I. Theoretical outlines and techniques [302679]

[anonimizat] a lot of time and precision when characterizing these systems. Nanostructured materials exhibit unique properties which analyzed properly and being well understood can be manipulated and utilized for a variety of applications from in situ to device implementation.

This work aims to describe the process of property determination and optimization for detection applications of both spherical and anisotropic gold nanoparticles.

The first chapter gives an overview of the theoretical outlines involved in the characterization and understanding of the systems. The analysis methods are also presented and their role is explained.

The synthesis methods for both spherical and anisotropic particles are described in the second chapter as well as their condition optimization. [anonimizat], the surface activation is presented in the third chapter of this study.

[anonimizat]: interaction with dye molecules in order to enhance/[anonimizat]-streptavidin complex.

The thesis ends in conclusions and further perspectives giving acknowledgements to all the parties involved in the process of this study to be performed.

I. Theoretical outlines and techniques

Metal colloids are nowadays very well studied having a [anonimizat] a [anonimizat].

[anonimizat] a [anonimizat]. The famous Lycurgus cup dating from the 4-5th century stands as the most important example. Depending on the illumination of the cup its colour changes: [anonimizat]. The last is due to the interaction of the light with a very small amount of colloidal gold. The same way were the glasses of Middle Age cathedrals made. The formation of different sized nanoparticles gives the glasses the remarkable colours. Also, their biological properties were exploit by the alchemists for various diseases.

I.1. Gold nanoparticles

Nanoparticles are structures with one or more dimensions in the nanometer scale. [anonimizat]. [anonimizat], [anonimizat]. [2]

A [anonimizat], their size and shape have a major influence in their characterization. These properties can be exploit in various fields of application if they are properly tuned, therefore a significant aspect is to gain great control over the nanoparticle’s size and shape.

I.2. Localised Surface Plasmon Resonance

Gold nanoparticle colloids exhibit all the colours of the visible spectrum, this property is highly depended on the size and shape of the nanoobjects. With regard to their geometrical attributes, the absorption spectrum contains from one to two absorption peaks. Generally, nanospheres exhibit a single peak around 500 nm depending on the diameter of the particles, while nanorods and bipyramids have two characteristic bands: a small one around 510 nm indirectly representing the width of the particles and an intense second one in the wavelength range 600-1000 nm depending on the length of the structures.

These bands absorption peaks are referred to as Surface Plasmon Resonance Bands. This phenomenon of surface plasmon resonance is defined as the collective oscillation of the conduction band electrons as a result of their resonant excitation with an incident light beam.

The theoretical model of Mie characterizes this phenomenon and observes that the extinction implies both absorption and scattering of light by a material. Maxwell’s equations allow the determination of the resonance frequency when solved for spherical particles. The total extinction cross-section is calculated with the relation:

where C is the extinction as the sum of the absorption and Rayleigh scattering contributions, c is the speed of light, ω is angular frequency of the exciting light, V is the volume of the particle, εm is the dielectric constant of surrounding medium of the nanoparticle, ε1 and ε2 are the frequencies dependent real and imaginary parts of the dielectric function of the material (εm= ε1 + iε2). [moni]

This equation clearly states the strong dependency of the SPR band of many factors: geometry of the particle, dielectric constant of the medium, particle distribution and presence of the ligand shell. With increasing the size, the absorption maximum is shifted to the red.

The SPR band is sensitive also to the microenvironment, property which makes the detection of temperature and pH induced changes, the ligand and solvent presence possible. Moreover, gold nanoparticles generate an amplified electromagnetic field which can be used in the fluorescence enhancement of a molecule in the close proximity of the particle and the Raman signal of a molecule on the metal surface. [Kanwarjeet, moni]

I.3. Fluorescence. Fluorescence life-time

Fluorescence is the emission of light following the transition of the molecule from the electronically excited singlet state to the ground state by the rapid emission of a photon. The fluorescence lifetimes are typically in the order of nanoseconds.

Usually, the exposure to a light source brings the electrons in an excited state. This process requires a certain wavelength of the light in order for it to be absorbed and the released at a different higher wavelength. This energy transfer implies energy loss. A schematic of these process is shown in Figure I.3.

In the Jablonski diagram above, S0 is the ground state, S1 and S2 are the second electronic singlet states, T1 is the triplet state. In each one of these electronic energy levels the molecules can exist in a number of vibrational energy levels, depicted by 0,1,2 etc. The vertical lines represent the transitions between the states. Light absorption takes place almost instantaneous. Light is used to induce fluorescence instead of heat, because the energy difference between S0 and S1 is too large for thermal population of S1.

Upon light absorption, several processes usually occur. A fluorophore is usually excited to a higher energy state after which molecules rapidly relax to the lowest vibrational level of the respective excited state, process known as internal conversion. Internal conversion terminates generally before the fluorescence emission. The relaxation to the ground state involves a transition on a higher vibrational level. A consequence of emission to higher vibrational levels is that the fluorescence spectrum may be a mirror image of the absorption from S0 to S1 spectrum.

In addition, molecules in the first excited state may undergo a spin conversion to the first triplet state, which corresponds to the phosphorescence emission. This type of conversion is called intersystem crossing.

Emission spectrum is a representation of fluorescence spectral data, a plot of the fluorescence intensity in arbitrary units versus wavelength in nanometers (nm). These data strongly depend on both the chemical structure of the substance and of the dissolution solvent. Usually, the energy of emission is less than that of absorption, the process involves energy loss. Fluorescence occurs at longer wavelengths causing a significant difference between the wavelength of the absorption maxima and the emission maxima, generally known as the Stokes shift.

The fluorescence emission has the property of independence regarding the excitation wavelength. This is the aftermath of the rapid relaxation of the molecules to the lowest vibrational energy level of S1, spreading to the surrounding medium the energy excess.

