WO2012028936A1 - Enhanced fluorescence of gold nanoparticles - Google Patents

Abstract

The present invention provides a method of increasing fluorescence of metal nanoparticles (NPs), as well as the metal nanoparticles with increased fluorescence themselves. The present invention relates to a system and method for simultaneously performing photothermal microscopy and fluorescence detection of metal, such as gold, nanoparticles (NPs).

Description

ENHANCED FLUORESCENCE OF GOLD NANO PARTICLES

Field of the Invention

The present invention provides a method of increasing fluorescence of metal nanoparticles (NPs), as well as the metal nanoparticles with increased fluorescence themselves. The present invention relates to a system and method for simultaneously performing photothermal microscopy and fluorescence detection of metal, such as gold, nanoparticles (NPs). Background of the Invention

Absorption and fluorescence properties of nanoobjects depend on their chemical structure, composition, dimensions and local environment. These properties determine the use of nanoobjects as labels for optical detection and sensing. Fluorescence microscopy provides a broad range of well-known methods for sensitive fluorescence detection. In absorption microscopy only a few existing methods are capable to directly detect absorption of individual nanoobjects.

Photothermal microscopy is a powerful method allowing for background-free detection of absorption of individual nanoobjects. Photothermal microscopy is a method based on interferometric sensing of optical absorption. A modulated heating laser beam provides energy absorbed by a nanoobject. The absorbed energy which is dissipated into heat gives rise to a temperature change around the nano-absorber. The maximum temperature rise in steady state on the surface of the particle is

where o_ab_S is the absorption cross-section of the NP, l_heat is the intensity (irradiance) of the heating laser, κ is the thermal conductivity of the medium around the NP, R is its radius. A consequent change in refractive index of the media around the nano- absorber scatters a probe laser, which is spatially overlapped with the heating beam. The scattered light interferes in the far-field with a reference wave, in practice the reflected or transmitted probe laser. The probe light is collected by a microscope objective and is recorded by a photodiode and a lock-in-amplifier. Demodulated small changes of the probe light give the photothermal signal. It is proportional to the heating and probe powers, as well as to the temperature derivative of the refractive index of the medium. Recent developments of photothermal correlation spectroscopy and wide-field photothermal detection open new perspectives for photothermal microscopy in a variety of applications. Combined with fluorescence detection it can directly correlate absorption and fluorescence characteristics. A simultaneous insight into absorption and fluorescence delivers understanding of the photophysics of labels, provides an additional detection dimension and expands the observation time and tracking capabilities when labels start to blink. High throughput molecular and cell sorting, tracking and imaging in live cells, as well as biosensing are a few examples of applications where durable labels and multidimensional detection are beneficial.

Most fluorescent labels, such as organic molecules, suffer from blinking and bleaching already at low excitation powers, which limits their use. Semiconductor nanoparticles and nanowires can be photostable alternatives. Still, different excitation regimes are needed for photothermal and fluorescence detection of a single semiconductor nanoparticle. This complicates their simultaneous registration.

Gold NPs of various sizes and shapes are commonly used for the calibration of the signal in photothermal microscopy and for measurements of absorption cross-sections. The synthesis of gold NPs is well established, and their absorption properties measured on a single NP level are in a good agreement with the results predicted by Mie theory. "Naked" gold NPs do not fluoresce to a useful degree. However, gold NPs decorated with fluorescent labels (dye molecules and fluorescent proteins) are frequently used in fluorescence microscopy to enhance the fluorescence of the labels. On the one hand, experimental results show the enhancement of dye fluorescence by large NPs (> 80 nm diameter) or by aggregates of smaller NPs. On the other hand, the photoluminescence of gold NPs itself is of great interest.

Currently, photoluminescence efficiency is reported to decrease when the size of NPs increases. Measured on the ensemble of NPs, an efficiency of 10^"6 has been reported for 4-80 nm (diameter) NPs in water. Even higher photoluminescence efficiencies, >10^"3 have been reported from small (28 atoms), molecular-like gold nanoclusters in ensemble measurements. Recently, a quantum yield of 3% has been measured for polymer stabilized gold clusters of 2.2-3.4 nm diameters where authors have systematically varied the polymer to gold ratios in order to tune the properties of fluorescence emission. The highest quantum yield of ~41 % has been reported for Au₈ NPs stabilized in dendrimer aqueous solution. However there is a desire to improve fluorescent properties of nanoparticles, such as gold.

It is amongst the objects of the present invention to obviate and/or mitigate one or more of the aforementioned disadvantages.

Summary of the Invention The present invention in part concerns observations by the present inventors that gold NPs can be modified to increase their fluorescence properties. It also covers the uses this may be put to.

In a first aspect, the present invention provides a method of increasing fluorescence, such as the fluorescence quantum yield, fluorescence intensity or fluorescence lifetime, of metal NPs, the method comprising the step of treating (e.g. by suitable heating) a sample of metal NPs in an organic solvent.

The size of the NPs may be between 5 nm and 100 nm, such as 10nm -90nm, such as 20nm - 80nm. The metal NPS can be obtained from commercial sources, such as BBInternational (Cardiff, UK). The metal NPs may be "naked" or unmodified, or the metal NPs may be modified so that the NPs are capable of be conjugated to one or more chemical entities, such as a biological molecule. Suitable biological molecules may be proteins, peptides, antibodies, carbohydrates and the like, as well as fragments thereof and may be obtained from BBInternational (Cardiff, UK).

Fluorescence is the emission of light from a molecule in which an electronically excited state has been populated. The emission of the light is usually in the ultraviolet to visible portion of the spectrum, sometimes in the near-infrared. The fluorescence quantum yield (OF) is the ratio of photons absorbed to photons emitted through fluorescence. In other words the quantum yield gives the probability of the excited state being deactivated by fluorescence rather than by another, non-radiative mechanism. Fluorescence lifetime refers to the average time a molecule stays in its excited state before emitting a photon. The present inventions allows for detection of the fluorescence properties of the metal NPs to be carried out at the level of the individual NP and the inventors have shown that it is possible to detect an increase in emission and /or quantum yield of greater than 20 times, such as at least 50 times and up to 100 times following modification. The organic solvent may be polar or non-polar and may typically be an alkane, such as pentane, cyclopentane, hexane, or cyclohexane; an aromatic hydrocarbon, such as benzene, or toluene; a halogenated hydrocarbon, such as chloroform, or dichloromethane; a hydrocarbon ketone, such as ethyl acetate, dimethylformamide and acetone; or an alkanol such as glycerol, isopropanol, ethanol and the like.

Treating of the metal NPs, typically by heatling and/or a photochemical effect in the solvent may be carried out by any suitable means including the use of an incident laser beam. For example, a visible or near-infra red laser with an appropriate power may be used to create a local temperature rise, such as between 40-400K (for example an increase of about 57 K (in glycerol) for 20nm gold NPs, leads to an increase in fluorescence. Moreover, a fluorescence enhancement effect has been observed using a wavelength at 514nm or 532nm at a power of from 0.25mW to 2mW focused into a diffraction limited spot of 250-300nm, on 20nm gold NPs. It is to be appreciated that the method of fluorescence enhancement may depend on the thermal properties of the solvent and may be different for NPs of different sizes.

