With the rapid development of colloidal science, nanoparticles (NPs) of different sizes, shapes, and materials are available and widely used in biology, medicine, and material science. One category of NPs, noble metallic NPs, such as gold, silver, or copper NPs, have stimulated significant interest, owing to their inimitable optical and electrical properties. They are widely utilized in diverse fields, such as surface-enhanced Raman scattering (SERS) [1–3], nanosensing [4,5], solar energy harvesting [6,7], nonlinear light generation [8,9], and nanolithography [10,11]. Characterization among applications requires the NPs to be drop-cast on a dry surface. The problem is that these NPs firmly adhere to the surface once they are dried and cannot be removed or transferred easily to other surfaces for further processing. Additionally, there has been significant interest in NP-based nanomanufacturing, primarily using laser-induced forward transfer (LIFT) [12–14] or laser-induced backward transfer (LIBT) [15,16] methods, which offer versatility in printing material and small feature sizes. However, these methods require costly femtosecond lasers and have limited control of the generated NPs. In contrast, we have demonstrated that a low-cost continuous wave (CW) laser can also transfer adhered particles in air with significantly reduced laser powers [17]. Unlike LIFT and LIBT, this technique transfers preexisting NPs dried on a soft substrate, which allows unlimited choice in NP size and shape for nanoprinting, as well as for characterization applications. However, besides the demonstration of NP release with a CW laser, there has, to our knowledge, been no investigation into how laser intensity affects the release process and the NPs’ quality after release. Additionally, future diverse applications require high repeatability; this demands an understanding of the stochastic nature of release for various parameters, such as laser intensities and NP size and charge, as well as substrate contaminants and mechanical properties.

In this study, we investigated the probabilistic release of gold nanospheres and gold nanoshells from a soft substrate with a CW laser over several experiments. With these new insights and the wide variety of commercially available NPs, this technology may find applications in NP characterization and nanomanufacturing.

For the first experiment, to determine how NP size affects release, four samples were prepared with gold nanospheres (BBI Solutions) of 100 nm, 150 nm, 200 nm, and 250 nm, drop-cast on a polymethyl methacrylate (PMMA) coated glass substrate and dried in ambient conditions. For the second experiment, gold nanoshells (nanoComposix) containing 100-nm gold cores and 20-nm silicon shells of variable surface charge were drop-cast on a PMMA substrate that underwent plasma cleaning (Plasma Cleaner MNT-PC-2) to analyze how surface properties can affect release. For the third experiment, the gold nanoshells were drop-cast on either a PMMA, polydimethyl siloxane (PDMS), or polyvinyl chloride (PVC) substrate to give insight into how the substrate’s mechanical properties affect release. Finally, we investigated damage to the released NPs for different laser intensities.

As an example of the process, Fig. 1(a) shows, schematically, a gold nanosphere being released from a donor substrate and transferred to a receiver substrate. The experimental setup is shown in Fig. S1 in the supplementary material. Figure 1(b) is an optical micrograph of a gold nanosphere with a diameter of 200 nm (the black dot in the red circle) on a donor substrate. This gold nanosphere is then moved to the position of the laser spot (marked as a red circle). The laser is then switched on and the gold nanosphere is released from the donor substrate, as shown in Fig. 1(c) (the gold nanosphere disappears). Figure 1(d) is an optical micrograph of the gold nanosphere (the black dot in the red circle) that is transferred to the receiver substrate from the donor substrate after the laser is switched on. The time scale of a particle release is of the order of hundreds of nanoseconds, based on our previous calculation [18], and the particle has been exposed to the laser for a sufficient time.

 

Fig. 1. (a) Transfer process of gold nanosphere. (b) Optical micrograph of gold nanosphere on PMMA donor substrate when laser is off. (c) Optical micrograph of a gold nanosphere on PMMA donor substrate when laser is on. (d) Optical micrograph of gold nanosphere that is transferred to receiver substrate when laser is on.

