Size-Selected Submicron Gold Spheres: Controlled Assembly onto Metal, Carbon, and Plastic Substrates.

Size-selected submicron spheres become very useful building blocks if the spheres could be synthesized and integrated at any desired position. In particular, spheres having a similar size to visible-light wavelength have attracted much attention. Here, we show the synthesis and assembly of size-selected submicron gold spheres using pulsed laser ablation of a gold plate in a supercritical fluid. Four findings were obtained in the study. Submicron spheres with a narrow size distribution were generated, and the polydispersity was ≈ 6%. The average diameter was controlled from 600 to 1000 nm. A thermodynamic condition for scalable synthesis was found. The assembly of spheres onto a metal, carbon, or plastic substrate was accomplished.

Introduction

Nobody doubts that gold as a material has attracted considerable attention in chemical and material sciences and industrial applications.[1] In particular, gold nanomaterials are utilized as substrates for single-molecule detection by surface-enhanced Raman scattering[2] and enhanced fluorescence,[3] as optomaterials for ultrasensitive biosensors and medical sensors,[4] as catalysts for chemical reactions such as CO oxidation,[5] and as photothermal materials for cancer therapy.[6] All of these properties are accomplished by nanomaterials with a size < 50 nm, which are synthesized by a conventional chemical synthesis method: reduction of gold (Au) ions in HAuCl4 aqueous solution.[7] However, it has been difficult to produce a size-selected metal particle larger than 100 nm. Specifically, submicron particles could become very useful building block materials if submicron-sized particles with a homogeneous size could be synthesized and integrated at any desired position, for example, photonic crystals matching the visible and near-infrared (NIR) light, nanogap electrodes with a special smooth surface, field enhancement driven by high scattering efficiency, and parts of a molecular circuit.

Pulsed laser ablation (PLA), a physical synthesis method, is a promising approach used to obtain stable nanoparticles. This method consists of a one-step process conducted at room temperature with a short duration, for example, a few minutes to an hour.[8−14] In addition, scalable synthesis of nanoparticles, several grams/hour as products, has been recently realized.[11d,11e] PLA of Au in solution has been extensively investigated in the last decade.[11a,11b] Recently, submicron-sized spherical Au particles were synthesized by PLA.[12e,13] We have developed a novel method for nanoparticle synthesis by conducting PLA in a supercritical fluid,[12a−12f] as shown in Figures and S1. Specifically, thermal and dielectric properties of a surrounding medium can be easily tuned by changing the fluid density and pressure. Thus, light-emitting silicon nanocrystals were generated with photoluminescence color (RGB[12b] and white-light continuum[12c]) that can be controlled by fluid pressure and/or density during PLA. The morphology of Au nanoparticles was also changed by the cooling rate or permittivity of fluids during PLA.[12d,12e] Furthermore, PLA in a supercritical fluid becomes a popular method to synthesize various nanomaterials, recently.[14]

Figure 1

Schematic diagram of the system used for nanoparticle generation by laser ablation in supercritical fluids. Photograph represents the high-pressure vessel for preparing supercritical fluids. Photograph courtesy of Saitow. Copyright 2019.

Here, we show submicron Au spheres with a narrow size distribution, that is, deviations of diameters ranging 5–8%, using PLA laser synthesis in a supercritical fluid. The average diameter of particles ranged from 600 to 1000 nm, whose size matches the light wavelength in the visible and NIR regions. In addition, a thermodynamic condition for scalable synthesis was found. Furthermore, the integration of gold spheres onto a metal, carbon, or plastic film was accomplished and optimized.

Results and Discussion

Figure shows scanning electron microscopy (SEM) images of submicron-sized Au spherical particles generated by PLA at three reduced densities ρr = ρ/ρc, where ρc is the critical density, of supercritical trifluoromethane (CHF3) (ρr = 0.7, 1.7, and 1.9). Many submicron-sized spherical Au particles are generated, and the particle size is very uniform. This phenomenon has not been observed by PLA in other fluids, for example, CO2 and SF6. The SEM analysis from each sample provided the average diameter D, of the spherical particles, and the standard deviation, σ. D ranged over 700–1000 nm: D(ρr = 0.7) = 890 ± 55 nm, D(ρr = 1.7) = 710 ± 40 nm, and D(ρr = 1.9) = 1050 ± 80 nm. Their size distributions are shown in Figure S2.[15,16] Here, the plus–minus values indicate the standard deviation σ from the average diameters, and the dispersity of particle size is evaluated as the value σ/average. Note that this dispersity of particle sizes is small and estimated as 6%, ranging from 5 to 8%. Here, let us discuss the particle size dependence on the fluid density during PLA, briefly. According to the transient absorption spectra monitoring PLA in a supercritical fluid,[12f] submicron-sized Au particles are generated within the time shorter than 300 ns after laser irradiation. Thermodynamic calculations indicated that the generation process is attributed to the solidification of hot liquid droplets, by which the lowest surface free energy is obtained with a spherical shape, within a few 100 ns.[12d] A cavitation bubble, in which nanoparticles are generated, emerges within submicroseconds on the irradiated surface of the target in liquids and high-pressure liquids, according to time-resolved shadowgraph and light scattering measurements.[17] On the basis of these experimental evidences in a supercritical fluid and the observations by in situ time-resolved spectroscopy, it was considered that a concerted process of evaporation and cooling of Au liquid droplets are responsible for the particle size. Specifically, the evaporation and cooling rates of the gold liquid droplet in which the cavitation bubble adiabatically expands and shrinks in the high-pressure fluid are key factors for the characterization of the particle size by PLA at high pressures.