Without wishing to be bound by theory, it is postulated that heating of the metal NPs in the organic solvent may lead to an increase in fluorescence due to addition of unsaturated groups on the surface of the NPs accompanied or not by a rearrangement of gold atoms at the surface of the particles.

Once the NPs are modified through treating, so as to render them more fluorescent, the present inventors have observed that the NPs retain their fluorescent properties even when removed from the initial solvent and placed in another medium such as an aqueous medium e.g. water. Thus, the "activated" fluorescent NPs of the present invention may be used in systems where the solvent used to modify the NPs would be undesirable. Thus, the method may further comprise removing the solvent, so as to provide the metal NPs of enhanced fluorescence in a substantially pure form and essentially free from the solvent in which they were initially suspended and to which the treatment was applied. Optionally, the substantially solvent free fluorescently enhanced NPs may be resuspended in a medium, such as an aqueous medium (e.g. water).

The present invention also provides fluorescent metal NPs obtainable by the method according to the first aspect. The NPs may be gold NPs.

Particles so produced may find application in biological assays, such as cell based assays, and other applications discussed in the background section. One particular area is in the field of photothermal microscopy which allows background-free detection of absorbers at room temperature and in the absence of residual absorption by the sample'with a sensitivity reaching the single-molecule level. Further developments of this technique, including photothermal correlation spectroscopy, photothermal tracking, wide-field photothermal detection and its combination with fluorescence microscopy makes photothermal microscopy a valuable tool in a large variety of applications.

In a further aspect, there are provided metal NPs in an organic solvent (as defined hereinabove), which may be treated so as to increase the fluorescence, typically the fluorescence quantum yield, of the NPs. There may also be provided a kit comprising metal NPs and a solvent, together with instructions describing how to increase the fluorescence of the NPs by treating them in the solvent provided.

Irradiation at a proper wavelength may be required to achieve the fluorescent enhancement

The present invention also provides use of the NPs prepared in accordance with the method described herein in cell (e.g. mammalian) based assays.

The present invention also provides a method for combining or performing simultaneously photothermal microscopy and fluorescence detection using metal (e.g. gold) NPs, thereby to provide a simultaneous measure of absorption and fluorescence.

Absorption and fluorescence properties of nanoobjects determine the use of nanoobjects as labels for optical detection and sensing. Fluorescence microscopy provides a broad range of well-known methods for sensitive fluorescence detection. Photothermal microscopy is a powerful method for detecting absorption of individual nanoobjects at room temperature. Photothermal microscopy combined with fluorescence detection can directly correlate absorption and fluorescence characteristics of single nanoobjects. Simultaneous detection provides insight into absorption and fluorescence, delivers understanding of the photophysics of labels, provides an additional detection dimension, and expands the observation time and tracking capabilities when labels start to blink. High-throughput molecular and cell sorting, flow cytometry, tracking and imaging in live cells, as well as biosensing are a few examples of applications where durable labels and multidimensional detection are beneficial.

Detailed description of the Drawings

Various aspects of the invention will now be described by way of example only with reference to the accompanying drawings, of which:

Figure 1(a) shows a first system for the combined simultaneous measurement of fluorescence and photothermal effects;

Figure 1(b) shows a second system for the combined simultaneous measurement of fluorescence and photothermal effects;

Figure 2 is a graph of the transmission characteristics of some of the components of the system of Figure 1(b);

Figure 3 shows various fluorescence and photothermal signals;

Figure 4 is a plot of normalised photothermal signal and normalised fluorescence signal for a single 20nm gold nano-particle, as a function of time;

Figure 5 is a plot of the photothermal signal and fluorescence signal as a function of axial position (perpendicular to the sample plane)

Figure 6 is an illustration of normalized absorption and fluorescence spectra of 20 nm diameter gold NPs;

Figure 7 shows simultaneously obtained photothermal and fluorescence images of 20 nm gold NPs;

Figure 8 is a graph showing variation of surface temperature as a function of illumination time and the heating power;

Figure 9 shows SE images of 20 nm gold particles prepared by drop casting of 50 μΐ_ of undiluted NP solution on the surface of an ITO-glass substrate;

Figure 10 shows AFM topography images of 20nm gold NPs prepared by spin coating on a cleaned glass surface; and

Figure 11 is a photothermal microscopy image of 20 nm gold NPs (42 NPs in total) prepared by spin coating on a cleaned glass surface, and includes a histogram showing a unimodal distribution of photothermal signals. Figure 12a) shows a Normalized optical absorption (photothermal) signal from gold NPs of different sizes spin coated on APTES-modified glass surface in glycerol as a function of their size, as measured by AFM. As we could not correlate AFM with photothermal contrast for 10 nm NPs, we used the average AFM diameter in the plot. The grey line represents a linear relationship between the photothermal signal and NP volume; and b) Normalized luminescence signals from gold NPs of different sizes spin coated on APTES-modified glass surface in glycerol as functions of their size (measured by AFM). The grey line represents a linear relationship between the luminescence signal and NPs volume.

Figure 13 shows Combined normalized luminescence and absorption

(photothermal) signals from gold NPs of different sizes spin coated on APTES-modified glass surface in glycerol. Grey lines (here and in B) represent linear relationships between optical signal and NPs volume. (B) Normalized luminescence and absorption (photothermal) signals from gold NPs of 5 and 20 nm diameters spin coated on APTES-modified glass surface in air. Dots represent values for individual NPs. The histograms of signals are shown in the top and right-side axes correspondingly. (C) Luminescence quantum yield (QY) as a function of measured diameters of gold NPs. Estimation of the QY is either based on absorption measurements of size of NPs via photothermal contrast (black dots), or on the size of NPs measured by AFM or specified (grey dots). Error bars in QY represent the spread of values for individual NPs. Error bars for NPs diameters for black and grey dots are defined by the spread of diameters estimated from photothermal microscopy experiments and measured in AFM experiments correspondingly. Brief description of the Drawings

Figures 1(a) and 1(b) shows experimental arrangements for simultaneous photothermal and fluorescence detection¹.

In Figure 1(a) the photothermal heating beam passes through an acousto-optical modulator AOM, which modulates the heating beam, beam expander optics, and is incident on a folding mirror M, which directs the light to a dichroic mirror DM, where the heating beam is reflected towards an iris diaphragm ID. Light passing through the iris diaphragm is reflected from a mirror M1 and passes through an objective lens, which focuses the heating beam on a sample area. The output from the photothermal probe laser passes through beam expander optics to expand the initial beam. The expanded beam passes through a polarizer cube, a quarter wave plate λ/4, and the dichroic mirror DM, where the heating and probe beams are overlapped and sent towards the objective. From there, the probe light follows the same optical path as the heating beam towards the sample. The grey boxes schematically indicate the positions of a telescope and beam expander optics. The telescope and beam expander optics expand the initial beams to approximately 20mm to overfill the entrance pupil of the microscope objective (approximately 10mm). The telescope also serves to adjust the convergence of the pump and probe beams, and thereby the respective positions of their foci, which can compensate for residual chromatic aberrations that may be present in the objective.

In use, the sample is excited with both the heating and probe beams. Probe light backscattered at the sample passes through the objective, hits the mirror M1 and passes through the iris diaphragm ID. At the dichroic mirror DM, the backscattered probe light is transmitted and is directed towards the quarter wave plate λ/4, after which it is incident on the polarizer cube. At this stage, the backscattered sample light is directed into a detection optical path. On this path are interference filters IF to block residual heating light, a photodiode PD, and a video camera CCD.

