Abstract
Optical and thermal activity of plasmon-active nanoparticles in transparent dielectric media is of growing interest in thermal therapies, photovoltaics and optoelectronic components in which localized surface plasmon resonance (LSPR) could play a significant role. This work compares a new method to embed gold nanoparticles (AuNPs) in dense, composite films with an extension of a previously introduced method. Microscopic and spectroscopic properties of the two films are related to thermal behavior induced via laser excitation of LSPR at 532 nm in the optically transparent dielectric. Gold nanoparticles were incorporated into effectively nonporous 680 μm thick polydimethylsiloxane (PDMS) films by (1) direct addition of organic-coated 16 nm nanoparticles; and (2) reduction of hydrogen tetrachloroaurate (TCA) into AuNPs. Power loss at LSPR excitation frequency and steady-state temperature maxima at 100 mW continuous laser irradiation showed corresponding increases with respect to the mass of gold introduced into the PDMS films by either method. Measured rates of temperature increase were higher for organic-coated NP, but higher gold content was achieved by reducing TCA, which resulted in larger overall temperature changes in reduced AuNP films.
Export citation and abstract BibTeX RIS
1. Introduction
The optical and thermal characteristics of localized surface plasmons in gold nanoparticles (AuNPs) are of growing importance in biomedical diagnosis and treatment [1–4], photovoltaics with improved efficiency [5–8], and chemical synthesis and purification [9–11]. Large optical absorption cross sections of subwavelength AuNPs at resonant frequencies produce local electromagnetic fields which are orders of magnitude larger than the corresponding incident irradiation [12]. Conductive dissipation of this energy by phonon–phonon transfer can efficiently produce high local temperature changes by low-power, continuous irradiation [12–16]. Using AuNPs as optothermal transduction elements allows focused localization of heating, which can be employed to selectively ablate cancerous tissue [17–23] or kinetically drive local chemical processes. For example, AuNPs formed in microporous cellulose acetate membranes [24] and polydimethylsiloxane (PDMS) gels and foams [25] were recently shown to enhance solvent flux and drug delivery, respectively. Incorporation of AuNPs into dense nonporous dielectrics and films such as PDMS has been demonstrated using gold salt reduction both by superficial permeation into cured substrates [26] and by mixing in polymer precursors [27–29]. However, direct incorporation of pre-formed NPs in PDMS films has not been explored. The ease of fabrication, chemical stability, and optical transparency of PDMS films allows facile integration into biomedical [30, 31], sensing [28], and microelectromechanical systems [32–37] (MEMS).
The correspondence between spectral and thermal properties of nanoparticles in suspension, coated on surfaces, or embedded in films and microporous media is of increasing interest for device applications. Films embedded with magnetic nanoparticles [38, 39] or surface-coated with AuNPs [26] have been characterized at a range of electromagnetic frequencies and laser powers [12–16, 40–44]. Glass transition temperatures of these films [38] have been characterized as well as an equation derived to calculate temperature distributions for various concentrations or arrangements of NPs based on a low-intensity xenon flash lamp [45]. But spectral and thermal properties due to LSPR of AuNPs embedded in dense dielectric films remain uncharacterized. Key figures of merit such as optical absorptivity or thermal conductivity of dense AuNP nanocomposite media remain to be analyzed to support the design and development of robust, predictable biomedical, sensing, and MEMS devices.
This work compares two facile methods to embed AuNPs into dense composite thin films of PDMS, and introduces novel optoplasmonic characterization tools to compare the corresponding optical and thermal properties induced via laser excitation of LSPR within the optically transparent dielectric. Table 1 in the supplementary data (available at stacks.iop.org/Nano/23/375703/mmedia) directly compares these methods and their characteristics. The two methods are direct addition of organic-coated 16 nm nanoparticles into polymer film precursors, a new approach, and reduction of hydrogen tetrachloroaurate (HAuCl4, TCA) within the film [27–29]. Ethanol suspensions of organic-coated AuNPs (oAuNPs) and aqueous solutions of TCA, respectively, at increasing gold concentration were each dispersed in a mixture of Sylgard®184 PDMS reagents before curing to produce transparent, flexible films. Previously, addition of KAuCl4 to PDMS reagents had yielded brittle gels and foam [25]. Each method used in this work yielded films with values of absorption at a resonant wavelength and maximum steady-state temperature increases upon laser irradiation that rose as the gold mass per cent in the film increased.