The fluorescence lifetime and quantum yield are the most used and the most important features of a fluorophore and the first to be analyzed when characterizing a system involving the presence of such luminescent molecules. Quantum yield represents the number of emitted photons with regard to the number of absorbed photons. Substances with the highest quantum yield display the most bright emission. The fluorescence lifetime is the short period of time the fluorescent molecules spends in the excited state before returning to the ground state. The best representation of the meanings of these features of fluorescence is the simplified Jablonski diagram presented in Figure 1.4.

Γ is the emissive rate of the luminescent molecule, knr is the rate of non-radiative decay to S0 and they both depopulate the excited state. The fraction of fluorophores that decay through emission, consequently the quantum yield, is given by:

The lifetime of the excited state is defined as:

Fluorescence lifetime and quantum yield can be modified if either rate constant is affected by external factors like pH or temperature and others. Also, the length of time fluorescent molecules remain in the excited state gives them a chance to interact with other molecules in the solution making their detection possible.

I.4. Characterization techniques

For this study, a series of techniques have been applied in order to characterize and analyse the comportment of the nanoparticles and the environment they find themselves. Spectroscopy methods are used for monitoring the LSPR response of the system, fluorescence emission and environmental changes, surface analysis microscopies offer an image of the systems making the determination of the size and self-assemblies visible and measureable.

I.4.1. UV-Vis Spectroscopy, Zeta-Potential and Dynamic Light Scattering

The extinction is defined as an attenuation of an electromagnetic wave due to both absorption and scattering. The UV-Vis spectroscopy measures the intensity of a light beam after it passes a sample, transmitted light, and compares it to the incident beam. This technique records extinction spectra with respect to the optical response of the medium the sample exists, reference. For instance, in this study for colloidal solution the reference is water and for the substrates glass. [Kanwarjeet]

The Zetasizer Nano ZS 90 instrument is used to determine the surface potential of the nanoparticles and dynamic light scattering measurements which establish the hydrodynamic diameter of the spherical nanoobjects. The zeta potential is the electric potential at the hydrodynamic plane of shear and does not only depend on the surface of the particle but also on the dispersing medium. Its determination suggest the dispersion stability and can be used as proof measurement of grafting processes.

I.4.2. Fluorescence Spectroscopy. Time resolved fluorescence

Fluorescence measurements can be classified into two types of measurements: steady-state and time-resolved. During steady-state measurements the sample is illuminated by a continuous beam of radiation, while time-resolved measurements imply exposure to a pulse of light where the pulse width is typically shorter than the decay time of the sample.

Steady-state is reached almost instantaneously after first exposure to light and the emission spectra are recorded. Due to the nanoscale timeframe of fluorescence, steady-state measurements are more widely used. However, in time-resolved measurements the intensity decay is recorded with a high-speed detection system that permits the intensity or anisotropy to be measured on the nanosecond timescale.

Fluorescent compounds have two characteristic spectra: an excitation spectrum (the wavelength and amount of light absorbed) and an emission spectrum (the wavelength and amount of light emitted). The main reason why fluorescence measurements are a highly specific analytical technique is because each compound has an unique fluorescence spectrum.

I.4.3. Raman and SERS Spectroscopies

Raman and Surface Enhanced Raman Scattering (SERS) Spectroscopies are based on the inelastic interaction between matter and a light beam.

SERS is a highly specific and sensitive analytical technique useful for investigating molecular composition in a non-destructive manner as it enables an ultrasensitive detection, down to a single molecule, and provides structural information about the molecular species from their unique vibrational Raman fingerprint [articol moni+eu].

I.4.4. Scanning and Transmission Electron Microscopies

Scanning Electron Microscopy (SEM) and Transmission Electron Microscopy (TEM) generate images of nanoscaled probes using an electron beam instead of light waves. The techniques are similar in both working principle and the resulting data, there are considered complementary analysis methods. TEM images are recorded based on the pattern the electrons create after passing through the sample, in the case of SEM, a detector collects the by the sample reflected electrons and generates the image. The advantage of TEM is the high resolution, while SEM can be used to reproduce cross sections of the samples. In this study, both techniques have been used for the determination of the nanoparticle sizes. TEM images have been also recorded in order to confirm the binding of the molecules to the nanostructures.[nidumolo thesis]

I.4.5. Atomic Force Microscopy

Atomic Force Microscopy (AFM) is a surface analysis technique which provides a three-dimensional image of the surface with high accuracy due to its high z direction resolution. The method implies a sharp tip attached to a cantilever that scans the surface of a probe. The cantilever oscillates at its resonance frequency, when the tip and the surface interact due to attractive and/or repulsive forces a deflection in the cantilever is observed and recorded. The series of deflections is convoluted into an image of the sample.[Kanwarjeet]

For the investigations required by this study, AFM is used to investigate the surface of gold nanosphere substrates, to image cross sections and determine the height of the particles on the substrate. The technique is used as a complementary to TEM images of the nanoparticles.

II. Fabrication of the Gold Nanoparticle Systems

The synthesis processes are both physical, “top down” techniques which imply the deterioration of the bulk material into defined patterns using mechanical techniques or lithography, and chemical, “bottom up”, which consist in the fabrication of nanoparticles using chemical reactions, nanoparticle aggregation or laser induced assembly. [2]

The most common fabrication method relies on the chemical reaction of the metal salt reduction using a citrate solution. The synthesis is performed in a medium which inhibits the tendency of the particles to aggregate, while the citrate exhibits a double role: as a reduction agent it is involved in the growth kinetics and defines the size of the particles, but often it behaves as a stabilizing agent too.

