*Proceedings* **Classification-Based Screening of Parkinson's Disease Patients through Graph and Handwriting Signals †**

**Maria Fratello 1,2, Fulvio Cordella 1,2, Giovanni Albani 3, Giuseppe Veneziano 3, Giuseppe Marano 4,5, Alessandra Paffi 6,7 and Antonio Pallotti 2,8,9,\***


**Abstract:** Parkinson's disease (PD) is one of the most common neurodegenerative diseases, affecting millions of people worldwide, especially among the elderly population. It has been demonstrated that handwriting impairment can be an important early marker for the detection of this disease. The aim of this study was to propose a simple and quick way to discriminate PD patients from controls through handwriting tasks using machine-learning techniques. We developed a telemonitoring system based on a user-friendly application for drawing tablets that enabled us to collect real-time information about position, pressure, and inclination of the digital pen during the experiment and, simultaneously, to supply visual feedback on the screen to the subject. We developed a protocol that includes drawing and writing tasks, including tasks in the Italian language, and we collected data from 22 healthy subjects and 9 PD patients. Using the collected signals and data from a preexisting database, we developed a machine-learning model to automatically discriminate PD patients from healthy control subjects with an accuracy of 77.5%.

**Keywords:** graph signal; handwriting signal; Parkinson's disease; machine learning; telemonitoring

#### **1. Introduction**

Parkinson's disease (PD) is one of the most common neurodegenerative disorders in the world, second only to Alzheimer's disease [1]. The differential diagnosis of PD is still an ongoing challenge for the scientific community: to this day, confirmation of the disease is available only postmortem and the rate of misdiagnosis is high; it has been estimated that 25% of diagnoses are incorrect [2]. The main cause of PD is the lack of dopamine production, and its main motor symptoms are bradykinesia, tremor, and rigidity [3]; neurologists rely on imaging techniques such as MRI (Magnetic Resonance Imaging), CT (Computed Tomography), or PET (Positron Emission Tomography), and patient clinical evaluations [3]. Machine-learning techniques have been studied to help the diagnosis of

**Citation:** Fratello, M.; Cordella, F.; Albani, G.; Veneziano, G.; Marano, G.; Paffi, A.; Pallotti, A. Classification-Based Screening of Parkinson's Disease Patients through Graph and Handwriting Signals. *Eng. Proc.* **2021**, *11*, 49. https://doi.org/ 10.3390/ASEC2021-11128

Academic Editor: Nunzio Cennamo

Published: 15 October 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

PD and have shown promising results. Pereira et al. presented a review on recent studies concerning computer-assisted methods to aid PD recognition [4], which included speech, gait, and handwriting analysis. This work is part of a home-monitoring project that aims to aid in PD detection through a combined analysis of graphological and vocal signals [5]. An accuracy of 98.5% was reached through the analysis of vocal data from 55 subjects: 18 healthy control subjects and 33 drug-free and newly diagnosed PD patients.

In this study, we focused on handwriting of PD subjects; handwriting requires a complex coordination of consecutive movements, and the motor symptoms of PD can provoke handwriting impairments in the size of letters, which is referred to as micrographia, and in the pressure and kinematics of the pen [6,7], together with a general difficulty in writing, which involves different graphological patterns. Since "graphology is a discipline that deals with the dynamic study of the graphic gesture" [8], we based our analysis on computational graphology. Several studies have investigated the most relevant writing features and tasks for the diagnosis of PD. Reference [9] presents the state of the art of these studies. It is possible to collect relevant information from drawings (Archimedean spiral [10–15], circles [16], meanders [13,14], etc.) and from handwritten words and graphemes. The drawing of an Archimedean spiral (spirography) is a common task for tremor and other movement disorder analysis [10]. Thanks to the development of digitizing tablet technologies, it is possible to analyze not only the offline images, but also the kinematic characteristics of the graphic signal and the pressure applied to the tablet [17,18]. "Online" data are those collected while the user writes, while "offline" data are those available after the writing is completed [19].