The optical and thermal responses of AuNP embedded PDMS films are indicative of the Au morphologies that result from the respective methods. Optical extinction maxima at LSPR wavelengths and steady-state temperature maxima at 100 mW continuous wave laser irradiation each increased with respect to the mass of gold introduced into the PDMS films by either method. For the AuNP evaluated, the LSPR wavelength favored laser excitation near 532 nm at higher energies than photothermal therapy which favors LSPR wavelengths in the mid-IR 'water window' which increases optical penetration depth to ∼ centimeters. Measured rates of temperature increase per mass of Au added were higher for oAuNPs, but higher gold content was achieved using TCA, which enabled larger overall temperature changes. In particular, films containing dispersed oAuNPs (oAuNP–PDMS) displayed higher equilibrium temperatures per gold mass added, but the maximum equilibrium temperature was limited to ∼44 °C because the total mass of gold added by the method was limited by the NP densities of commercially available preparations and the solubility of ethanol in Sylgard® reagents. In contrast, the composite films containing AuNPs reduced from TCA (rAuNP–PDMS) reached a maximum temperature of ∼94 °C using just 10 μl of TCA (25% Au by mass) solution, because reduction of TCA allowed a higher total gold content to be achieved. This yielded higher overall steady-state temperature values at a given irradiation power. rAuNP–PDMS composite films thus appear useful for applications that use remote, optical induction of local thermalization such as facilitated heat transport [24, 46–48].
2. Experimental methods
2.1. Materials
PDMS (Sylgard® 184 silicone elastomer kit #4019862) was purchased from Dow Corning (Midland, MI). Hydrogen tetrachloroaurate (III) (TCA; HAuCl4⋅3H2O, G4022) was purchased from Sigma-Aldrich (St. Louis, MO). Organic spherical gold nanoparticles (oAuNPs (20-NS-20-50, 16 nm) were purchased from Nanopartz (Loveland, CO). Distilled, de-ionized water (Milli-Q, Billerica, MA) was degassed (DDD) for use in preparing aqueous solutions. Thermal Equipment: 532 nm green solid state laser (MXL-H-532) was purchased from CNI (Changchun, China). Power meter (PM310D), shutter (SH05), and shutter controller (SC10) were purchased from Thorlabs (Newton, NJ). Infrared thermal imaging camera (ICI 7320, P-Series) was purchased from Infrared Cameras Inc. (Beaumont, TX). Spectral equipment: 6 V tungsten microscope light source, a series of lenses for collimating and focusing, and a polarization crystal (GT5) were purchased from ThorLabs (Newton, NJ). A 3-axis micropositioner for film alignment, a 100 × microscope objective (NA= 0.70), additional focusing lenses, and a fiber optic collimator were used to collect the light and deliver it to a fiber optic spectrometer (AvaSpec 2048), purchased from Avantes Inc. (CO). A Lumenera Infinity 1–5 microscope was purchased from Nikon Instruments Inc. (Brighton, MI).
2.2. PDMS film fabrication with reduced TCA
All processes were performed at ambient temperature and pressure, unless otherwise noted. rAuNP–PDMS films were fabricated by first mixing the Sylgard® 184 base polymer and curing agent in a ratio of 10:1. A speed mixer (DAC 150SP, FlackTek, Inc., Landrum, SC) was used to mix the PDMS film solution for 2 min at 3035 rpm. Solid HAuCl4⋅3H2O was dissolved into DDD water to yield a ratio of 25% gold by mass. The 25 wt% HAuCl4 solution was further diluted with DDD water to eight different gold concentrations (figures 1(a) and (b)) which resulted in final film gold mass percentages ranging from 0.0073 to 0.1896%. Ten μl of each respective diluted HAuCl4 solution was added to 2.0 g PDMS film solution, then mixed for 4 min at 3035 rpm in the speed mixer. The resulting mixture was degassed for 3 min and then placed in an oven at 70 °C for 6 min to remove water. The mixture was then degassed for 10 min and mixed in the speed mixer for an additional 4 min to eliminate concentration gradients in the film solution. It was then poured onto a glass curing surface and distributed evenly using a micrometer film applicator (AP-99500303, Gardco, Pompano Beach, FL) to obtain consistent thickness of 0.68 mm. The film was then cured on a hot plate at 150 °C for 10 min. Each film was fabricated once and was sectioned into three 5.0 mm squares for analysis in triplicate.
Figure 1. (a) Optical images of 5 mm × 5 mm samples of PDMS into which preformed organic-coated 16 nm Au NP were mixed (at gold mass percentages shown above each image) prior to curing (oAuNP–PDMS). (b) Optical images of 5 mm × 5 mm samples of PDMS into which tetrachloroaurate solution was mixed (at gold mass percentages shown above each image) prior to curing (rAuNP–PDMS). Optical microscope images of oAuNP–PDMS (c) and rAuNP–PDMS (d) composite films at 0.005 mass% and 0.1896 mass%, respectively. Images (from left to right) were obtained in brightfield reflected mode, darkfield reflected mode, and brightfield transmitted mode, respectively. See text for details.