For gold spherical, isotopic structures the first method was described by Turkevich et al. in 1951. His approach consists in the reduction of gold chloride using sodium citrate in order to produce monodispersed nanoparticles in an aqueous solution. In the case of anisotropic structures, the synthesis is called the seed-mediated growth approach and is divided into two steps: a nucleation followed by the controlled growth of the already existing structures. Mainly, the first step is to create small spherical particles using the citrate reduction and then they are used to grow more complicated structures with the help of a mild reducing agent, which facilitates the growth but does not induce further seed formation. In this case the size and the shape of the nanoparticles are controlled by the variation of the type and concentration of the reducing agent. [3]

II.1. Gold Nanoparticle Synthesis

In the fabrication of nanoparticles, gold makes a good candidate of choice due to is stability and ability to shape the nanoobjects. [bp to nanojav] Irrespective of the synthesis method, for the development of the nanostructures, Gold(III) chloride trihydrate (99.9%), 1 HAuCl4 to 3 H2O, has been used.

As suspension medium, the nanoparticles will be fabricated in the presence of milli-Q water, aqueous cetyl trimethylammonium bromide (C19H42BrN, CTAB), which is a surfactant which covers the surface of the nanoparticles and reacts as a stabilizing agent and provides a convenient environment for the nanorod seeds, and aqueous cetyl trimethylammonium chloride (C19H42ClN,CTAC), the latter is similar with CTAB but it is suitable for the bipyramid seed solution.

The seed preparation is based on the reduction of the gold ions using a strong reduction agent. In this study, sodium borohydride (NaBH4) is used for that matter, even though it has the down side of degradation in water. In order to avoid this issue, sodium hydroxide (NaOH) is used and the storage conditions are adapted. In the case of the seeds needed to grow bipyramids, some acids like nitric acid (HNO3) and citric acid (C6H8O7) are used.

In the growth process, a decisive role is assigned to silver nitrate (AgNO3). The full process of the growth is not well understood yet, but it involves a dynamic process in which the silver atoms place themselves in certain arrangements and isolate the spot from gold controlling this way the growth direction.

The surface modification of the anisotropic nanoparticles involves the replacement of the surfactant with TWEEN 80 and polyvinylpyrrolidone (PVP), both polymers which aim to increase the reactivity of the gold surface and ensure particle stability.

II.1.1. Gold nanospheres

The most common synthesis method is described by Turkevich et al. for small sized gold nanospheres [teza moni], but there are some more complicated procedures which allow the fabrication of nanoparticles with diameters up to 150 nm. Such methods imply more than one step, are successive seed-mediated growth approaches and need more complex growth mediums. [successive seed growth]

For the current study, it was essential to have m1onodispersed homogeneous colloid solution. Colloidal AuNSs were synthesized according to the Turkevich method by chemical reduction of HAuCl4 with sodium citrate. Mainly to 100 ml of 10-3 M HAuCl4 aqueous solution is heated until boiling point during continuous stirring. When the first bubbles are to be seen as a boiling sign, 10 ml of 38.8 mM sodium citrate are added. The mixture is left for 15minutes under heat and continuous stirring. For the last 15 min the heating o the solution is stopped. The reduction reaction can be observed by the color changing process: once the reducing agent is added the yellowish solution turns transparent, then black and starts to brighten until it reaches a reddish colour.

The as-prepared AuNSs in solution exhibit a plasmonic resonance at 519 nm and an average diameter of 18 ± 2 nm, as presented in Figure II.2., parameters which are consistent with the measured hydrodynamic diameter of 20 ± 2 nm and zeta potential of -41 mV, which implies high stability of the colloidal solution.

II.1.2. Gold bipyramids

Anisotropic structures present a higher interest in the investigation of gold nanoparticles due to their interesting features exhibited by their shape respectively the sharpness of their tips. Gold bipyramids present a strong localized electromagnetic field at their tips, moreover due to their shape they present two plasmon resonance bands: the first one, at lower wavelengths is assign to the transversal resonance modes and the second one, at higher wavelengths corresponds to the longitudinal resonance modes. [from gol bp to nanojv ] Also, edgy structures tend to be highly sensitive to local changes in the surrounding medium. [syn,te and single part]

Firstly, bipyramidal shaped structures have been observed as by-products in the fabrication of nanorods. Literature presents great attempts to increase the yield of bipyramids and gain control over their size. [syn,te and single part]

The process of fabrication is based on the seed-mediated growth approach consisting of two main steps: the synthesis of the seeds, which are defining the shape and size of the structures, and the growth of the particles.

In order to obtain a high yield of diamond like structures, polycrystalline seeds are desired. This step is very important and least reproducible since parameters as addition speed, solution temperature variations, infirm volumes of impurities can have a great influence in the process. [from gol bp to nanojv,dispersion np sol gel hybrid]

The first step is to produce a sodium borohydride/ sodium hydroxide stock solution which will be kept in cold medium (4șC), 100 µl of 2.5 M NaOH is added to a cold 5 ml 5 M NaBH4 solution. Before de addition a 100 times dilution is made using cold water and 190 µl of 2.5 M NaOH is added in order to obtain the ratio 1 NaBH4 : 1 NaOH, after the dilution the final solution has a concentration of 50 mM.

2.72 g of 25%w CTAC is added to 32 ml milli-Q water during vigorous stirring, 320 µl 1 M gold salt is then brought to the aqueous CTAC solution. The mixture changes colour from translucent to yellow. The addition of 296 µl of 0.25 M HNO3 does not modify the solutions appearance. With a constant addition speed for 2-3 seconds 400 µl of the 50 mM NaBH4/NaOH mixture are dropped, as expected the reduction takes place and the solution gets brownish. After 1 minute, 320 µl of 1 M citric acid is added to the final seed solution, which is put in a 80șC warm water bath for 60 to 90 minutes. After the last addition the stirring is stopped and as soon as the heat treatment is ended, the seed solution is kept 1 week for aging, the time needed for the spherical particles to be stable and well defined.