In the past decade, important databases have been constructed in order to study handwriting impairments in PD: the PaHaW database [11], which includes real-time data (pen position, pen pressure, and pen inclination) collected from 38 PD patients and 37 control subjects, and the HandPD [13] and NewHandPD [14] databases, which include offline images collected by Pereira et al.

Dròtar et al., analyzing the PaHaW database, obtained an accuracy of 85.61% [20]; they demonstrated the relevance not only of the on-tablet movements, but also of the in-air movements, i.e., the variation of the pen position while the pen is not touching the table. Considering only the spiral task, they obtained an accuracy of 62.8% [12].

The aim of this work was to analyze handwriting signals from both PD patients and control subjects and to propose a way to automatically distinguish these two classes. In order to collect the necessary data, we developed a telemonitoring system based on a userfriendly application for drawing tablets that enabled us to collect real-time information about the digital pen during the experiment and, simultaneously, to supply visual feedback on the screen to the subject. Through this system, data can be collected remotely in order to allow patients to execute tasks in the comfort and safety of their home, reducing the demand on hospital services. We decided to propose a protocol to explore writing and drawing impairments, including specific tasks for subjects who declared Italian as their first language, since, to our knowledge, the literature is lacking in automatic handwriting classification studies with Parkinson's subjects whose first language is Italian.

#### **2. Methods and Materials**

In this study, we collected data from 22 healthy subjects and from 9 PD patients. The data from the PD subjects were collected thanks to a collaboration with the Casa di Cura Le Terrazze institute. All participants were right-handed except for one PD patient, with an age in the range of 60 ± 25 years. Information about subjects' age, gender, dimensions of the hand, and level of education are collected in Table 1. The educational level was classified according to UNESCO's ISCED 2011 (International Standard Classification of Education) [21]. This classification distinguishes nine levels of education, from early child education (level 0) to doctoral or equivalent level (level 8). These levels can be aggregated into three categories: low (0–2), medium (3–4), and high (5–8) [22]. The hand dimension was quantified by measuring the distance between the wrist and the top of the distal phalanx of the dominant hand's middle finger.

**Table 1.** Subject data. For each group (healthy control (C) or subject with Parkinson's disease (PD)), the age and the number of males and females are specified. Information about the anatomical measurements of the hand and the level of education is indicated only for the control group.


Information about the PD patients is collected in Table 2. For every patient, the Hoehn and Yahr Scale level is indicated. The Hoehn and Yahr Scale is a clinical scale used to describe the progressive motor impairment of subjects with PD [23]. This scale goes from 1 to 5 in order of severity of the motor symptoms; the first level corresponds to only unilateral involvement with minimal motor dysfunction, while the fifth and last stage corresponds to a more serious level of motor dysfunction wherein the subject is confined to bed or wheelchair unless aided. Table 2 also indicates the side of the body affected by the motor dysfunction (left, right or bilateral if both sides are involved), years since the diagnosis of PD, and the levodopa-equivalent daily dose (LEDD) corresponding to each PD patient.

**Table 2.** This table shows, for every PD patient, the Hoehn and Yahr Scale level (H&Y), the side of the body affected, the years since the PD diagnosis, and the levodopa-equivalent daily dose (LEDD) assumed by the subject.


A commercial Wacom One drawing tablet with a screen was used for this test in order to be able to extract both "online" and "offline" features. Wacom tablets are widely used in handwriting movement analysis, as they offer high spatial and temporal resolution [9].

An application was developed by our team using the development platform Unity, which allowed us to collect information about pen position (*x*, *y*), pressure, and inclination with a frequency of 133 Hz and, simultaneously, to supply visual feedback on the tablet's screen to the subjects. For the protocol, the Wacom tablet was connected to a Lenovo Thinkpad T495 computer with Windows 10 as the operating system. The "Duplicate" modality was selected in order to have the same screen shown on the computer and the Wacom One tablet, as shown in Figure 1; Figure 1a shows the point of view of the operator, while Figure 1b shows the point of view of the participant.

**Figure 1.** Experimental setup: (**a**) shows the operator's point of view, while (**b**) shows the participant's point of view. The subject uses the Wacom tablet to complete the protocol tasks, while the operator follows the experiment in real time from the monitor of his computer.