Fluorescence light is collected by the same microscope objective. It passes the mirror M1 , reflected at the dichroic mirror DM, and is directed back to the beam splitter BS. At the BS 70% of the light send towards spectral separation element, bandpass filters IF, and detected by the avalanche photon counting module (APD). The system is software controlled and the data are collected by a data acquisition card.

Raster scanning of the sample is performed using a 3-axis piezostage (not shown). A spectrograph is used to obtain the fluorescence spectra. Polarizers and spatial filters are not shown in Figure 1(a), but may be included. For experiments, the heating beam was provided by a laser diode, for example a diode that emits at 532 nm (Shanghai Lasers, power 150 mW). The probe beam at 790 nm was provided by a Ti:sapphire laser (Mira, Coherent, pumped with Coherent Verdi V10). The apparatus was based on home-build optical microscope, equipped with an Olympus 60x oil immersion objective with NA = 1.4. The photodiode was a Si photodiode (DHCPA-100-F, Femto). The bandpass filters were AHF 615/150 and Omega 595/100 filters, and the avalanche photon counting module was an SPM-AQR-16 module. The experiments were controlled using dedicated LabView software routines, and data was collected using a data acquisition card (ADWin Gold, Germany). The 3-axis piezostage was either a NanoCube or a MARS II, Physik Instrumente. The spectrograph was a USB4000, Ocean Optics.

An alternative arrangement, based on an inverted optical microscope Olympus 1X71 , equipped with an Olympus 60x oil immersion objective with NA = 1.45 is shown in Figure 1(b). AOM - acousto-optical modulator, M - mirrors, FM - flip mirrors, BS - beam splitter, DM - dichroic mirrors, ID - iris diaphragm, λ/4 waveplate, IF - interference filters, APD - avalanche photodetector (optionally spectrograph), PD - photodiode, CCD - video cameras. Positions of beam expanding telescopes are indicated with dashed boxes. Spatial filters are not shown in the schematic representation. The heating beam was provided by an Ar-lon laser (Coherent Innova 300), and the probe beam at 800 nm was produced by a Tksapphire laser (S3900s, Spectra Physics) pumped with the Ar-lon laser. The dichroic mirror for fluorescence spectral filtering was an AHF z532/NIR mirror. The transmittance of the microscope objective, dichroic mirror and optical filters used in the experimental setup are shown in Figure 2.

The arrangement of Figure 1(b) operates in a similar fashion to that of Figure 1 (a). Probe light backscattered at the sample, and fluorescence light produced by the sample, is collected by the microscope objective and directed back via the various optical components in the optical path to the photodiode or avalanche photon counting module.

The arrangements of Figure 1(a) and Figure 1(b) were used for various experiments, which will now be described. The results illustrated in Figure 5 were obtained using the arrangement of Figure 1(b), whereas most of the results illustrated in Figures 3 and 5 were obtained using the arrangement of Figure 1(a). However similar experiments can be performed using either arrangement.

Commercial 20 nm diameter gold NPs were used for the calibration of the spatial overlap in simultaneous photothermal and fluorescence detection. The fluorescence of these commercial NPs is weak (with the <J>F less than 6.4· 10^"9 for a single NP). We demonstrate a simple and efficient way to enhance it (up to the OF about 10^"6) by illumination at moderately high power in different solvents (glycerol, hexane, pentane). Samples of colloidal suspensions of gold NP's with diameters of 20 nm (British Biocell International, EM.GC20) were prepared by dilution in ultra-pure water at volume ratios of 1 :8. Approximately 50 pL of the suspension were deposited on the surface of cleaned glass immediately after filtration through a 450 nm porous membrane and spin coated at 2000 rpm for 5 s, followed by drying at 4000 rpm for 90 s. Glass coverslides (Menzel, Germany) were cleaned in several steps by sonication for 20 min in: 2 % Hellmanex (Hellma) solution in water, acetone, ethanol (both 96 % purity) and in deionised distilled water after each cleaning step. Experiments were performed in a cell (approx. 50-150 pL volume) made from a rubber o-ring or a top of plastic Eppendorf tube attached to the coverslip. Glycerol (>99.5 %, spectrophotometric grade), pentane and hexane (all AR grade) were used as photothermal transducing fluids for our experiments.

The results of simultaneous photothermal and fluorescence imaging using the arrangement of Figure 1(a) are shown in Figure 3. This shows photothermal (A, D, G) and corresponding fluorescence (B, E, H) raster scan images simultaneously obtained on individual 20 nm diameter gold NPs on glass surface in glycerol. (C, F, I) shows the overlap of the two signals. Relative macro-times at which the images were taken (since the start of the experiment) are indicated next to each row. Vertical lines (along fast scan axis) in fluorescence images originate from diffusing fluorescent impurities in glycerol. There is no detectable fluorescence observed from single colloids in (B), while (E) demonstrates bright fluorescence from the treated central NP (Figure 3A) and others remain non-fluorescent. Finally all three nanoparticles have been made fluorescent, as shown in (H). Experimental parameters for raster scans are P_heanng = 0.26 mW (532 nm) and P_pr0be = 40 mW at the sample, At = 1 ms, AT_surf is about 23 K.

Considering the results shown in Figure 3 in more detail, time-marks next to each row of images indicate the relative time of each experiment. Three single 20 nm diameter gold NPs are observed in Figure 3 A, situated more than 500 nm apart from each other. The photothermal signal-to-noise ratios are higher than 300 with 0.255 mW heating power (532 nm) and 45 mW probe power (800 nm) at the sample, and integration time (Δί) of 1 ms. The corresponding temperature elevation at the NP surface caused by the heating light is about 23 K. The full width at half maximum (FWHM) of the Gaussian fit of the shape of photothermal signals are 220 nm and 260 nm (for vertical and horizontal directions respectively). The overall performance of the photothermal detection is as reported previously. The simultaneously acquired confocal fluorescence image (Figure 2B) shows only the background signal in glycerol and no detectable fluorescence signal from gold NPs. The vertical lines observed in the fluorescence image (along the fast scan axis) are due to diffusing fluorescent impurities in glycerol. The lowest detectable fluorescence quantum efficiency of a 20 nm gold NP in these particular experiments was estimated to be 6.4x 10^"9. Calculations are based on the background counts (2.2 kcounts/10 ms) and detection efficiency estimations of 5 %. The overlap of the simultaneously detected photothermal (A) and fluorescence (B) signal is shown in (C) with the red and green colour representing the photothermal and fluorescence signal, respectively.

The laser beams were then focused on the central 20 nm gold NP shown in Figure 3 A. Figure 4 shows a time trace for the single 20 nm gold NP. This illustrates the appearance of the fluorescence signal from the single NP. The graph shows simultaneously recorded photothermal (red) and fluorescence (green) time-traces normalized to the heating power. An increase of the heating power (top, black) at the first axis break (98 s) leads to the temperature rise at the NP. A significant change in fluorescence signal and its fluctuations are observed at the same time (green). At the second axis break (182 s) the heating power is reduced to its starting value. The normalized fluorescence signal remains above its background level at the start (horizontal dashed green line). Due to the mechanical drift in the setup (about 20 nm/min) the photothermal signal (red) is about 20% smaller after 230 s.