The growth process is easier to reproduce once the seeds are obtained. 40 µl of 25 mM HAuCl4 are added to 4 ml 45 mM CTAB solution, which turn the transparent solution into a dark yellow one. To this mixture 30 µl of 5 mM AgNO3 is added, this step does not imply any changes in color. 60 µl 0.4 M hydroxyquinoline is added and the reduction of the gold ions takes place slowly, the process can be followed by the colour change (dark yellow to translucent). When the solution perfectly transparent is, 20 µl of seed solution is added to the final mixture. In order for the growth to take place the sample is placed in the oven 50minutes at 45șC. After each addition, the sample is gently mixed to avoid the formation of undesired compounds.

By varying the seed concentration during the growth process, one can gain a great control over the size of the bipyramids they need and desire. As seen in the Figure II.3., with increasing the added volume of seeds the size decreases and structures with plasmon resonance bands in the range 600 – 1000 nm are achieved.

In the Figure II.4.SEM microscopic images are presented, the nanoparticle colloid is homogenous in both size and shape and the high yield of fabrication can be seen in the left image, there are just a few by-products (nanospheres). Moreover, using the analysis toolkit ImageJ [] the dimensions could be determined for the bipyramids used in this study, which have an longitudinal LSPR band at 800 nm: 86 ± 5 nm in length, 27 ± 5 nm in width; for this determinations 25 structures have been analysed.

II.1.3. Gold nanorods

Gold nanorods have been fabricated in order to have a reference to compare the bipyramids responses. They are a good candidate of choice due to their similar properties: two characteristic LSPR bands, comparable shapes and behaviours.

The most widely used procedure is the seed-mediated growth approach developed by Murphy et al. [17,18 hqrodschemmater] and El-Sayed [19 same article]. This method implies the reduction of the gold ions using sodium borohydride for the nucleation process, which is added to a Au(III)- CTAB solution reduced with ascorbic acid in the presence of silver nitrate. The down sides of this method are the sensitivity to small environmental changes which make the surfactant replacement almost impossible and the low yield of the nanorod-like structures and high yield of by-products (~15% of the gold ions are converted to metallic gold)[23same article].

Leonid Vigderman and Eugene R. Zubarev have developed a synthesis method which is based on Murphy’s approach but introduces a new reducing agent, hydroquinone, to replace the ascorbic acid.[High-Yield Synthesis of Gold Nanorods with Longitudinal SPR Peak Greater than 1200 nm Using Hydroquinone as a Reducing Agent] This change in the growth process overcomes the limitations of the initial procedure and allows a greater control over the size of the structures making it possible to fabricate nanorods with a LSPR response in the near-infrared (~1000 nm) with a high reproducibility rate.

The seed-mediated growth approach consists of two steps: the nucleation and the growth process. For the seed fabrication, a solution of 5 M sodium borohydride and sodium hydroxide has been prepared and kept at 4șC. A flask containing 10 ml of milli-Q water is cooled down to 4șC. In the synthesis a 10 mM NaBH4/ NaOH is needed but since sodium borohydride easily degrades in water, the dilution will be made before the addition by dropping 20 µl 5 M mixture into the 10 ml cold water. The glassware has been properly cleaned and dried. To a 0.1 M 5 ml CTAB solution 100 µl of 25 mM HAuCl4 has been added under mild stirring to avoid the creation of gold – CTAB complexes, the solution went from translucent to dark yellow in terms of colour. When the solution becomes clear, 250 µl of 10 mM NaBH4/ NaOH are added under rapid stirring. The colour changes from dark yellow to light brown, the stirring is continued for 25 seconds and the seed solution is let to rest for 1 hour at 27șC. Note that the seeds are useable for only 24 hours.

The second step represents the actual growth of the nanorods which is very similar to the growth of the bipyramids. For instance, 100 µl of 25 mM HAuCl4 are added to 5 ml 0.1 M CTAB solution expecting the same colour change as for the seed solution. To this mixture 30 µl of 20 mM AgNO3 is added, this step does not imply any changes in color. 25 µl 0.5 M hydroquinone is added and the reduction of the gold ions takes place slowly, the process can be observed by the slow change in colour from dark yellow to translucent. When the reaction is finished 240 µl of seed solution is added to the final mixture. In order for the growth to take place the sample is placed in the oven for at least 6 hours at 45șC. After each addition, the sample is gently mixed to avoid the formation of undesired compounds.

Figure II.. Extinction spectra of nanorods of different sizes depending on the seed concentration used in the growth process

In order to control the size of the produced nanorods, one can vary one or more parameters in the synthesis as shown in Figure II.5., by decreasing the amount of added seed solution the longitudinal LSPR band is shifted to the red which means that with smaller concentrations of seeds the size of the particles increases.

As seen in the SEM images represented in Figure II.6., this method allows the fabrication of gold nanorods in a high yield with a low concentration of by-products. In this case, some nanospheres are observed. Using the program ImageJ[], the size of the particles has been determine to be: 33 nm ± 5 nm in length and 6 nm ± 1 nm in width (25 structures have been analysed) for the nanorods with an extinction maximum at 800 nm which are going to be used for further investigations.

II.2. Surface modification

The surface modification implies the removal of the surfactant used in the synthesis process and replace it with another one in order to gain higher reactivity and stability of the nanoparticles.

The synthesized nanospheres are negatively charged on the surface, which makes them very reactive and usable immediately after the fabrication. For a colloidal solution to be stable the zeta potential should be in the intervals (-∞, -20] and [+20,+∞), zeta measurements determine a potential of -41.1 mV. [art moni+eu] One can conclude that a surface modification is not needed.

In the case of the anisotropic nanostructures, the surfactant change is recommended. The fabrication occurs in the presence of CTAB, which both stabilizes the system and covers the surface of the nanoobjects.

