal energy is negligible compared to the electric energy. As shown in Figure 3, the Brownian motion of the particles is dominant at an electric field strength below 4 V/cm as no directed movement toward one electrode is observed. The phenomenon that nanoparticles do not show a directed movement in an electric field at lower electric field strength has been previously reported in literature.44 In
Optimization of process parameter determinants.— In order to characterize the influence of the applied electric field on the colloid itself and on the electrophoretic deposition rate, the outflow of the EPD flow-through chamber was analyzed online via UV-Vis-spectroscopy. The size characterization was achieved by observing the shift of the material specific surface plasmon resonance peak (SPR) (Figure 4a), which is located at a wavelength of approximately 520 nm for AuNP. A shift of the peak maximum to higher wavelengths is an indication for the presence of bigger particles or agglomerates.48 As the results demonstrate (Figure 4), the volumetric flow rate and the applied electric field strength show an influence on the colloidal properties. Lower flow rates (< 2 mL/min) lead to an increased SPR shift due to longer retention times in the electric field (Figure 4a) which may cause a reduction of the electrostatic stabilization. Accordingly, higher electric field strengths (> 22 V/cm) inside the chamber caused a bathochromic shift of the SPR-peak as well. It can be assumed that the hydrodynamic particle size of nanoparticle is increased due to agglomeration causing plasmon coupling of the primary particles. In contrast, low electric field strengths and shorter retention times tend to cause no significant change in the size of nanoparticles as verified by the UV-Vis-spectra. To determine whether the hydrodynamic size of the nanoparticles increased by an agglomeration process the primary particle index (PPI) was analyzed as an indicator for the stability of the colloid (Figure 4b). The primary particle index decreases at low flow rates and high electric fields strengths in the deposition chamber (from a PPI of 5.2 to a PPI of 1.1). This validates that high electric field strengths lead to more agglomeration. This also means that the increased hydrodynamic size of the colloid is due to an agglomeration process of nanoparticles. It can be observed that the PPI exhibits considerable fluctuations at high electric field strengths ≥20 V/cm in deionized water. In order to avoid agglomeration of the colloid, higher electric field strengths should be avoided during EPD of ligand-free metal nanoparticles in water. It should be noted that a decrease in the PPI may also be caused by a selective adsorption of smaller particles, while larger particles remain
Journal of The Electrochemical Society, 162 (4) D174-D179 (2015)
Figure 4. (Color online) Influence of electric field strength and volumetric flow rate on the colloid a) characterization of SPR-Peak, b) characterization of primary particle index, c) determination of deposited mass per second, d) sketch of the used set-up. Green areas mark preferred conditions.
in solution. However, as Figure 3b clearly reveals that the particle size distributions of the colloid prior and subsequent to electrodeposition do not differ significantly, the impact of size selective adsorption is probably negligible. It was shown that the electric field can interfere with the colloid, in case it is exposed to the field for a long time. This could be explained by the high electric charge of the laser generated nanoparticles, which is also the reason for electrostatic stabilization. The electrophoretic mobility increases at higher electric field strengths (Figure 3), which also leads to a deformation of the electrostatic double layer around the nanoparticles reducing the stability of the colloid. Since a controlled and reproducible deposition of the ligand-free nanoparticles with a defined substrate coating thickness is desired it is important to characterize the deposition rate of the nanoparticles (Figure 4c). The mass deposited on the substrate depends on the volumetric flow rates of the colloid and the electric field strength. At a lower electric field strength (1.8 mL/min, which corresponds to a nanoparticle mass flow of 2.7 μg/min for the given set-up. Since negative effects are less pronounced at higher flow rates the optimized process window broadens in this regime. At a flow rate of 5 mL/min a range of electric field strengths between 16.5–22.5 V/cm can be chosen to deposit a high mass without forcing a negative impact on the nanoparticles’ colloidal stability, allowing to recycle the flux. Based on this optimization, a flow rate of 3 mL/min and an electric field strength of 20 V/cm were chosen for consecutive experiments. These parameters are well within the boundaries of the defined optimal process window (Figure 5d). Deposition rate of nanoparticles using DC-EPD.— The deposition kinetics of a gold colloid was determined during DC-EPD via online UV-Vis-spectroscopy inside a cuvette. The experiments were carried out with three different initial gold concentrations (3.8 μg/mL, 13.2 μg/mL, 37.2 μg/mL, Figure 6). It was observed that the gold concentrations, estimated from the absorbance measured at the isosbestic point (380 nm) of gold, decreases over time (1 ng/s, 5.1 ng/s, 7.4 ng/s). The normalized UV-Vis-spectrum at 380 nm (Figure 6c) shows that the deposition does not change the measured particle size in the colloid and that the colloid does not tend to agglomerate. Additionally, exemplary SEM-images (Figure 6d) also reveal that no agglomeration was initiated by the applied current and that more nanoparticles get deposited on the surface with increasing deposition time. This proves that the electric field has no negative influence on the colloid in this defined parameter window. The experiment confirms that the deposition kinetic of the nanoparticles depends on the concentration of nanoparticles in the colloid. As expected, the deposition rate increases with the initial concentration. Although, the deposition rate depends on the nanoparticle concentration it has to be noted that the particle concentration decreases linearly in the observed time frame (Figure 6a). In theory deposition rates could be independent of time when first order kinetics are assumed (if the deposited mass is negligible compared to the colloid concentration). Since the used nanoparticles in
Journal of The Electrochemical Society, 162 (4) D174-D179 (2015)
Figure 5. (Color online) Definition of the EPD process window by combining preferred parameter areas (marked green) regarding a) colloidal stability (SPR), b) colloidal stability (PPI), c) deposition rate per second, resulting in d) optimized process window for EPD of ligand-free metal nanoparticles in water.
