Babes-Bolyai University [311666]

Babes-Bolyai University

Faculty of Physics

Specialization Technological Physics

Master Thesis

Development of SERS biosensor

(detection of Ochratoxin A)

Scientific advisor

Prof. Univ. Dr. [anonimizat]. [anonimizat]

2015

Development of SERS biosensor

(detection of Ochratoxin A)

Thesis outline

Chapter 1

Nanostructures – [anonimizat] 1 Nanostructures – [anonimizat] 1 to 100 nm. Many institutes of research have been working to improve the fabrication methods for developing new nanometric scale practical materials. [anonimizat] [1], soft lithography [2], film deposition [3], etching [4], bonding [5], [anonimizat] [6], electrically induced nanopatterning [7], colloidal monolayer [8] and focused ion beam lithography [9]. [anonimizat], including novel medication delivery systems and biosensors. [anonimizat] a [anonimizat], saving money and time.

Fabrication methods of nanostructures

Nanofabrication can be achieved in different ways depending on the approach. These nanofabrication techniques are normally ordered into two principle classes: (1) bottom-up and (2) top-down [1]. [anonimizat].

[anonimizat]. [anonimizat] (IC) creation that has been standard since the 1970s [anonimizat] ([anonimizat]) or chemical processes ([anonimizat]) are used. [anonimizat] ([anonimizat]), ions ([anonimizat]) or photons ([anonimizat]). Additionally, a [anonimizat] (NIL), a procedure which doesn't [anonimizat] a hard stamp that is utilized to engraving nanoscale highlights onto a polymer film. [anonimizat] a [anonimizat].

[anonimizat], [anonimizat]. [anonimizat]. Liquid phase methods are also numerous. [anonimizat]. A [anonimizat] (NLS) [1], [anonimizat] scale or nano-spheres on solid layers followed by the deposition of a metallic thin film of wanted thickness and consequent evacuation of the polystyrene spheres. Atomic layer deposition is a mechanical procedure that is fit for covering any material, regardless of size, with a monolayer or to a greater extent a slim film.

The principle advantages of the bottom-up technique is that desired nanoscale structures are delivered in a less difficult, less expensive way and offer high throughput. Disadvantages include establishment of long-range order. Both top-down and bottom-up manufacture methodologies may be utilized to realize a range of all-around characterized nanostructured materials with alluring physical and chemical attributes. Among these, the bottom-up self-assembly procedure offers the most realistic solution toward the manufacture of next-generation useful materials and devices.

Optical properties of nanostructures

In the most recent decade analysts done a serious work with a specific end goal to comprehend the plasmonic fundamentals and creating of plasmonic device based on nanostructures. Researchers studied that surface plasmons can be found in diverse structures, for example, propagative electron density waves at the interfaces between a dielectric and a metallic layer or localized electron oscillation in metallic nanostructures or nanoparticles. Plasmonics represent a relative new bearing that has as a ground state properties of collective electron excitation, which are referred to likewise as surface plasmons, in films or nanostructure of noble metals (Au, Ag, Cu) [2], [3]. Key examination of plasmonic uncovers that their interesting properties can be interpreted in different useful applications, from identification at single molecule level to high resolution optical imaging beneath diffraction point or upgrade transmission through a variety of subwavelength holes.

Plasmons are collective oscillations of the electron cloud of a metal, that is to say plasma oscillations; The latter are quantified in the same way as photons or phonons, which leads to defining them as quasi-particles. When these plasmons are excited at the plane interface between a metal and a dielectric, this is called surface plasmon. Another interesting case is that of the metallic nanoparticles for which the oscillation of the plasma takes place throughout the volume of the nanoparticle rather than strictly at the metal / dielectric interface.

At a plane interface between a metal and a dielectric, such as a mirror, it is possible to excite with photons a plasmon propagating like an electromagnetic wave on the surface of the metal (Figure 1) And which is then called delocalized surface plasmon or propagating surface plasmon (PSP).

Hy

Figure 1: Representation of the propagation of a surface plasmon at the interface between a metal and a dielectric of respective permittivity εm and εd, the field lines induced by the propagative wave of the oscillation of the local charges are presented under Curved arrows. The projection of the different components of the plasmon field is presented on the right, the electric field E is in the plane (x, z) and the magnetic field H is along the y-axis.

In the event that the wavelength of the incident light is larger than the dimension of the nanoparticle, the electrical element of the radiation occurrence results in a dipolar distribution of the electronic cloud. Because of its localized character, the electrical charge is going to oscillate at a specific frequency. Despite what might be expected, if the frequency of the incident light matches the frequency of the surface electron oscillations, photons begin to interact resonantly with the nanoparticle. As a consequence of this interaction, metallic element nanoparticles will show a strong optical phenomenon called, absorption band, in the UV-Vis region, that isn't present within the bulk metal spectrum [4],[5],[6].

Surface plasmon resonance occurs when polarized light strikes an electrically conducting surface at the interface between two media. The resonance condition is built up once the frequency of incident photons matches the natural frequency of oscillatory surface electrons against the restoring force of positive nuclei. LSPR excitation leads to wavelength-selective absorption with extraordinarily giant molar extinction coefficients ~10-11 cm-1 [8] resonant Rayleigh scattering [9],[10] with an associate degree to that of 106 fluorophores [11] and increased local electromagnetic fields close to the surface of the nanoparticle [12].

1.3 Applications of nanostructures in developing of sensors

The excitation states of surface plasmons are in very sensitive changes within the dielectric constant of surrounding medium. In these condition optical biosensors bases on plasmonic transducers represent a powerful alternative in non-labeling analysis of biomolecular interaction [13]. Notwithstanding the colossal aforementioned favorable circumstances, use of Kretschmann pure mathematics for excitation of SPR makes an arrangement of inconveniences, for example, limitation of the SPR sensors to detection of a little array of biochemical response and therefore the chance of creation in smaller than usual scale.