In Figure II.7. the fabrication process is illustrated, the nanoparticles are covered in the surfactant. CTAB has a longitudinal structure and forms a double-layer at the surface of the structures both bipyramids and nanorods. This behaviour has a major influence on the reactivity of the surface to other changes due to the strength of the bonds between the two CTAB molecules, which are difficult to break in order to functionalize the nanoobjects for further applications. The bond strength provides on the other hand a very good stability.

The desire of high reactivity and stability reduces the candidate options to two different polymers: TWEEN 80 and PVP. TWEEN 80 is a neutral surfactant with a branched geometry which has a milder interaction with the gold surface therefore is suitable for increasing the reactivity of the particles. PVP has similar properties but differentiates itself by the longitudinal geometry. [] This aspect of geometry is crucial because it determines how the polymers binds to the gold surface and is represented in Figure II.8.

A 10% in weight TWEEN 80 and a 5% in weight PVP solutions have been prepared. The replacement protocol has been followed:

1. the colloidal nanoparticle solution has been centrifugated for 15 minutes at 8000 rpm and the supernatant has been removed;

2. 2 ml of the polymer has been added and the sample is put in ultrasound bath for 10 minutes

3. 2 ml fresh milli-Q water has been added and to ensure homogeneity, samples have been mixed with the vortex for a few seconds.

After the replacement the extinction spectra have been recorded and analysed as seen in the Figure II.9. The bipyramids covered with PVP exhibit a broad band in the range 700 – 1100 nm, which indicates that the nanoparticles have aggregated, while the particles in TWEEN 80 replicate the narrow sharp band of the colloidal solution, only a small part of the particles have aggregated as a small bump is recognizable around 1000 nm. By normalizing the spectrum of the fabricated nanoobjects and the TWEEN 80 covered particles, a 5 nm shift to the red is observed. This implies the fact that the environment of the nanostructures has changed leading to the conclusion that the polymer has replaced the CTAB at the gold surface.

These results are sustained also by the zeta potential measurements. The nanoparticles have a positive zeta potential after the fabrication of +40 mV, after the replacement the nanoparticles have a potential of -3.07 mV for the TWEEN 80 coverage and -7.8 mV in the case of PVP. This values are in the range of instability, behaviour observed for the PVP covered nanoobjects, which are very unstable in time and tend to aggregate. The polymer tends to cover the surface and arrange itself as a layer, one reason is its geometry. The TWEEN 80 behaves differently even if the zeta potential does not indicate stability, this feature is ensured by its geometry. TWEEN 80 can bind to the gold surface in a variety of ways, but in almost all of them there is a high probability that at least one branch remains free. In this way, the stability of the system is given by the steric hindrance between the free branches of different nanoparticles.

For further applications and investigations, the gold nanoparticles, nanorods and bipyramids, are covered in TWEEN 80, which is the best candidate of choice allowing a good compromise between higher reactivity and stability. In the discussions, when the nanoparticles are mentioned, it is referred to the nanoparticle solutions which underwent the surfactant replacement process.

III. Detection Applications of Gold Nanoparticles

III.1. Dye-to-Particle Interaction

Gold nanoparticles are often used as signal amplifiers in many application fields, one of the most desired phenomenon to be enhanced is the fluorescence in order to detect certain molecules or to evidence their presence in both spectroscopy and microscopy measurements.[art fluorescenta] This feature assigned to the nanoparticles and their capability to amplify the fluorescence is verified, characterized and evaluated when a chromophore is grafted on their surface and the interaction between the two is observed.

III.1.1. Lucifer Yellow Polymer

In this study, the interaction between anisotropic gold nanoobjects and a luminescent polymer, Lucifer Yellow polymer (LYP), is investigated. Lucifer Yellow Polymer is a longitudinal polymer with units of Lucifer Yellow chromophore attached along the chain (Figure III.1.). The polymer is commercially available and has been chosen due to its chemical and optical properties: it is soluble in water and its absorption spectrum does not overlap with the optical response of the nanoparticles, meaning the spectrum of the nanoparticles can be investigated without the results to be influenced by the chromophore. Moreover, this is a thiolated polymer, at one end the polymer has a thiol-group (SH-group) which makes it susceptible to the gold surface.

Polyethyleneglycol (PEG) has been used in order to stabilize the system, PEG is also o thiolated polymer of a much lower size than LYP. The chemical structure is shown in Figure III.2. Moreover, since both PEG and TWEEN 80 are PEGylated systems the change of the environmental medium is negligible (no major shifts are presented in the extinction spectra) ensuring that only the modifications induced by the grafting of the LYP are obtained.

Previous studies underline the fact that the distance between the chromophore and the particle is of major importance for the enhancement of the fluorescence emission. The same polymer has been investigated in similar conditions, the main difference was its structure. Navarro et al. observed fluorescence enhancement when the LYP had a two-block structure: two unit polymer, only one unit had chromophore molecules attached along the chain. The chromophore was at least at one unit polymer far away from the gold surface. []

The purpose of this study is to see what happens when the chromophore molecules are in the close proximity of the gold surface and does the shape of the nanoparticles influence the behaviour. To do so the study has been performed on both nanorods and bipyramids with a LSPR band at 800 nm and both having the surfactant replaced by TWEEN 80.

III.1.2. Grafting Methods

The coupling of the luminescent polymer is performed in aqueous solution. A 15 mM LYP and a 15 mM PEG stock solution have been prepared using milli-Q water. In order to graft the LYP on the gold surface, two strategies have been pursued and will be presented individually as follows.

1. The simultaneous addition of the luminescent polymer and the stabilizing agent

To a 1 ml colloidal solution 2 ml of milli-Q water has been added in order to obtain an absorbance of 1 a.u. After gently mixing to achieve homogeneity, LYP and PEG are simultaneously added to the nanoparticle solution in a ratio of 1:3 (25% LYP and 75% PEG calculated in such a manner that to a mol of gold atoms correspond a mol of PEG, LYP respectively; 100% equals a volume of 170 µl 16.5 mM polymer solution). The sample undergoes heat treatment for 1 hour and 30 minutes at 45șC. The polymer excess is removed by centrifugation at 8000 rpm for 15 minutes and redispersion in 3 ml milli-Q water.

