Development of nanostructured thin films for humidity and temperature sensors

fevereiro 2021 João Paulo Carmo da Silva Development of nanostructured thin films for humidity and temperature sensors UMinho | 2021 João Paulo Carmo da Silva Development of nanostructured thin films for humidity and temperature sensors fevereiro 2021 João Paulo Carmo da Silva Development of nanostructured thin films for humidity and temperature sensors Dissertação de Mestrado Mestrado em Física Trabalho efetuado sob a orientação de Doutor Armando Ferreira Doutora Daniela Correia ii Direitos de autor e condições de utilização do trabalho por terceiros Este é um trabalho académico que pode ser utilizado por terceiros desde que respeitadas as regras e boas práticas internacionalmente aceites, no que concerne aos direitos de autor e direitos conexos. Assim, o presente trabalho pode ser utilizado nos termos previstos na licença abaixo indicada. Caso o utilizador necessite de permissão para poder fazer um uso do trabalho em condições não previstas no licenciamento indicado, deverá contactar o autor, através do RepositóriUM da Universidade do Minho. Licença concedida aos utilizadores deste trabalho Atribuição CC BY https://creativecommons.org/licenses/by/4.0/ iii Acknowledgements During these past 18 months, many were those who contributed in some way to the development of these master’s thesis, and for those people, which I must express my gratitude for all that they represented. To Doctor Armando Ferreira and Dra. Daniela Correia for the amazing help, counselling, patience and friendship that they were able to provide me. To the Electroactive Smart Materials Group where I was integrated for making me feel welcomed and for the help that they never denied me and specifically to engineer Nelson Pereira for all the help that he gave me. To Professor Felipe Vaz and to Professor Marcio Correa for the contributions given in the publication of the articles that served as the background of this document. To the GRFVAZ for providing the PVD equipment and the means to produce the thin films presented on this thesis. To the University of Minho, especially to the Centre of Physics for providing means to produce the work presented on this thesis, on such challenging times. To my family for the love and the support they gave along the years and through this thesis development. To my friends, for their friendship since one cannot do this path alone. And last, but not least, to all those who helped me get through some personal problems that appeared along with this master thesis development, some of those people already mentioned here. To all of you, a sincere Thank you iv Statement of integrity I hereby declare having conducted this academic work with integrity. I confirm that I have not used plagiarism or any form of undue use of information or falsification of results along the process leading to its elaboration. I further declare that I have fully acknowledged the Code of Ethical Conduct of the University of Minho. v Resumo Este trabalho visa apresentar todo o processo de produção e resposta de um conjunto de filmes finos multifuncionais com capacidade de sensorização de temperatura e/ou de humidade relativa produzidos através da técnica de deposição física em fase de vapor ( Physical Vapor Deposition , PVD) com auxílio do sistema Glancing Angle Deposition - GLAD. Neste sentido, foram preparados dois conjuntos de filmes finos: filmes de nitreto de crómio (CrN) e oxinitreto de crómio (CrNxOy). Durante a preparação dos filmes nanoestruturados em ziguezague, foi variada a quantidade de oxigénio e azoto com a finalidade de se obter uma variação na resposta elétrica quando sujeitos a uma variação de temperatura ou humidade relativa. Os filmes produzidos foram sujeitos a diferentes técnicas de caracterização com o intuito de avaliar as propriedades estruturais, morfológicas e elétricas. Para os sensores de temperatura foram usados os filmes finos de CrNx, nos quais, as técnicas de difração raio-X e de microscopia eletrónica de varrimento permitiram a avaliação das propriedades estruturais e morfológicas das amostras de CrNx em função do fluxo de N2 durante a deposição. Efetuou-se um estudo sistemático do efeito termorresistivo dos filmes finos de CrNx, de modo a determinar o coeficiente de resistência de temperatura (TCR). A resposta termorresistiva foi avaliada medindo a resistividade elétrica em função da temperatura através do método de duas pontas. Os filmes finos depositados com um fluxo de N2 entre os 4 e os 8 sccm apresentaram um TCR negativo, com uma variação de 9.18×10-4 ± 2.47×106 ºC-1 até 1.48×10-2 ± 1.74×10-5 ºC-1 e uma resposta elétrica estável ao longo do tempo para diferentes estágios de temperaturas. O modelo teórico de fronteira de grão foi usado para descrever os resultados experimentais obtidos. Para os sensores de humidade foram utilizados os filmes finos de CrNxOy. Os resultados obtidos demonstraram que com o aumento de oxigénio, à medida que o fluxo da mistura N2+O2 aumenta, ocorre uma transição de fase de nitreto de crómio para oxido de crómio. Ao mesmo tempo ocorre uma mudança na estrutura cristalina de cubica de corpo centrado do crómio (BCC-Cr) para uma estrutura cubica de face centrada (FCC-CrN). A quantidade de O2 presente na composição dos revestimentos faz com que exista um aumento da resistência elétrica, variando de 1.27×10-6 ± 5.02×10-8 .m para a amostra com uma percentagem atómica de O2 de 16 % até 4.50×109 ± 4.20×107 .m para a amostra com uma percentagem atómica de O2 de 60 %. A maior resposta do transdutor à variação de humidade relativa vi (RH) entre 40 % e 80 % foi de 1.87×10-2 ± 7.00×10-5 (RH%)-1 . Os resultados obtidos demonstram o potencial de CrNx e de CrNxOy para aplicações de sensorização. Palavras chave: Revestimentos duros multifuncionais; pulverização catódica em corrente continua; Coeficiente de Resistência de temperatura; propriedades térmicas; sensorização de temperatura; nitreto de crómio; Oxinitreto de crómio; Humidade relativa; sensorização de humidade. vii Abstract This work aims to present the entire production and response process of a set of multifunctional films capable of sensing temperature or relative humidity produced using the Physical Vapor Deposition, (PVD) with the assistance of the Glancing Angle Deposition technique - GLAD. In this work, two sets of thin films were prepared (chromium nitride (CrN) and chromium oxinitride (CrNxOy) grown with a zigzag like nanostructure, varying the amount of nitrogen and oxygen in order to obtain a variation in the electrical response when subjected to a variation in temperature or relative humidity. The films produced were analyzed by different characterization techniques in order to evaluate their structural, morphological and electrical properties. In order to obtain thin films with the capability to sense variations in temperature, a systematic study of the thermo-resistive effect of chromium nitride (CrNx) thin films with negative temperature coefficient of resistance (TCR) has been carried out. A systematic study of the thermoresistive effect of chromium nitride (CrNx) thin films with negative temperature coefficient of resistance (TCR) has been carried out. CrNx nanostructures were grown by reactive magnetron sputtering, with oblique angle deposition, under distinct Ar+N2 conditions. This experimental setup enables to confer a zigzag columnar morphology to the CrNx thin films and thus modify the structural, morphological and physical properties of the samples. X-ray diffraction and Scanning Electron Microscopy allowed to evaluate the evolution of the structural and morphological properties of the samples as a function of the N2 flux during the deposition. The thermoresistive response was evaluated by measuring the electrical resistivity as a function of temperature by the two-point method. The thin films with N2 flux between 4 and 8 sccm presented a negative TCR with values ranging from 9.18×10-4 ± 2.47×10-6 ºC-1 to 1.48×10-2 ± 1.74×10-5 ºC-1, respectively, and a very stable time response at a given temperature. The grain-boundary model was used to describe the experimental results and, in particular, the negative TCR values of the samples. In order to obtain thin films with relative humidity sensing capability, thin films based on chromium oxynitride (CrNxOy) were prepared with zigzag nanostructures and varying the amount of nitrogen and oxygen in order to enhance the humidity response. The obtained CrNxOy thin films show an increasing amount of oxygen as the N2+O2 mixture flux increased, leading to a transition from chromium nitride to chromium oxide. Further, a change from a BCC-Cr to a pure FCC-CrN phase is obtained as the oxygen in the sample increases. The amount of oxygen present in the composition of the coating leads to high xiv List of abbreviations A – Ampers Å – Angstrom 𝐴𝑐 – Area of contact BCC – Body-centered cubic Cr – Chromium CrN – Chromium nitride CrNXOY – Chromium oxynitride Cr2O3 – Chromium oxide 𝐷𝑅 – deposition rate dc – Direct current EDS – Energy-dispersive x-ray spectroscopy FCC – Face-centered cubic GLAD – glancing angle deposition K – Kelvin NTC – Negative temperature coefficient (of resistance) PVD – Physical vapor deposition xv PTC – Positive temperature coefficient (of resistance) RTD – Resistance temperature detector RBS – Rutherford backscattering spectrometry SEM – Scanning electron microscopy sccm – Standard cubic centimetre per minute TCR – Temperature coefficient of resistance V – Volts XRD – X-ray diffraction 1 1. Introduction The introduction of this work aims to contextualize and indicate reasons for the importance of this study, the entire process which involve the production of thin films by PVD techniques and their characterization in terms of physical, chemical and thermal properties, in order to assess its applicability and functionality. Since the industrial revolution, the industry has had a growing need to control production variables, such as temperature, pressure and humidity [1]. This necessity was further increased with the beginning of the fourth industrial revolution, denominated by Industry 4.0 [2]. Industry 4.0 is inherently a symbol of connectivity between things and people throughout the manufacturing process. Some of the goals for Industry 4.0 are quite ambitious, the hope is that it can evolve into artificial intelligence and automated decision-making, near-perfect mechanical automation and symbiotic human integration, and have manufacturing facilities that are completely interconnected and ‘smart’, from raw materials to finished goods. While this change may seem overwhelming, it can be achieved by working incrementally, starting small by identifying pain points of facilities and scaling when success is shown [2]. Manufacturers are able to make strides toward Industry 4.0 through three pillars of thought: digitization, sensorization and optimization. This work is based on the second industry pillar of thought: the sensorization. Sensors are cost-effective ways to measure and control production variables, in this new production standardization. Control devices and industrial machinery should be able to make decentralized decisions in order to control the production in a semi-independent way [2]. These decisions are adapted to the manufacturing process variables since the quality and the economic viability of the product will be affected by the production conditions [3]. Therefore, sensors should not only have a good sensibility to the intended variable, but also, they should have stability [4], durability [4], reduced size [4], easily transferable results [4], low price and be able to be applicable in different environments [4]. For that, the development of a skin-like thin film needs to be optimized according to the basic requirements of each kind of sensing [5,6]. For the development of functional and multifunctional thin films, Physical Vapor Deposition (PVD) is one of the preferred techniques as it allows the control of the process parameters and stoichiometry of the deposited coatings [7]. The synergy between PVD technologies and Glancing Angle Deposition (GLAD) has attracted increasing interest due to the precise control of the microstructural (isotropic and anisotropic) and physicalchemical properties [8]. The GLAD takes the advantage of the shadowing effect and diffusion of the particles on 2 the growing film [9]. The basic features of this deposition technique consist in depositing thin films under an oblique incident flux of the sputtered particles, using a mobile substrate (Fig. 1). In the columnar films, the substrate holder was static, while for the zigzag coatings the substrate was periodically rotated in the Φ direction. The growth of thin films through GLAD enables to develop advanced materials and devices with unique morphologies aiming to high-performance technological applications [10,11], with tailored functionalities [12,13]. Figure 1: Schematic diagram of the Oblique Angle Deposition set-up.  is the applied angle of the substrate relative to the Cr particle flux,  is the column growth angle, and  is the angular direction. 