Keeping in mind the end goal to conquer these detriments, the enthusiasm of scientists in optical nano-sensors in light of LSPR in metallic nanostructures has expanded [14],[15],[16],[17]. The instrument at the base of LSPR detecting is the alteration of the electromagnetic field decay length after the absorption of molecules on the metallic surface. More particular on account of metallic nanoparticles, the electromagnetic field has a working mode fundamentally the same with the engendering of SPPs (Surface plasmons polariton) on a level metallic film and empowers the transduction of chemical binding events into a quantifiable wavelength shift of the extinction peak. Regardless of exhibited proficiency of LSPR biosensors in light of nanoparticles in laboratory [18],[19],[20], the manufacture of biosensor devices based on nanoparticle and LSPR still remains a troublesome errand and requires essential enhancements with respect to spatial resolution, reliableness and monodisperse in size and form of metallic nanoparticles.

Surface plasmons have been used for some time, to enhance the surface sensitivity of many spectroscopic analyses together with fluorescence and Raman scattering. The increase in force of the Raman signal on specific surfaces happens as a result of an upgrade in the electric field gave by the surface. At the point when the incident light in the experiment strikes the surface, localized surface plasmons are excited. The field enhancement is greatest when the plasmon frequency is in resonance with the radiation. For scattering to occur, the plasmon oscillations must be perpendicular to the surface, in the event that they are in-plane with the surface, no scattering will occur. The subsequent spectrum of Raman scattering comprises in a series of bands that are comparing to vibrational or rotational transition particular to the structure of particle, subsequently giving a certain chemical "unique mark" extremely valuable for analyze recognizable proof. In spite of the immense point of interest of high specificity, the Raman scattering procedure is an extremely weak phenomenon as simply approximation of 1 to 106-1010 photons are inelastically scattered [21],[22],[23].

Therefore, by utilizing plasmonic nanostructures, for example, gold and silver nanoparticles, with plasmon resonances situated inside of the wavelength extent used to energize Raman modes, that are the visible and near-infrared regions of the spectrum, scientists had the capacity distinguish chemical “fingerprint” of a solitary molecule because of increase the Raman signal up to 14 or 15 order of magnitude [24],[25].

Surface Enhanced Raman Scattering (SERS) is a phenomenon that was discovered in 1974 by M. Fleischmann et al. [9], who observed a very intense Raman signal of pyridine adsorbed on the surface of a rough silver electrode. In the first place, the enhancement of the signal was attributed to the simple increase in specific surface area of ​​silver under the laser beam, due to the surface roughness of the electrode. However, such a hypothesis was not enough to explain such a large increase in the signal, and it was only later that the phenomenon was explained electromagnetically [10,11] and chemically [12,13]. Thus, it can be seen that any molecule adsorbed on a nanostructured metal surface has its Raman signal enhanced by a very important factor, typically of the order of 106 and possibly up to 1012. Such amplification factors have made it possible to detect and to observe single molecules. [14,15,16]

The power of SERS lies in its ability to identify chemical species and obtain structural information in a wide variety of fields including polymer and materials science, biochemistry and biosensing, catalysis, and electrochemistry. The enhancement takes place at a metal surface which has nanoscale roughness, and it is molecules adsorbed onto that surface which can undergo enhancement. Typical metals used are gold or silver – preparation of the surface can be through electrochemical roughening, metallic coating of a nano-structured substrate, or deposition of metallic nanoparticles (often in a colloidal form). Many researchers create their own SERS substrates, but commercially available kits offer a more routine approach. SERS is a highly sensitive and selective technique for use in the detection of biological samples. SERS it’s a vast topic but in terms of biosensing has been reviewed in great detail buy other researchers [El-Ans ary, A., and Faddah, L. M., Nanotechnol Sci Appl (2010) 3, 65. Hudson, S. D., and Chumanov, G., Anal Bioanal Chem (2009) 394, 679]. SERS biosensors are used in detection of various biological samples and diseases, including various cancers, glucose detection and other. SERS biosensing has been used to detect and identify small molecules, nucleic acids, lipids, peptides, and proteins, as well as for in vivo and cellular sensing. SERS detection of biomolecules has been accomplished in both intrinsic and extrinsic formats. In intrinsic SERS biosensing, the molecular signature for the analyte of interest, such as small molecule, DNA strand, or protein, is acquired directly. In extrinsic SERS, the analyte or interaction of interest is associated with a molecule with an intense and distinguishable Raman signature, traditionally a commercially available fluorescent dye, and it is the SERS spectrum of the tag that is used for sensing or quantification. In either format, SERS has unique advantages for biosensing.

Chapter 2

Ochratoxin A – Introduction

Chapter 2 Ochratoxin A – Introduction

Ochratoxin A {N-[(3R)-(5-chloro-8-hydroxy-3-methyl-1-oxo7-isochromanyl) carbonyl]-L-phenylalanine} is a quite stable nephrotoxic molecule, able to resist to most food processing steps. Aspergillus ochraceus and Aspergillus carbonarius are considered as the most frequent fungal contaminants of several agricultural products. They are able to produce several polyketide-derived secondary metabolites including mycotoxins such as Ochratoxin A (OTA), which has been found as a common contaminant of cereals, coffee beans, cocoa, nuts, beer, spices, dried fruits, animal organs and also in certain wines, particularly those from Mediterranean area. The International Agency for Research on Cancer, knowing the teratogenic and immunotoxic effects in humans [3, 4 Cruzaguado], has classified OTA as a potential carcinogen (group 2B) for humans. At present, regulatory limits for OTA exist in many countries, and testing of products is carried out at central testing laboratories. European Union introduced limits for the levels of OTA in raw cereal grains (5 mg/kg), products derived from cereals (3 mg/kg), dried fruits (10 mg/kg), roasted coffee and coffee products (5 mg/kg) and grape juice (2 mg/kg) (EC No 123/2005). The European Food Safety Authority has also set a maximum level for OTA of 2 mg/kg for all types of wine.

Ochratoxin has been known for a long time as nephrotoxic4,5,6 and the absorption of contaminated food or prolonged exposure to a contaminated environment can lead to kidney failure in children, man. This toxin is also genotoxic and is therefore potentially carcinogenic7. Moreover, its elimination is very slow, resulting in an accumulation effect in the case of repeated exposures, which increases the associated risks.

Ochratoxine A Mw = 404 Da

Figure 2.1: Chemical formula for ochratoxin

The structure of ochratoxin A (OTA) is shown in Figure 2.1. We can see the presence of two aromatic rings as well as a carboxylic group, an amide group and a ketone. All these groups will offer an intense spectral signature because they have a strong Raman scattering cross section.