Figure 6. (Color online) Deposition kinetics measured by online-UV-Vis-spectroscopy a) electrophoretic deposition rate of AuNP colloid in correlation with the initial concentration of the gold colloid, b) UV-Vis-spectra of an exemplary colloid during EPD with decreasing absorbance, c) UVVis-spectrum from b) normalized to the interband wavelength of 380 nm, indicating constant particle size and stability, d) exemplary SEM-images of the coated surfaces with increasing deposition time.
Journal of The Electrochemical Society, 162 (4) D174-D179 (2015) this study are ligand-free there is no barrier formation in front of the electrode that would hinder nanoparticle deposition and lead to a decreased deposition rate. This barrier formation can be caused by excess ligands and may occur if the nanoparticle solution is not properly prepared.37 Therefore, the ligand-free laser generated nanoparticles may be used as a model system for the characterization of unhindered and ideally linear deposition rates. Conclusions Despite the impact of the EPD process, there is still a need for a better understanding regarding the process itself. This study investigated the electrophoretic deposition of ligand-free metal nanoparticles in liquid flow. For best results a process window between electric field strengths of 16–22 V/cm and a volumetric flow rate of the colloid higher than 2 mL/min was defined. This parameter set yielded a high deposited mass without having a negative effect of the electric field on the colloidal stability. This enabled the recycling of the colloid for further passages through the continuously operating deposition chamber. Within the borders of this process window the deposition rate proved to be constant over a defined time frame using direct current EPD and ligand-free nanoparticles. This finding suggests that laser-generated bare nanoparticles do not lead to a barrier formation in front of the electrodes as it may occur for ligand-capped particles. As expected, increased nanoparticle concentration lead to higher deposition rates, indicating a first order kinetic for EPD. Therefore, the use of ligand-free nanoparticles enables the implementation of a continuous process to coat several substrates simultaneously. With this continuous process more of the colloid can be used, potentially rendering the EPD-process more cost efficient and time-saving. Acknowledgments We thank the Deutsche Forschungsgemeinschaft (DFG) for the funding within the project BA 3580/8-1. We also like to thank the workgroup of Prof. Dr. Christian Mayer (University of DuisburgEssen), Christoph Groß-Heitfeld who provided us with the dark field microscope used in our studies, and Christoph Rehbock for proofreading. References 1. B. Neirinck, O. Van der Biest, and J. Vleugels, J. Phys. Chem. B, 117, 1516 (2013). 2. I. Corni, M. P. Ryan, and A. R. Boccaccini, J. Eur. Ceram. Soc., 28, 1353 (2008). 3. P. Sarkar and P. S. Nicholson, J. Am. Ceram. Soc., 79(8), 1987 (1996). 4. O. Van der Biest and L. J. Vandeperre, Annu. Rev. Mater. Sci., 29, 327 (1999). 5. J. Jakobi, A. Men´endez-Manj´on, V. S. K. Chakravadhanula, L. Kienle, P. Wagener, and S. Barcikowski, Nanotechnology, 22, 145601 (2011). 6. L. Besra and M. Liu, Prog. Mater. Sci., 52, 1 (2007). 7. X. Chen, N. Li, K. Eckhard, L. Stoica, W. Xia, J. Assmann, M. Muhler, and W. Schuhmann, Electrochem. Commun., 9, 1348 (2007).