The extinction spectra (Figure III.3.) does not present any major changes, the LSPR band representing the hybrid system is broader which sustains the formation of aggregates, but no significant shift neither to the red nor to the blue are noticed. This results coud be explained by two hypotheses: a. the grafting of the LYP did not occur, no shifts imply no changes in the refractive index of the system meaning the surrounding medium is the same as before the grafting; b. the LYP is grafted on the faces of the particles and not at the tips and edges which give rise to the longitudinal LSPR band. An ideal representation of the system is shown in the Figure III.4. The latter hypothesis has a higher probability of occurrence, PEG is way smaller in size compared with LYP which makes it more rapid to graft and the LYP covers only the remaining free spaces.

2. The gradual addition of the luminescent polymer and the stabilizing agent

The first steps of this strategy coincide with the previous one: 1 ml of colloidal solution is diluted with 2 ml of milli-Q water and gently mixed to achieve better homogeneity. To the nanoparticle 42.5 µl (25%) 16.5 mM LYP is added and left undisturbed for 30 minutes for the reaction to take place. 127.5 µl 16.5 mM PEG is then brought to the mixture and the sample undergoes heat treatment for 1 hour and 30 minutes at 45șC to give the polymer time to stabilize the system. The reaction is then interrupted by the removal of the supernatant while centrifuging the samples for 15 minutes at 8000 rpm and redispersing them into 3 ml milli-Q water.

In this case, the grafting of the LYP is confirmed by a 15 nm shift to the red of the longitudinal LSPR band, which rises the hypothesis: LYP most probable is grafted at the tips of the bipyramids and at the edges of the basis. An ideal representation of the system is shown in the Figure III.6.. The addition of the PEG after a period of time allows the LYP to graft on the gold surface.

For both strategies the assumption is that the first spots for the polymers to graft are the tips and edges of the nanoparticles. This supposition is sustained by the fact that the sharp formations of the nanoobjects are more sensitive to changes and tend to oxidize more quickly. On the other hand, the high electromagnetic field at the tips could enhance the spots susceptibility to graft molecules. For further investigations, 1.5 ml of each sample have been kept undisturbed at room temperature.

III.1.3. Dissolution of the Gold Core

To evidence the grafting and the properties of the LYP, the gold core has been dissolved using sodium cyanide (NaCN). A 0.1 M stock solution of NaCN has been prepared using milli-Q water. For this step, the remaining 1.5 ml of the samples have been used.

For the system produced by simultaneous addition, the extinction spectra were recorded after the addition of 50, 100 and 150 µl of NaCN in order to observe the process and determine the needed volume of NaCN for the total dissolution. In the left graph one can notice how with increasing volume of the NaCN both the transversal and the longitudinal bands decrease in intensity. For the last addition, the spectrum consists of a straight line that signifies the disappearance of the gold nanoparticles. Same behaviour is observed in the case of the gradual addition strategy (right graph), but in this case the dissolution involved one single step of adding 150 µl of the cyanide solution.

Taking a closer look to the spectrum in the range of 400 – 600 nm, a small band is observed when the spectra are magnified. This band is assigned to the LYP and appear in both grafting strategies. For instance, in Figure III.8. the spectrum before and after the gold core dissolution are compared. The band is present even after the dissolution, behaviour that confirms the grafting process. The intensity of the band in both grafting methods is decreased after the nanoparticles are removed, fact that suggest a low quantity of LYP has been grafted and that before the dissolution even if the supernatant has been removed some free/ not grafted LYP has remained present in the sample.

This step underlines the fact that the grafting processes have been performed with a LYP excess and that a small amount of polymer was grafted on the surface of the nanoparticles. Until this point, one can state that using this two strategies it is possible to also tune where the luminescent polymer attaches on the nanoobjects.

The same procedure has been performed on the nanorods, which present the same characteristics and behaviours during the process.

III.1.4. Optical Study and Discussion

In order to achieve the main purpose of investigating the interaction between the two entities, luminescent polymer and the nanoparticle system, an optical study has been performed, mainly the emission of fluorescence has been monitored and the life-time of the molecules has been analysed.

Previous studies show a fluorescence enhancement when using a luminescent polymer at the distance of one spacer from the gold surface. In this case, the polymer unit with the chromophore attached along the chain is directly grafted on the bipyramids surface, interestingly the obtained signal for the hybrid system has a lower intensity than the free luminescent polymer. The fluorescence is partially quenched (Figure III.9.), this can be explained by the fact that the polymer unit has chromophore molecules in the very close proximity of the gold core which are totally quenched and the existing fluorescence is given by the chromophore molecules attached close to the other end of the unit at a bigger distance with regard to the nanoparticle. Moreover, the position of the molecules influence the emission of fluorescence, only the chromophore molecules which are on the perpendicular direction with regard to the nanoobject contributes to the fluorescence emission.

Moreover, fluorescence life-time measurements confirm the phenomenon of quenching, …..

This results consolidate the idea that the distance between the surface of the particle and the chromophore molecules is very important when talking about their behavior. Irrespective of the grafting strategy or the nanoparticles used, the phenomenon of fluorescence quenching is present and the results overlap. Still, an interesting observation is that there are no to very small changes in the lifetime curves, result which arises some interesting questions like : is this behavior influenced by the tips or is there a different cause of this comportment.