In most of the cases, to detect this specific molecule (OTA) involves conventional techniques for analysis, such as high-performance liquid chromatography (HPLC), thin-layer chromatography (TLC), and gas chromatography (GC), coupled to ultraviolet/visible, fluorescence or mass spectrometry (MS) (Turner, Subrahmanyam, & Piletsky, 2009). All of these procedure steps are very expensive and take a lot of or too much time. Another problem is that protocols are still laborious and require trained personnel.

In the past decades, immunochemical assays combined with new technologies have been proposed for rapid quantitative or semi quantitative analysis of OTA in food and beverages (Van der Gaag et al., 2003). They include enzyme immunoassays (De Saeger, Sibanda, Desmet, & Van Peteghem, 2002; Radoi, Dumitru, Barthelmebs, & Marty, 2009), fluorescence polarization immunoassays (Zezza, Longobardi, Pascale, Eremin, & Visconti, 2009), and immunosensors (Alarcón et al., 2006; Prieto-Simón, Campàs, Marty, & Noguer, 2008; Radi, Muñoz-Berbel, Cortina-Puig, & Marty, 2009; Ricci, Volpe, Micheli, & Palleschi, 2007). In terms of stability and cost for analysis, the use of antibodies remains a disadvantage even though it has good sensitivity and selectivity against OTA. The creation of antibodies is linked to animal immunization and is a very laborious process that can take several weeks. Another big problem is the dependence of temperature of antibodies, which in many cases leads to degradation upon contact with surfaces, producing inconclusive data and results.

At present, ochratoxin is detectable using an ELISA9 assay with a detection limit of 0.5 ppb and quantitation of the toxin concentration up to 10 ppb. This test is therefore perfectly suited to quality control in the food industry. Such tests are available on the market and are commonly used. However, ELISA tests take several hours to complete. Therefore, we propose to develop a SPRi / SERS protocol that would be faster and that would allow in the near future simultaneous detection of other toxins. In the ELISA test, the probe used to capture ochratoxin is a monoclonal antibody. These antibodies are expensive and their stability is limited over time. In addition, these antibodies have a very large size (15 nanometers) and cannot therefore be used in SERS because in this case the target would be outside the zone of exaltation which is only a few nanometers. We therefore decided to focus on a probe system called aptamer and consisting of an oligonucleotide.

Objectives for SERS specific Detection of Ochratoxin A (OTA)

Substrate Au 5nm + 2) single strand DNA + 3) Ochratoxin A rough surface

2.1 Specific Aptamer for Ochratoxina A

An aptamer is a single strand of DNA (ssDNA), also called an oligonucleotide. Aptamers are developed in order to maximize their interaction with a target molecule. They can thus be substituted for other probes and in particular antibodies or their fragments. The DNA (deoxyribonucleic acid) consists of bases (Adenine, Cytosine, Guanine, Thymine (Figure 2)), linked by sugar, deoxyribose, and by phosphate groups. A single strand of DNA is therefore characterized by a sequence of ATGC bases, shown in Figure 2.2.

Adenine Thymine Guanine Cytosine

Figure 2.2: Structures of the different bases composing the DNA

These bases are complementary and bind in pairs by forces of Van Der Walls. Adenine is complementary to Thymine and Guanine of Cytosine. By virtue of these electrostatic forces, each strand of DNA can bind to another complementary strand thus forming a double helix, discovered by Watson and Crick. There is a chirality in the structure of DNA, which makes it possible to define a canonical sense of reading. A sequence begins at the free phosphate group located 5 'from the deoxyribose which begins the sequence and ends with the free alcohol located on the 3' carbon of the terminal deoxyribose. In order that the two helices can interlock and form a double helix, each base is associated with its conjugate base and the order of the bases is reversed from 5 'to 3'. Thus, on the example shown in Figure 2.3, the complementary sequence of ACTG becomes CAGT.

Phosphate

Desoxyribose

Figure 2.3: Structure of a single strand of DNA of sequence ACTG (left) and pairing with its complementary sequence CAGT giving a double strand of DNA (right).

For the detection of ochratoxin, an aptamer was selected by SPR method. During these tests, several sequences were used and the one with the greatest affinity with the target molecule was chosen to be used within the sensor. This sequence is as follows and has been proposed in references [25],[26]:

5’ GAT-CGG-GTG-TGG-GTG-GCG-TAA-AGG-GAG-CAT-CGG-ACA 3’ (36 nucleotides mw 11.5 kDa) Thiol en 5’ (C6).

Composition : A: 8/36 = 22% G: 17/36 = 47% T: 6/36 = 17% C: 5/36 = 14%. A chain of six carbons, terminated by a thiol, is attached to the 5 'end of the aptamer. This end can thus cling to the surface of the gold to functionalize our sample.

Chapter 3

SERS Detection of Ochratoxin A on rough Gold 5 nm

Chapter 3 SERS Detection of Ochratoxin A on rough Gold 5 nm

3.1 Fabrication of rough gold 5nm

Firstly, we made a SERS substrate, which does not require any lithography technique and consists of a standard glass microscope slide on which we deposited 5 nm gold by thermal evaporation. As the gold badly wets the glass, nano-islets of gold are formed on the surface of the lamella (Figure 3.1) forming a SERS substrate. The reason for using such substrates is that they are easy to produce, that their manufacturing cost is low and that they are very simple to use. Thus, these substrates are adapted to the numerous tests necessary for the definition of the functionalization protocols. We have therefore turned to this choice in order not to have to use many lithographed samples which are longer to manufacture.

3.2 Specific SERS detection on rough Gold 5nm

In a second step, the surface of the rough gold film was functionalized with the aptamer, by dropping a drop of concentrated solution at 10μM overnight before rinsing with distilled water. A schematic diagram is presented in Figure 3.1.

a) b)

Figure 3.1: (a) SEM image of a rough gold film used as SERS substrate. (b) Schematic diagram of the surface functionalization of our SERS substrate. The aptamers will allow the capture of the target molecule on the surface.

