Technologies

SERS

Our approach is based on Surface Enhanced Raman Spectroscopy (SERS) and Electrochemical Impedance Spectroscopy (EIS).

Why these technologies?

High sensitivity and specificity
Portable format
Quantitative and qualitative analysis

The SERS analyses are based on the Raman Effect, consisting of quantification and characterization of Raman Scattering. All molecules scatter light when excited by a laser beam. We can observe Raman Scattering when the scattered light has a different wavelength than the incident laser beam. After characterising this phenomenon, we can study the chemical nature of the components analysed.

This technique is based on the analyses of vibrational states expressed via ‘Raman Shift’. It represents the difference between incident wavelength and scattered wavelength. To determine the value of the ‘Raman Shift’ this equation can be used:

Raman Scattering is an unlikely phenomenon and has a very low intensity when it happens. Using a particular conductivity surface we can enhance the Raman signal. This enhancement is based on the plasmon properties of metals. In particular noble metals are often used due to their chemical stability. In our project, we have used a surface made of gold nanoparticles (Au-NPs).

Image adapted from [2]

In this image, we can observe the Scattering mechanism. The light scattered is divided into:

Rayleigh Scattering: same scattering wavelength;
Stokes Raman Scattering: scattering wavelength greater than incident wavelength;
Anti-Stokes Raman Scattering: scattering wavelength greater than incident wavelength;

In SERS analyses, Rayleigh wavelength is removed by a holographic filter. We also consider Stokes Raman Scattering because it is more probable and less disturbed as well.

Electrochemical impedance spectroscopy (EIS)

Electrochemical impedance spectroscopy (EIS) allows for the analysis of the interface between the electrode and the solution. In applying this technique, the system is subjected to a frequency spectrum, and the impedance is measured.

The need to analyse this parameter arises mainly because, unlike ideal systems, real systems deviate from purely ohmic behaviour and no longer adhere to the canonical law
R = V(t) / I(t)
due to the presence of additional capacitive and inductive components. To obtain this value, the system is first polarised at a specific Vo, referred to as the ‘working point,’ and then a sinusoidal perturbation dV (usually with an amplitude in the order of a few mV) is applied.
This perturbation generates a current disturbance dI, with which the impedance of the system can be calculated, defined as
Z(ω) = δv(ω) / δi(ω).
This impedance is characterised by a real component with zero phase (the resistance of our system) and an imaginary component that also considers the non-ideal behaviours of the system - j / (ωC) with a phase of
π / 2.
This technique is considered non-perturbative because the currents generated within the system are sufficiently small.
For this reason, this measurement could be performed using only two electrodes while still obtaining good results. However, in our measurements, we preferred to use three electrodes to better appreciate the system’s characteristics. The frequency spectrum analysed is generally broad and can range from mHz to MHz, depending on the system being measured and the specific requirements.

Surface functionalization

SERS and EIS are based on surface-sample interactions; hence, they require the presence of sample molecules on the conductivity surface.
Using the bare surface made by Au-NPs we saw that the molecule of PFAS does not interact. A functionalization was needed in order to attract the PFAS molecule to the surface and proceed with our analysis.
Both techniques used are more efficient when the analyte is attached or really close to the surface.

Materials and methods

LAL (Laser Ablation in Liquid)

Laser Ablation in Liquid (LAL) for gold nanoparticles production is based on the expulsion of material by means of a laser pulse that irradiates a solid target immersed in a liquid solution, generally water.

This procedure is based on a sequence of processes:

laser ablation
plasma induction
exchange of energy from the plasma to the liquid
generation of the cavitation bubble
release of particles from the bubble to the aqueous solution

Image adapted from [3]

When the laser beam passes through the liquid solution only one part reaches the surface, most of the energy is spent on the excitation of external electrons and consequently heating. For this reason, the ablated material is converted to plasma phase so that due to the solution incompressibility reaches a very high density. After the interaction between the cavitation bubble and the surrounding solution, the plasma cools down and expels nanoparticles.
As a result, we obtain a red solution, due to the presence of gold nanoparticles.
In our experiments, we used LAL to get a gold nanoparticle (AuNPs) solution and then used it to create the conductivity surface.

