Photothermal Heating of Plasmonic Nano-antennas: Influence on

Subscriber access provided by Monash University Library

Article

Photothermal Heating of Plasmonic Nano-antennas: Influence on Trapped Particle Dynamics and Colloid Distribution Steven Jones, Daniel Andrén, Pawel Karpinski, and Mikael Käll ACS Photonics, Just Accepted Manuscript • Publication Date (Web): 21 May 2018 Downloaded from http://pubs.acs.org on May 21, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Photonics

Photothermal Heating of Plasmonic Nano-antennas: Influence on Trapped Particle Dynamics and Colloid Distribution AUTHOR NAMES: Steven Jones*, Daniel Andrén, Pawel Karpinski, and Mikael Käll* AUTHOR ADDRESS: Department of Physics, Chalmers University of Technology, 412 96 Göteborg, Sweden. KEYWORDS: Nanoscale Thermometry, Plasmonic Optical Tweezers, Inelastic Light Scattering, Thermoplasmonics, Interfacial Thermal Resistance, Thermophoresis. ABSTRACT: Plasmonic antennas are well-known and extremely powerful platforms for optical spectroscopy, sensing, and manipulation of molecules and nanoparticles. However, resistive antenna losses, resulting in highly localized photothermal heat generation, may significantly compromise their applicability. Here we investigate how the interplay between plasmonenhanced optical and thermal forces affect the dynamics of nano colloids diffusing in close proximity to gold bowtie nanoantennas. The study is based on an anti-Stokes thermometry technique able to measure the internal antenna temperature with an accuracy of ~5 K over an extended temperature range. We argue that Kapitza resistances have a significant impact on the local thermal landscape, causing an interface temperature discontinuity of up to ~20% of the total photothermal temperature increase of the antenna studied. We then use the bowties as plasmonic optical tweezers and quantify how the antenna temperature influences the motion and distribution of nearby fluorescent colloids. We find that colloidal particle motion within the plasmonic trap is primarily dictated by a competition between enhanced optical forces and enhanced heating, resulting in a surprising insensitivity to the specific resonance properties of the antenna. Further, we find that thermophoretic forces inhibit diffusion of particles towards the antenna and drive the formation of a thermal depletion shell extending several microns. The study highlights the importance of thermal management at the nanoscale and points to both neglected problems and new opportunities associated with plasmonic photothermal effects in the context of nanoscale manipulation and analysis.

TOC Graphic

ACS Paragon Plus Environment

1

ACS Photonics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 34

Corresponding Authors *E-mail: [email protected] *E-mail: [email protected] Author Contributions All authors have contributed and given approval to the final version of the manuscript. Funding Sources The Swedish Research Council (VR), The Knut and Alice Wallenberg Foundation.


2

Page 3 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Photonics

INTRODUCTION Plasmonic nano-antennas are an important tool for sensing and analysis of nanoscale objects. Their utility is based on the large electric field enhancement generated at plasmon resonance, which amplifies far-field optical coupling to nano-objects, and the simultaneous confinement of the optical interaction volume to deeply subwavelength scales. This enhancement/confinement provides greater selectivity and contrast in optical sensing techniques. Past works have utilized this phenomenon in, for example, molecular sensing applications based on refractive index variations1,2, surface-enhanced Raman scattering3,4 and surface-enhanced fluorescence5. Plasmon resonances in metal nanoantennas can also be used to induce forces that directly act on the nanoobjects to be sensed. This is the concept behind plasmonic optical tweezers, which utilizes field-enhancement and confinement to dramatically amplify optical gradient forces and the depth of the associated optical potential well acting on diffusing nanoobjects6,7. In this way, optical trapping and manipulation technology can be extended to the nanoparticle regime, which is typically out of reach for conventional optical tweezers. Plasmonic excitation also enhances photothermal heat generation due to the finite conductivity of metals at optical frequencies. Indeed, plasmonic nanoparticles have been utilized as extremely efficient heat sources, remotely controllable through far-field illumination, in a variety of applications8,9. Plasmonic heating can be expected to increase the thermal Brownian motion of nanoobjects diffusing in the vicinity of a resonantly excited nanoantenna, thereby directly counteracting plasmon-enhanced optical forces. Moreover, because of the highly localized nature of the heat generation, large temperature gradients are expected near resonantly excited nanoantennas. Such gradients can contribute to directed motion and even manipulation of nearby nano-objects through thermophoretic forces,10 as recently demonstrated experimentally11–13. It is