3.2.1 Direct detection on rough Gold 5nm

Before carrying out a detection test, we wanted to characterize the target molecule in Raman and SERS. The first test was taken by Raymond Gillibert a colleague of mine from University Paris 13. We have therefore performed a Raman measurement of the solid ochratoxin powder and we have also deposited a drop of 1 mM ochratoxin solution on a rough gold substrate before rinsing the latter and performing a SERS measurement. The spectra are shown in Figure 3.2, 3.3. The SERS spectrum and the Raman spectrum have common bands and the bands are widened in SERS.

Figure 3.2: Raman on powder.

The SERS and Raman spectra were acquired on a spectrometer from Horiba Scientific: a Xplora at 660nm. The SERS signals were collected using a 100× objective with a numerical aperture of 0.9. Acquisition time for one spectrum was 60 s with 200 μW power. Measurements were done ten times on different points to analyze the reproducibility of the signal. SERS signals were then averaged.

Figure 3.3: SERS on rough gold

We deposited an ochratoxin solution of 10 μM concentration on the rough gold surface functionalized with the aptamer. The ochratoxin was diluted in 1M KCl solution, so that the solution imposed high ion pressure on the aptamer, leading to its unfolding. An incubation time of one hour has been applied so that the ochratoxin molecules have time to cling to the aptamer. The sample was then thoroughly rinsed with distilled water to remove any non-specific adsorption of ochratoxin molecules on the substrate, as well as saline buffer. The sample was then dried and characterized in SERS.

After that we tried to see if changing the dilution of ochratoxin A can deliver changes to the spectra in any way. In addition to KCl, we used other buffers to solve the ochratoxin, we tried H2O and TBS. In terms of concentrations used same the same parameter of 10 μM. The spectras can be seen in figure 3.4

Figure 3.4: Spectra’s of OTA directly on Au 5nm surface. Black: Raman on powder, Red: SERS from literature, Green: OTA in H2O buffer, Blue: OTA in TBS buffer.

The SERS OTA and Raman OTA Powder are spectra’s obtained by my colleague and coordinator Raymond Gillibert in Paris. He was already studying the detection when I arrived in Paris in 1 November 2017. I will be using them few more other times as a comparison factor with my results and his. We can see that the SERS spectra’s look almost the same. SERS of Raymond was also done in water buffer.

The result shows us that the molecule can be detected on the gold 5nm rough surface. We can see the bands of ochratoxin A from 1562 cm-1 and 1268 cm-1. We can see also the bands from 566 cm-1 and 960 cm-1 but are much lower.

3.2.2 Specific detection on rough Gold 5nm

Step 1. Detection DNA different buffers

The spectra plotted in Figure 3.5 are derived from quantum molecular simulations using the Density Functional Theory (DFT). They were carried out by Shivalika Tanwar, PhD student in the CSPBAT laboratory, working on molecular modeling and crystallography. We used these spectra as a basis for interpretation of the Raman bands of aptamers. Another bibliographic source corresponds to the work of C. Otto et al. [27] (1986) which propose a detailed assignment of the different bands of the DNA bases visible in Raman and SERS.

Figure 3.5: Raman spectra simulated by DFT of the bases adenine, cytosine, guanine, thymine.

Finally, the work of B. Prescott et al. [28] allowed an analysis of the spectra of strands of DNA in different conformations. We used these results to perform an allocation of the most intense bands of our spectra (Table 3.1).

Table 3.1: Attribution of the main Raman bands of our aptamers

If the SERS and Raman spectra of the ochratoxin aptamer are compared, a large number of spectral differences can be observed. Indeed, we see in SERS the disappearance of certain bands. To explain such differences, it can first be assumed that the DNA strand undergoes a conformational change when it is dried and covalently attached to the gold surface. It can thus be understood that there is a slight shift of the Raman bands as well as a change in their relative intensity. But this hypothesis can not explain so great a difference between spectra. We know that our DNA strand making 36 nucleotides has an unfolded length of 12 nm (about one third nanometer per base[29]). However, the SERS intensity decreases exponentially when the molecule moves away from the surface over a distance of the order of a few nanometers. Thus, if the DNA strand is orthogonal to the surface, only the first bases of the strand will be visible, which explains the reduced number of Raman bands observable on the SERS spectrum. This behavior can also be explained by the fact that the selection rules in SERS are different from those in conventional Raman scattering.

Figure 3.6 First set of buffers for the DNA

In figures 3.6, 3.7 We did two sets of tests of different buffers for the aptamer connection to the surface. In the first set we used H2O, PBS (phosphate-buffered saline), HEPES (4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid), TBS (tris-buffered saline) and KCl (potassium chloride) and in the second test H2O, PBS (phosphate-buffered saline) and DSB. We used 20µl drops to cover the gold rough surface. We let it sit for 1 hour and after that we rinsed with water and dried the surface with NO2. We did this to see which helps better the aptamer to bind to the gold better. But we came to the conclusion that just using water is ok, because the buffers included some major modifications in the spectra of the aptamer.

Figure 3.7 SERS spectrums of the second set of buffers for the DNA

Step 2. Specific detection OTA

The next step in the detection of Ochratoxine A using rough gold surface was the addition of the ochratoxin over the sets of aptamer funcionalised samples. We used 20µl drops to cover all the surface of the rough surface. We let it sit for 1 hour and after that we rinsed with water and dried the surface with NO2. On this step we tried OTA in HEPES and H2O buffers. Figure 3.8, 3.8

Figure 3.8 SERS spectra of Ochratoxine in HEPES added over the DNA

We see that in the second step we have better signal. We were thinking that maybe the cause of this is because the DNA and OTA from the first step was removed from frige -20 for more than a month. We know that DNA degradates in time if it is not kept in the fridge. So the second set aquisition came from this idea that we should try to use a new fresh set of aptamer and OTA.

Figure 3.9 SERS spectra of Ochratoxin in H2O added over the DNA

The SERS spectra were acquired on the spectrometer from Horiba Scientific, an Xplora at 660. The signals were collected using a 80× objective. Acquisition time for one spectrum was 60s with 200 μW power. For each diameter, measurements were done nine times on different points to analyze the reproducibility of the signal. SERS signals were then averaged

In Figure 3.10 we have the comparison of the spectra’s obtained by me in H2O solvent in comparison with the ones already obtained by my colleague. We have the spectra of Ochratoxine direct on surface, and the two subsequent of DNA on substrate and DNA plus OTA added. This way we can see which bands are from the DNA and which are from ochratoxin in the final spectra.