Download Full Size | PDF

A gold nanosphere is released from the PMMA donor substrate because of the thermal expansion of the PMMA, which is heated by the nanosphere [17,18]. The amount of deformation of the PMMA depends on the laser intensity. Figure 2 shows optical micrographs of the PMMA substrate before and after the release of a gold nanosphere (150 nm) at three different laser intensities: 10 mW/µm² (Figs. 2(a) and 2(b)), 20 mW/µm² (Figs. 2(c) and 2(d)), and 40 mW/µm² (Figs. 2(e) and 2(f)), respectively. The PMMA donor substrate is partially damaged at the location of the nanosphere at lower laser intensity, as shown in Figs. 2(b) and 2(d). However, the damage is invisible at higher laser intensity, as shown in Fig. 2(f). The release probability shown in Fig. 2(g) indicates that nearly all the nanospheres (150 nm) cannot be released under a laser intensity of 10 mW/µm². However, while not being released, the nanosphere continues to absorb and transfer heat to the substrate, causing deformation and eventual melting of the substrate beneath it. On the contrary, when the intensity is above 40 mW/µm², a nanosphere has a high probability of being released from the substrate, as shown in Fig. 2(g) and it causes less substrate damage, as shown in Fig. 2(f). Therefore, a higher laser intensity is preferred to release the nanosphere, with minimum damage to the donor substrate.

 

Fig. 2. (a, b) Optical micrograph of PMMA deformation at a laser intensity of 10 mW/µm² for 150-nm gold nanosphere. (c, d) Optical micrograph of PMMA deformation at a laser intensity of 20 mW/µm². (e, f) Optical micrograph of PMMA deformation at a laser intensity of 40 mW/µm². (g) Release probability of gold nanospheres with diameters of 100 nm, 150 nm, 200 nm, and 250 nm from a PMMA substrate.

Download Full Size | PDF

The laser intensity also affects the release probability of a nanosphere from the donor substrate. Here, the release probability is defined as the chance of successfully releasing a nanosphere from a PMMA substrate at a particular laser intensity. The release probability of a nanosphere can be measured experimentally and can be defined as the ratio of the number of nanospheres that are successfully released from a donor substrate to the total number of nanospheres targeted in the experiment.

Figure 2(g) shows the release probability of gold nanospheres with diameters of 100 nm, 150 nm, 200 nm, and 250 nm from a PMMA substrate. For each data point, 100 nanospheres are targeted for the release experiment. Each nanosphere is aligned to the laser focus with the laser off and then illuminated at the specified intensity. By manually counting the number of nanospheres that are successfully released from the substrate, the release probability (y axis) can be calculated at each laser intensity (x axis). The release probability of the four sizes of nanospheres first increases rapidly with an increase in the laser intensity and then saturates after reaching a release probability of around 90%. The minimum laser intensity with 90% release probabilities for nanospheres with diameters of 100 nm, 150 nm, 200 nm, and 250 nm are around 50 mW/µm², 30 mW/µm², 16 mW/µm², and 10 mW/µm², respectively. The smaller the particle size, the higher the laser intensity that is required to saturate the release probability, which is a result of the smaller absorption cross section of the smaller size of the gold nanosphere [19,20]. A gold nanosphere is released from the substrate, owing to the thermal expansion of the substrate caused by the optical heating of the gold nanosphere by the laser beam. The amount of thermal expansion is determined by the amount of heat generated on the gold NP from the laser beam, which is in turn determined by the optical absorption cross section of the NPs. Therefore, the smaller the gold NP, the smaller the optical absorption cross section, the less heat generated on the NP, and the larger the laser intensity required to release the particle. The highest release probabilities for nanospheres with diameters of 100 nm, 150 nm, 200 nm, and 250 nm are 96%, 98%, 98%, and 98%, respectively. Therefore, a higher laser intensity leads not only to less damage to the donor substrate but also to a higher release probability.