The top part of the graph of Figure 4 illustrates how the heating power is varied in the experiment. The temperature rise on the surface of the nanoparticles (AT_surf) and the photothermal signal are linearly dependent on the heating power. At approximately 98 s, the heating power has been increased and, with it, the temperature of the NP. The increase of the heating power (514 nm) from 0.26 mW up to 0.85 mW leads to a temperature rise from 17 K up to 57 K on the surface of the NP. At the same time an increase of the fluorescence signal is observed, as well as fluorescence fluctuations. The fluorescence signal remains after the heating power is reduced back to 0.26 mW. The sudden enhancement of fluorescence described in the preceding paragraph seems to be caused by heating induced by the green light as discussed in more detail below. A similar effect is observed when a comparable temperature rise at the surface of NPs is achieved due to the absorption of the probe light only (the power is larger than 150 mW at the sample). As a result after such an exposure to the probe light, gold NPs show same intense fluorescence when illuminated with heating light (514 nm, 0.26 mW). No enhancement is seen with probe light alone for powers below 100 mW at the sample, also for illumination times longer than 100 s. Consequent simultaneous photothermal and fluorescence raster scans reveal no change in the photothermal picture (Figure 3D).

The unaltered absorption of light suggests that no defragmentation of the particle has occurred. A defragmentation of similar size NPs has previously been reported when intense pulsed laser light has been applied and has caused temperature rises above 1000 K. The absence of defragmentation of NPs is supported in independent experiments by SEM imaging after the studies of the vibrations of individual gold NPs, for temperature elevations of about 300 K. In contrast to the unchanged absorption, the confocal fluorescence image (Figure 3 E) shows a clear change. A fluorescent spot with a signal-to-background ratio of 1.4 and a full width at half-maximum (FWHM) of 300 and 340 nm (for vertical and horizontal directions correspondingly) appears. The spatial overlap of the two images is shown in Figure 3F with the yellow colour encoding the overlap of photothermal and fluorescence signals. The same effect is observed on other NPs in the sample (about 50 NPs tried) and further demonstrated in (G, H, I), where the other two NPs in the image are made fluorescent in the exact same manner as the middle one. A single gold NP with the calculated absorption cross-section of 630 nm² in glycerol would absorb 1.5*10¹² photons/s with a heating power of 0.26 mW. Thus, the quantum yield of fluorescence is calculated to be about 1.410.5X10^"6, assuming one-photon excitation with the heating beam and with our estimated detection efficiency. The obtained fluorescence persists for a long time, although its intensity fluctuates. In the particular experiment presented in Figure 4 it was also detected after 148 min. The observed stability of the fluorescence signal provides enough time for the alignment of fluorescence detection, and we demonstrate a good spatial overlap between photothermal and fluorescence signals in the scanning plane (lateral overlap). The brightness of NPs is about 0.9 counts/(J/cm²), which is an order of magnitude less than the brightness of closest in size (diameter of 3.6 nm) fluorescence labels, CdSe/ZnS quantum dots (QD) with fluorescence quantum yield of 40 %. However, QD show fluorescence intermittence with 1 ms - 10 s characteristic times, have fluorescence saturation intensities of 10-80 kW/ cm², and photobleach.

To probe the spatial overlap in the axial direction, we perform a one dimensional scan on a NP in the direction perpendicular to the sample plane and simultaneously record photothermal and fluorescence signals from the NP. Figure 5 shows the results of this for a Z-scan (perpendicular to the glass surface) on the fluorescent 20 nm gold particle. A red solid line shows a Gaussian fit to data points of photothermal signal and a green dashed line shows the fit for the fluorescence. Central positions and the width of fits are 1.516±0.624 pm and 1.764±0.779 pm for photothermal and for fluorescence, respectively. A shift (250 nm) between the maxima of two signals is observed. A single fluorescence maximum and a single photothermal maximum are observed in contrast with results reported on photothermal measurements of semiconductor wires (optical absorption). We take this single maximum of the photothermal signal as a convenient criterion for the proper relative alignment of the heating and probe foci. The FWHM of the fluorescence signal along z-axis is 780 nm, which is slightly larger than the FWHM of the photothermal signal (620 nm). These values are in a good agreement with expected diffraction-limited values. The position of the maximum of the fluorescence signal (focused position of the heating beam) and the position of the maximal photothermal signal (the overlap of the heating and probe) are found to be shifted by 250 nm. The shift is well explained by difference in the spatial modes between the incident and scattered fields, which theoretically predicts a discrepancy in z-axis between fluorescence and photothermal signals. This result demonstrates that the model of Hwang and Moerner (Optical Comm., 2007, 280, 487) for paraxial rays may also hold for large numerical apertures at the focus of an immersion objective.

In order to understand the origin of the fluorescence of NPs, a fluorescence spectrum was recorded for a single NP in the spectral region between 560 nm and 640 nm. From the featureless shape of the fluorescence spectrum in the limited spectral range (due to the limitations of the present setup) it is difficult to judge if the spectrum resembles the absorption of gold NP. Thus, we cannot give any solid conclusion as to whether the fluorescence is due to NP's plasmon emission only, or whether other mechanisms are involved.

The experiments described above demonstrate that the illumination by the laser light causes the surface modification of gold NPs and leads to their fluorescence. The fluorescence enhancement could be explained by the modification of the surface of NPs, caused by thermal processes, photochemical processes or a combination of such processes. The fluorescence enhancement can be assigned to a modification of the surface of NPs due to a transient temperature elevation at the surface of NPs upon illumination with laser light. Heating may lead to fluorescence due to a modification of the surface electronic properties of gold caused by the removal or addition of ligands and the rearrangement of gold atoms at the surface. A temperature-mediated mechanism is supported by the observation of enhancement of fluorescence after exposure to only the probe light with a power larger than 150mW at the sample. In contrast, no enhancement was seen with heating light power kept as low as 0.26mW for times longer than 200s, or with the probe light only with powers below 100mW for times longer than 100s. The possibility of this mechanism is supported by the appearance of fluorescence also in pentane and hexane, organic solvents that are chemically different from glycerol. This fluorescence enhancement effect is not observed in water, although fluorescent gold NPs prepared in glycerol remain fluorescent when glycerol is exchanged with water. The simplest activated chemical reactions in the above-mentioned solvents would first lead to saturated and therefore non-fluorescent products. A high enough local temperature elevation at the surface of NPs, however, can lead to the formation of more and more complex unsaturated products, which may decorate the NP and generate fluorescence.

An additional factor is the possible catalytic role of the gold surface, which may act in more specific ways than just a local "frying pan". Possible surface chemistry effects of NPs in glycerol include the following: (i) thermal decomposition of glycerol or residual fatty acids in glycerol; (ii) formation of a reactive acrylic acid aldehyde (acrolein, H2C=CH-CHO) obtained in thermal decomposition of glycerol by dehydration in the presence of a catalyst. Reversible and irreversible temperature mediated reactions of glycerol and citric acid in the temperature range 20°C-220°C may produce irreversible esterification and glycerol citrate polyesters formation. Also, a selective etherification of glycerol to linear, branched and cyclic diglycerols, and triglycerol in the presence of alkaline earth metal oxides may occur.

Further results are now described in relation to Figures 6 to 1 1.