Figure 3.10 SERS spectra’s of OTA and DNA alone in comparison with SERS of ochratoxine over DNA final spectra in H2O solvent.

Chapter 4

SERS Detection of Ochratoxin A on flexible substrate of gold nanocups

Chapter 4 SERS Detection of Ochratoxin A on flexible substrate of gold nanocups

4.1 Fabrication of gold nanocups on PDMS

We create a new flexible substrate of gold nanocups on a PDMS base. In terms of fabrication of the nanocups we used 2 steps. Firstly, we created the polystyrene monolayer through the well-known method of fabrication convective self-assembly, after that we poured the PDMS liquid phase and we let it dry. Next we pealed of the PDMS and the result was the nanohole imprint on which we deposited the 50 nm layer of gold with physical vapor deposition method.

4.1.1 Convective self-assembling fabrication method

The initial phase in the manufacture procedure is the self-assembling of micro or nano-spheres in a monolayer on a solid substrate. Fischer and Zingsheim reported in 1981 surprisingly the manufacture of a monolayer of 312 nm polystyrene spheres by deposing a colloidal solution on a solid substrate and letting it to evaporate. Amid the dissipation process just a little region of monolayer was formed. Keeping in mind the end goal to create bigger range of monolayer of polystyrene spheres the spin coating deposition method was used. The spin coating parameters, like rotation speed and time, are vital and delicate in self-assembling particles in a controllable way. Therefore, Nagayama et al. utilized a convective self- assembling method, namely the dip coating, for creating a monolayer on a large area [27],[28].

During the convective assembly, the dispersed particles are brought together and crystallized in thin wetting films. By dragging the meniscus of a fluid colloidal suspension over a hydrophilic surface at constant speed, the assembly is produced. A compensation for the water evaporation at the contact line area is created by the continuous flux of solvent from the thicker part of the layer. Convective transport of particles toward the already ordered array causes the growth of an ordered layer. This convective flow carries the suspended particles toward the 3 phase contact line. As the layer of continuous 2D lattice grows, the substrate is withdrawn together with the already formed layer. A schematic representation is shown in Figure 4.1:

Figure 4.1: Schematic of the drying region of a thin evaporating film. Polystyrene monolayer grows at the 3 phase contact line where water evaporates.

A homogeneous layer grows continuously if the withdrawal rate is equal to the rate of layer formation. At fixed evaporation rate and volume fraction, a uniform crystalline monolayer with a defined thickness is obtained. Raising the substrate velocity will produce a monolayer or diminishing the volume fraction of the suspension results in an incomplete layer. Conversely, diminishing the substrate speed alternately expanding the volume fraction will result in the formation of multiple layers. Besides substrate velocity and volume fraction, the growth velocity is influenced by the evaporation rate which depends on the temperature of the colloidal suspension and can therefore be controlled via the substrate temperature.

Another important aspect in creation of the monolayer is the hydrophilicity of the glass surface. So before deposition we treated the glass in ultraviolet ozone in a cleaning system (PSDP-UVT, Novascan) for 20 minutes.

4.1.2 Gold nanocups substrate fabrication method

To obtain the polystyrene nanolayer we used 4 different colloids of PS with sizes: of 527 nm (PS 527), 600 nm (PS 600), 719 nm (PS 719) and 836 nm (PS 836) diameters. The depositions were obtained at room temperature as we did not have any cavity to close over the deposition machine. We only controlled the speed of the deposition with the velocity of 20-30µm/s of the bottom glass and the temperature of the bottom layer (20-25˚C). We fluctuated the speed to obtain the best largest surface of monolayer polystyrene.

The next step was to pour over slowly the PDMS elastomer mix in liquid form over the glass with polystyrene nanolayer filling all the voids in the monolayer. We used a ratio of 10:1 of polymer precursor and curing agent and we cured in an oven at 60˚C for 1 hour. After that we cut the edges of PDMS and peeled it of gently so that only the PDMS comes off and the polystyrene remains attached to the glass surface. The result was a nanocups array surface engraved in PDMS. You can see the steps of production in Figure 4.2:

Figure 4.2: Schematic illustration of fabrication steps of the flexible 3D gold nanocups platform, Steps: (A) Deposition of PS on glass substrate, (B) Pouring PDMS onto the deposited PS, (C) Peeling-off the PDMS imprint, (D) Obtaining the flexible PDMS substrate with impregnated nanocups, (E) Coating the as-prepared flexible nanocups with gold film to obtain flexible plasmonic substrate.

Third and final step was the gold coating of the top side of the PDMS mold with a thick layer of 50nm gold. The procedure was by thermal evaporation in high vacuum using a Prevac deposition equipment (Prevac, Poland).

As a result, we obtained a large-area flexible ordered gold nanocups array platform. In Picture 4.3 you can see how the nanocups reflects and transmits light under the optical foto camera. You already can see the dispersion of light on thin films.

Picture 4.3: Left photo: nanocups on PDMS see through, Right photo: bending of PDMS and beautiful light dispersion.

4.2 Characterization

4.2.1 Atomic Force Microscopy (AFM)

The next step was to see the results of our work. We chose first the AFM technique where with help of my colleagues from ICEI we obtained images of two different samples 600nm and 719nm. The images were obtained with an Atomic Force Microscope system (Alpha 300 A, Witec). We can observe that the grating remained the same of the polystyrene nanolayer in Picture 4.4. The distance between 2 centers of holes is the polystyrene size, so 600nm and 719nmfrom center to center. We also can see that we have some imperfect lines that are higher which is due to the PS deposition, which we know that it deposes in sectors which has some zones where the deposition has gaps

AFM 719nm AFM 3D 719 nm AFM 600nm

Picture 4.4: AFM images of the two sizes of nanocups (719 and 600nm)

In terms of gold deposition, we concluded that the gold deposited on the walls of the holes and it thickened them, but also deposited in the bottom of the hole. A SEM image of a slice should have been done so that we can see also the real depth of the wholes and the height of the walls of the nanocups.