1.1. Thin films based on chromium Coatings based on transition nitride materials have received a great technological attention in the field of the cutting tools or components subjected to wear and corrosion as they provide high protections to the materials on which they are coated [14,15]. Among various nitrides, and to explore the possibility of developing hard coatings with sensing capabilities, chromium nitride (CrNx) is among the most interesting ones based on its good mechanical properties [16], corrosion resistance [17], excellent wear behavior [18], electrical conductivity [19] and is widely used in bearings, machine parts and polymer injection molds [6]. To extend the range of applications of the CrNx materials, and to turn possible the implementation of coatings with multifunctional capabilities, it has been increasingly recognized that hard coating materials with sensing capabilities are needed [6,20–23] being temperature the most relevant parameter to be monitored in a variety of applications and processes [24]. Additionally, among the various chromium oxides, Cr2O3 is the most stable under ambient conditions, and it is characterized by its chemical inertness, stability, mechanical strength and relatively high hardness [25]. Previous research shows that the hardness of chromium oxide coatings strongly depends on the stoichiometric 3 polycrystalline Cr2O3 phase present in the coating [26–29], with high-quality Cr2O3 stoichiometric coatings reaching nearly 30 GPa combined with good scratch resistance. As chromium oxide is an insulating antiferromagnetic material it is also suitable as a tunnel junction barrier [30]. Depending on its stoichiometry CrNx shows a metalliclike (ρCrN ≈ 6.4×10-4 Ω cm) or semiconducting behaviour (ρCrN > 1×10-2 Ω cm) [31] while chromium oxide is a wide-bandgap semiconductor (Eg ≈ 4 eV) [32–34]. Combining both nitrogen and oxygen (oxynitride) opens the possibility to tune the energy bandgap and hence the electronic properties in a wide range. In this scope, it is to notice that few studies report on the relation between the relative humidity and temperature sensibility of chromium oxynitride (CrNxOy), the deposition process and the possible morphologies to obtain relative humidity sensors. CrNxOy is an interesting material system since it can combine the properties of chromium nitride and chromium oxide, being harder than CrN [35,36] and corrosion resistant like Cr2O3 [37]. Moreover, it is to notice that a nanostructured material with high porosity and large surface area will facilitate the adsorption of water vapor on its surface giving great scope for enhancement in the sensitivity of a relative humidity sensor. 1.2. Main goal of the work Thus, the present work focuses on the investigation of the relationship between the: i) the thermoresistive response of the CrNx thin films and ii) the relative humidity response of the CrNxOy thin films prepared under different processing conditions. The variation of the film's geometry and deposition parameters allows to modify not just the structural properties of the Cr-based thin films, but also, the thermoresistive properties and the relative humidity response, leading to interesting multifunctional characteristics for a variety of technological applications [21–23]. 1.3. Thesis structure This thesis is divided into 6 different chapters: Introduction, State-of-the-Art, Materials and Methods, Results and discussion, Conclusion, and finally Future work. 4 This thesis addresses two separate types of transducers based on CrNx and CrNxOy thin films to be used as temperature and humidity sensors. Because of that, we chose to present the results and discussion divided into two subchapters: temperature sensors and relative humidity sensors. The first chapter, “Introduction”, aims to present the problem which we, in some way, pretend to contribute to solve, the main goal and the thesis structure. The second chapter, “State-of-the-art”, aims to give the reader a general knowledge about temperature sensors and relative humidity sensors, the sensors that are available in the market and some of the sensors that were already made or explored. The third chapter is devoted to the experimental details. The basic principles of the technique selected for the thin film’s preparation are explored, in a first stage. Then, an introduction of the methods used to produce and characterize the samples and a presentation of the conditions in which these methods were performed are given. The fourth chapter is assigned to the results of the characterization introduced in the third chapter and the discussion of these results. In this chapter, the chemical, structural, morphological, electrical, thermoresistive and humiresistives properties are analysed. The fifth chapter states the conclusions achieved by the reported characterization and evaluates the quality of the achieved sensors. The sixth chapter aims to present future steps to increase the knowledge about this area of work and to increase the functionality of these sensors. 5 2. State-of-the-Art This chapter aims to give to the reader the state-of-the-art on sensors to measure relative humidity and temperature. The chapter is divided into 2 subchapters: one on temperature sensors and the other on relative humidity sensors. 2.1. Temperature sensors In terms of temperature control, temperature is one of the most important factors to human comfort, development and survival. Since the first device, developed by Galileo Galilei in the year of 1592 [38], there has been an increasing evolution to control the temperature measurements [39] due to the fact that temperature control is a critical aspect in several processes such as chemical reactions [40,41], energy measurement [41], combustion [40], extrusion processes [40], separation [40] and storage. Nowadays there is a wide range of sensors to be used in industry control such as thermocouples [42], resistance temperature detectors (RTDs) [42], thermistors [42], infrared pyrometers [42], diode temperature sensors [42], bi-metallic thermometers [42] and liquid thermometers [42], among others [42,43]. Independently of the application, for the industry, the most commonly used temperature sensors are thermocouples and RTDs making around 99% of temperature sensors in use [40]. The transduction mode of the thermocouples is based on the Seebeck effect and are made with two metals of different electrical conductivities [39,44]. When a temperature gradient is applied over the metal, electrons will migrate from the hotter region to the colder region producing a voltage between both regions. The difference between the voltage measured will be proportional to the temperature gradient [39,44]. The most common thermocouples reported are based on SiC-SiC, W-CNi, PtRh-Pt, C-C, C-W, W-Mo, and SiC-C [44,45]. The RTDs are made from one metallic element material such as platinum, copper or nickel [39,44]. Metallic materials possess a positive resistive variation with temperature due to the increase in vibrations amplitude of the metal atoms which will decrease the electrons mobility [44]. The metal most used in RTDs is platinum since it is resistant to high temperatures, is chemically inert and the electrical response are practically linear with the temperature [39,44]. Another attractive temperature sensors are the thermistors [44]. Thermistors are commonly made by semiconductors materials such as metal oxides and may have a positive resistance variation with temperature, called PTC (positive temperature coefficient) thermistors, or have a negative resistance variation, called NTC (negative temperature coefficient) thermistors [44]. Between this two, the most appealing thermistors are the NTC