The application has a start page where the participant's ID can be entered and which includes a menu from which the user can choose which task to take. The data are saved locally in different.tsv files for every acquisition.

In order to analyze the data, we used the software MATLAB.

The protocol was divided into four parts: drawing an Archimedean spiral, writing the bigram "le" six times and two Italian sentences, drawing ten concentric circles, and writing seven lines of free text. For each part of the protocol, a different screen was shown to the subject: firstly, an image of an Archimedean spiral was shown and the subject was asked to trace it at a comfortable speed; secondly, a blank screen was shown and the subject was asked to write six times in cursive the bigram "le" and the two Italian phrases: "I fiori sono sul prato" and "Nel cielo ci sono le stelle". On the third screen, a circle was shown and the subject was asked to draw ten concentric circles inside it. Lastly, a blank screen was shown and the subject was asked to write seven lines of free text in cursive. The overall duration of the test varied between 10 and 15 min from subject to subject. The subjects were given the opportunity to try the tablet before the test. During the execution of the tasks, the subjects were seated in a comfortable position on a chair without armrests, and the tablet was positioned on a table in front of them.

Features were extracted separately from each task.

Data were separated into components, i.e., lines that are traced without lifting the pen from the tablet. In order to do that automatically, indices of the samples where pressure went from positive to zero and vice versa were saved in a vector of markers. Both in-air and on-tablet features were extracted.

Figure 2 shows an example of the "le" bigram task, where the different components, automatically detected, are represented in different colors and the "in-air" points of the pen

position are represented as blue points. For each component, the velocity was calculated as follows:

$$v = \sqrt{v\_x^2 + v\_y^2},\tag{1}$$
 $v\_{i,+1} - x\_i \quad \dots \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \quad \tag{1}$ 

$$\text{where } \upsilon\_{\mathbf{x}} = \frac{x\_{i+1} - x\_i}{t\_{i+1} - t\_i} \text{ and } \upsilon\_{\mathbf{y}} = \frac{y\_{i+1} - y\_i}{t\_{i+1} - t\_i}.$$

**Figure 2.** Image of the "le" bigram task written by a healthy control subject from our dataset. The "in-air" points of the pen position are represented in blue. Different components of the "on-tablet" pen position are represented in different colors.

Where *xi*, *yi*, and *ti* correspond to the pen position along *x*, pen position along *y* and time at a specific index *i* of the recorded data, respectively.

Acceleration and jerk of the components were also calculated. To analyze the spiral, the angular and radial velocity were calculated. Furthermore, the distance of the drawn spiral from the spiral guide was calculated using the following algorithm:


$$d\_{l} = \sqrt{(\mathbf{x}\_{l}^{2} - s\mathbf{x}\_{l}^{2}) + (y\_{l}^{2} - sy\_{l}^{2})} \tag{2}$$

3. We found the parameter

$$p = \sum\_{i} d\_{i}^{\;} \, ^{2}\prime\tag{3}$$

that describes how much the drawn spiral is distant from the spiral guide. A smaller value of *p* meant a higher precision.

Furthermore, the power spectral density between 4 Hz and 9 Hz of the absolute velocity of the pen during the spiral task was calculated as a feature.

In order to develop a model for automatic classification of PD, we used data from 36 PD subjects and 35 healthy control subjects from the PaHaW dataset and data from the 22 healthy control subjects and 9 PD patients that we collected in our database. Two tasks that the two databases have in common were analyzed: the guided spiral and the bigram "le". The features that were considered are reported in Table 3.

After the feature extraction, we proceeded with the selection of the most significant features and the generation of the classification model. In Figure 3, a scheme of the workflow is presented.

**Table 3.** Features extracted: *1* if the feature was extracted from the spiral task analysis, *2* if the feature was extracted from the "le" bigram task analysis.


**Figure 3.** Workflow diagram of the process.

#### **3. Results and Discussion**

In order to discriminate between PD patients and healthy control subjects, three models were constructed: one using only data from the spiral task, one using only data from the "le" bigram task, and one using data from both of them.