Download figure:
Standard image2.3. PDMS film fabrication with dispersed AuNPs.
oAuNP–PDMS films were fabricated similarly by first mixing the Sylgard® 184 base polymer and curing agent in a ratio of 10:1 for 2 min at 3035 rpm. Suspended oAuNPs were diluted in ethanol to 1013 NP ml−1. Volumes ranging from 97 to 241 μl were then mixed into the 1.0 g PDMS film solution for 10 min to obtain four different AuNP concentrations (figures 1(a) and (b)) in the film solution, with mass percentages in the film solution ranging from 0.0020 to 0.0050%. Higher gold content required longer mixing times to achieve full dispersion of oAuNPs into the PDMS film solution. For example, the 0.0050% required a mixing time of 15 min to reach full dispersion. The oAuNPs–PDMS film solution was degassed for 3 min and then placed in an oven at 70 °C for 6 min to remove ethanol, which otherwise formed bubbles in the film during boilout while curing. The viscous mixture was then degassed for 10 min and mixed in the speed mixer for an additional 10 min. The polymer solution was then distributed and thermally cured using the method discussed for rAuNP–PDMS.
2.4. Sample imaging
Samples were imaged with a digital camera (Lumenera Infinity 1–5; Nikon) in brightfield reflection mode. Features of composite thin films in images were characterized using image analysis software (Infinity Analyze). Images were stored as .pdf files, uploaded into Microsoft PowerPoint for cropping, and stored as .ppt files.
2.5. Thermal data collection
The PDMS films were aligned vertically with the laser and the infrared thermal imaging camera. The laser spot was centered in the 5.0 mm film square using a micropositioner attached to the film holder. The infrared thermal imaging camera was placed in line with the laser and film for optimal temperature recording. The laser was turned on at a power near 100 mW with a wait time of 35 min before experiments were conducted to obtain stable power output while running the thermal data collection. A neutral density filter was used to adjust the laser power to the desired power of 100 mW. The laser power was recorded before and after the experiment of each film using the power meter. The transmitted power was also recorded for each film using the power meter. The experiments consisted of a 3 min heating with the laser at 100 mW and a 2 min cooling time. The IR camera recorded the temperatures in a field of view containing each 5 mm× 5 mm film sample during the heating process. The laser-film-camera setup was enclosed in an environmental chamber, which remained closed during experiments to eliminate forced convection. Films with low gold content underwent the thermal characterization in triplicate exhibited minor differences, so a single film was evaluated at higher gold content. Further experiments show <5% differences in optical extinction across a larger range of gold content (data not shown).
2.6. Spectral evaluation
Spectral measurements were made using a custom designed transmission UV–vis microspectrometer [49]. Briefly, the spectrometer setup consisted of a 6V tungsten microscope light source, a series of lenses for collimating and focusing, a polarization crystal, a 3-axis micropositioner for sample alignment, a 100 × microscope objective, additional focusing lenses, and a fiber optic collimator to collect the light with a fiber optic spectrometer. Spectra were analyzed using Matlab, including savfilt, a generalized Savitzky–Golay smoothing algorithm, and peakdet, a maxima detection algorithm (both available from Matlab Central).
3. Results and discussion
3.1. Embedding AuNP in PDMS films
Dispersing preformed oAuNPs into PDMS resulted in a higher mass fraction of gold present as optically active AuNPs but a lower overall mass of AuNPs relative to reducing TCA in PDMS due to curability limitations. The magnitude and efficiency of optothermal conversion were observed to rise in proportion to the content of optically active AuNP. Figure 1(a) shows gold nanoparticle–poly(dimethylsiloxane) composite films formed by dispersing ethanol suspensions of organically coated 16 nm AuNPs (oAuNPs) in Silgard® precursor (oAuNP–PDMS); and figure 1(b) rAuNP–PDMS composite films formed by reduction of Au from aqueous solutions of tetrachloroaurate (TCA) into viscous PDMS precursor solutions at increasing Au mass percentages. Brightfield reflectance images in figure 1(a) show a pink intensity in oAuNP–PDMS composite films increasing as the gold mass percentage rises from 0.0020 to 0.0050%. Reported diameters for oAuNPs are 16 nm, which would strongly absorb and weakly scatter green light at the excitation wavelength of ∼532 nm.
In the images in figure 1(b) for rAuNP–PDMS composite films, the intensity of the pink color increases notably slower as the gold mass percentage increases from 0.0073 to 0.0368% than for oAuNP–PDMS composite films at corresponding mass percentages. Both the 0.0073% and 0.0141% gold samples in figure 1(b), for example, exhibit less color than the bright pink sample at 0.0050% in figure 1(a), indicating a smaller fraction of Au is present in a resonantly absorbent form. Comparing the two samples under light microscopy at 1000 × , as illustrated in figures 1(c) and (d), showed a higher incidence of microscopic gold inclusions and agglomerates in the rAuNP–PDMS samples. Greater aggregate content and reduced color intensity from resonant AuNP scattering suggested that a large fraction of dissolved TCA had been diverted away from AuNP formation and into microscopic aggregation. Both colorimetric and microscopic results indicate that the absorbent AuNP content per total mass of gold added appears lower in reduced Au films relative to dispersed Au films.