III.2. Biotin – Streptavidin Detection

The main purpose of biosensors is to recognize molecules and particles, which can be then further linked to certain diseases. Most of this molecules do not exhibit by themselves a detectable signal making the use of gold nanoparticles of great interest. Developing a system composed of such structures and proteins enables efficient immobilization without any chemical interactions between the two and the undistorted detection of the biological molecules. For the determination of these two capabilities of the nanoparticles, the biotin-streptavidin complex is ideal because of is high binding affinity. Each streptavidin has four binding sites for biotin to be positioned giving the system a high specificity. [nidumolu]

III.2.1. Gold Nanospheres

Gold nanospheres are good candidates of choice for a simple and efficient strategy to improve the sensitivity of localized surface plasmon resonance shift-based biosensors using biotin-streptavidin recognition interaction. Specifically, biotin molecules are immobilized on a low-cost plasmonic LSPR biosensor based on annealed self-assembled spherical gold nanoparticles and successively incubated with increasing concentrations of streptavidin, achieving a limit of detection (LOD) of 5 nM.

III.2.2 Grafting methods

First, to improve the reactivity of the glass solid surfaces and to remove any organic remaining contaminants, the glass substrate was cleaned with chromic acid solution, washed thoroughly with ultrapure water and further treated in ozone cleaner (PSDP-UVT – Benchtop UV Cleaner) for 20 min. Second, the as-cleaned glass substrate was silanized through immersion in a 1% (v/v) MPTES methanol solution for 1 h. In order to remove any loosely adsorbed MPTES molecules, the substrate was rinsed with methanol and ultrapure water several times. The as-functionalized substrate was subsequently immersed into a solution of pre-synthesized AuNSs for another 3 h to allow the assembling of the NPs. The obtained AuNSs substrate was dried and transferred into a high-temperature oven (Barnstead Thermolyne Type 47900 Furnace) to conduct thermal annealing at 500°C for 90 min. The annealed AuNSs substrate was left to cool at room temperature and subsequently used as platform in the construction of our sensitive LSPR immunosensors.

The proposed biosensing protocol using the annealed AuNSs substrate is schematic illustrated in Figure III.11. For the better immobilization of biotin-NHS, the metallic surface of the annealed substrate was chemically modified with two self-assembled monolayers (SAM) (i.e. MPTES and APTES) deposited via the layer-by-layer method [Zhang, 2008]. The first step of the label-free biosensing protocol consisted in functionalizing the annealed AuNSs substrate through immersion in a 1% MPTES methanol solution for 30min, resulting in the formation of a silane-terminated MPTES SAM on the plasmonic surface. After carefully rinsed with methanol and water, the modified substrate was silanized in 4 % APTES ethanol solution for 20min. After attaching the linkers, biotin-NHS was anchored onto the metallic surface of functionalized-AuNSs by immersing the substrate into 50 μg/mL biotin solution for 24 h. Afterwards, a large amount of phosphate buffered saline (PBS) solution was used to flush away the unreacted biotin-NHS from the metallic surface. Then, in order to eliminate the influence of nonspecific adsorption, the surface was then blocked with BSA by immersing the substrate in a solution of 50 μg/mL BSA in PBS for 1 h. After rinsing with water, the biotin-functionalized annealed AuNSs substrate was used as LSPR biosensor for the specific detection of streptavidin. To allow the capture of streptavidin, the biotin-functionalized substrate was incubated with streptavidin solution at 4 oC for 5 h. In order to determine the limit of detection of the proposed LSPR biosensor, different streptavidin concentrations (i.e. 5 nM, 10 nM, 20 nM, 35 nM and 50 nM in PBS, pH 7.4) were tested.

III.2.3. Results and Discussion

Typical UV-Vis-NIR extinction spectra of the self-assembled AuNSs on the functionalized glass substrate before and after thermal annealing process are presented in Figure III.12.A. Before annealing, the plasmonic substate displays two LSPR bands, at around 535 nm and 710 nm, assigned to the isolated AuNSs and to the partial aggregated (due to existing assemblies) AuNSs [Ozhikandathil and Packirisamy, 2014], respectively, in good agreement with AFM and TEM imaging observation (Figure. III.12.B and Figure III.12.E). From the recorded AFM cross-section we find that the height of the AuNSs is around 16-20 nm (Figure III.12.D – black spectrum), similar to the value obtained for colloidal AuNSs [Suarasan et al., 2013], thus showing that the AuNSs preserve their size. Interestingly, the thermal annealing at 500 °C for 90 min leads to dramatic changes in the extinction spectrum of the self-assembled AuNSs. Specifically, the broad asymmetric peak from NIR region completely disappears and a new sharper well-defined single LSPR band emerges at around 536 nm due to the well-dispersed newly formed AuNSs. The morphological changes of the substrate induced by the thermal annealing treatment (Figure III.12.C and III.12.F) confirm the LSPR response. Specifically, at this high temperature the height of the new formed AuNSs is around 50-60 nm (Figure III.12.D – red spectrum). Furthermore, according to the TEM image analysis produced by using an image processing Image J toolkit[], the average number of the NSs decreases from 1578 (Figure III.12.E) to 259 NSs per TEM area (Figure III.12.F) and the distance between them became almost twice larger than their diameters. However, it should be noted that the annealed AuNSs substrates are stable, as the plasmonic response was almost constant after 1 month. The assumption is that the formation of higher AuNSs could possible involve two mechanisms operating concurrently: i) coalescence of AuNSs -process most common for a high density of clusters, the small AuNSs merging to form a larger AuNS and ii) Ostwald ripening process which occurs by removal of atoms from one AuNSs via substrate surface diffusion and transfer to another one [Bechelany et al., 2010; Wang et al., 2009].