We selected the most discriminant features using a Mann–Whitney test (*p* < 0.05). Table 4 presents the features with the lowest *p*-scores, divided per task.

**Table 4.** Feature ranking. The left column presents the most discriminant features for the spiral task, and the right column presents the most discriminant features for the "le" task.


A 10-fold cross-validation was conducted. Results are reported in Table 5. Accuracy, specificity, sensitivity, F1 score, and precision were calculated in terms of *TP* (true positive), *FP* (false positive), *TN* (true negative), and *FN* (false negative), as follows:

$$Accuracy = \frac{TP + TN}{TP + TN + FP + FN} \ast 100\tag{4}$$

$$Specificity = \frac{TN}{FP + TN} \ast 100\tag{5}$$

$$Sensitivity = \frac{TP}{FN + TP} \* 100\tag{6}$$

$$F1\ score = \frac{TP}{TP + \frac{1}{2}(FP + FN)} \ast 100\tag{7}$$

$$Precision = \frac{TP}{TP + FP} \* 100\tag{8}$$

**Table 5.** Model used and classification accuracy, specificity, sensitivity, F1 score, and precision of the two tasks analyzed separately and of the two tasks combined.


Considering the two tasks separately, we obtained a higher accuracy for the "le" bigram tasks than for the spiral tasks.

Moreover, considering the spiral and the "le" bigram task separately, the accuracy that we obtained for the spiral (71.6%) was higher than the accuracy obtained for the spiral by Dròtar et al. (62.8%) and, similarly, considering only the "le" bigram task, the accuracy that we obtained (75.5%) was higher than the accuracy obtained by Dròtar et al. for this task (71%).

The confusion matrix for each one of the three models (spiral, "le" bigram and combined tasks) are reported on the Figure 4.

**Figure 4.** Confusion matrix of the three models generated. In (**a**), (**b**), and (**c**), PD' and C' indicate, respectively, subjects that were predicted to be subjects with PD and control subjects, while PD and C indicate the true categories of the subjects. PD corresponds to PD patients and C corresponds to healthy control subjects.

The highest accuracy (77.5%) and sensitivity (77.8%) were obtained by combining the two tasks, while the highest specificity was obtained when using only the data from the spiral task. The machine-learning algorithms that were employed were the support vector machine (SVM) for the spiral task and the "le" bigram task, and the medium k-nearest neighbors (Medium KNN) for the combined tasks.

#### **4. Conclusions**

In this study, an application is presented that allowed us to register data from tablets with a frequency of 133 Hz, in order to aid the recognition of PD through handwriting impairments. The tool that is proposed is simple and easy to use, allowing subjects to complete the test in the comfort of their home.

Data from 22 healthy subjects and 9 PD patients were collected and added to the PaHaW database [11,20], a pre-existing dataset that includes data from PD patients and healthy control subjects. Using only two of the eight tasks that the PaHaW database includes, an accuracy of 77.5%, was obtained, close to the 85.61% accuracy that Dròtar et al. obtained when considering all the eight tasks together [20]. We could not compare the other tasks because the first language declared by our subjects (Italian) was different from the first language of the PaHaW database's subjects (Czech).

The major limitations of this study are linked to the limited number of subjects involved and to the fact that we compared data from different databases, collected under different experimental conditions, such as the position of the subject during the tasks, and acquired with different devices, which could lead to a bias in the measurements. Moreover, subjects' characteristics such as age could lead to misclassification; for example, a control subject could present some tremor or bradykinesia not linked to Parkinson's disease, and for this reason could be misclassified as a PD subject.

However, the protocol that we developed can be used in future studies to collect more data from Italian PD subjects, in order to be able to create a model using only data from our protocol, using the combination of six tasks proposed here (drawing of an Archimedean spiral, writing the bigram "le", writing two Italian phrases, drawing ten concentric circles, and writing seven lines of free texts.

Moreover, this work is part of a home-monitoring project that aims to aid in PD detection through a combined analysis of graphological and vocal signals [5].. The sets of subjects tested for the vocal tasks and the graphological tasks were different from each other, so we could not create a classification model using combined vocal and graphological data, but the aim of this project is to continue to collect both vocal and graphological data in order to create a single, more complete, classification model.