The release probability of a gold NP from a donor substrate highly depends on the surface property of the donor substrate. The electrostatic force between the NPs and the substrate could increase adhesion and reduce the chance of a successful release. In this experiment, the release probability of gold nanoshells prepared with opposite surface charges (negatively charged nanoshell and positively charged nanoshell) from a PMMA donor substrate was investigated. Additionally, plasma cleaning was used to learn whether environmental contaminants affect NP adhesion and release. The black and red solid curves in Fig. 3(a) show the release probability of negatively charged nanoshells on a PMMA substrate and a plasma-cleaned PMMA substrate, respectively. The black and red dashed curves show the release probability of positively charged nanoshells on a PMMA substrate and a plasma-cleaned PMMA substrate, respectively. The results shown in Fig. 3(a) indicate that plasma cleaning has a major effect on the release probability. In both cases, plasma cleaning the substrate reduces the release probability of gold nanoshells. A possible reason is that a cleaner substrate makes the nanoshells adhere more strongly, therefore, making it harder to release compared with those from a donor substrate without the plasma cleaning under the same laser intensity.

 

Fig. 3. (a) Release probability of positively and negatively charged gold nanoshells with a diameter of 140 nm from a PMMA substrate. (b) Release probability of gold nanoshells with a diameter of 140 nm from PDMS, PMMA, and PVC substrates.

Download Full Size | PDF

The experimental result indicates that the net charge of the gold nanoshell has a negligible effect on the release probability. While the nanoshells were charged in solution (ethanol), their charges would neutralize after drop casting and drying, which would eliminate the effect on adhesion from the electrostatic attraction. The release probability of a NP from a substrate also depends on the mechanical property of the substrate. Figure 3(b) shows the release probability of gold nanoshells from PDMS, PMMA, and PVC substrates. The release probability of a gold nanoshell from a PVC substrate follows a similar trend to that from PDMS and PMMA substrates. But the release probability is the smallest for PVC under the same laser intensity; this is due to the different elastic properties of the substrates. Young’s moduli for PDMS, PMMA, and PVC are around 4 MPa, 2.5 GPa, and 7 GPa, respectively, which means that the PDMS and PMMA are more easily deformed than PVC when the gold nanoshell is heated by the laser under the same laser intensity. Therefore, a more flexible substrate is preferred for particle release. However, the release probability for PDMS and PMMA does not differ significantly, despite a large difference in Young’s modulus. This may be an indication that Young’s modulus is not the only parameter that affects this process. For example, the thickness of the polymer film may also play a part. In this experiment, the PDMS and PMMA are spin-coated on a glass substrate with a thickness of 5 µm and 0.4 µm, respectively. The PVC substrate has a thickness of 170 µm.

We have also observed damage to the silicon shell from the laser interaction in the release process. In the experiment, an indium tin oxide (ITO) coated glass is placed above the PMMA substrate so that the gold nanoshells can be released from the PMMA substrate and transferred to the ITO glass. Here the ITO-coated glass is used for the sole purpose of making a conductive surface to facilitate scanning electron microscopy. Once the gold nanoshells are transferred to the ITO glass, their micrographs are evaluated. Figures 4(a) and 4(b) show the size distribution of the silicon shell and the gold core before and after the release, based on measurements of the scanning electron micrograph. The gold core size does not undergo significant changes under a laser intensity of 50 mW/µm². However, the size of the silicon shell shrank by 10%. Figure 4(c) is the micrograph of a gold nanoshell before release. Figures 4(d) to 4(f) are micrographs of a gold nanoshell after it has been transferred to the ITO glass, with increasing laser intensity. The thickness of the silicon shell decreases with increasing laser intensity. The silicon shell can even be peeled off the gold core at high laser intensity.

 

Fig. 4. (a) Core size distribution before and after transfer. (b) Shell size distribution before and after transfer. (c) Scanning electron micrograph of gold nanoshell before transfer. (d–f) Micrographs of gold nanoshell after transfer for laser intensities of 20 mW/µm², 40 mW/µm², and 80 mW/µm². Scale bar, 100 nm.

Download Full Size | PDF

In conclusion, we studied a NP transfer process under CW laser illumination. The release probability of gold NPs under different conditions was studied. For example, plasma cleaning of the substrate, a procedure widely used in surface treatment, will decrease the release probability. The net surface charges of the NPs will not significantly affect the release probability. A more flexible substrate will facilitate the release of the particle from its surface. It is possible to keep the shape of the transferred NPs under lower laser intensities, but a high laser intensity might damage the particles. These results will provide useful guidelines for the application of this NP transfer technology.