Figure 2

SEM images of submicron spherical Au particles generated by PLA in supercritical CHF3 at a reduced temperature Tr = T/Tc = 1.02. The Au particles are generated at a reduced density ρr = ρ/ρc of (a) 0.7, (b) 1.7, and (c) 1.9. Diameter D ranges over 600–1200 nm: (a) D(ρr = 0.7) = 890 ± 55 nm, (b) D(ρr = 1.7) = 710 ± 40 nm, and (c) D(ρr = 1.9) = 1050 ± 80 nm. The plus–minus values denote the standard deviation, σ, from the average diameters. The deviations of particle sizes are in the 5–8% range of the average diameters.

The amounts of particles produced as a function of fluid pressure/density during PLA were investigated. Figure a–d shows the typical SEM images of Au spherical particles fabricated with four representative densities. A significant dependence on the fluid density is observed; a large amount of gold nanospheres was generated by PLA at a high fluid density, whereas that generated at a fluid density lower than ρr = 0.7 was significantly decreased. To quantify the dependence of the amount of particles on the fluid density, the number of particles was estimated both for the number density of particles and the particle-deposited area in the SEM image. Figure e shows the number of particles as a function of reduced densities during PLA. Many particles were generated by PLA at ρr ≥ 0.7, whereas few particles were generated at ρr < 0.7. Thus, Figure e reveals that the thresholds of fluid density and pressure for the generation of particles are ρr = 0.7 and P = 5.3 MPa, respectively. From the Supporting Information, it is noted that many gold nanonetworks, composed of smaller nanospheres with 30 nm diameter, are generated by PLA at a lower density of supercritical CHF3 but disappear at a higher density of ρr = 0.7 (Figures S3 and S4). The density for the emergence of submicron particles is in good agreement with that for the disappearance of the nanonetworks, that is, liquid gold droplets as precursors are fragmented to form smaller nanospheres at lower densities giving the gold nanonetwork, whereas at higher densities, the solidification of liquid droplets produces submicron-sized particles. This precursor model of gold liquid droplets has been confirmed by measuring the amounts/laser pulse of both spherical particles and nanonetworks generated by PLA in supercritical CO2.[12d] Briefly, the amount of gold nanonetworks increased, whereas that of spherical particles with diameter ≅ 800 nm simultaneously decreased, as the number of laser pulses for PLA was increased. The density dependence of the amount of spherical particles is attributed to the branching ratio, which determines whether the liquid gold droplet solidifies or fragments to the nanonetwork.[12e] Thus, the threshold density for solidification via cooling was ρr = 0.7.

Figure 3

SEM images of submicron spherical Au particles generated by PLA in supercritical CHF3 at a reduced temperature, Tr = T/Tc = 1.02. Particles were fabricated at a reduced density ρr = ρ/ρc of (a) 0.3, (b) 0.7, (c) 0.9, and (d) 1.9. Corresponding pressures are denoted in the upper axis. The particles were integrated into the micropores. The number of particles increases with the fluid density during PLA. (e) Numbers of Au particles produced at various ρr.

Another distinct feature evident in Figures and S3 is that almost all the synthesized particles are not present on a flat region of the substrate but are selectively collected in a pore of size ≈ 100 μm on the substrate. Briefly, let us describe the mechanism of collection of particles into the pore, and the details are given elsewhere (vide infra). As a first step, particles are synthesized in supercritical CHF3. In this step, the particles disperse in the supercritical fluid. As a second step, the pressure of supercritical CHF3 is decreased to retrieve the synthesized particles from the high-pressure vessel to the air. In this process, large particles with the size of 600–1000 nm do not disperse in the gas but sediment into the liquid region of the substrate, which is located at the bottom of the vessel. As a third step, all liquid region evaporates by decreasing further pressure, and the concentration of particle in the liquid becomes higher. The final liquid involving many spherical particles drops into a pore, which is located at the lower position of the substrate.