thermistors since they possess a much larger sensibility than both RTDs and thermocouples. Additionally, 6 they are also smaller and have a higher electrical resistance than RTDs making it less affected by the problems associated with the connection between the RTD and the wires [44]. The principal characteristics of the available commercial temperature sensors are presented in Table 1. Table 1: Characteristics of the most widely applied temperature sensors, platinum RTDs, thermocouples and thermistors, adapted from [40] . Quality RTD Thermocouple Thermistor Sensitivity 0.001 K 0.05 K 0.0001 K Minimum size 2 mm 0.4 mm 0.4 mm Repeatability 0.02K – 0.5K 1K – 8K 0.1K – 1K Temperature range 73K – 1123K 73K – 2273K 173K – 573K Drift 0.01K – 0.1K 1K – 20K 0.01K – 0.1K Signal output 1V – 6V 0V – 0.06V 1V – 3V Thermistors are widespread in the market with a great variety of shapes and sizes [43,46] as demonstrated in Fig. 2. However, thermistors have some weakness related to the corrosion, wear and oxidation [43,46,47]. To reduce these drawbacks, they need to be encapsulated to have a good chemical, physical and electrical stability. Yet, one of the advantages of the thermistors and also the RTDs is the fact that they can be produced into the form of thin film increasing the functionality and the wide range of applications. If the metal-based material used has good properties to chemical, physical and electrical properties, they could be a good choice to implement as a sensor in several harsh applications. Figure 2: Some thermistors shapes and sizes [43] . 7 2.2. Relative humidity sensors Water vapor is a natural component of air, and it plays an important role in a wide range of practical measurement situations [48,49]. Hygrometry, measurement of water vapor content of a gaseous atmosphere, is in fact a “branch of applied physics in which the multitude of techniques is an indication of complexity of the problem, and of the fact that no one solution will meet all requirements at all times and in all places” [50]. Currently, humidity sensors have been used intensively in several areas such as medicine, agriculture, meteorology, cuisine, storage, civil engineering and naturally in industry processes [4,51]. The need to control humidity has been promoting a continuing development of new and improved humidity sensors to enhance the selectivity, sensibility and stability, while also striving to decrease the response time, the size and the price of the sensors [51]. Humidity sensors can be separated into five categories: optical, electrochemical, electrical, thermometric, and mass sensitive sensors [4]. From these, electrical humidity sensors, particularly capacitive and resistive humidity sensors, have received a great deal of attention due to their properties. Table 2 shows the advantages and disadvantages of capacitive vs resistive humidity sensors. Apart from these two, there is also thin film and fieldeffect transistors, hetero junction-based humidity sensors and Kelvin probe humidity sensors [52]. Table 2: Capacitive and resistive humidity sensors comparison [52] . Humidity sensor type Advantages Disadvantages Resistive humidity sensors Easier to use, cheap, small, mass reproducible, high sensitivity, Non-linear, limited range, poor stability, drift, high-temperature dependence, Capacitive humidity sensors Low maintenance, cheap, small, mass reproducible, tolerate condensation, fast response, linearity, broad humidity range Loses accuracy at low humidity, high humidity and high temperature, need for electronics, drift However, there are a wide variety of resistive humidity sensors as reported in [4,48,52]. For both capacitance and resistive humidity sensors, ceramics have been widely explored due to their promising results such as their high mechanical strength, high corrosion resistance and thermal stability [53,54]. Table 3 shows the advantages and disadvantages of each material base used as a humidity sensor. 8 Table 3: Advantages and disadvantages of resistive humidity sensors made from different materials [4,48,52] Material Advantages Disadvantages References Polymer Low cost, small, no calibration needed, long life, fast response, high accuracy, broad range, and excellent reproducibility Chemical contamination may cause failure, temperature dependency [55–58] Metal Oxides High sensitivity, low cost, fast response, broad range, high mechanical strength, high thermal capability, physical stability and high corrosion resistance, high versatility Need for periodic cleaning, drift, temperature dependency, hysteresis, contamination See Table 3 Black phosphorus Ultra-sensitive, short response time, broad range Short-term stability, low repeatability, temperature dependency [59–62] Carbon-based sensors Flexible, elastic, high hardness, broad range, may have fast response and recovery Temperature dependency, high hysteresis, may have a slow response and recovery times [63–67] Chalcogenides Fermi-level at the surface can be changed, very fast response Low selectivity, drifts at high temperatures [68–72] Like temperature sensors, thin films for relative humidity sensors have the advantage of improving the functionality of the component surface with a very reduced size and cost. Chromium oxide, Cr2O3, is a very hard coating and has high corrosion, oxidation, and wear resistance. In this sense, Cr2O3 is a perfect candidate to be used in humidity sensor applications. Chromium-oxide based materials such as MgCr2O4 [73], ZnOCr2O4 [74], PANI–Cr2O3 [75], PPy-Cr2O3(PCO) [76], Cr2O3-WO3 [77] have also already been proved to be an excellent alternative to the commercial humidity sensors. Table 4 summarize the commonly used materials to produce resistive humidity sensors. 15 3.1.2.3. X-ray diffraction measurements The structure of the films was characterized by X-ray diffraction (XRD). This technique is used to evaluate the structure, crystallite size, crystallite deformation and crystallite growth preferential orientation. The X-ray diffraction apparatus is composed of 3 components: The X-ray source, the sample holder and the x-ray detector. The X-ray source and the detector will move in relation to the sample holder at the same speed maintaining the same angle with the sample holder (fig. 6). Figure 6: XRD schematization, the detector and X-ray source move by – 𝜃 and by 𝜃 providing the 2 𝜃 measurement. The interplanar distance of the sample was calculated by the Bragg Law [134]: 2𝑑∙𝑠𝑒𝑛(𝜃)=𝑛∙𝜆, (6) where 𝑑 is the interplanar distance of the sample, 𝜃 is the angle obtained by the crystallography, 𝜆 is the wavelength of the radiation source and 𝑛 is a positive integer [134]. This equation may be understood as the left side of the equation being the path-length difference (figure 7 b)) and the right side of the equation as the path length needed in order to make constructive interference happen, in other words, a multiple of the wavelength (fig 7). 