Macroscopic Au particles visible in figure 1(c) may result from coagulation of 16 nm oAuNPs within ethanol droplets in the organic PDMS mixture. However, the increased color intensity in figure 1(a) with increased gold content suggests a large fraction of sub-100 nm oAuNP remain in the PDMS, indistinguishable at the magnification shown in figure 1(c) (1000 × ). Increasing the mass per cent of oAuNPs blue-shifts the spectra from 554 to 532 nm, which suggests a higher proportion of smaller oAuNPs remain in samples with higher mass percentages. Comparing figures 1(c) and (d) results in similar conclusions for AuNPs formed from TCA, but the spectra do not show a monotonic shift in resonant wavelength as mass per cent TCA was increased.
At gold mass ≥0.0540%, the red hue of rAuNP–PDMS films in figure 1(b) deepens considerably. The corresponding brightfield transmission image of the 0.1896% sample in figure 1(d) echoes this red coloration, in contrast to the brightfield transmission image of the 0.0050% oAuNP–PDMS sample in figure 1(c), which lacks coloration. Darkfield transmission images (not shown) were indistinguishable from brightfield transmission images. oAuNP–PDMS samples at gold mass >0.0050% failed to cure, remaining gel-like even after increasing incubation time at curing temperature and being left at ambient conditions for more than a week. Failure to cure and substantially increased resonant coloration shows that, despite a larger fraction of aggregation, composite films with higher AuNP content were fabricable by reduction rather than by dispersion. Figures 1(c) and (d) shows AuNP reflection brightfield samples in which some inclusions appear yellow, indicating metallic reflectance. Occasionally such an inclusion will appear red in reflection darkfield and transmitted brightfield modes (see arrows). Red is an indication of absorption and/or scattering of resonant light. Other inclusions remain gold or turn dark. These changes likely reflect distinct local optical environments, because macroscopic inclusions are not expected to resonantly absorb. For similar reasons, the apparent size of some inclusions changes between the various modes, underscoring the heterogeneous nature of the inclusions, their local environment, and indicating the sample depths probed by the respective modes.
3.2. Fabrication of oAuNP–PDMS composite films
Dispersing preformed AuNPs into PDMS composite film required developing a fabrication method to address solvent outgassing and uneven particle distribution of AuNPs observed during preparation. Figure 2(a) illustrates the appearance of PDMS and oAuNP mixtures after each successive step in the finalized fabrication process, as well as the final composite film. The initial PDMS solution with the crosslinker was transparent. Ethanol containing dispersed oAuNPs added to the solution is initially suspended on the top of the PDMS solution without any visual dispersion. Speed-mixing oAuNPs into the PDMS the solution resulted in a pale pink color. Following the mixing step, the solution is placed in the oven for a pre-cure step to drive off ethanol, which results in the solution becoming a light purple.
Figure 2. Diagram of the steps and color changes that occur during fabrication of (a) oAuNP–PDMS and (b) rAuNP–PDMS. The images to the right are of the AuNP–PDMS samples with the highest gold content, showing the final color after curing.
Download figure:
Standard image3.2.1. Solvent outgassing.
Initial composite films into which oAuNP-containing ethanol suspensions were dispersed contained visible gas voids. Ethanol-free preparations lacked such voids, suggesting that ethanol boiling out of the film during the PDMS film cure step at 150 °C was the root cause of the voids. Increasing the degassing time of the salmon-colored solution containing oAuNPs mixed into Sylgard® reagents from 10 to 25 min before thermally curing the film was not sufficient to eliminate the voids. However, adding a low-temperature, intermediate heating step at atmospheric pressure before the PDMS film was cured eliminated a majority of the bubbles. The time and temperature of this step, which yielded the lavender solution shown in figure 2(a), were set at 6 min and 70 °C, respectively. These values maximized the optical activity of the oAuNP–PDMS composite films obtained by dispersion.
3.2.2. Uneven NP distribution.
Eliminating the vapor bubbles revealed that subsequent dispersal of the NPs in the film was not uniform. The pre-cure intermediate heating step introduced a top-to-bottom increase in NP content which inverted when the mixture was deposited onto the glass plate beneath the film applicator. The concentration of AuNPs increased in the direction in which the film applicator was moved to spread the PDMS mixture, creating a visible gradient from light to dark pink. Increasing the duration of mixing after adding the AuNPs to 15 min reduced, but did not eliminate, the gradient. However, a second mixing step added after degassing the pre-heated mixture just prior to pouring and casting the composite PDMS film eliminated the gradient. The duration of mixing for this step was 10 min. Adding intermediate heating and degassing steps and a second mixing step resulted in oAuNP–PDMS films with uniformly dispersed NPs that lacked gas voids.