Figure 6 shows normalized absorption and fluorescence spectra of 20 nm diameter gold NPs. Solid black line: Calculated absorption spectrum in glycerol according to Mie theory. Dashed black line: Measured ensemble absorption spectrum in a glycerohwater mixture (3:1 ). The broader shape of the spectra arises from the distribution of NP sizes and shapes. Solid green line: Fluorescence spectrum from a single NP made fluorescent after moderate heating on the surface of glass in glycerol. The steep decrease of the fluorescence spectrum in the shortwavelength region is due to the optical transmission of dichroic mirrors and interference filters (Figure 2). Dashed green line: Fluorescence spectrum measured on a solution of gold NPs in glycerohwater mixture (3:1 ). This spectrum did not change after illumination with 514 nm laser light for >2 hours with the laser intensity >100 W/cm². The vertical solid green line indicates the fluorescence excitation at 514 nm.

Figure 7 shows simultaneously obtained photothermal (left, in a.u.) and fluorescence (right, in counts/10ms) images of 20 nm gold NPs. Top row: Images in glycerol.

Particles were made fluorescent according to the procedure described in the main text.

The fluorescence quantum yield is about 7.6x10^"7. Bottom row: The same particles are imaged in water. As expected, the photothermal signal decreases in water [Gaiduk, A.;

Ruijgrok, P.V.; Yorulmaz, M.; Orrit, M. Chem. Science 2010, 1 , 343]. The fluorescence signal is detectable after solvent exchange. The fluorescence quantum yield is about

2.6-10^"7.

Figure 8 illustrates the effect of the appearance of fluorescence of 20 nm gold NPs, as a function of illumination time and the heating power. The right axis shows the corresponding temperature elevation. No detectable appearance of fluorescence is observed at 0.26 mW heating power {AT_suff = 17 K) for > 200 s illumination time. A spread of times is found for heating powers of 0.5 mW and 0.85 mW (AT_surf = 33 K and 57 K. Figure 9 comprises SEM images of 20 nm gold particles prepared by drop casting of 50 μΙ_ of undiluted NP solution on the surface of an ITO-glass substrate. The images (labelled A, B and C) shown in Figure 9 are taken at different magnifications. Figure 9 A shows a large area of the sample and illustrates different aggregates of particles in this area. Figure 9 B is the magnified image of a large NP cluster from Figure 9 A. Figure 9 C shows the image resulting from a zoom into an area of Figure 9 A that contains smaller clusters and also where individual NPs can be resolved. The SEM images show that the drop casting preparation method described in this paragraph does not result in a homogeneous sample containing individual NPs separated from each other at distances required to resolve them using optical microscopy.

Figure 10 comprises AFM topography images of 20 nm gold NPs prepared by spin coating on a cleaned glass surface, the same method used to obtain the samples that were subject to the measurements whose results are illustrated in Figures 3 to 8. The images (labelled A, B and C) shown in Figure 10 are taken at different magnifications. The images of Figure 10 A (of area 50x50 prn²), Figure 10 B (of area 10x10 pm²) and Figure 10 C (of area 2x2 pm²) show no aggregates. Most of the gold NPs are separated from each other at distances greater than 300 nm, a large enough separation to resolve single particles with optical imaging.

Figure 11 shows a photothermal microscopy image of 20 nm gold NPs (42 NPs in total) prepared by spin coating on a cleaned glass surface, the same method used to obtain the samples that were subject to the measurements whose results are illustrated in Figures 3 to 8. The image shows homogeneously distributed NPs with no aggregates. Most of the individual gold NPs are separated from each other at distances greater than 300 nm, large enough to be resolved with optical imaging. The histogram included in Figure 11 shows a unimodal distribution of photothermal signals, well offset from the background, with a mean value of 2.53±0.53 (a.u.). The variation of photothermal signal of about 21 % is in a good agreement with expected variation value of 24 % (based on manufacturer specifications of the diameters of the NPs of 19.9 nm with less than 8 % variation coefficient).

Additional examples concerning metal nanoparticles of varying sizes, are described below: Absorption (photothermal, PTA) microscopy

The absorption of individual NPs is measured by photothermal contrast. Photothermal detection is described in detail elsewhere (D. Boyer, P. Tamarat, A. Maali, B. Lounis, M. Orrit, Photothermal Imaging of Nanometer-Sized Metal Particles Among Scatterers, Science, 2002, 297, 1160) and the experimental setup for simultaneous absorption and fluorescence microscopy was home-built around an inverted optical microscope, Olympus 1X71 , equipped with the Olympus 60x oil immersion objective (NA = 1.45) as reported previously (A. Gaiduk, P. V. Ruijgrok, M. Yorulmaz, M. Orrit, Making gold nanoparticles fluorescent for simultaneous absorption and fluorescence detection on the single particle level, Phys. Chem. Chem. Phys., 2011 , 13, 149).

In photothermal microscopy the modulated heating laser beam causes absorption of a variable amount of the energy by the nano-absorber under study. The heating beam at 514 nm is provided by an Ar-lon laser (Coherent Innova 300) and passes an acousto-optical modulator which modulates the heating light at a frequency of 740 kHz. The dissipated energy gives rise to a time-dependent temperature gradient around the nano-absorber. The maximum temperature rise in steady state on the surface of the particle is AT_sur/ = _{ab heat} Ι{4πκΚ) , where a_abs is the absorption cross- section of the NP, l_hea, is the intensity (irradiance) of the heating laser, κ is the thermal conductivity of the medium around the NP, and R is its radius. The probe beam is scattered by the subsequent change in refractive index of the medium around the nano-absorber. The probe and heating beams are spatially overlapped at a dichroic mirror and both sent through the same objective to the sample. The probe beam (at 800 nm) is produced by a Ti:sapphire laser (S3900s, Spectra Physics) pumped with the Ar-ion laser. Sets of spatial filters and telescopes expand the initial beams to - 20 mm to overfill the entrance pupil of the microscope objective (- 10 mm). The telescopes also serve to adjust the convergence of the heating and probe beams, and the respective positions of their foci, compensating for residual chromatic aberration in the objective.

The interference of the probe scattered light with a probe reference (reflected or transmitted) wave is detected in the far-field. The probe light is collected by the microscope objective in the backward configuration. It is recorded by a Si photodiode (DHPCA-100-F, Femto) and the photothermal signal is demodulated by a lock-in- amplifier (SR830, Stanford Research). The photothermal signal is proportional to the absorption cross section of the nano-absorber, as well as to the heating and probe powers and to the temperature derivative of the refractive index of the medium.

Fluorescence microscopy

The fluorescence detection is performed in the backward direction by a spectral separation of the fluorescence signal at the dichroic mirror (AHF z532/NIR). The fluorescence signal is additionally spectrally filtered by a set of bandpass filters

(AHF 615/150 and Omega 595/100) and detected by an avalanche photon-counting module (SPCM-AQR-16).

The experiment is controlled with home-written LabVIEW software, and the data are collected by an acquisition card (ADwin Gold, Germany). The raster scanning of the samples is performed with a 3-axis piezo-stage (MARS II, Physik Instrumente).

The overlap of the absorption (photothermal) and fluorescence signals is achieved by enhancing the fluorescence of a few selected individual gold NPs, as reported previously (A. Gaiduk, P. V. Ruijgrok, M. Yorulmaz, M. Orrit, Making gold nanoparticles fluorescent for simultaneous absorption and fluorescence detection on the single particle level, Phys. Chem. Chem. Phys., 2011 , 13, 149). These highly fluorescent NPs are then used as fiducial markers in a reference sample used for alignment of simultaneous photothermal and photoluminescence detections.