4.2.2 Reflectivity

We measured also the optical reflection spectra of the gold nanocups obtained for the sizes: PS 527 and PS 600 which are shown in Figure 4.5. I also represented the spectra of pure gold obtained with the help of 100nm gold surface. We can see that by changing the size of the nanocups we can change the response of the surface and to amplify the SERS signal. With increasing the size of nanocups the peak shifts to higher wavelengths. This could be very beneficial to make Nano surfaces in accordance with the wavelength of the SERS microscope so you have better enhancement for your specific laser. Spectra’s were acquired with Spectrophotometer UV-VIS (V-530) with reflectivity Module (SLM-468S).

Figure 4.5: Reflectivity spectra of Au100nm and nanocup size of 527nm and 600nm

4.2.3 Amplification of SERS signal of thiophenol according to the nanocup size

On my first week in Paris I tested the detection of thiophenol on my 4 sizes of nanocups substrates to see if it can detect it and to see maybe which is better in amplification. As a result, we can see in Figure 4.6 the spectra is almost the same. We get all the thiophenol major bands in all 4 spectra’s which are an average of 10 acquisitions.

Figure 4.6: SERS spectra of thiophenol on the 4 nanocup surface sizes (527,600,719,836) baseline substracted.

The SERS spectra were acquired on a spectrometers LabRam 300 at 633 nm. The signals were collected using a 80× objective. Acquisition time for one spectrum was 60 s with 200 μW power. For each diameter, measurements were done ten times on different points to analyze the reproducibility of the signal. SERS signals were then averaged and in terms of signal amplification, we can see that the best amplification we get with the 527nm size surface and the 719nm one. The SERS signal strength in dependence of the size of nanocups can be seen in Figure 4.7:

Figure 4.7: SERS signal according to the size of nanocups

4.2.4 Mercathophenil boronic acid detection test

At the beginning of my experiments, after I obtained the gold nanocups, me and my professor wanted to do a detection test to see if the surface is good for detection. We decided to use a solution of 1mM MBA (4-mercathophenil boronic acid) as is a good molecule that is hard to detect through Raman, but can be easily detected with SERS and has already been studied in our laboratory. As we can see in Figure 4.8 after the addition of the MBA the higher peaks of PDMS disappeared and we could see the most important bands of MBA at 1074 cm-1 and 1573 cm-1 as it is shown in literature. We used the sample with nanocups 836 nm.

Figure 4.8: Raman spectra PDMS(red) and MBA Mercathophenil boronic acid(black) on nanocups type 836nm

The band at 1573 is visible through Raman also, but the band of 1074 can be seen only through SERS. So we concluded that we have surface amplification and that the substrates are what we wanted to create. The only problem that we did not consider is that this molecule is one that gives good distinctive peaks and they have high intensities. But for molecules as ochratoxine and DNA where there appear a significant number of additional peaks with lesser intensity we had a much difficult way to identify the specific peaks of our molecules. Spectra’s were acquired with a Raman mobile spectrometer Micro Raman System (R-3000, Ocean optics) with a 785 nm laser.

4.3 Specific SERS detection of Ochratoxin A on gold nanocups

4.3.1 Direct detection on gold nanocups

Before carrying out a detection test, we wanted to characterize the target molecule in Raman and SERS with the gold nanocups on PDMS. We have therefore performed a SERS measurement, a drop of 1 mM ochratoxin A solution was deposited on a gold PDMS substrate. An incubation time of one hour has been applied so that the ochratoxin molecules have time to cling to the gold surface. The sample was then thoroughly rinsed with distilled water to remove any non-specific adsorption of ochratoxin molecules on the substrate. This spectra + SERS of rough 5nm gold are shown in Figure 4.9. The result shows us that the molecule can be directly detected on the gold nanocups. We can see the bands of ochratoxin A from 1562 cm-1 and 1268 cm-1 with high intensity.

Figure 4.9: Spectra of 1mM-OTA on nanocups, compared with reference spectra of OTA on gold 5nm

The SERS spectra were acquired on the spectrometers from Horiba Scientific: Xplora at 660. The signals were collected using a 80× objective. Acquisition time for one spectrum was 60 s with 200 μW power. For each diameter, measurements were done 10 times on different points to analyze the reproducibility of the signal. SERS signals were then averaged.

4.3.2 Specific detection of Ochratoxin A on PDMS

After we saw that the detection with gold nanoholes can be done, we turned to the second part of the experiment. We first treated the substrate in UV ozone, after that we functionalized the surface first with a solution of 10µM of singe strand DNA in water. After a time-lapse of 1 hour we rinsed with ultrapure water and did the spectras on Horiba microscope. Then we added the ochratoxine for another hour. In Figure 4.10 we have the spectra of the sample before and after the addition of ochratoxine. I also added the spectra of OTA directly on nanocups surface and the spectra obtained on Au 5nm rough surface. This way we can easily compare the peaks of ochratoxine.

Figure 4.10: Comparison between: Red: SERS spectra of OTA direct on surface – Green: SERS of DNA on substrate – Blue: SERS OTA after the DNA on nanocups – Black: SERS of OTA on Au 5nm as reference

The result shows us that the molecule can be detected with the nanocoups. We can see the bands of ochratoxine from 1562 cm-1 and 1268 c. We can see also the bands from 566 cm-1 and 960 cm-1 but are much lower.

4.3.3 Influention of the nanocup sizes sizes in DNA detection

As we wanted to precisely see if the sizes of PS spheres influence the detection of DNA and to be able to better detect the DNA we did an additional experiment with all the 4 sizes. We could determine that the bands of DNA could be seen in all 4 samples, but we decided to continue in experiments first with 719 size because I had more samples and also with 836nm. Compared to the first approach with the Au 5nm we had better intensities and the specific peaks of DNA 1093cm-1, 1327cm-1 and 1573cm-1 could be easily identified in Figure 4.11.

Figure 4.11: SERS spectra of DNA on all the 4 types of nanocups in comparison to reference.

Chapter 5 Conclusion

In this paper we have seen that ochratoxin is a highly dangerous mycotoxin whose concentration must be regulated and controlled in any product that can contain it. Despite the existence of an ELISA test, we have chosen this target because it represents a major interest for quality control in agriculture-food and we wish to validate the instrument developed for a project in future for the detection of toxins. We propose a detection with an aptamer, both for reasons of stability and cost and for its small size allowing a better response in SERS. We have characterized the chosen aptamer in classical Raman and SERS.