3.3. Fabrication of rAuNP–PDMS composite films
Dispersing as-formed AuNPs into PDMS composite film required developing a fabrication method to address film brittleness and solvent outgassing observed during preparation. Figure 2(b) illustrates the appearance of PDMS and TCA mixtures after each successive step in the fabrication process, as well as the final composite film. As with the oAuNP procedure, the initial PDMS solution with the crosslinker was transparent. TCA solution added to the PDMS solution remains suspended without apparent dispersion. As the two solutions are rotated in the speed mixer, the droplet of TCA follows the direction of motion. Once the TCA is mixed into the PDMS the solution becomes a pale gold color. Following the mixing step the solution is placed in the oven for a pre-cure step to drive off moisture, which results in the solution becoming a pink or purple with intensity ranging from light to dark depending on the concentration of the TCA added.
3.3.1. Film brittleness.
Addition of KAuCl4 to PDMS had previously resulted in brittle films and foams [25]. Adding up to 300 μl of TCA to a 2 g mass of PDMS mixed with curing agent similarly resulted in either a brittle film or failure of the film to cure. Reducing the solution volume of TCA added to a 2 g mass PDMS/curing agent mixture from 300 to 10 μl while maintaining constant gold content by using more concentrated TCA yielded AuNP–PDMS composite films that were fully cured, transparent, and not brittle. The texture and transparency of resulting films allowed laser experiments to be conducted and spectral data to be collected.
3.3.2. Solvent outgassing.
Formation of vapor bubbles during thermal curing of rAuNP–PDMS initially produced visual anomalies that precluded unambiguous microscopic and spectroscopic analysis of composite films. This problem was similar to the vapor bubbles observed in curing oAuNP–PDMS. So a similar low-temperature, intermediate heating step at atmospheric pressure before the PDMS film was cured was added in order to eliminate a majority of the bubbles. The time and temperature of this step, which yielded the reddish solution shown in figure 2(b), were also set at 6 min and 70 °C, respectively. A final mixing step with the same duration as the initial mixing step (4 min) was also added after the thermal pre-cure, as in the steps for oAuNP–PDMS film fabrication, even though there was no gradient issue with the TCA films. This resulted in the final procedures for the two fabrication approaches being similar, facilitating more direct comparison of their respective microscopic, spectroscopic, and thermal properties. The primary differences were that the final oAuNP procedure used a smaller mass of PDMS film solution (1.0 g versus 2.0 g), a larger added volume of gold-containing solution (200 μl versus 10 μl), and longer mixing times (15 and 10 min, respectively, versus 4 min). These values improved the spectroscopic and microscopic characterization of rAuNP–PDMS composite films.
3.4. Characterizing AuNP in thin polymer films
Once development of the procedures for the two methods of producing AuNP composites was complete, optothermal and spectroscopic properties were analyzed and compared for oAuNP–PDMS and rAuNP–PDMS films. Overall, increases in gold content resulted in approximately proportional increases observed in the spectral extinction near the surface plasmon resonance, in the power consumed from an incident laser beam, and in the temperature increase of the nanocomposite polymer film. The optothermal response of oAuNP–PDMS films was more efficient than rAuNP–PDMS, based on the mass of gold added. However, the maximum attainable temperature was higher for the latter films due to the larger gold content possible in rAuNP–PDMS composites.
3.4.1. Optothermal and microspectroscopic analysis.
In figure 3, thermal topographies of 25 mm2 samples of oAuNP–PDMS (a) and rAuNP–PDMS (b) nanocomposite films are mapped during irradiation at 532 nm with 100 mW of power at ambient conditions. The laser spot of ≤2 mm was centered on each sample, resulting in a centrosymmetric parabolic temperature distribution. While the power distribution profile of the fiber optic laser used was not measured, the beam was expanded to several inches to confirm the power distribution was uniform rather than Gaussian. This indicates that the thermal profile in figure 3 results primarily from radial conduction away from excited AuNP in the spot, rather than from the profile of the laser spot. We have previously modeled spatiotemporal thermal distributions for closed and open Au-ceramic nanocomposite systems [13–15]. In order to perform a similar detailed analysis of more refined Au–polymer nanocomposites, this report begins by correlating the maximum thermal responses of Au–polymer nanocomposites fabricated by two methods under open, ambient conditions with corresponding optical and microscopic features.
Figure 3. Thermal profiles for (a) oAuNP sample at 0.005% Au (highest gold content) and (b) rAuNP sample at 0.1896% Au (highest gold content). Note the maximum temperature of 44 °C attained in (a) is 50 °C lower than the 94 °C maximum in (b).
Download figure:
Standard imageThe optothermal images in figure 3 are from samples shown in figures 1 and 2 which have the highest attainable mass percentages of Au for oAuNP–PDMS (0.0050%) and rAuNP–PDMS (0.1896%), respectively. Identical irradiation parameters were used to obtain optothermal images for each oAuNP–PDMS and rAuNP–PDMS sample across the range of mass percentages shown in figure 1. The difference between maximum and ambient temperature in each image was compared with the measured power absorption and the value of spectral extinction at 532 nm obtained in a separate transmission UV–vis microspectroscopy measurement, as shown in figure 4.