Micro-Raman

The micro-Raman used in our experiments is a Renishaw inVia equipped with three laser sources and connected to a Leica DM-LM optical microscope.
The laser sources used for the characterization of the samples are a Renishaw RL633 He-Ne with emission at 633 nm and a Renishaw RL785 diode with emission at 785 nm.
We can see in the image a simplified diagram of the operating system.

Image adapted from [4]

The sample is initially presented under the microscope with many objectives, from 5x to 100x.
Thanks to the magnification, we can focus the incident laser radiation on a small portion of the sample, allowing the detection of very small amounts of material despite the weakness of Raman scattering.
The spectrometer has a series of holographic filters, which eliminate the corresponding Rayleigh component.
We then have a diffraction grid that divides the spectrum obtained from Raman scattering into its monochromatic components, sending them individually to a CCD detector, which exports the spectral data obtained.
The results are then processed by a software called Wire 4.4, which allows us to carry out quantitative and qualitative analyses through the graphical analysis of the spectral function.

Wire 4.4 Renishaw Software

Wire 4.4 Renishaw is a software that permits us to control the micro-Raman system and to analyse the SERS data. The collected data can be downloaded as a text file (.txt) and processed with another software. Our graphics images are created using MATLAB.

EIS hardware

Image adapted from www.palmsens.com

The EmStat Pico Development Kit features a reliable, tested, and calibrated potentiostat module that enables rapid prototyping without requiring any programming expertise. With this instrument, you can perform a wide variety of electrochemical measurements using your own sensor.
Some important features of this instrument that make it a perfect candidate to perform the analysis needed for our PFAS detection are:

Possibility of being connected and powered via USB
PSTrace software can be used to complete the measurements
Wireless connection with mobile devices

These characteristics allow us to work with the objective of creating a portable, affordable and fast sensor system for PFAS revelation.

PSTrace software

This software enables easy management of PalmSens instruments. Its interface allows you to effortlessly set the necessary parameters for measurements and then perform data fitting to identify the equivalent electrical components of your electrochemical system. The collected data can be downloaded and processed; in our case, we used MATLAB to create graphs that help better interpret the data.

Image adapted from www.palmsens.com

Functionalization

To get the correct interaction between PFOA and the gold nanoparticles, we’ve used a molecule composed of a carbonious chain (C9) to mimic the shape of our analyte, a Tiole (-SH) for the affinity with gold on one end and a Positive Guanidinium group for the affinity with PFOA on the other end. In our project, we called this molecule GV38.
The functionalization solution was composed MeOh+GV38 (1mM)

GV38 is really promising because it has dimensions similar to PFOA and permits insertions between the functionalization molecules.
The functionalization strategy is shown in this figure.

The solutions that we have tested were composed of:

milliQ water + NaCl 10 mM + PFOA
milliQ water + Hepes 10 mM +PFOA / PFOS / PFBA 1mM

Hepes stabilises the pH of the solution at 7.55 and makes AuNPs-GV38-PFOA interactions more efficients. It also serves as an electrolyte to boost the conductivity of the solution in order to allow EIS analysis.

EIS parameters

Three different EIS measurement cycles were performed using a three-electrode setup. For more information visit Engineering Sensor.
The electrodes used were:

Working Electrode: Au electrode
Reference Electrode: Ag/AgCl
Counter Electrode: Pt

In the first cycle, the functionalization molecule used was 11-Mercapto-1-undecanol, while NaCl was chosen as the electrolyte, however, this setup did not lead to significant results.

In the second cycle, the new functionalization molecule GV38 was used. In all measurements, a 10mM Hepes solution was used, and the measurements on the surface were as follows:

Measurement of the metallic surface without functionalization
Measurement of the metallic surface after functionalization (1 hour of incubation)
Measurement of the functionalized metallic surface with PFOA solution (at 0 minutes after adding the PFOA solution)
Measurement of the same metallic surface from (3) after 30 minutes
Measurement of the same metallic surface from (3) after 60 minutes

The parameters used for the measurements were:

Vdc = 0V
Vac = 15 mV
Hold time = 5 seconds
Frequency range = 100 kHz – 1 Hz

In the third cycle (where only the most significant results will be presented for practical reasons), five different prototypes were built to test the following conditions:

GV38 + 10mM HEPES
GV38 + 1mM PFOA
MeOH + 1mM PFOA
GV38 + 1mM PFBA
GV38 + 1mM PFOS

The presence of prototype 1 (where no PFAS were added) and prototype 3 (which was not functionalized) allows us to have a negative control on the measurements to confirm that the impedance variations we observe are due to the interaction between the functionalization and the PFAS, and not other external factors. The measurement protocol chosen is similar to the one used in the second cycle, with the difference that we chose to wait 40 minutes between measurements with the PFAS solution. The parameters used for the measurements were:

Vdc = 0V
Vac = 15 mV
Hold time = 10 seconds
Frequency range = 100 kHz – 1 Hz

We first performed a measurement on all the prototypes after functionalization, as well as on prototype 3, which was not functionalized. The second measurement was performed immediately after the addition of the PFAS solution, and the subsequent three measurements were taken at 40-minute intervals for all prototypes.

As the final experiment, we performed a test using ferri/ferrocyanide as redox species, hoping to amplify the phenomenon. We carried out some EIS measurements following the same protocol used in previous measurements. The parameters used for these measurements were as follows:

Vdc = 22 mV
Vac = 15 mV
Frequency = 1-100 kHz
Hold time = 10 s

The decision to use a non-zero Vdc was driven by the need to polarise the solution to allow the redox reaction to occur properly.

SERS parameters

After carrying out every EIS measurement cycle we continued using the SERS technology. This allowed us to confirm the results obtained via EIS. For more information visit Engineering Sensor.

In the first cycle, SERS technology confirmed to us that 11-mercapto-1-undecanol does not interact very well with PFOA molecules.
We analysed PFOA 1 mM + NaCl 1 mM using the laser wavelength of 633 nm.

The microscope of micro-Raman showed us the presence of aggregates, as we can see in the picture below.

The presence of aggregates means that the functionalization does not work well as it does not hold properly with the PFOA molecules.

In the second cycle and in the third cycle, SERS measurements showed us the presence of PFOA in the solution analysed.

We analysed only PFOA 1 mM + Hepes 10 mM solution.

The Raman spectra has two characteristic C-F bond peaks due to the attachment of PFOA molecules to the functionalized surface. For more information visit sensor results.

The parameters using for this measurements are the following:

Laser wavelength: 633 nm
Power: 0.1 - 10 %

Results

Sensor design and creation

For the sensor, we chose a particular design that allows us to perform SERS analysis and EIS analysis as well. The presence of a containment chamber is essential to being able to carry out analysis on solutions.
The sensor assembled is made by a conductivity surface placed over a glass plate and surrounded by a containment chamber. Two metal stripes are placed at the extremity of the electrode to monitor its conductivity.

We can see a 3D model of the sensor designed.

To create the electrode we’ve combined the technologies of Ablation Laser in Liquid for getting gold nanoparticles and spray coating to deposit it.
Between the creation of the solution and its deposition, we have carried out some tests to verify the quality of the nanoparticles. We’ve used a UV-Vis Spectroscopy to analyse the spectrum of the AuNPs solution. As we can see in the graphic, the solution reaches 0.4 absorbance intensity at 520 nm, which is the wavelength of the plasmon peak for gold nanoparticles [1]. That confirms the high stability of the solution.

After the stability verification, we proceed with the deposition using spray coating. We used a geometrical mask to follow a rectangular shape.

Finally, we have assembled the sensor using a bicomponent glue that permits high stability and heat resistance.

References

Weihong Shi et al. ≪Synthesis and Characterization of Gold Nanoparticles with Plasmon Absorbance Wavelength Tunable from Visible to Near Infrared Region≫. In: International Scholarly Research Notices 2012.1 (2012). DOI: 10.5402/2012/659043.
Mithun N et al. ≪Raman Spectroscopy for Detecting Neurological Disorders: Progress and Prospects≫. In: Theranostic Applications of Nanotechnology in Neurological Disorders. Springer, Singapore, 2023, pp. 219–250. DOI: 10 . 1007 / 978 - 981 - 99 - 9510 -310.
Enza Fazio et al. ≪Nanoparticles Engineering by Pulsed Laser Ablation in Liquids: Concepts and Applications≫. In: Nanomaterials 10 (23 nov. 2020), p. 2317. DOI:10.3390/Nano10112317.
Beatrice Campanella, Vincenzo Palleschi e Stefano Legnaioli. ≪Introduction to vibrational spectroscopies≫. In: ChemTexts 7 (19 gen. 2021). DOI: 10 . 1007 / s40828 - 020 -00129-4.

Sensor