16 Figure 7: Bragg law a) X-rays diffracting on the sample atoms, b) path-length difference. By the relation for the cubic structure, the lattice constant, 𝑎, is calculated through [134]: 𝑑2=𝑎2 ℎ2+𝑘2+𝑙2, (7) where ℎ, 𝑘 and 𝑙 are the Miller indices [134]. The grain size was calculated by the Scherrer formula [134] 𝜏= 𝐾∙𝜆 𝛽∙cos (𝜃), (8) where 𝜏 is the grain diameter, 𝐾 is the shape factor, which is considered to be 0.9 since spherical crystallites are assumed, 𝛽 is the full width at half maximum (FWHM) in radians [134]. In order to quantify the preferential crystallite orientation, i.e., the preferred growth texture of a particular plane, the texture coefficient 𝑇𝑐(ℎ𝑘𝑙) was calculated after Eq. 9 [135]: (a) (b) 17 𝑇𝑐(ℎ𝑘𝑙)=𝐼(ℎ𝑘𝑙)𝐼𝑟(ℎ𝑘𝑙) ⁄ 1 𝑁∑𝐼(ℎ𝑘𝑙)𝐼𝑟(ℎ𝑘𝑙) ⁄ 𝑛 𝑘=0 , (9) where 𝐼(ℎ𝑘𝑙) is the intensity of a particular reflection, 𝐼𝑟(ℎ𝑘𝑙) is the intensity of the reference peak obtained from the database (ICSD), and 𝑁 is the total number of reflections considered in the obtained diffractogram. In our study, the structure of the films was characterized by an XRD apparatus using a Bruker D8 Discover diffractometer (Bruker, Billerica, Massachusetts, EUA), operating in a θ-2θ configuration and a step of 0.02° per 0.2 s from 30 to 75°(Cu λKα1 =1.54060 Å). 3.1.3. Electrical, thermoresistive and relative humidity measurements 3.1.3.1. Electrical resistive measurements The electrical resistivity of the samples was measured by a Four-Point Probe System (Ossila Ltd, Solpro Business Park Windsor Street, Sheffield, UK). The probe head uses spring-loaded contacts instead of sharp needles, with 1.27 mm spacing between the probes, 0.48 mm probe diameter, and 60 g spring pressure. 3.1.3.2. Thermoresistive measurements The thermoresistive properties were obtained from the electrical resistance vs temperature measurements with the setup presented in Figure 8. For that, the electrical resistivity of the samples was measured in two cycles over the range of 305 K to 473 K, regulated with a Linkam’s LTS420 stage at a rate of 10 K min-1 and the electrical resistivity measured with a multimeter Keithley serie 2700. 18 Figure 8: Schematic representation of the thermoresistive setup. Commonly, the electrical resistance of a material varies with temperature because temperature affects the number of carriers in semiconductors or their mobility in conducting materials. In some materials, the electrical resistance will decrease with increasing temperature, due to the charge carriers excitation by thermal energy, while for others the electrical resistivity will increase as mobility of the carriers will decrease with increasing temperature due to the scattering effect of lattice vibrations. Thus, when the temperature increases, the resistivity of the semiconductors may decrease (negative temperature coefficient of resistance - NTCR) or increase (positive temperature coefficient of resistance - PTCR), depending upon the change in the number of carriers and their mobility. For semiconductor temperature sensors, the temperature dependence of electrical resistance is generally represented in the following form [136]: 𝑅𝑇=𝐴∙𝑒𝑥𝑝(𝛽𝐺 ∆𝑇) (10) where 𝑅𝑇 is the electrical resistance at temperature 𝑇 (K), 𝐴 () and 𝛽𝐺 (K) is the thermal indexes, which are used to evaluate the sensitivity of the thermo-resistive effect in thermistors, and 𝑇 (K) is the absolute temperature. The relationship between 𝛽𝐺 and TCR can be written as [136]: 𝑇𝐶𝑅=Δ𝑅 𝑅0∙1 ∆𝑇=𝛽𝐺 𝑇2 (11) In Equation 11, Δ𝑅 () is the variation of the film’s resistance, 𝑅0 () is the initial resistance of the film at 25 °C (room temperature) and ∆𝑇 (K) is the variation of temperature. Equations 10 and 11 are only valid over small temperature ranges, where the slope of the Ln(𝑅𝑇) versus 1/𝑇2 approximates to a linear relationship [6]. 19 3.1.3.3. Relative humidity measurements The relative humidity measurements were done using a homemade setup using a hermetic vapor chamber (PSelecta) with temperature control as described in [137] and represented in Fig. 9. Figure 9: Schematic representation of the humidity chamber setup using a hermetic vapor chamber (P-Selecta) with temperature control. The humidity was created by an ultrasonic humidifier beaker with distilled water. The sensors were placed in the humidity chamber along with a commercial temperature and relative humidity datalogger (Amprobe TR200-A). The temperature within the test chamber was kept at 24 °C during the experiments. The variation of the electrical resistance of the developed sensors was measured using an HP 34401 digital multimeter connected to a PC for data acquisition. The resistance of the sensors was obtained for a relative humidity variation from approximately 40% to 90% with an accuracy of ± 5% RH. The relative humidity dependence with the electrical resistance is generally represented in the following form [138]: 𝑆𝑅=𝑆 𝑅0=∆𝑅 𝑅0×1 ∆𝑅𝐻, (12) where SR is the sensor response, ∆𝑅 is the resistance variation, 𝑅0 is the resistance at room temperature and ∆𝑅𝐻 is the relative humidity variation. 20 4. Results and discussion In this chapter we present the results obtained for the two systems. First, the chapter presents the work focuses on the investigation of the relationship between the microstructure and the thermoresistive response of CrNx. Then, the relationship between the microstructure and the relative humidity sensing response of CrNxOy thin films under distinct amounts of nitrogen (N2+O2) in a mixture relation of 85% N2 and 15% O2, both deposited by DC-reactive magnetron sputtering. 4.1. Temperature sensor results 4.1.1. Structural and Morphological Characterization The crystal structure and texture evolution of the CrNx thin films have been evaluated by XRD, and the obtained diagrams are presented in Fig. 10, which shows the phase evolution of the CrNX thin films deposited onto Si substrates. The XRD diagrams (Fig. 10) suggest that the CrNx thin films are characterized by a polycrystalline structure. In general, with the increase of the N2 flux, it is observed an increase of the CrNx (111) (ICSD card no. 00–900– 8619) peak intensity [139]. The observed changes in the peak intensities and full width at half maximum (FWHM) as N2 flow rate vary from 0 up to 8 sccm can be attributed to differences in N2 content in the CrNX, as reported in Tab. 6 and verified in [139]. This feature indicates a clear transition from a Cr-type with base-centered cubic, BCC, structure and space group Im-3m (ICSD card no. 00 – 001–1261) to a CrN-type with face-centered cubic, FCC, structure and space group Fm-3m (ICSD card no. 00–900–8619). At N2 flow rates of 2 and 4 sccm, a coexistence of FCC cubic (ICSD card no. 00–900–8619) and hexagonal (ICSD card no. 00–003–1191) phases associated with CrN-type, is observed: while at N2 flow rate of 4 the shoulder at the right of the CrN (111) peak can be attributed to hexagonal Cr2N (110) (ICSD card no. 00–003–1191), the peak situated at 2  45º can be a reminiscence of the Cr BCC phase. 