Finally, we have demonstrated that it is possible to detect ochratoxin by SERS method over two types of substrates. This detection has been demonstrated on gold rough film substrates on glass for reasons of simplicity and availability of substrates. The other method implied a more laborious method but the gold nanocups were formed on a flexible PDMS substrate that can have many biological applications. We can also see that we do not have a real identification of the target molecule but that its detection is done by observation of modification of the spectrum of the aptamer.

Appendices

Appendix 1. Raman spectroscopy

Raman spectroscopy is a useful technique for various chemical analyses. In Calcutta on February 28, 1928, Professor C. V. Raman was the first to observe the scattering of light at different wavelengths not present in the incident light source when a monochromatic incident light beam was applied on a sample. The Raman spectroscopy is a spectroscopy technique used to observe vibrational, rotational and other low – frequency modes of the system. Electric and magnetic components of a radiation cause the electrons to be in the intense field, at the point when electromagnetic radiation touches the molecule, and interacts with the electron cloud and the bonds of the molecule. In this way the electrons of molecule/particle will oscillate with the same frequency as the incident radiation.

The emanated radiation is elastic scattered radiation and is known as Rayleigh scattering. When electrons of a molecule are excited by a low-energy incident beam to a virtual excited state below the next-highest electronic state and fall back to another vibrational sate different from the incident one, takes the Raman scattering place. Common are the transitions from a vibrational state to another. Thus, the photons emitted by relaxation of the electrons have completely different energies and therefore different wavelengths compared to those of incident excitation beam.

In Raman spectroscopy, the sample is irradiated with an intense monochromatic beam (laser). The incident wavelength does not have to be one that is absorbed by the molecule, although it can be, as in the case of resonant Raman scattering. Excitation is considered to take place to a virtual level, which is not the actual level of the molecule (Figure A1.1) and corresponds to the laser frequency.

Figure A1.1: Energy-level diagram showing the states involved in Raman signal. The line thickness is roughly proportional to the signal strength from the different transitions Reproduced from [43].

The basic components of Raman system include a light source (laser), collection optics to gather the Raman – scattered light, and detection system (Spectrometer, CCD). The advantages of confocal microscopes are: increased resolution, by using spatial pinhole to eliminate out – of – focus light in specimen that is thicker than the focal plane, thus the light rays from outside the focal plane will not be recorded; possibility of recording three dimensional data sets; focus on sample. The combined Raman spectrometer with confocal microscope presents the advantages of taking complete Raman spectrum at every image pixel, extraction of relevant information (peak intensity, position etc.), and the display of taken information as image. Figure A1.2 presents the main components of the confocal Raman system used in our lab, from Witec, Alpha 300 R model, and the picture of the system.

Figure A1.2: The main components of the confocal Raman microscope 300 R and the beam path (right) and picture of the confocal microscope (left) Reproduced from [44],[45].

Appendix 2. Atomic Force Microscopy

The Atomic Force Microscopy (AFM) was invented in 1986 by Binning, Quate and Gerber. Since then the AFM has evolved into an invaluable surface analysis technique on micro- and nanoscales and even on atomic and molecular scales. With the help of AFM it is possible to image the molecular structures of real-space surfaces and even additional surface properties such as adhesion, stiffness, magnetic properties, conductivity and many more.

The operating principle of the AFM is based on scanning the surface of a sample using a sharp tip. In order to trace the surface topography are used various interactions between tip and sample as feedback mechanisms. AFM’s most widespread operating modes are the contact mode (CM) and the tapping mode (AC). In CM, a sharp tip (less than 10 nm across) is brought into contact with the sample while the repulsive force between tip and sample bends the cantilever. With the help of a highly focused beam deflection system the bending of the cantilever can be measured. (see Figure A2.1)

Figure A2.1: Principle of AFM contact mode. Reproduced from [46].

By keeping the bending of the cantilever constant, a constant force is applied to the sample while scanning the tip across the surface. The up and down movement of the scan stage is recorded as surface topography. These images highlight the edges of various topographic levels. Beside these, invaluable information is also provided by recording the torsion of the cantilever. The disadvantage of contact mode is in samples which are too soft or weakly bound to the substrate. In AC Mode this can be overcome where the cantilever is oscillated at its resonant frequency with free amplitude. By keeping the damping of the amplitude constant, the surface topography can be imaged.

Appendix 3. Related to Thesis

A. Sabau, C. Leordean, Monica Potara, Sanda Boca-Farcau, C. Farcau, S. Astilean, Fabrication of parallel plasmonic stripes through convective self-assembling of gold nanoparticles -Advanced Spectroscopy on Biological and Nanostructured Systems, 7-10th of September 2014, Cluj-Napoca (Romania)

References

[1] T Adrega, SP Lacour, Stretchable gold conductors embedded in PDMS and patterned by photolithography: fabrication and electromechanical characterization. IOP Publishing Ltd, 2010

[2] Dong Qin, Younan Xia & George M Whitesides, Soft lithography for micro- and nanoscale patterning, Nature Protocols 5, – 491 – 502 (2010)

[3] S. Paul , C. Pearson , A. Molloy , M. A. Cousins , M. Green, Langmuir−Blodgett Film Deposition of Metallic Nanoparticles and Their Application to Electronic Memory Structures, Nano Letters, 2003

[4] Eun Chul Cho†, Jingwei Xie, Understanding the Role of Surface Charges in Cellular Adsorption versus Internalization by Selectively Removing Gold Nanoparticles on the Cell Surface with a I2/KI Etchant, Nano Lett., 2009

[5] D. Matthey, J. G. Wang, S. Wendt, Enhanced Bonding of Gold Nanoparticles on Oxidized TiO2(110), Science 2007

[6] Anjeanette D. Ormonde ,† Erin C. M. Hicks, Nanosphere Lithography:  Fabrication of Large-Area Ag Nanoparticle Arrays by Convective Self-Assembly and Their Characterization by Scanning UV−Visible Extinction Spectroscopy, Langmuir, 2004

[7] Minhwan Lee, Ryan O’Hayre, and Fritz B. Prinz, Electrochemical nanopatterning of AgAg on solid-state ionic conductor RbAg4I5RbAg4I5 using atomic force microscopy, Appl. Phys. Lett. 85, 3552 (2004)

[8] Takayuki Okamoto, Ichirou Yamaguchi, and Tetsushi Kobayashi, Local plasmon sensor with gold colloid monolayers deposited upon glass substrates, Optics Letters, 2000

[9] John Melngailis, Focused ion beam lithography, Elsevier, 1993.