Atomic force microscopy

We use an atomic force microscope (Digital Instruments) in tapping-mode equipped with cantilevers with a resonance frequency of 280 kHz. We record AFM images in air to characterize the topography of our samples.

Sample preparation

Samples of colloidal suspensions of gold NP's with diameters of 80, 60, 50, 40, 30, 20, 10 and 5 nm (British Biocell International) were prepared by dilution in ultra-pure water at volume ratios of 1 :1 , 1 :4, 1 :4, 1 :8, 1 :20 1 :20, 1 :150 and 1 :1000, respectively. Approximately 50 μΙ_ of the suspension were deposited on the surface of cleaned glass immediately after filtration through a 450 nm porous membrane (except for the 80 nm diameter NPs) and spin coated at 2000 rpm for 5 s, followed by drying at 4000 rpm for 90 s.

Glass coverslides (Menzel, Germany) were cleaned in several steps by sonication for 20 min, successively in a 2% Hellmanex (Hellma) solution in water, in acetone, in ethanol (both 96% purity), and in Milli-Q water after each cleaning step. The surfaces were modified with 3-aminopropylthriethylethoxysilane (APTES, Sigma- Aldrich) by immersing clean coverslides in 2% APTES solution in acetone overnight. Experiments were performed in a fluid cell (approx. 50-150 μΙ_ volume) made from a rubber o-ring attached to the coverslip. We used glycerol (> 99.5%, spectrophotometry grade) as the photothermal transducing fluid in all experiments. The background fluorescence of glycerol was too high to detect the luminescence of the smaller gold beads (5 nm). Their fluorescence and photothermal signals were measured in air and calibrated with those of larger NPs (20 nm) in the same conditions.

2. Results and Discussion

2.1. AFM characterization of gold NPs

The results of our AFM studies of commercial gold NPs of various sizes are shown in Table 1. The mean values of measured heights are in good agreement with the specified diameters. The spreads of heights for 20, 40, 50, 60, and 80 nm diameter individual NPs are about 11% (rms), which is close to 8-10% specification values. However, larger variations are observed for NPs with nominal diameters of 5, 10, and 30 nm. In particular, for the NPs with specified diameter 30 nm, a distinct population of smaller NPs leads to a 19% variation of the measured height. The difference between the actual mean values and the nominal diameters can be explained by the non- spherical shapes of the NPs. Because our AFM images were recorded at a relatively low resolution over large sample areas, in order to correlate them with optical images, they did not allow us to retrieve the detailed non-spherical shape of NPs in the present experiments.

2.2. Optical microscopy on gold NPs

2.2.1 Correlation of AFM and optical microscopy results on gold NPs

To correlate the AFM images with those obtained in optical microscopy, we used a cross as a position reference visible in the two types of microscopy techniques. This cross was made by scratching the glass coverslip's surface prior to cleaning and to surface modification. It is straightforward to find patterns of perfect overlap, enabling unambiguous identification of the topographical signals of individual nanoparticles (white dots in the AFM image) and of their optical absorption signals (red dots). The overlap of the two images is not perfect over the whole scanned area, however, due to the hysteresis and to the thermal drift of the AFM scanner during the data acquisition time (about 30 min). 2.2.2 Correlation of absorption and luminescence of individual gold NPs

We have simultaneously measured absorption (photothermal) and luminescence signals of individual gold NPs (diameters 10, 20, 40, 60 and 80 nm) on an APTES-modified glass surface in glycerol. The AFM and photothermal absorption results are presented in Fig. 12 as a scatter plot, where particles of different nominal sizes are represented by dots of different colors. As Fig. 12 shows, the photothermal absorption signal scales linearly with the volume of NPs deduced from the AFM height. This linear dependence was expected and reported previously (D. Boyer, P. Tamarat, A. Maali, B. Lounis, M. Orrit, Photothermal Imaging of Nanometer-Sized Metal Particles Among Scatterers, Science, 2002, 297, 1 160). The linear correlation is well obeyed for particles smaller than 40 nm. We also note a larger spread of photothermal signals for 80 nm NPs. We attribute this spread to a distribution of the local heat transfer parameters (heat conductivities, contact areas) around each individual particle. Figure 13 presents a scatter plot of luminescence intensity correlated with absorption signal. Particles of different nominal sizes are represented by dots of different colors. The powers of the heating and probe lasers were adjusted in experiments to keep a comparable temperature rise, less than 20 K, at the surface of NPs of different diameters. For all particle sizes, the photothermal signal-to-noise ratio was more than 10 with 3 ms integration time. For luminescence, the signal-to- background ratio was larger than 2 and the signal-to-noise ratio higher than 10. The luminescence signal of smaller NPs (5 nm diameter) was too weak to measure against the fluorescence background of glycerol. Therefore, these particles were measured in air (Fig. 13B), where their luminescence was detected with a signal-to-background ratio of 1 and a signal-to noise ratio of 5 to 10. The photothermal signal-to-noise ratio of 5 nm diameter NPs in air was more than 10 with 100 ms integration time. The collection efficiencies for the luminescence and the photothermal properties differ in air and glycerol. Therefore, we repeated these optical microscopy experiments for a number of 20 nm diameter NPs in air as a reference to calibrate the luminescence and photothermal signal of 5 nm NPs. For these 20 nm NPs, the luminescence signal-to- background ratio was 5, the signal-to noise ratio was larger than 10, and the photothermal signal-to-noise ratio larger than 30 with 3 ms integration time.

As shown in Fig. 13A, the luminescence roughly scales linearly with photothermal signal, as reported previously by Dulkeith (E. Dulkeith, T. Niedereichholz, T.A. Klar, J. Feldmann, G. von Plessen, D.I. Gittins, K.S. Mayya, F. Caruso, Plasmon emission in photoexcited gold nanoparticles, Phys. Rev. B, 2004, 70, 205424).

. This linear relation is well followed from 5 nm to 20 nm. However, the luminescence signal appears somewhat larger for larger diameters and tends to decrease for very large particles, 80 nm and above. The scaling of the luminescence signal with the volume of the particles instead of their surface indicates that this luminescence does not arise from surface states or from impurities at the NP's surface and supports the mechanism of particle plasmon emission proposed by Dulkeith et al. E. Dulkeith, T. Niedereichholz, T.A. Klar, J. Feldmann, G. von Plessen, D.I. Gittins, K.S. Mayya, F. Caruso, Plasmon emission in photoexcited gold nanoparticles, Phys. Rev. B, 2004, 70, 205424). Thus, this luminescence is different from the illumination- induced fluorescence (A. Gaiduk, P. V. Ruijgrok, M. Yorulmaz, M. Orrit, Making gold nanoparticles fluorescent for simultaneous absorption and fluorescence detection on the single particle level, Phys. Chem. Chem. Phys., 2011 , 13, 149), which was attributed to impurities produced at the particle surface by (photo-) chemical reactions. Furthermore, the good agreement of measurements for 5 nm and 20 nm particles in glycerol and in air is also consistent with a luminescence mechanism involving volume instead of surface states. The signals of 80 nm diameter NPs (Fig. 13A) appear to deviate from the general linear trend of the absorption-luminescence correlation. The luminescence intensity seems to reach a saturation value.