3

ACS Photonics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 34

worthwhile to note the experimental similarities between plasmonic optical trapping and thermophoretic manipulation platforms; and yet, the fundamentally different forces involved. The thermal aspects of plasmonic excitation are of well-known critical importance in works that involve biological analytes, for which elevated temperatures can cause denaturation14,15, conformational changes16,17, or influence the interaction affinity of the biological particles being studied. However, the interplay between plasmon-enhanced optical and thermal forces is much less studied and understood. To reach such an understanding, a first step is obviously to be able to accurately describe the nanoscale thermal landscape surrounding a plasmonic antenna structure. Unfortunately, this task is far from trivial. Fluorescence based thermometry techniques, in which temperature dependent fluorophores are used as local probes for temperature mapping18, can be strongly affected by plasmonic enhancement and quenching effects and have diffraction limited spatial resolution, which is also the case for temperature measurements based on phase transitions in surrounding materials19. Thermometry methods based on plasmon resonance shifts caused by temperature induced variations in the refractive index of the surrounding medium,20,21 on the other hand, requires an a-priori assumptions on the temperature distribution outside the antenna. An additional difficulty in predicting and simulating the temperature profile around a plasmonic antenna comes from poorly defined nanoscale thermal boundary effects. In particular, imperfect phonon coupling between two dissimilar materials results in increased thermal resistance, also known as the Kapitza resistance, which produces a discrete temperature discontinuity at the interface22. The magnitude of this effect is highly material dependent and difficult to predict quantitatively, but it can be expected to have significant impact on the temperature profile around plasmonic antennas supported by a substrate.


4

Page 5 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Photonics

In this article, we attempt to address some of the issues discussed above based on experiments and simulations on one of the most important types of nanophotonic antenna structures, the plasmonic bowtie23,24. We first address the question of obtaining an accurate temperature determination, where we use a variant of a recently developed and non-invasive far-field optical thermometry technique based on ratiometric analysis of anti-Stokes (AS) inelastic light emission25. This method directly probes the electron-hole energy distribution and, consequently, the internal temperature of the antenna itself. We validate the AS thermometry method using a temperature-controlled stage and then apply the technique to quantify photothermal heating of individual nano-antennas situated in an aqueous environment. The results are compared to finite element simulations, which allows us to discuss the influence of Kapitza resistances on the local temperature profile. We next investigate and discuss the interplay between optical and thermal forces acting on colloidal nanoparticles diffusing near an optically excited bowtie antenna. Two length scales are investigated; submicron distances from the antenna, where optical gradient forces dominate but thermal effects are found to have pronounced impact on particle motion; and micron length-scales, where thermophoretic forces are found to push particles away from the plasmonic heat source, leading to a large depletion region surrounding the antenna. Altogether, the results highlight the complex interplay between optical and thermal effects in plasmonic systems and point towards new avenues for active particle manipulation and novel plasmonic sensing techniques.

RESULTS AND DISCUSSION Anti-Stokes thermometry of plasmonic nano-antennas. Plasmonic particles excited by a laser are well known to produce broad and rather featureless emission spectra (Figure 1a), often


5

ACS Photonics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 34

referred to as a surface-enhanced Raman scattering (SERS) background26–29. Several different explanations for the origin of this feature have been proposed, including photoluminescence due to hot carrier recombination following plasmon excitation30,31, surface-enhanced fluorescence from contaminant molecules28, and Raman scattering from electronic transitions in the metal28,32– 34

. While it is likely that these factors will all affect the observed spectra under certain

conditions, several works have been conducted to determine the extent of their influence. Portales et al. has shown similar spectra (in terms of Stokes shift) for the same plasmonic particles under different excitation wavelengths, indicating an inelastic scattering component of the signal34. Clean Au samples were tested to determine the effects of surface contaminants, finding that the emission background persisted without any adsorbate molecules26,35. Past reports have shown that the background spectrum also persists even for incident photon energies lower than the onset of interband transitions, indicating they are not the sole contributor28. One explanation that has gained considerable traction in the community, is through an emission process where electrons are excited from just below the Fermi level to a virtual state, and instantaneously relax back into a vacant state28,32,33,36. This process, sometimes referred to as inelastic light scattering (ILS), has been shown to fit well with the spectrum on the anti-Stokes side of the excitation wavelength28,32.

Figure 1. Demonstration of anti-Stokes thermometry. a) Typical experimental emission spectrum from a single gold nano-antenna excited at 633 nm. b) Semi-log plot of AS spectra,


6

Page 7 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Photonics

normalized to the emission intensity at a Stokes shift of -500 cm-1, demonstrating the increasing portion of the signal at higher photon energies as the temperature is increased (temperature legend given in c). c) Ratio of each spectra in b) to a reference spectrum obtained at 300 K. Dashed black lines are numerical fits to the ratio of two Boltzmann distributions, Eq. (3). d) Extracted temperatures from numerical fitting in c) plotted against the stage temperature. Violin plots are used to show the distribution of extracted temperatures from several measurements for a given stage temperature. Regardless of the exact origin of the emission spectra, it has been shown that the anti-Stokes side of the spectrum, (, , ), is well described by 36,37: (, , ) ∝ (, , ) (, ), #(1) where is the frequency of the elementary excitation, is the frequency of the excitation laser, is the temperature of the plasmonic particle, and ℎ( − ) (, , ) = exp − 1 , #(2)

is the Bose-Einstein thermal population factor, with ℎ and being elementary constants. The factor (, ) describes the combined spectral response of the plasmonic particle and the detection system at the excitation and emission frequencies. We first validated the AS thermometry method by investigating several different types of gold nanoantennas with uniform temperature set through a temperature-controlled microscope stage. Figure 1b-d illustrate the measurement procedure and the results for the case of an array of rectangular Au antennas fabricated on fused silica by electron beam lithography and measured in air. The antennas were designed to resonate at the probe laser wavelength 633 nm to maximize the AS emission yield, but the laser spot size was defocused over ~20 particles to prevent heating from the probe laser. The integrated anti-Stokes signal from this ensemble was then recorded as a function of stage temperature as shown in Figure 1b, where we have normalized the signal to highlight the expected relative increase of the AS emission for increasing stage temperature.