**Author Contributions:** Conceptualization, A.P. (Antonio Pallotti), A.P. (Alessandra Paffi) and G.M.; methodology, A.P. (Antonio Pallotti); software, M.F.; validation, M.F., A.P. (Alessandra Paffi), G.M., G.A., G.V. and A.P. (Antonio Pallotti); formal analysis, M.F. and F.C.; investigation, M.F. and A.P. (Antonio Pallotti); resources, M.F., G.A., G.V. and A.P. (Antonio Pallotti); data curation, A.P. (Antonio Pallotti); writing—original draft preparation, M.F. and A.P.(Antonio Pallotti); writing—review and editing, M.F. and A.P. (Antonio Pallotti); visualization, M.F. and A.P. (Antonio Pallotti); supervision, A.P. (Antonio Pallotti), A.P. (Alessandra Paffi) and G.M.; project administration, A.P. (Antonio Pallotti); funding acquisition, A.P. (Antonio Pallotti). All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** The protocol was approved by the Consorzio Parco Scientifico e Tecnologico Pontino Technoscience, with the code PST/342/P/130422.

**Informed Consent Statement:** Informed consent was obtained from the subject involved in the study.

**Data Availability Statement:** The datasets underpinning this work are available from the corresponding author upon request (agreement signing).

**Acknowledgments:** We thank Joint Research Laboratory "Computational Graphology"(Consorzio Parco Scientifico e Tecnologico Pontino Technoscience and San Raffaele University of Rome)for the support and contribution in the research work.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


## *Proceeding Paper* **Optical Coatings: Applications and Metrology †**

**Paola Zuppella 1,\*, Paolo Chioetto 1,2, Chiara Casini 1,2, Simone Nordera 1, Nunzio Cennamo 3, Luigi Zeni <sup>3</sup> and Vania Da Deppo <sup>1</sup>**


**Abstract:** The development of optical coatings has experienced rapid growth in the last few decades for a wide range of applications. The strong demand is motivated by the progress of new-generation sources, large-scale facilities, new lithography arrangements, innovative methods for materials science investigation, biosensors, and instruments for space and solar physics observations. The research activities carried out at the Padova branch of the Institute for Photonics and Nanotechnologies of the National Research Council range from the design and characterization of optical components for space activities to the development of nanostructured coatings for tools, such as biosensors and surface plasmon resonance devices. In recent years, we have dealt with the optical characterization of 2D materials in order to explore the feasibility of innovative optical elements designed and optimized to cover wide spectral ranges. In this manuscript, we show the results on the optical characterizations of MoS2 and graphene samples, both monolayers, deposited on thick SiO2 film. We present the preliminary and comparative analysis of the samples in question, showing a direct comparison with the optical performance of the pristine SiO2 over the visible spectral range.

**Keywords:** optical coatings; metrology; thin films; MoS2; graphene

#### **1. Introduction**

The Institute for Photonics and Nanotechnologies (IFN) of Italy's National Research Council (CNR) carries out pioneering research in several fields of photonics. The Padova branch stands out in the technological activities related to the development of optical devices. Applications range from space instrumentation to sensors platforms, including optical metrology, and are strongly oriented to applied physics and technology transfer [1].

Thin films and optical coatings are transversal topics, common to all activities just mentioned. In the field of biosensors development, nanostructured films find a very interesting application in the use of innovative metals for surface plasmon resonance (SPR) platforms based on prism and fiber [2,3]. The scope is to improve sensitivity, detection accuracy, dynamic range, and application fields of this type of biodevice [4,5]. In space optics, high-performance optical coatings are optimized both to fulfill the scientific requirements of the instruments and to survive harsh operation environments [6]. Furthermore, in some spectral regions, such as vacuum ultraviolet and soft X-ray, the structures of the optical films become particularly complex, requiring design and fabrication of multilayer stacks [7].