To further investigate the collection in the pore, we analyzed the amounts of particles by preparing a brass substrate with pores of different depths. Figure a,b show the top and side views of the substrate with the micropores. The pore diameter was set to 300 μm and the depth ranged from 50 to 1000 μm. The typical cross sections of pores measured by a laser microscope are also displayed in Figure c,d. PLA of a gold plate substrate in supercritical CHF3 was conducted at a pressure of 5.8 MPa, which corresponds to ρr = 1.25, and the generated particles in the pores were observed using a scanning electron microscope. The amount of particles was counted and the results are shown in Figure d as a function of pore depth. These values correspond to the summation of the numbers of particles synthesized by the experiments performed five times.[18] The maximum integration of particles into a pore is observed when the pore depth is 400 μm. Here, we describe the integration processes of submicron particles into the pore in detail (vide supra). Supercritical CHF3 with a pressure of 5.8 MPa was released to retrieve the substrate from the high-pressure vessel after the synthesis, during which the fluid temperature decreased from 305 to 285 K with the adiabatic expansion of the high-pressure fluid. Thus, the temperature T of CHF3 became lower than the critical temperature[19] (Tc = 299.3 K), that is, tentative temperature ≈ 10 °C, so that the supercritical CHF3 became a liquid. Liquid CHF3 was successively evaporated until its pressure was equal to the pressure of the atmosphere. These successive phase transition processes caused liquid CHF3 to settle at the bottom of the vessel. The final liquid, including many Au particles, was collected in the pores of the substrate, and many particles were integrated in the bottom of the pores. Multiple concerted factors, such as viscosity and surface tension depending on the pressure and temperature of CHF3, wettability of the substrate, mass transfer driven by capillary flow,[20,21] and pinning and depinning at the interface between the liquid and the substrate[22] can also govern the collection efficiency. Note that the integration into the pore was also established using the substrate made of other materials, such as stainless steel (SUS), copper, carbon, and polyethylene terephthalate (PET). The data are displayed in Figure . Thus, the pores in these substrates can also collect many submicron particles.

Figure 4

Laser microscope images of the (a) top and (b) side of the brass substrate used to collect submicron spherical Au particles. There are eight different pores with the size of 500 μm and different depths. Laser microscope images of the cross sections of pores with the depths of (c) 50 μm and (d)1000 μm. (e) Number of submicron spherical Au particles in pores as a function of pore depth. Particles were fabricated at ρr = ρ/ρc = 1.25.

Figure 5

Submicron spherical Au particles generated by PLA in supercritical CHF3 at a reduced temperature Tr = T/Tc = 1.02. The Au particles are generated at a reduced density ρr = ρ/ρc of 1.25 and gathered at the micron-sized pore of each substrate. SEM images of the particles on the substrates of (a) carbon tape, (b) copper, and (c) SUS. Optical microscope image of the particles on the substrate of (d) PET. The scale bar in (d) is 50 μm.

Conclusions

In summary, nanosecond PLA of an Au plate was conducted in supercritical CHF3. Many spherical Au particles with diameters from 600 to 1000 nm were synthesized over the threshold density and/or pressure, that is, ρr = 0.7 and P = 5.3 MPa. The particle diameter was controlled according to the density and/or pressure used during PLA, and the polydispersity was as low as 6%. The selective integration of particles into various substrates was accomplished by both PLA in a supercritical fluid and successive phase changes from a high-pressure fluid.

Experimental Section

An instrument to fabricate particles was constructed, each equipment of which is illustrated in Figures and S1. A high-pressure vessel, a high-performance liquid chromatography (HPLC) pump, and a gas cylinder were used to prepare the supercritical state, and optics with a Q-switched frequency-doubled Nd:YAG laser were used to conduct PLA. The temperature of the fluid in the vessel was controlled using a set of heaters, a proportional–integral–derivative controller, and a thermocouple. The pressure of the fluid was increased with the HPLC pump. The Nd:YAG laser serves as the PLA light source and was operated at an excitation wavelength of 532 nm, an energy of 19 mJ/pulse, a repetition rate of 20 Hz, a fluence of 0.8 J cm–2, and a pulse width of 8 ns. A gold plate (99.95%, Tanaka Co.) immersed in supercritical fluid (CHF3, 99.995%) is irradiated with the laser for 10 min at the isotherm corresponding to a reduced temperature Tr = T/Tc = 1.02, where Tc is the critical temperature. The pressure for PLA ranged from 3.81 to 14.9 MPa. The fluid density was calculated from the empirical equations of state, using the measured values of P and T. The density ranged from 0.158 to 1.00 g cm–3 and is expressed as 0.3 ≤ ρr = ρ/ρc ≤ 1.9, where ρr and ρc are the reduced and critical densities, respectively, as shown in Figure S5. The Au particles generated were deposited on a substrate immersed in supercritical CHF3, SUS, copper, carbon, and PET. After the sedimentation of the particles in the fluid, the substrate was retrieved from the vessel and examined using a field emission scanning electron microscope (Hitachi N-3400) and a laser microscope (Shimadzu OLS-4000). The sedimentation time was calculated using the pressure-dependent viscosity of the fluid[12d]where v is the sedimentation velocity depending on the fluid density, z is the depth, t is the time, ρ is the density of Au, ρ0 is the fluid density, D is the particle diameter, g is the gravitational constant, and η is the fluid viscosity depending on the fluid density. The sedimentation time was adjusted to the amount of time required for a 100 nm diameter sphere to sink to a depth of 1 cm. The same sedimentation time was applied to every experiment to prepare SEM samples for the evaluation of the amount of Au nanoparticles according to the pressure and/or density during PLA. The critical constant of CHF3 is reported to be Tc = 299.3 K, Pc = 4.83 MPa, and ρc = 0.527 g cm–3. We represent the density of CHF3 by the reduced density ρr = ρ/ρc, as shown in Figure S3.