7

ACS Photonics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 34

To obtain an accurate temperature estimate from the AS spectra, it is necessary to remove the spectral response function (, ) from the measured signal, thereby isolating the spectral modulation arising from temperature variations. Several methods have been proposed in the literature to achieve this. In the initial work by Hugall et al.,28 plasmonic substrate with a very broad spectral response, such that (, ) can be assumed effectively constant over the spectral range of interest, was used. This technique was then extended to particles with sharp plasmonic resonances by Xie et al.,25 who utilized one spectrum at a known temperature as a reference. That is, by taking the ratio of the AS spectra with the reference, the spectral response (, ) is cancelled out (assuming it is temperature independent). Recently, Carattino et al.38 used gold nanorods as thermometers by approximating their dark field spectra as a Lorentzian profile and using this analytical form of (, ) to extract the temperature from the observed AS spectra. In this work, we use a method similar to the technique described by Xie et al. but allow for

(, ) to incur slight amplitude variations, either due to temperature or environmental factors, while assuming that the spectral shape is strictly temperature independent. We found that this method of temperature extraction resulted in better agreement with the stage temperature and smaller variation in the extracted temperatures than what was achieved without the extra fitting parameter. The approach is justified in that it is the relative spectral shape of the anti-Stokes emission that is most sensitive to temperature changes, rather than the overall change in AS amplitude. The temperature of the antenna is thus extracted from the ratio: ℎ( − ) −1 exp (, , ) (, , ) ! R(, ! , , ) = ∝ = , #(3) ℎ( − ) (, , ! ) ! (, , ! ) exp −1 Figure 1c shows the ratios of the spectra from Figure 1b, using the signal at ! = 300 K as a reference, together with numerical fits to Eq. (3). The antenna temperatures extracted using


8

Page 9 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Photonics

this methodology are compared to the set stage temperatures in Figure 1d and found to be in excellent agreement, thus validating the methodology.

Measurement of plasmonic heating in single nano-antenna. The AS thermometry technique was next applied to quantify laser induced heating of individual bowtie nano-antenna fabricated on a fused silica substrate, a system which is suitable for plasmon-enhanced optical trapping and sensing. A 1064 nm laser beam was focused onto a single gold bowtie, situated in water, (Figure 2a-e) to act as the photothermal heat source (and subsequently the trapping source) while the low-intensity 633 nm probe laser simultaneously excited the anti-Stokes emission used for thermometry. Heat generation and thermometry are thus de-coupled and independent. The lengths L of the individual nano-antennas elements were varied to change the degree of interaction between the corresponding longitudinal dipolar plasmon resonance and the 1064 nm heating laser (Figure 2f, g) while the widths W were kept constant such that the transverse dipole plasmon always resonate near the probe laser wavelength at 633 nm (Figure 2h,i). The gap widths varied between ~30 nm for L = 120 nm to ~60 nm for L = 230 nm, which is suitable for optical trapping of small beads (see Methods section in Supplementary Information for details on fabrication and antenna characterization). The scattering spectra of the bowties, shown in Figure 2f and h, were found to agree well with electrodynamics simulations.


9

ACS Photonics 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 34

Figure 2. Optical properties of bowtie nano-antennas. a) – e) SEM images of Au bowtie nano-antennas of varying length # = 120 − 240 nm but fixed width W = 70 nm and height H = 30 nm (scale bar = 100 nm). f) Representative dark-field scattering spectra for various L (lines – simulated, shaded – experimental, color code as in a-e)) for incident polarization parallel to long axis. The dashed vertical line indicates the heating laser wavelength & = 1064 nm. g) Simulated (FEM) internal heat dissipation density for the L = 180 nm antenna excited at 1064 nm. h) Measured scattering spectra for incident polarization along the short axis (color code as in a-e)). The dashed vertical line indicates the probe laser wavelength & = 633 nm. i) Simulated (FEM) electric field enhancement inside the L = 180 nm bowtie for 633 nm excitation polarized along the short axis. Photothermal heating of plasmonic antenna results from Ohmic losses and can consequently be described by '(() =

/0! 1 Re)J ∗ ⋅ -. = Im)0(/).|-(()|4 , #(4) 2 2


10

Page 11 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Photonics

where '(() (Figure 2g) is the local dissipated heat density, 5∗ is the complex conjugate of the local induced current density, -(() is the local electric field and 0(/) = 0 6 (/) + 80′′(/) is the frequency dependent relative permittivity of the metal. The total heat produced in the antenna, :;?@A ) through: :;

Photothermal Heating of Plasmonic Nano-antennas: Influence on

Recommend Documents