Over the years, the CNR-IFN focused on the design and characterization of nanostructured thin films to be used as sensitive layers for biosensors [8,9], mirrors [10], filters, phase

**Citation:** Zuppella, P.; Chioetto, P.; Casini, C.; Nordera, S.; Cennamo, N.; Zeni, L.; Deppo, V.D. Optical Coatings: Applications and Metrology. *Eng. Proc.* **2021**, *11*, 50. https://doi.org/10.3390/ ASEC2021-11137

Academic Editor: Saulius Juodkazis

Published: 15 October 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

retarders [11], and polarizers. We have collaborated on the fabrication of a palladium/gold bilayer designed for an SPR sensor based on D-shaped optical fibre (POF). The novel SPR-POF platform was optimized to work in the 1.38–1.42 refractive index range, where it exhibits excellent performances in terms of sensitivity and signal to noise ratio [9].

Another interesting application we dealt with was the development of innovative biochips for Kretschmann SPR tools [8]. The new chips are based on palladium thin films deposited on plastic substrate. The plastic support is low cost and commercially appealing, and the palladium is interesting from the scientific point of view, showing inverted surface plasmon resonance response. The biochips were tested for the detection of DNA chains, selected as the target of the experiment, since they can be applied to several medical early-diagnosis tools, such as different biomarkers of cancers or cystic fibrosis [8].

With regard to characterization and metrology capability, an ellipsometry system dedicated to the study of optical properties, composition, and contamination of materials has been recently developed in the CNR-IFN laboratories [12]. One of the recent applications for this system was the study of the optical performance of a few layers graphene at hydrogen Lyman-alpha (121.6 nm) [13]. We determined the optical constants of such a material at this spectral line and observed the optical anisotropy and the effects induced on the substrate performances as a shift of the pseudo-Brewster angle [13].

Graphene is interesting for several reasons [14]: for example, its excellent chemical and thermal stability [15]. The arrangement of the carbon atoms makes it inert and impermeable to all atomic species, including helium [16]. It is an outstanding candidate as protective layer of optical coatings designed to operate in hostile conditions.

However, graphene is not the only 2D material under investigation. Among the others, MoS2 is appealing for biosensing and optical sensors development [17]. The bioapplications, such as DNA, cancer, and COVID-19 detection, are the most relevant [18], and are supported by a study that shows MoS2 is compatible with human bodies, while graphene is still under study for this aspect. Whatever the potential in optics, any feasibility assessment can be made after a careful evaluation of the optical performance of the materials. We therefore sought to measure the effects of monolayer MoS2 deposited on the top of a widely used substrate. The substrate chosen for this comparative analysis was a thick SiO2 film. The SiO2 is well known due to excellent properties, such as anti-resistance, hardness, corrosion resistance, dielectric, and optical transparency [19–21].

Starting from this scenario, we characterized the optical response of MoS2 in the visible spectral range under light polarization control and compared its optical throughput with that of graphene, the most popular of the 2D materials, and pristine SiO2.

We describe the preliminary results in this manuscript, that includes a section dedicated to the experimental equipment and samples, and a section dedicated to the experimental achievements and discussion. The conclusion reports a summary of what is presented in the paper by envisioning a future experimental campaign.

#### **2. Experimental Arrangement and Samples**

The samples whose measurements we report in this paper are:


The thicknesses of the graphene and MoS2 were not experimentally determined at the time of this manuscript, but this is planned as part of the measurement campaign we have planned. We want to use Raman spectroscopy for this purpose, as we have already for the study of the three-layer graphene in [13]. Then, we are referring to "monolayer", according to what it is declared by the suppliers.

The nominal thickness of SiO2 is 300 nm. The reflectivity measurements described hereafter make it possible to estimate the value from the experimental measurement test.

A proper thickness of the SiO2 substrate allows the direct observation of graphene and MoS2 by optical microscope. In mono layered graphene and MoS2 materials, few layers, discontinuities, and wrinkles are directly detectable in a very simple way [22–25]. The first measurement performed on the samples under examination was an observation by optical microscope. The images were acquired at 100× magnification.

We also investigated the morphology of the monolayer graphene sample by atomic force microscope (AFM) in no-contact mode operation (Park System XE-70).