Figure 4. Laser power loss, change in temperature, and spectral extinction comparison (inset (2)) of the oAuNP–PDMS films (a) and rAuNP–PDMS films (b). Inset (1) shows the diagram of where the spectral extinction values at 532 nm came from.
Download figure:
Standard imageFigure 4(a) (inset (1)) shows transmission UV–vis spectra measured using custom microspectroscopy apparatus [43] applied to the samples analyzed in figure 3. The difference between the spectral extinction at 532 nm and the interpolated spectral baseline was recorded as peak height at 532 nm from the baseline for each sample. This difference corresponded to the measured laser power loss and change in maximum temperature obtained from optothermal images and is illustrated in figure 4 as 'Extinction (AU)'. The maximum temperature change, power loss, and extinction at 532 nm were utilized in calculations discussed below. The wavelength of 532 nm was selected because the laser from the optothermal experiments has a wavelength of 532 nm and the plasmon peak of the AuNPs is near 532 nm. Previous work suggested that plasmon absorption occurs in proportion to incident power absorbed at the laser wavelength. These measures allowed direct comparison of spectroscopic and optothermal measures between the two methods for fabrication of Au–polymer nanocomposite films.
3.4.2. oAuNP–PDMS composite films.
oAuNP–PDMS composite films (figure 1(a)) generally contained lower gold content than rAuNP–PDMS films (figure 1(b)). Commercially available oAuNP solutions had variable, batch-dependent concentrations that ranged from 4 to 7 × 1013 NP ml−1. These solutions were then diluted with ethanol to a common concentration of 1013 NP ml−1. At this concentration, film Au content was limited by the miscibility of oAuNP solution in PDMS precursors. Resulting films either did not cure or the oAuNPs did not fully disperse into the PDMS solution. It may be possible to increase oAuNP concentration about 4–7-fold using the method described herein, by not performing the initial dilution of the commercial solution and using them as received. This assumes the consistency of available solutions and that particle aggregation will not become a barrier at higher concentrations. However, this approach would also proportionally increase the cost of the composite material, which could significantly reduce its practicality. These oAuNPs were the only commercially available particles that could be found that were suitable for this study in terms of morphology, concentration, and availability. If other, more concentrated particles were found or fabricated reproducibly, consistently higher levels of Au content could potentially be reached.
3.4.3. Thermoplasmonic response of oAuNP–PDMS at 532 nm.
The correspondence between spectroscopic, optothermal, and resonant laser absorption characteristics measured in oAuNP–PDMS composite films as a function of gold content is summarized in figure 4(a). The maximum temperature (left axis) increased linearly with gold content. Laser power loss also increased linearly with gold content, except at 0.0020 mass%. Spectral extinction at 532 nm increased nearly in proportion to gold content (inset (2)). Specifically, at a mass gold percentage of 0.0020% in the oAuNP–PDMS composite films, the maximum temperature reached approximately 33 °C—a 10 °C temperature change increase from the initial ambient value of 23 °C—while spectral extinction reached 0.0195. This corresponded to 5000 °C and 9.75 A U per% mass gold, respectively. At a mass gold percentage of 0.0050%, the maximum temperature reached approximately 44 °C—a 21 °C temperature increase—while spectral extinction reached 0.0545. This corresponded to 4200 °C and 10.9 A U per% mass gold, respectively. It appears that increasing gold content in oAuNP–PDMS films yields nearly proportional increases in spectral extinction and power loss at 532 nm. This suggests that Au-specific absorptivity remains about constant. These trends appear consistent with previously reported increases in temperature change and laser power extinction measured for AuNP–silica composites [14, 15]. But as gold content increases, the maximum temperature in oAuNP–PDMS films per gold content decreases while spectral extinction per gold content increases. This could result from increasing composite film thermal conductivity or decreasing optothermal transduction efficiency (or both) at higher gold content. Steady-state temperatures of isolated gold NPs of various shapes (e.g. nanostars [50], rods, dimers, disks and rings [52]) or regular assemblies (chains and square arrays) and their immediate environment have been modeled. Uniform molecular coatings and local thermal properties have been considered [51]. These studies relate temperature distributions quantitated at nanoscales to optical and physical properties of fundamental thermoplasmonic elements. However, complexities of the size, shape, and distribution of AuNPs and the surrounding PDMS matrix in the present work, as well as temperature measurements made at micro-scales, prevent direct application of these model results. Further quantitation of these complexities is underway to support modeling of dynamic and steady thermal temperature conditions in nanocomposite polymer films.
3.4.4. rAuNP–PDMS composite films.