Figure III.0.12 (A) Extinction spectra of the AuNSs assemblies before (black spectrum) and after annealing (red

spectrum) and FDTD simulated extinction cross section (red dotted spectrum) of the substrate after

annealing; AFM and TEM images of the AuNSs self-assembled on glass surface before (B and E) and after

(C and F) thermal annealing; (D) Cross-sections before (black spectrum) and after (red spectrum) thermal

annealing

The LSPR spectra corresponding to different biofunctionalization steps involved in the fabrication of plasmonic platform for streptavidin detection are presented in Figure III.13.A. As previously showed, the annealed self-assembled AuNSs substrate exhibits a well-defined LSPR band at 536 nm. After incubation with biotin, the plasmonic extinction peak of the substrate red-shifts 3 nm, indicating the success biofunctionalization (Figure III.13.A – spectrum b). Such spectral modifications are commonly observed in the case of LSPR biosensors and can be assigned to the increase of the local refractive index, transduced by plasmonic tuning of the resonance band -including here resonant wavelength shift or/and maximum optical density changes [Jia et al., 2014]. However, in this case, following the APTES modification, an amino-containing SAM is formed, which will facilitate the chemical binding to biotin-NHS via an amination reaction [Niu et al., 2013]. The covalent bonds of biomolecules to Au surfaces are -in general- preferred than electrostatic or adsorption interaction which could lead to the desorption of biomolecules. Subsequently, when streptavidin targets 10 molecules are immobilized onto the biotin-functionalized plasmonic substrate (Figure xA – spectrum c and inset), a large red-shift of 10 nm is recorded, which is attributed to the change of the local refractive index generated by the specific detection of streptavidin.

Figure III.0.13 (A) Extinction spectra of annealed AuNSs platform before (a) and after incubation with biotin-NHS (b) and streptavidin (50 nM) (c) solutions. The inset shows the LSPR peak shift after biotinstreptavidin

binding; (B) LSPR response of the plasmonic platform as a function of streptavidin

concentration. The error bars represent the standard deviation resulted from three independent

measurements. (C) FDTD simulated extinction cross section obtained for the considered morphology before

and after coating with biotin and streptavidin.

The LSPR response of the plasmonic biosensor as a function of streptavidin concentration (from 5 to 50 nM) is presented in Figure III.13.B. The LSPR maximum is shifted to longer wavelengths upon streptavidin binding, with the magnitude of the red-shift increasing from 1 to 10 nm with streptavidin concentration. To note that the LSPR response gets saturated for a concentration higher than 35 nM. The simulated optical response of the substrate after the binding of biotin and streptavidin Figure III.13.C shows LSPR shifts of 3 nm and 12 nm after interaction with biotin and streptavidin, respectively, in good agreement with experimental results.

Beyond LSPR spectroscopy, the capture and adsorption of streptavidin onto the surface of biotin-functionalized AuNSs platform was also proved by SERS analysis. Comparative SERS measurements were performed on biotin-AuNSs platform before and after the adsorption of streptavidin. Figure xA- spectrum b shows a representative SERS spectrum collected from biotin-AuNSs substrate. For comparison, the Raman spectrum of solid biotin is also shown in Figure III.14.A spectrum a. The presence of the same spectral features in both Raman and SERS spectra clearly proves the adsorption of biotin molecules onto the surface of AuNSs substrate. The assignment of the major vibrational bands is given in Figure III.14.B.

Figure III.0.14 (A) Normal Raman spectrum of solid biotin-NHS multipled by 10 (a), SERS spectra of biotinfunctionalized

AuNSs platform (b) and streptavidin adsorbed onto the biotin-functionalized AuNSs platform

(c); (B) Assignment of the major SERS bands in the spectra of biotin and biotin/streptavidin complex

illustrated in A. [articol eu +moni]

For example, the prominent band at 1131 cm-1 observed in the SERS spectrum corresponds to the Trp13, C-N stretching vibrations (Bi et al., 2013). The vibrational bands at 1615, 1530 and 1364 cm-1 are assigned to C-C wagging, amide II vibration due to N-H bending and C-H stretching and Trp 7, respectively, 12 according to the literature Bi et al., 2013; Galarreta et al., 2011). A dramatic change of SERS fingerprint is observed after the adsorption of streptavidin molecules. As shown in Figure III.14.A- spectrum c and Figure III.14.B, the vibrational bands of biotin (marked by *) are present in the spectrum together with the characteristic bands of streptavidin molecules. In particular, the bands at 1642, 1468, 1244, 1097 and 965 cm-1 are assigned to amide I vibration due to C=O stretching, CH2 ring bending, amide III (r. coil), Glu, Thr and Trp, Val vibrations of streptavidin, respectively.

The reproducibly of the SERS signal is an important requirement for any analytical applicability of a SERS platform. The SERS reproducibility of the fabricated biotin-AuNSs platform was tested by recording several SERS spectra from random multiple sites on the substrate surface.

All these results are part of the article written by Monica Focsan, Andreea Campu, Ana-Maria Craciun, Monica Potara, Cosmin Leordean, Dana Maniu, Simion Astilean; “A Simple and Efficient Design to Improve the Detection of Biotin-Streptavidin Interaction with Plasmonic Nanobiosensors”, which is under review at the moment.

IV. Conclusions and Future Considerations

Acknowledgements

Firstly, I would like to acknowledge my supervisors, Professor Dr. Simion Astilean and Scientific Researcher I Dr. Monica Focsan, for giving me the opportunity to work on this project.

Thanks to our collaborators, MCF. Dr. Frederic Lerouge and Prof. Stephane Parola from the Laboratoire de Chemie , Ecole Normale Superieure, Lyon, France for receiving me during my 5 month Erasmus+ internship and for all the given insights into the subject matter.

References

Declarație

Prin prezenta declar că Lucrarea de licență cu titlul “Titlul complet al lucrării” este

scrisă de mine și nu a mai fost prezentată niciodată la o altă facultate sau instituție de

învățământ superior din țară sau străinătate. De asemenea, declar că toate sursele

utilizate, inclusive cele de pe Internet, sunt indicate în lucrare, cu respectarea regulilor de

evitare a plagiatului.

Cluj-Napoca,

22. Iunie 2016

Absolvent:

Câmpu Andreea-Maria