After the observation of the surface, the samples were characterized in terms of optical performance.

The reflectometers equipment available at the CNR-IFN laboratories located in Padova cover a wide spectral range extending from the extreme ultraviolet (EUV) to visible wavelengths. An additional tool for infrared measurements has also recently been acquired.

The optical characterizations at shorter wavelengths (30–400 nm) require the use of normal incidence Johnson–Onaka reflectometer operating in vacuum. Figure 1 exhibits the sketch of the device [13].

**Figure 1.** The figure used under CC-BY 4.0 depicts the EUV-VUV-UV Johnson–Onaka reflectometer.

The dispersion element mounted on the reflectometer is a Pt-coated toroidal grating with 600 lines/mm. The main radius is 0.5 m and the subtended angle between the entrance and the exit slits is 25◦. A toroidal mirror working at 45◦ incidence angle focuses the monochromatic radiation on the sample. In the experimental chamber, the samples are hosted on a holder that can be rotated to the desired incidence angle. We have recently implemented this facility with a rotating linear polarizer optimized for the VUV. In this way, the reflectometer is suitable for ellipsometry measurements [12,13]. The upgraded system has recently been used for the characterization of phase retarders for EUV-VUV wavelengths [12,13].

The optical characterization in the visible spectral range is accomplished by using the VIS reflectometer depicted in Figure 2. The system designed for testing the optical response of samples at variable incidence angles, working in a θ-2θ configuration, was recently assembled. It consists of a compact, stabilized, broadband light source (360–2600 nm), a rotator stage to hold the sample at a desired working angle, and a spectrometer coupled with a cosine corrector for the detection.

The measurements, in reflection and transmission modes, can be performed for any type of sample with no restrictions in size. The optical response can be tested with polarized and unpolarized light in order to investigate the polarization response of the specimen of interest. A commercial polarizing filter was used in the measures described in this paper [25]. The filter has a p-polarized transmission close to 90% in the spectral range

450–900 nm. A calibrated sample of protected aluminium was used as a reference specimen (Filmetrics KLA Corporation).

**Figure 2.** The design and the photograph exhibit the reflectometer available at CNR-IFN in Padova and optimized for the optical characterization in the visible range.

#### **3. Results**

In the sample under analysis, the nominal thickness (300 nm) of the pristine SiO2 is tuned to enhance the optical visibility of the 2D materials on the top [22–25]. The top surface quality of graphene and MoS2 can be easily determined by using an optical microscope, that allowed us to estimate the linear dimensions of the superficial defects. The presence of film discontinuities and wrinkles can be assessed in a qualitative and semi-quantitative way over a wide region. In case of regions of the same sample with a different number of layers, we usually observe areas with different contrast. An illustrative example is depicted in Figure 3 [24]. The image by Y. Stubrov et al. [24] shows a graphene plate on a Si substrate, covered with 300 nm-thick SiO2 layer.

**Figure 3.** The figure used under CC-BY 4.0 depicts the optical microscopy image of an investigated graphene sample located on Si substrate, covered with 300 nm-thick SiO2 layer.

For the specimens characterized in this paper, we cannot determine how many layers there are, but we can say that the number of layers is the same in all regions of the sample. The images (see Figure 4), acquired by a camera connected to the optical microscope, show a good quality of the samples surface, good homogeneity for both monolayers, and no relevant defects.

Several areas of the specimens were observed; we report the representative ones.

Figure 5 depicts the surface of the graphene sample (left) analysed by AFM over 3.5 μm × 3.5 μm area. At the time of writing, we have only analysed the AFM measurements of graphene, but we plan to perform the characterization on all samples under study. The surface quality of the graphene is good even at the nm scale, showing some wrinkles and defects mainly due to the transfer process of graphene, grown by chemical vapor deposition

and then transferred onto SiO2 substrate. The roughness estimated on the profiles (right) corresponding to the white cross in Figure 4 is Ra = 0.6 nm and Rq = 0.75 nm.