Unlike the less concentrated oAuNP samples (figure 1(a)), the rAuNP–PDMS composite films (figure 1(b)) generally contained higher gold content. Increasing gold content further required more concentrated solutions of TCA not readily available due to the dilution of the solution to a specific gold concentration. Greater AuNP solution volumes required to produce higher gold content required the addition of more than the 10 μl volume established for the AuNP films. This resulted in the film either not curing or being too fragile to conduct optothermal experiments. For example, since a volume of 10 μl was used for 2.0 g of PDMS in the baseline process, to increase the amount of gold in the PDMS, 15 μl was added to a 2.0 g PDMS film. This resulted in a gel-like film. Such resulting films remained gel-like even after increasing the final cure temperature above 150 °C, extending the final cure beyond 10 min or allowing ambient incubation for weeks. Laser experiments were not conducted for gel-like films because the stability of the film was insufficient to allow stable vertical alignment in the laser measurement system.
3.4.5. Thermoplasmonic response of rAuNP–PDMS at 532 nm.
Correspondence between spectroscopic, optothermal, and resonant absorption characteristics measured in rAuNP–PDMS composite films as a function of gold content is summarized in figure 4(b). The film with the lowest gold content (0.0073%) reached a maximum temperature of about 34 °C, corresponding to an 11 °C temperature change, while spectral extinction reached 0.0092. This corresponded to 1507 °C and 1.26 A U per% mass gold. These values are 3.3- and 7.7-fold lower than the corresponding values of 5000 °C and 9.75 A.U. per% mass gold for 0.0020% oAuNP–PDMS films, respectively. In addition, they are 2.8- and 8.6-fold lower than those for 0.0050% oAuNP–PDMS films. Thus while the temperature rise per mass gold for rAuNP–PDMS films appears about three-fold lower, the measured optothermal efficiency (temperature increase per extinction unit at 532 nm) is 2.3–3.1-fold higher. The maximum temperature reached for the film with the highest gold content (0.1896%) was approximately 94 °C—a 71 °C increase from the starting temperature of the film—for a spectral extinction of 0.1503. This corresponded to 374 °C and 0.79 A U per% mass gold. Thus, as gold content increases in rAuNP–PDMS films, the maximum temperature per gold content decreased much faster than the extinction at 532 nm per gold content.
Overall, the temperature profiles in figure 4(b) show that as the gold content increases in rAuNP–PDMS nanocomposites, the maximum temperature will increase less than linearly until the maximum curable gold content is reached. For the samples tested, aqueous dissolution of TCA to 25 wt% resulted in a 0.1896 mass per cent of gold in the film when 10 μl of the stock solution was added to the polymer precursor. Gold content was increased past the 0.1896% but the resulting films barely cured and were too fragile to undergo the laser experiment. These trends differ from previously reported increases in temperature change and laser power extinction measured for AuNP–silica composites [14, 15], and thus warrant further study.
The thermal dynamics of TCA reduction to AuNPs on PDMS precursor were significant in achieving the maximum temperature increases per gold content. The time of the thermal pre-cure (which was originally intended to eliminate bubbles in the film upon subsequent final curing) proved to be important in obtaining the maximum optothermal response. A notable increase in viscosity of the rAuNP–PDMS solution after the pre-cure suggested that polymerization occurred to some extent in this step. Visual colorimetric observations indicated that the reduction of Au also began during pre-cure. Increasing the pre-cure from 6 to 10 min lowered the maximum temperature that two films reached during the laser experiment by about 10 °C. On the other hand, a 6 min pre-cure did not provide an energy sufficient for complete reduction of Au to NP, because colorimetric Au reduction did not appear complete until the final cure was finished.
3.4.6. Thermoplasmonic response related to AuNP content.
In general, the maximum temperature changes and extinction values measured at 532 nm per unit mass of gold added to organic-coated NP films were both higher than for gold reduced onto polymer films. As examples, at 0.0073% gold content in rAuNP–PDMS films, the ratios of maximum temperature increase and extinction at 532 nm per% mass gold were 3.3- and 7.7-fold lower than those for 0.0020% oAuNP–PDMS films, respectively, as shown above. They were also 2.8- and 8.6-fold lower than those for 0.0050% oAuNP–PDMS films. As a result, essentially doubling the gold content in oAuNP–PDMS films from 0.0026% to 0.0050% increased the temperature by 8.4 °C, at a rate of 3360 °C per cent Au; whereas a comparable 8.5 °C temperature change obtained by doubling the gold content in rAuNP–PDMS films from 0.0073% to 0.0141% was produced at a 2.8-fold lower rate of 1214 °C per Au%.
On the other hand, the higher gold content achievable by reducing TCA in polymer composite films resulted in larger overall temperature increases. The maximum temperature (43.9 °C) attained in the most concentrated oAuNP–PDMS composite film (0.0050%) was 1 °C higher than the maximum temperature (42.9 °C) in an rAuNP–PDMS film with about three times the Au content (0.0141%). Figure 3 shows the maximum temperature reached by any film tested was approximately 94 °C for rAuNP–PDMS at 0.1896%. This temperature could be increased by increasing TCA concentration to its solubility limit in H2O to increase the Au content in the final film, without increasing the volume of solution added to the polymer precursors. On the other hand, increasing Au content in such oAuNP–PDMS samples was limited by the unstable, gel-like properties of more concentrated films. For example, approximately 241 μl in a 1.0 g film was required to reach 0.0050% of gold in an oAuNP–PDMS film. This larger solution volume required 5 additional minutes for mixing into polymer precursors than lower volumes. Still larger volumes would disperse less freely and produce uneven gold content.