21 Figure 10: X-ray diffraction diagrams for the Cr and CrN samples. The peaks are indexed by using the ISCD car nos. 00–900–8619 and 00–003–1191 for CrN and Cr2N, respectively. Additionally, EDS measurements allow to identify a low concentration of O2 in the samples which is expected due to some superficial contamination, as well as from the annealing protocol, but not confirmed in the XRD diffractograms, in terms of any oxygen containing crystalline phase. The atomic percentages of Cr are reduced with increasing N2 content, while oxygen contents are enhanced during the annealing protocol, Tab. 6, indicating the occurrence of surface oxidation around the coatings as the temperature increases up to 523 K. However, we observe the absence of the peaks located at 41.4° (113) and 51.3° (024), associated with Cr2O3 phase (ICSD card no. 00 – 38 – 1479). The formation of Cr2O3 during annealing in air has been discussed in [140] and, from the results, the formation of Cr2O3 phase needs an air-annealing temperature higher than 873 K. The samples could also include an amorphous phase, but it has not been observed within the detecting precision of the XRD. Table 6: Elemental composition of the CrN samples. N2 Flow rate (sccm) [Cr] (±5 at%) [N] (±5 at%) [O] (±5 at%) 0 96.5 0.0 3.5 2 59.0 34.9 6.2 4 49.9 42.6 7.5 6 46.5 45.5 8.0 8 47.1 45.0 7.9 22 The texture coefficient calculated for the highly oriented peaks is represented in Fig. 11 (a). For a sample with randomly oriented crystallites 𝑇𝑐(ℎ𝑘𝑙)= 1, while higher values indicate the abundance of grains oriented along the (hkl) direction. Fig. 11 (a) shows that the highest Tc(hkl) value has been found for CrNx samples growth with an N2 flux of 6 sccm (𝑇𝑐(111)= 1.934) and 8 sccm (𝑇𝑐(111)= 1.430). On the other hand, for CrNx samples growth with an N2 flux of 2 and 4 sccm, the texture coefficient shows a value near the unity, Fig. 11 (b) (𝑇𝑐(111)= 1.032) and (𝑇𝑐(111)= 0.960) respectively, which is related to the coexistence of the cubic and hexagonal phases. Additionally, in order to determine the crystalline size (Ds) of the studied films, the Debye–Scherrer method [141] was used (Figure 11 (c)), demonstrating that increasing N2 flux leads to a crystallite size increase from  20 nm up to  23 nm. Figure 11: (a) Texture coefficient Tc(hkl) variations of the CrN thin films at different N2 concentrations. (b) Grains size of the films as a function of N2 concentrations obtained from the XRD results. a) b) 23 SEM micrograph images for the different films are represented in Fig. 12. Showing the characteristic zigzag columns of the Cr and CrNx thin films. Increasing N2 flux leads to a variation of the morphological features, namely in the columnar arrangement (type, density, and surface aspect). For the Cr samples (Fig. 12 (b)), the shape of the column looks like an elliptical profile and some porosity alignment is observed. By increasing the N2 flux up to 8 sccm, the zigzag-like structure is maintained. On the other hand, from the top view micrographs for the sample with an N2 flux of 4 sccm, Fig. 12 (c), it is observed that the surface of the films seems to grow without a preferential alignment, as also indicated by XRD results and confirmed by a texture value near the unity. Increasing, the N2 flux to 8 sccm (Fig. 12 (d)) leads to a reduction of the definition of the zig-zag columns when compared with the Cr samples, but the porosity increase between columns and columnar gaps are easily detected on the surface of the films. Figure 12: Representative SEM micrographs of the Cr samples, (a) Cr cross section, (b),(c) and (d) top view of the Cr, CrN-4sccm, and CrN-8sccm samples, respectively. However, the results of the crystallite size (𝐷𝑠) of the films, calculated by the Debye–Scherrer equation, show an increasing trend with the increase of N2 flux, leading to a maximum average crystallite size near of 23 nm, using Cr layer (a) (b) Zigzag structure (c) (d) 24 the (111) reflection. In fact, the use of the Scherrer's approach assumes symmetric crystallites and the SEM micrographs show that it is not the case in the studied Cr samples, Fig. 12 (b). Actually, the micrographs show that the CrNx thin films are formed by short segments of columnar grains for Cr films and that these segments decrease with the increase of the N2 flux and only the approximate size and not the shape can be inferred using the Scherrer's technique. Therefore, the sizes obtained by using Scherrer's method should be considered only for comparison purposes among samples. 4.1.2. Electrical and thermo-resistive measurements Temperature-dependent measurements were performed to obtain quantitative information about the sensitivity of the CrNx thin films and to evaluate their potential as temperature sensing coatings. Before that, and as explained above, the samples were submitted to an annealing protocol at 523 K for 120 min in order to obtain a stable thermosensitive response due to the structural refinement [142]. Figure 13 shows the variation of the electrical resistivity as a function of temperature for the CrNx thin films prepared with an N2 flux in the range of 4 up to 8 sccm, Measurements are presented in the temperature range of 300-473 K with a holding step of 600 seconds at temperatures of 323 K, 373 K, 423 K and 473 K to evaluate the stability for the response. The thermo-resistive response of the samples is very stable for the different N2 flux conditions in which they were prepared. When the samples prepared with an N2 flow rate of 4 sccm, Fig. 12 (a) shows a linear response with a hysteresis lower than 0.14%. Increasing the N2 flow rate from 4 to 6 and 8 sccm, Figs. 12 (c) and (e), leads to a sharp decrease in the linearity and the hysteresis disappears for the sample prepared with 6 sccm, being less than 0.22% for the sample prepared with 8 sccm. Figures 12 (b), (d) and (f) present a magnified time view in the range between 500 s and 2300 s, where it is observed the stability of the electrical signal as a function of temperature and time. The resistance variation rate at a constant temperature of 373 K after 600 s is very low for the CrNx thin films. Thus, the samples show very good temporal stability at a given temperature as corresponding to a metallic conduction mechanism [143]. Further, the high resistivity and the negative TCR values of CrNx thin films indicate high scattering caused by a large amount of disorder due to the amorphous structure of the CrNx thin films [144], the latter not being confirmed by the XRD results. 