[1] Colson, P., Henrist, C., and Cloots, R.. Nanosphere Lithography: A Powerful Method for the Controlled Manufacturing of Nanomaterials. J. Nanomater. 2013, e948510, 2013.

[2] Kreibig, P.D.U., and Vollmer, P.D.M. Introduction. In Optical Properties of Metal Clusters, Springer Berlin Heidelberg, pp. 1–12, 1995.

[3] Raether, H. Introduction. In Surface Plasmons on Smooth and Rough Surfaces and on Gratings, (Springer Berlin Heidelberg), pp. 1–3. 1988.

[4] Haynes, C.L., and Van Duyne, R.P. Nanosphere Lithography: A Versatile Nanofabrication Tool for Studies of Size-Dependent Nanoparticle Optics. J. Phys. Chem. B 105, 5599–5611, 2001.

[5] Hulteen, J.C., Treichel, D.A., Smith, M.T., Duval, M.L., Jensen, T.R., and Van Duyne, R.P. Nanosphere Lithography: Size-Tunable Silver Nanoparticle and Surface Cluster Arrays. J. Phys. Chem. B 103, 3854–3863., 1999.

[6] Link, S., and El-Sayed, M.A. Spectral Properties and Relaxation Dynamics of Surface Plasmon Electronic Oscillations in Gold and Silver Nanodots and Nanorods. J. Phys. Chem. B 103, 8410–8426., 1999.

[7] http://nanocomposix.com/pages/plasmonics-and-nanophotonics ;

[8] Jensen, T.R., Malinsky, M.D., Haynes, C.L., and Van Duyne, R.P. (2000). Nanosphere Lithography: Tunable Localized Surface Plasmon Resonance Spectra of Silver Nanoparticles. J. Phys. Chem. B 104, 10549–10556., 2000.

[9] Michaels, A.M., Nirmal, M., and Brus, L.E. Surface Enhanced Raman Spectroscopy of Individual Rhodamine 6G Molecules on Large Ag Nanocrystals. J. Am. Chem. Soc. 121, 9932–9939., 1999.

[10] Schultz, S., Smith, D.R., Mock, J.J., and Schultz, D.A. Single-target molecule detection with nonbleaching multicolor optical immunolabels. Proc. Natl. Acad. Sci. 97, 996–1001.., 2000.

[11] Yguerabide, J., and Yguerabide, E.E. Light-scattering submicroscopic particles as highly fluorescent analogs and their use as tracer labels in clinical and biological applications. Anal. Biochem. 262, 157–176., 1998.

[12] Zhang, X., Yonzon, C.R., and Van Duyne, R.P. Nanosphere lithography fabricated plasmonic materials and their applications. J. Mater. Res. 21, 1083–1092, 2006.

[25] Galarreta, B. C., Tabatabaei, M., Guieu, V., Peyrin, E. & Lagugné-Labarthet, F.

Microfluidic channel with embedded SERS 2D platform for the aptamer detection of

ochratoxin A. Anal. Bioanal. Chem. 405, 1613–1621 (2013).

[26] Cruz-Aguado, J. A. & Penner, G. Determination of Ochratoxin A with a DNA

Aptamer. J. Agric. Food Chem. 56, 10456–10461 (2008).

9. M. Fleischmann, E. et al. Experimental verification of the SERS electromagnetic model beyond the| E| 4 approximation: polarization effects. J. Phys. Chem. C 112, 8117–8121 (2008).

10. Wokaun, A., Gordon, J. P. & Liao, P. F. Radiation Damping in Surface-Enhanced Raman Scattering. Phys. Rev. Lett. 48, 957–960 (1982).

11. Gersten, J. I. The effect of surface roughness on surface enhanced Raman scattering. J. Chem. Phys. 72, 5779–5780 (1980).

12. Grand, J. et al. Role of localized surface plasmons in surface-enhanced Raman scattering of shape-controlled metallic particles in regular arrays. Phys Rev B 72, 033407 (2005).

13. Su, K.-H. et al. Interparticle Coupling Effects on Plasmon Resonances of Nanogold Particles. Nano Lett. 3, 1087–1090 (2003).

14. Le Ru, E. C. & Etchegoin, P. G. Quantifying SERS enhancements. MRS Bull 38, 631– 640 (2013).

15. Le Ru, E. C. & Etchegoin, P. G. Rigorous justification of the |E|4 enhancement factor in Surface Enhanced Raman Spectroscopy. Chem. Phys. Lett. 423, 63–66 (2006).

16. Le Ru, E. C., Blackie, E., Meyer, M. & Etchegoin, P. G. Surface Enhanced Raman Scattering Enhancement Factors:  A Comprehensive Study. J. Phys. Chem. C 111, 13794– 13803 (2007).

[27] 15. Otto, C., Tweel, T. J. J. van den, Mul, F. F. M. de & Greve, J. Surface-enhanced

Raman spectroscopy of DNA bases. J. Raman Spectrosc. 17, 289–298 (1986).

[28] Prescott, B., Steinmetz, W. & Thomas, G. J. Characterization of DNA structures by

laser Raman spectroscopy. Biopolymers 23, 235–256 (1984).

[29] Mandelkern, M., Elias, J. G., Eden, D. & Crothers, D. M. The dimensions of DNA in

solution. J. Mol. Biol. 152, 153–161 (1981).

[27] Dimitrov, A.S., and Nagayama, K. Continuous Convective Assembling of Fine Particles into Two-Dimensional Arrays on Solid Surfaces. Langmuir 12, 1303–1311., 1996.

[28] Rossi, R.C., Tan, M.X., and Lewis, N.S. Size-dependent electrical behavior of spatially inhomogeneous barrier height regions on silicon. Appl. Phys. Lett. 77, 2698–2700, 2000.

[43] https://en.wikipedia.org/?title=Raman_spectroscopy#/media/File:Raman_energy_levels.svg

[44] http://www.witec.de/products/raman/alpha300-r-confocal-raman-imaging/

[45] http://spie.org/x17652.xml

[46] http://www.physics.ncsu.edu/wang