3.4.7. Optothermal efficiency.
Interestingly, comparable rAuNP–PDMS films exhibited higher apparent optothermal efficiencies determined using spectral extinction, i.e. the maximum temperature increase produced per extinction unit at 532 nm. While the temperature rise per mass gold for 0.0073% rAuNP–PDMS films was about 3-fold lower than that for comparable oAuNP–PDMS films, measured values of temperature increase per measured extinction at 532 nm (optothermal efficiency) were 2.3–3.1-fold higher. A higher optothermal efficiency could result from increased absorption relative to scattering at 532 nm or from decreased thermal conductivity in reduced Au–polymer films, or from some combination of both. One possibility could be that organic-coated AuNPs are likely sequestered in PDMS films due to hydrophobic interactions, while TCA may reduce directly onto active crosslinker sites. This could result in substantially different local particle environments for the two film types, which could play an important role in their thermoplasmonic response. The medium adjacent to AuNPs in either type of film significantly affects the optical and thermal behavior. Gaseous space adjacent to particles reduces their effective local refractive index, blue-shifts the LSPR peak, and changes thermoplasmonic absorption. Adjacent gas also insulates particle surfaces, slowing thermal conductivity. Adjacent condensed matter generally has the opposite effect [15]. The proprietary polymer coating of oAuNPs may further influence the optical and thermoplasmonic properties. Quantitative microscopy and characterization of environments adjacent to embedded AuNPs in PDMS for each type of film is the subject of future work.
Figure 5 compares the change in the maximum temperature and total power loss versus gold content for the range of oAuNP–PDMS films tested with three comparable rAuNP–PDMS films. The latter are the three lowest gold-containing rAuNP–PDMS films of those evaluated. The oAuNP–PDMS film with 0.0050% gold gives roughly the same temperature increase and power loss as the rAuNP–PDMS film with 0.0141% gold, even though the latter sample has approximately one-third the gold content of the former. Thus, oAuNP–PDMS films intrinsically evolve more heat with less gold, even though the temperature increase per extinction at 532 nm appears smaller. The thermal response of oAuNPs per increase in gold content, 3.6 × 103 °C/Au%, is ∼4.7-fold larger than that of rAuNPs, 7.6 × 102 °C/Au%. Overall, the increase in incident power absorbed at 532 nm relative to gold content essentially tracked the increase in maximum temperature change. On a mass per cent basis, the oAuNP–PDMS nanocomposite films convert laser power loss into heat more efficiently than rAuNP–PDMS films. Laser power loss tracks temperature increases relative to gold per cent more accurately for both o- and r-AuNP-PDMS films. The magnitude and efficiency of optothermal conversion appears related to the degree to which gold content corresponds to absorbent AuNPs. Brighter red color in oAuNP–PDMS films relative to rAuNP–PDMS films at comparable gold content indicates a larger fraction of gold occurs as resonant absorbent AuNP in the former, while the latter contain a larger fraction of non-nanoscale particulate inclusions.
Figure 5. Comparison of the slopes for the change in temperatures and laser power loss values for the rAuNP–PDMS and oAuNP–PDMS films.
Download figure:
Standard image4. Conclusions
In this work, microscopic, optical, and optothermal properties of composite gold nanoparticle (AuNP)–polydimethylsiloxane (PDMS) films produced by direct addition of organic-coated 16 nm AuNPs (oAuNP–PDMS) and reduction of hydrogen tetrachloroaurate (rAuNP–PDMS) were compared. Methods for embedding and reducing gold into PDMS film were developed and refined, respectively, to obtain films with thermal and optical properties useful for photovoltaic, optoelectronic, or bio/medical applications. The maximum temperature in both oAuNP–PDMS and rAuNP–PDMS composite films increased as gold content increased, with corresponding measurable laser power losses. The oAuNP–PDMS films could be heated to comparable temperatures at about one-third less gold content than rAuNP–PDMS films. But the concentrations of commercially available organic AuNPs limited gold content within oAuNP–PDMS films. As a result, oAuNP–PDMS film with about 1/40th the gold content of curable rAuNP–PDMS films reached maximum temperatures of 44 °C compared with maximum temperatures near 94 °C in the latter.
Acknowledgments
K Berry performed the literature research and lab work and drafted the data and text for the manuscript. A Russell and P Blake designed and constructed the thermal and spectroscopic systems and developed methods. D K Roper directed the work and was the primary author of the manuscript. The authors thank J Hooker for early participation in development and J Hestekin for use of equipment. This work was supported in part by NSF CMMI-0909749, NSF ECCS-1006927, NSF CBET 1134222, the University of Arkansas Foundation, and the Walton Family Charitable Support Foundation.