The luminescence quantum yield (QY) from individual NPs is calculated taking into account the excitation (heating) laser power, luminescence signal, the absorption cross section of a NP, and the detection efficiency, estimated to 5% for our setup. A more accurate measurement of our collection efficiency could lead to small changes of the quantum yield. The absorption cross section in turn is estimated in two ways based on: (i) the absorption (photothermal, PTA) signal of NPs which is proportional to the volume in the approximation of Mie scattering, or (ii) the height of NPs measured in AFM experiments, which gives the diameter of NPs, assuming a perfect spherical shape. The particles in the optical scans which were out of the AFM scans were also included in the plots, but were assigned the average height measured by AFM. The results of the QY estimation are shown in Fig.13C and, in Table 1. The luminescence QY is found to be almost independent on the size of NPs and estimated to be a few 10^{' 7} Note that the quantum yields deduced from photothermal values and from AFM values are slightly different for particles larger than 40 nm. Table 1. The results of the AFM and optical microscopy measurements on commercial gold NPs spin-coated on APTES-modified glass surface. The height of NPs is measured in the tapping-mode in air. Absorption (photothermal) and fluorescence microscopy experiments are performed in glycerol, unless specified.

CONCLUSION

In this work, we have correlated three observation methods (AFM topography, optical photothermal absorption and optical photoluminescence) on individual gold nanoparticles of different sizes, from 5 to 80 nm. The AFM height signal gives a very good indication of the particle's volume, as deduced from absorption properties, at least for large enough particles. Photothermal detection is much easier than AFM for small particles (5 nm and lower), particularly when those have to be found over large sample areas. However, we found that the photothermal signal can be sensitive to the local conditions for heat transport around individual particles, especially for large particles (80 nm diameter).

Our measurements confirm that the photoluminescence signal scales roughly as the volume of the NPs, and that the luminescence yield is therefore roughly independent of size, from 5 to 60 nm. The value of the yield is about 3* 10^~7 (subject to a possible small calibration error), in good agreement with previous determinations. We also find that the yield appears to decrease for even larger particles, larger than 80 nm in diameter.

Independently of AFM measurements, the combined optical absorption and fluorescence microscopy of single NPs is a convenient new method to study luminescence of gold particles with various shapes and compare to theory. Gold - nanorods, -platelets, -stars, -cubes, -pyramids and various organic or hybrid NPs are obvious candidates for future investigations.

REFERENCES

1 X. Huang, S. Neretina, M.A. El-Sayed, Gold Nanorods: from Synthesis and Properties to Biological and Biomedical Applications, Adv.Mater., 2009, 21, 4880-

4910.

2 T.K. Sau, A.L. Rogach, F. Jackel, T.A. Klar, J. Feldmann, Properties and Applications of Nonspherical Colloidal Noble Metal Nanoparticles, Adv.Mater., 2009, 21, 1805-1825.

3 C.J. Murphy, T.K. Sau, A.M. Gole, C.J. Orendorff, J. Gao, L. Gou, S.E. Hunyadi, T.Li, Anisotropic Metal Nanoparticles: Synthesis, Assembly, and Optical Applications, J.Phys.Chem.B, 2005, 109, 13857-13870.

4 Z. Wang, J. Zhang, J.M. Ekman, P.J. Kenis, Y. Lu, DNA-Mediated Control of Metal Nanoparticle Shape: One-Pot Synthesis and Cellular Uptake of Highly Stable and Functional Gold Nanoflowers, Nano Lett., 2010, 10, 1886-1891.

5 X. Hong, E.M.P.H. van Dijk, S.R. Hall, J.B. Gotte, N.F. van Hulst, H. Gersen, Background-Free Detection of Single 5 nm Nanoparticles through Interferometric Cross-Polarization Microscopy, Nano Lett. 2011 , 11, 541-547.

6 S. Chong, W. Min, X.S. Xie, Ground-State Depletion Microscopy: Detection Sensitivity of Single-Molecule Optical Absorption at Room Temperature,

J.Phys.Chem.Lett., 2010, 1, 2216-3322.

7 L. Xiao, Y. Qiao, Y. He, E.S. Yeung, Wavelength-Dependent Differential Interference Contrast Microscopy: Selectively Imaging Nanoparticle Probes in Live Cells, Anal. Chem., 2010, 82, 5268-5274.

8 A. Tcherniak, J.W. Ha, S. Dominguez-Medina, L.S. Slaughter, S. Link, Probing a Century Old Prediction One Plasmonic Particle at a Time, Nano Lett., 2010, 10, 1398.

9 W.S. Chang, J.W. Ha, L.S. Slaughter, S. Link, Plasmonic Nanorod Absorbers as Orientation Sensors, Proc. Natl. Acad. Sci. USA, 2010, 107, 2781.

10 A.L. Tchebotareva, M.A. van Dijk, P.V. Ruijgrok, V. Fokkema, M.H.S. Hesselberth, M. Lippitz, M. Orrit, Acoustic and Optical Modes of Single Dumbbells of Gold Nanoparticles, Chem. Phys. Chem., 2009, 10, 1 11.

1 1 T. Zuchner, A.V. Failla, A. Hartshuh, A.J. Meixner, A novel approach to detect and characterize the scattering patterns of single Au nanoparticles using confocal microscopy, J. Microscopy, 2008, 229, 337-343. E. Absil, G. Tessier, D. Fournier, M. Gross, M. Atlan, Full field imaging and spectroscopy of individual gold nanoparticles, Eur. Phys. J. Appl. Phys., 2008, 43, 155-158.

F.V. Ignatovich, L. Novotny, Real-time and background-free detection of nanoscale particles, Phys. Rev. Lett, 2006, 96, 013901.

A.V. Failla, H. Qian, H.H. Qian, A. Hartschuh, A.J. Meixner, Orientational imaging of subwavelength au particles with higher order laser modes, Nano. Lett., 2006, 6,

1374-1378.

M.A. van Dijk, A.L. Tchebotareva, M. Orrit, M. Lippitz, S. Berciaud, D. Lasne, L.

Cognet, B. Lounis, Absorption and scattering microscopy of single metal nanoparticles, Phys. Chem. Chem. Phys., 2006, 8, 3486-3495.

V. Jacobsen, P. Stoller, C. Brunner, V. Vogel, V. Sandoghdar, Interferometric optical detection and tracking of very small gold nanoparticles at a water-glass interface,

Opt. Express., 2005, 14, 405-414.

M.A. van Dijk, M. Lippitz, M. Orrit, Far-field optical microscopy pf single metal nanoparticles, >Acc. Chem. Res., 2005, 38, 594-601.

D. Boyer, P. Tamarat, A. Maali, B. Lounis, M. Orrit, Photothermal Imaging of

Nanometer-Sized Metal Particles Among Scatterers, Science, 2002, 297, 1 160. E. Dulkeith, T. Niedereichholz, T.A. Klar, J. Feldmann, G. von Plessen, D.I. Gittins,

K.S. Mayya, F. Caruso, Plasmon emission in photoexcited gold nanoparticles, Phys.

Rev. B, 2004, 70, 205424.

O.P. Varnavski, T. Goodson III, M.B. Mohamed, M.A. El-Sayed, Femtosecond excitation dynamics in gold nanospheres and nanorods, Phys. Rev. E, 2005, 72,

235405.