31 Figure 16 shows that increasing N2+O2 flux during the PVD-GLAD growth process leads to morphological feature variations of the CrNXOY thin films, namely the columnar arrangement, including type, density, and surface characteristics. For pristine Cr thin films, well-defined zigzags structures were produced, Fig.16 a). By increasing the amount of reactive gas flux from 0 to 4 sccm, the zigzag columns become not well-defined and the surface gaps reveal a highly porous material, as detected on the films surface, Fig. 16 d). Increasing, even more, the reactive flux until 10 sccm leads to a vanishing of the zigzag structure and the porous structure is considerably reduced. With the successive increase of the reactive N2+O2 flux from pure Cr to 10 sccm of N2+O2, the thickness of the produced columns decreases, from 138 nm to 78 nm, leading to a decrease of the grain boundaries, as observed in the surface images Fig. 16 b) and d). Figure 17 a) shows the deposition rate of the samples processed by PVD-GLAD. For samples produced with a reactive flux from 2 to 8 sccm, the deposition rate is practically the same, around 15 nm/min, decreasing to approximately 10 nm/min for the samples produced with 10 sccm. This change in the deposition rate seems to be correlated with the zigzag growth since for higher deposition rates there is a better zigzag like structure and higher porosity, which disappear for low rates. Moreover, RBS results of the composition show a high level of O2 (16,5 at %) in the samples produced without reactive gas, Figure 17 b). This fact can be attributed to the processing of the samples where the annealing protocol in-air from room temperature to a temperature of 200 ºC can induce an oxidation layer on the surface of the samples [152]. Excluding the Cr sample, the addition of N2+O2 reactive gas leads to an increase in the amount of O2. Besides, the N2 also shows initially the same trend than the O2 up to a maximum of 4 sccm. After that, the presence of N2 starts decreasing as the mixture flux increases. It is worth noting that although the mixture was composed by a ratio of ~0,18 O/N, increasing the amount of N2+O2 from 2 sccm to 10 sccm leads to an increase of the O/N ratio from 0.55 to 7.27. This behaviour is expected since oxygen shows a higher reactivity than nitrogen due to the chromium oxide higher binding enthalpy (-564 kJ/mol) when compared with chromium nitride (-125 kJ) [153,154], occupying the nitrogen position in the CrNX lattice. 32 a) 0 2 4 6 8 10 6 9 12 15 18 21 Deposition rate (nm/min) N2+O2 Flux (sccm) b) 0 2 4 6 8 10 0 20 40 60 80 100 Composition (at.%) N2+O2 Flux (sccm) Cr N O Figure 17: a) Deposition rate and b) elementary composition variation as a function of the N2+O2 flux. In order to evaluate whether CrNXOY combines the properties of chromium nitride and chromium oxide like Cr2O3, an XRD analysis was performed (Fig. 18). The results suggest that the Cr(N, O) thin films are characterized by a polycrystalline structure. For the sample deposited without reactive flux, the diffractograms show a very intense Cr (210) peak located at 2 ≈ 44.6º and a smaller one at 2 ≈ 64.8º (310), related with a Cr-type with BCC cubic structure (ICSD card no. 00–001–1261). For the intermediate fluxes (2 and 4 sccm) the diffractograms show an amorphous structure with a broad band at 2 ≈ 38.4º and 2 ≈ 44.8º, related to CrN (110) and CrN (002) diffraction peaks, respectively. For these fluxes, there seems to be a presence of both FCC and hcp CrN 33 phases that can be attributed to a Cr2N ditrigonal scalenohedral crystalline structure (βCr2N). For higher fluxes (6 to 8 sccm) an intense peak at 38,1º, FCC-CrN (111), and a smaller one at 64.0º, FCC-CrN (022) are observed. This transition from a BCC-Cr phase, a mixture of β-Cr2N and FCC-CrN phase, are in agreement with the results related in [154,155]. 30 35 40 45 50 55 60 65 70 75 10 sccm 8 sccm 6 sccm 4 sccm 2 sccm [310] Cr [002] Intensity (a.u.) 2 (deg.) CrN [110] 0 sccm [111] [200] [220] ¨ ¨ ¨ ¨ ¨ ¨ ¨ * * * + + ¨ * [220] + [210] + CrO + * ª ª Cr2N ª + [111] * Figure 18: X-ray diffraction spectra of the CrNxOy samples. By increasing the amount of N2+O2 to 10 sccm, the presence of the O2 in the composition can be confirmed in the XRD results with the presence of a CrO crystalline structure, identified at 2 ≈ 38.6º and 2 ≈ 44.8º as well as the peak tail at 2 ≈ 65º (ICSD card no. 01–078–0722). The intense Cr peak located at 2 ≈ 44.6º and the weak CrO peak at 2 ≈ 44.8º probably consist of a mixture of both Cr and CrO phases [156]. The oxygen may be integrated into the CrN cubic structure partially substituting the N atoms at low percentages since CrO is only known to crystallize at a temperature above 400 ºC or at oxygen atomic percentages above 40 at.% [36,157,158] which are in agreement with our results. The fact of the Cr has been prepared in a metal phase by the reactive sputtering method, the addition of N2+O2 in a first step induces a nitrification process in the produced coatings and, in a second step, the amount of O2 becomes most reactive than nitrogen producing chromium oxide. 34 4.2.2. Electrical properties The electrical resistivity of the films (Fig. 19), a key factor for the suitability of the developed materials for sensing applications, was determined at room temperature. The obtained electrical resistivity values allow to categorize the CrNxOy coatings into three different regimes, as represented in Fig. 19. The thin films prepared without reactive flux (absence of O2) revealed low electrical resistivity values (1.27×10-6 ± 5.02×10-8) (Zone I), typical of a common metallic-like behavior [159], even when the RBS results demonstrate a presence of O2 around 16.5 at.%, as a result of the annealing protocol [152]. The Zone II correspond to the coatings prepared with intermediate reactive flux and is characterized by intermediate electrical resistivity values, increasing from 2.43×10-6 ± 4.54×10-8 .m to 3.80×10-2 ± 6.20×10-6 .m, which can be indexed to a semiconductor type of response. Finally, the samples produced with the highest oxygen contents (Zone III) show higher resistivity values (in the order of 4.50×109 ± 4.20×107 .m), typical of an insulator-type behavior. 0246810 10-7 10-4 10-1 102 105 108 1011 1014 Resistivity (.m) Resistivity N2+O2 Flux (sccm) 150 165 180 195 210 225 240 255 270 285 300 Grain size (nm) 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 160 180 200 220 240 260 280 300 Zone I Zone II Zone III 0 10 20 30 40 50 60 70 80 O2 at.% O2 (at.%) Figure 19: Room temperature resistivity, grain size and amount of O2 as a function of the N2+O2 reactive flux. The electrical resistivity is the result of various scattering processes that can be interpreted in terms of the Mayadas-Shatzkes theory [149]. Thin films often present an island or grain-like structures, showing thus a discontinuous morphology. When these grains or islands show dimensions on the order of the conduction electron mean free path, the scattering at the grain boundaries leads to a very high resistivity. In fact, and using the Debye–Scherrer method [141] to determine the crystalline size of the coating from XRD results, it is shown that by shifting from the lacking to the overload oxygen content zone (Fig. 19): a) the grain size increases from  191 nm up to  249 nm, b) the amount of oxygen in the coatings increases from  16.5 at.% to  59.1 at.