**Figure 4.** The figure depicts regions of the samples: SiO2 provided by Graphenea (**left**, top); graphene provided by Graphenea (**right**, top); SiO2 provided by 2D semiconductors (**left**, bottom); MoS2 provided by 2D semiconductors (**right**, bottom).

**Figure 5.** The figure depicts AFM measurement (**left**) of the graphene sample provided by Graphenea (**left**, top); the horizontal and vertical profiles (**right**) correspond to the white cross.

Once the samples were observed by optical microscope, their optical performance was measured. The experimental results are reported in Figure 6. All tests have been performed at 8◦ angle of incidence by using p-polarized light as probe.

In this test campaign, we used a known reference sample dealt by Filmetrics KLA Corporation for determining the experimental reflectance of any sample, without measuring the direct beam and without moving the experimental arrangement.

**Figure 6.** The plot depicts the experimental p-reflectance of the four samples under investigation together with the simulation of a SiO2 (315 nm)/Si.

It is required to determine the factor *F* (Equation (1)):

$$F = \frac{R\_{ref}}{R\_{cal}}\tag{1}$$

which is the ratio between the value of the light reflected by the reference sample (*Rref*) and the reflectance of the same sample measured and certified by the manufacturer (*Rref*).

Given the experimental value of the light reflected by the specimen that we want to characterize (*Rexp*), the experimental reflectance, *R*, of such as sample is given by the following relationship:

$$R = \frac{R\_{exp}}{F} \tag{2}$$

Figure 6 shows the measured reflectance according to Equation (2) for the four samples we are analysing. It is worth to note that the performances of the two SiO2 provided by Graphenea, Inc. (see Figure 5, "SiO2 [R-G] exp") and 2D semiconductors (see Figure 5, "SiO2 [R-MoS2] exp"), even if they come from two different suppliers, are both in good agreement with the simulation of the structure SiO2 (315 nm)/Si. Then, the actual thickness of the SiO2 that we can estimate from the reflectance measurements is 315 nm.

The reflectance of the SiO2 is the reference with respect to we want to observe the optical effects of 2D materials. The sample with graphene on the top (see Figure 5, "graphene exp") reflects sensitively less and shows a blue shift of the minimum, that occurs around 605 nm against 610 nm of SiO2. On the contrary, MoS2 (see Figure 5, "MoS2 exp") induces a red shift of the minimum, which is observed around 640 nm. The reflectance is higher than that of SiO2 up to 580 nm, then becomes significantly lower at longer wavelengths.

The present study is qualitative and shows how reflectance measurements with polarization control are sensitive to the presence of 2D materials on the surface of a SiO2/Si substrate, despite the sub-nanometric structures (nominal thickness of monolayer graphene is 0.34 nm and nominal thickness of monolayer MoS2 is 0.72 nm). For a quantitative analysis, we plan to perform experimental thickness measurements and a full characterization on samples with different thicknesses of graphene and MoS2.

Combined with experimental determination of thicknesses, the reflectance response can be used to retrieve the optical constants of such a material at the wavelength of interest. For graphene, this analysis has been addressed by several authors; for MoS2, there are still interesting studies that could be assessed.

#### **4. Conclusions**

In this manuscript, we analysed the optical response of four samples based on the structure SiO2/Si, two of them capped by MoS2 and graphene, respectively. The experimental results show a good sensitivity of the reflectance to 2D materials, offering great potential for their characterization in view of many application scenarios.

**Author Contributions:** Conceptualization P.Z., methodology P.Z. and P.C., validation P.Z. and P.C., formal analysis P.Z., investigation, resource, data curation P.Z., P.C., C.C., S.N., N.C., L.Z., V.D.D., writing-original draft preparation P.Z., writing-review and editing P.Z., P.C., C.C., S.N., N.C., L.Z., V.D.D., visualization, supervision P.Z., project administration P.Z., funding acquisition V.D.D. All authors have read and agreed to the published version of the manuscript.

**Funding:** This activity has been financially supported by the contracts n. 2020-4-HH.0 and n. 2019- 34-HH.0 between Agenzia Spaziale Italiana (ASI) and the Istituto Nazionale di Astrofisica (INAF).

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