M.R. Beversluis, A. Bouhelier, L. Novotny, Continuum generation from single gold nanostructures through near-field mediated intraband transitions, Phys. Rev. B, 2003, 68, 1 15433.

Z. Wu, R. Jin, On the Ligand's Role in the Fluorescence of Gold Nanoclusters, Nano Lett, 2010, 10, 2568-2573.

C. Zhou, C. Sun, M. Yu, Y. Qin, J. Wang, M. Kim, J. Zheng, Luminescent Gold Nanoparticles with Mixed Valence States Generated from Dissociation of Polymeric Au (I) Thiolates, J. Phys. Chem. C, 2010, 114, 7727-7732.

N. Schaeffer, B. Tan, C. Dickinson, M.J. Rosseinsky, A. Laromaine, D.W. McComb, M.M. Stevens, Y. Wang, L. Petit, C. Barentin, D.G. Spiller, A.I. Cooper, R. Levy, Fluorescent or not? Size-dependent fluorescence switching for polymer-stabilized gold clusters in the 1.1-1.7 nm size range, Chem. Commun., 2008, 34, 3986.

J. P. Wilcoxon, J.E. Martin, F. Parsapour, B. Wiedenman, D.F. Kelley, Photoluminescence from nanosize gold clusters, J. Chem. Phys., 1998, 108, 9137. A. Gaiduk, P. V. Ruijgrok, M. Yorulmaz, M. Orrit, Making gold nanoparticles fluorescent for simultaneous absorption and fluorescence detection on the single particle level, Phys. Chem. Chem. Phys., 2011 , 13, 149.

L. Cognet, S. Berciaud, D. Lasne, B. Lounis, Photothermal Methods for Single Nonluminescent Nano-Objects, Anal. Chem., 2008, 80, 2289.

S. Berciaud, D. Lasne, G.A. Blab, L. Cognet, B. Lounis, Photothermal heterodyne imaging of individual metallic nanoparticles: Theory versus experiment, Phys. Rev. B, 2006, 73, 045424.

S. Berciaud, L. Cognet, P. Tamarat, B. Lounis, Observation of intrinsic size effects in the optical response of individual gold nanoparticles, Nano Lett., 2005, 5, 515-518. S. Berciaud, L. Cognet, G.A. Blab, B. Lounis, Photothermal heterodyne imaging of individual nonfluorescent nanoclusters and nanocrystals, Phys.Rev. Lett., 2004, 93, 257402.

M. Tokeshi, M. Uchida, A. Hibara, T. Sawada, T. Kitamori, Determination of Subyoctomole Amounts of Nonfluorescent Molecules Using a Thermal Lens Microscope: Subsingle-Molecule Determination, Anal. Chem., 2001 , 73, 21 12.

S.E. Bialkowski, Photothermal Spectroscopy Methods for Chemical Analysis, Wiley, 1996.

A. Gaiduk, P.V. Ruijgrok, M. Yorulmaz, M. Orrit, Detection limits in photothermal microscopy, Chem. Sci., 2010, 1, 343.

A. Gaiduk, M. Yorulmaz, P.V. Ruijgrok, M. Orrit, Room-Temperature Detection of a Single Molecule's Absorption by Photothermal Contrast, Science, 2010, 330, 353.P.M.R. Paulo, A. Gaiduk, F. Kulzer, S.F.G Krens, H.P. Spaink, T. Schmidt, M. Orrit, Photothermal Correlation Spectroscopy of Gold Nanoparticles in Solution, J. Phys. Chem. C, 2009, 113, 1 1451.

R. Radunz, D. Rings, K. Kroy, F. Cichos, Hot Brownian Particles and Photothermal Correlation Spectroscopy, J. Phys. Chem. A, 2009, 113, 1674.

V. Octeau, L. Cognet, L. Duchesne, D. Lasne, N. Schaeffer, D.G. Femig, B. Lounis, Photothermal Absorption Correlation Spectroscopy, ACS Nano, 2009, 3, 345. 38 D. Lasne, G.A. Blab, S. Berciaud, M. Heine, L. Groc, D. Choquet, L. Cognet, B. Lounis, Single Nanoparticle Photothermal Tracking (SNaPT) of 5-nm Gold Beads in Live Cells, Biophys. J., 2006, 91, 4598.

39 E. Absil, G. Tessier, M. Gross, M. Atlan, N. Warnasooriya, S. Suck, M. Coppey- Moisan, D. Fournier, Photothermal heterodyne holography of gold nanoparticles, Opt. Express, 2010, 18, 780.

40 X. Wu, T. Ming.X. Wang, P. Wang, J. Wang, J. Chen, High-photoluminescence-yield gold nanocubes: for cell imaging and photothermal therapy, ACS Nano, 2010, 4, 1 13-120.

41 See Electronic Supporting Information (ESI) available free of charge.

A skilled person will appreciate that variations of the disclosed arrangements are possible without departing from the scope of the invention. Accordingly, the above descriptions of specific embodiments are made by way of example only and not for the purposes of limitation. It will be clear to the skilled person that minor modifications may be made without significant changes to the operation and features described.

Claims

Claims

1. A method of increasing fluorescence, such as the fluorescence quantum yield, fluorescence intensity or fluorescence lifetime, of metal nanoparticles (NPs), the method comprising the step of treating a sample of metal NPs in an organic solvent.

2. The method of claim 1 wherein the treating step is by applying heat to the metal NPs.

3. The method according to claims 1 or 2, wherein the NPs are gold NPs.

4. The method according to either of claims 1 or 2 wherein the NPs may be between 5 nm and 100 nm, such as 10nm -79nm in diameter.

5. The method according to any preceding claim wherein the organic solvent is polar or non-polar and may be an alkane, such as pentane, cyclopentane, hexane, or cyclohexane; an aromatic hydrocarbon, such as benzene, or toluene; a halogenated hydrocarbon, such as chloroform, or dichloromethane; a hydrocarbon ketone, such as ethyl acetate, dimethylformamide and acetone; or an alkanol such as glycerol, isopropanol, ethanol and the like.

6. The method according to any preceding claim wherein the treatment is carried out by carried out by use of an incident laser beam.

7. The method according to any preceding claim further comprising isolating the NPs from the organic solvent.

8. The method according to claim 8 further comprising re-suspending the isolated NPs in an aqueous solution or in a different organic solution.

9. The method according to any preceding claim, further comprising detecting, at the level of an individual metal NP, the metal NP displaying an increased fluorescence.

10. Metal NPs obtainable by the method according to any preceding claim.

11. Use of the NPs according to claim 10, or prepared in accordance with any one of claims 1-9 in cell (e.g. mammalian) based assays.

12. Use of NPs according to claim 10 or prepared in accordance with any one of claims 1-9 in photothermal microscopy and fluorescence detection.

13. A kit comprising metal NPs and a solvent, together with instructions describing how to increase the fluorescence of the NPs by treating them in the solvent provided.

14. A method for combining or performing simultaneously photothermal microscopy and fluorescence detection using metal NPs according to claim 10 or NPS produced according to a method of claims 1-9 such that a simultaneous measure of absorption and fluorescence is be obtained.

15. The method according to claim 14 wherein the NPs are heated by way of an incident laser beam of a microscope assembly.