8. K. Wu and I. Zhitomirsky, Int. J. Appl. Ceram. Technol., 8(4), 920 (2011). 9. V. Ozhukil Kollath, Q. Chen, R. Closset, J. Luyten, K. Traina, S. Mullens, A. R. Boccaccini, and R. Cloots, J. Eur. Ceram. Soc., 33, 2715 (2013). 10. P. Maarten Biesheuvel and H. Verweij, J. Am. Ceram. Soc., 82(6), 1451 (1999). 11. W. F. Pickard, J. Electrochem. Soc., 115, 105C (1968). 12. H. C. Hamaker and E. J. W. Verwey, Trans. Faraday Soc., 35, 180 (1940). 13. A. Neumeister, D. Bartke, N. B¨arsch, T. Weing¨artner, L. Guetaz, A. Montani, G. Compagnini, and S. Barcikowski, Langmuir, 28, 12060 (2012). 14. W. Guo and B. Liu, ACS Appl. Mater. Interfaces, 4, 7036 (2012). 15. R. Drevet, A. Viteaux, J. C. Maurin, and H. Benhayoune, RSC Adv., 3, 11148 (2013). 16. W. B¨ucking and T. Nann, Proc. Nanobiotechnol., 153(3), 47 (2006). 17. S. Ibrahimkutty, P. Wagener, A. Menzel, A. Plech, and S. Barcikowski, Appl. Phys. Lett., 101, 103104 (2012). 18. V. Amendola and M. Meneghetti, Phys. Chem. Chem. Phys., 15, 3027 (2013). 19. V. Amendola and M. Meneghetti, Phys. Chem. Chem. Phys., 11(20), 3805 (2009). 20. H. Zeng, X.-W. Du, S. C. Singh, S. A. Kulinich, S. Yang, J. He, and W. Cai, Adv. Funct. Mat., 22(7), 1333 (2012). 21. S. Barcikowski, F. Devesa, and K. Moldenhauer, J. Nanopart. Res., 11, 1883 (2009). 22. S. Barcikowski and G. Compagnini, Phys. Chem. Chem. Phys., 15, 3022 (2013). 23. J. P. Sylvestre, S. Poulin, A. V. Kabashin, E. Sacher, M. Meunier, and J. H. T. Luong, J. Phys. Chem. B, 108, 16864 (2004). 24. A. Schwenke, P. Wagener, S. Nolte, and S. Barcikowski, Appl. Phys. A: Mater. Sci. Process., 104, 77 (2011). 25. W. T. Nichols, T. Sasaki, and N. Koshizaki, J. Appl. Phys., 100, 114912 (2006). 26. J. Jakobi, S. Petersen, A. Menendez-Manjon, P. Wagener, and S. Barcikowski, Langmuir, 26, 6892 (2010). 27. T. Sasaki, C. Liang, W. T. Nichols, Y. Shimizu, and N. Koshizaki, Appl. Phys. A: Mater. Sci. Process., 79, 1489 (2004). 28. V. Amendola, P. Riello, and M. Meneghetti, J. Phys. Chem. C, 115, 5140 (2011). 29. V. Merk, C. Rehbock, F. Becker, U. Hagemann, H. Nienhaus, and S. Barcikowski, Langmuir, 30, 4213 (2014). 30. C. Rehbock, V. Merk, L. Gamrad, R. Streubel, and S. Barcikowski, Phys. Chem. Chem. Phys., 15, 3057 (2013). 31. H. Muto, K. Yamada, K. Miyajima, and F. Mafune, J. Phys. Chem. C, 111, 17221 (2007). 32. A. Einstein, Ann. D. Phys., 17, 549 (1905). 33. P. Wagener, A. Schwenke, and S. Barcikowski, Langmuir, 28, 6132 (2012). 34. C. He, T. Sasaki, Y. Shimizu, and N. Koshizaki, Appl. Surf. Sci., 254, 2196 (2008). 35. S. Strauß, A. Neumeister, S. Barcikowski, D. Kracht, J. W. Kuhbier, C. Radtke, K. Reimers, and P. M. Vogt, Plos one, 8, 53309 (2013). 36. W. Schuhmann, C. Kranz, H. Wohlschl¨ager, and J. Strohmeier, Biosens. Bioelectron., 12 (12), 1157 (1997). 37. M. A. Islam, Y. Xia, D. A. Telesca, M. L. Steigerwald, and I. P. Herman, Chem. Mater., 16, 49 (2004). 38. M. A. Islam, Y. Xia, M. L. Steigerwald, M. Yin, Z. Liu, S. O’Brien, R. Levicky, and I. P. Herman, Nano Lett., 3 (11), 1603 (2003). 39. B. Ferrari, R. Moreno, and J. A. Cuesta, Key. Eng. Mater., 314, 175 (2006). 40. S. K. Balasubramanian, L. Yang, L.-Y. L. Yung, C.-N. Ong, W.-Y. Ong, and L. E. Yu, Biomaterials, 31 (34), 9023 (2010). 41. P. Nachev, D. D. van’t Zand, V. Coger, P. Wagener, K. Reimers, P. M. Vogt, S. Barcikowski, and A. Pich, J. Laser Appl., 24 (4), 1 (2012). 42. C. Finder, M. Wohlgemuth, and C. Mayer, Part. Part. Syst. Charact., 21, 372 (2004). 43. L. Gamrad, C. Rehbock, J. Krawinkel, B. Tumursukh, A. Heisterkamp, and S. Barcikowski, J. Phys. Chem. C, 118 (19), 10302 (2014). 44. A. Men´endez-Manj´on, J. Jakobi, K. Schwabe, J. K. Krauss, and S. Barcikowski, JLMN, 4 (22), 95 (2009). 45. S. Stotz, J. Colloid Interf. Sci., 65 (1), 118 (2010). 46. T. L. Doane, C. Chuang, R. J. Hill, and C. Burda, Accounts Chem. Res., 45 (3), 317 (2012). 47. C. Streich, S. Koenen, K. Peneva, M. Lelle, and S. Barcikowski, Appl. Surf. Sci. (2015). 48. U. Taylor, S. Klein, S. Petersen, W. Kues, S. Barcikowski, and D. Rath, Cytom. Part A, 77A, 439 (2010).