%, and c) 35 the grain boundaries become denser, Fig. 16 f). It can be assumed that the contribution to the electrical resistivity by the grain boundaries and oxygen concentration is larger than the contribution of the grains size. 4.2.3. Relative Humidity Sensing Characteristics Taking into account the main goal of this work, the CrNxOy coatings prepared with different amounts of N2+O2 were evaluated through the resistance variation as a function of relative humidity (RH) variation from  40% to near 90% at a constant temperature of 25 °C. The repeatability is addressed in Fig. 20, which shows the measured electrical resistance plotted versus relative humidity cyclic variation. The black dots represent the raw data. These data represent four cycles of ultrasonic humidifier on-off switch beaker with distilled water. The purple line represents the raw data from a commercial sensor acquired from a commercial datalogger (Amprobe TR200-A). a) 3000 4000 5000 6000 7000 8000 9000 10000 127.6 127.8 128.0 128.2 Resistance Humidity Time (s) Resistance () 50 60 70 80 Humidity (%) b) 4600 4800 5000 5200 5400 5600 5800 6000 6200 127.6 127.8 128.0 128.2 Resistance Humidity Time (s) Resistance () ~60 s 50 60 70 80 Humidity (%) ~345 s c) 3000 4000 5000 6000 7000 8000 9000 10000 1.0x109 1.5x109 2.0x109 2.5x109 3.0x109 3.5x109 4.0x109 4.5x109 Resistance Humidity Time (s) Resistance () 40 50 60 70 80 90 Humidity (%) d) 4600 4800 5000 5200 5400 5600 5800 6000 6200 1.0x109 1.5x109 2.0x109 2.5x109 3.0x109 3.5x109 4.0x109 4.5x109 Resistance Humidity Time (s) Resistance () ~28 s ~345 s 40 50 60 70 80 90 Humidity (%) Figure 20: Sensor response characterization by the electrical resistance variation with the relative humidity for the samples deposited with a flux of N2+O2 of: a) 4 sccm, c) 10 sccm and the corresponding magnifications b) and d) for a single cycle. 36 Fig. 20b and d show the time response and recovery of the CrNxOy thin films and the signal response of the commercial humidity sensor. The response times are defined here as the time it takes for the signal to reach from 52% of the initial humidity value to 75% of the final humidity value. The recovery times are defined as the time it takes for the signal to reach from 80% of the initial humidity value to 60 % of the final humidity value. The regions are marked with dashed green and red lines in Fig. 20, respectively. The measured response times are 345 s for both systems produced with a flux of N2+O2 of 4 sccm, Fig. 20 b), and 10 sccm, Fig. 20 d). The recovery time decreases from 60 s to 28 s with increasing amount of O2 in the CrNxOy system. The combination of these measurements shows that the response and recovery times of the devices are on the order of seconds and are possibly much faster than observed due to the limited response time of the experimental setup. However, if we compare the electrical evolution of the samples with increasing N2+O2 fluxes, Fig. 20 a) and c), the CrNxOy thin films react to the humidity, decreasing the electrical resistance with increasing relative humidity and the time response decreases with increasing O2 in the composition, which suggests higher sensibility to the humidity. Thus, Eq. 12 allows to determine the sensor response of the CrNxOy films in the presence of humidity. The results are presented in Fig. 20. a) 0 2 4 6 8 10 -20 -15 -10 -5 0 SR Resistivity N2+O2 flux (sccm) SR (x10-3) (%) 10-8 10-2 104 1010 Resistivity (.m) b) Figure 21: a) Electrical resistivity and 𝛥 RH as a function of the N2+O2 flow rate applied during the reactive sputtering of the CrNxOy samples, b) adsorbed layers on a high humidity environment . Figure 21 shows that the lowest SR values are obtained for the lower N2+O2 flow rates (0 and 2 sccm). For chromium samples (0 sccm), the sensibility shows a value of SR = 2.36 ×10-5 ± 3.00×10-7 (RH %)-1, in the same order than the sample prepared with 2 sccm of N2+O2 flux (SR = 1.68×10-5 ± 1.30×10-6 (RH %)-1The low value can be attributed to the low electrical resistance and due to the fact that Cr was prepared in a metal phase. For N2+O2 37 flow rates of 4 to 10 sccm, the electrical resistivity increases and the highest negative SR values of 8.56×10-5 ± 6.00×10-7 to 1.87×10-2 ± 7.00×10-5 (RH %)-1 are obtained, confirming the highest time response of the system with higher O2 concentration. The relative humidity sensitivity is closely related to the concentration of water adsorption sites and since CrN is hydrophobic and oxides are hydrophilic, the relative humidity sensitivity increases in the presence of O2 [4] as verified in our study. The behaviour of the humidity in the samples will influence the number of adsorbed layers of water [4,48], as represented in Fig. 21 b). For low humidity environments, only a chemisorbed layer of hydroxyl ions will be present at the oxide surface, resulting from the water ionization reaction of the water vapor, in which the water loses one H+ atoms forming surface hydroxide ions (OH). The presence of water molecules and OHinto the surface, covering the surface, lead to proton H+ charge carriers, occurring the proton migration by hopping between adjacent OHsites, which promotes a decrease in the electrical resistivity of the sample as a result of the quantum tunnelling of electrons between neighbouring water molecules [4]. After the hydroxyl group formation, and with increasing humidity, the next water molecule layer is physisorbed (Fig. 21) by hydrogen double bonds on hydroxyl groups in which a proton is transferred from a hydroxyl group forming an H3O+ ion. For even higher humidity levels, water will be able to condense in the capillary sites of the surface, increasing the electrolytic conduction by proton hopping, being the protons able to tunnel between adjacent water particles [4]. In this sense, the sensibility of the films should be attributed not only to the presence of oxygen in the CrNxOy coatings but also to the porosity present in the surface of the coatings, Fig. 16, which will facilitate the adsorption of water vapor on its lower layers giving great scope for enhancement in the sensitivity of the relative humidity sensors. As a conclusion, coatings based on chromium oxynitrides have been developed for humidity sensing applications by tailoring thin film microstructure and composition. Increasing the amount of O2 in the composition allows to increase the sensitivity to the humidity response of CrNxOy and, consequently, the SR values. Thus, the results show a way to explore the CrNxOy system with negative SR for the development of multifunctional highperformance coatings with humidity sensing characteristics. 38 5. Conclusions Thin film chromium nitride samples and chromium oxynitride samples were successfully synthesized by GLAD assisted DC reactive magnetron sputtering. The samples of chromium nitride allowed a set of conclusions: • Nitrogen increased on the samples as the nitrogen flux increased on the deposition process. • As the flux of nitrogen increases the structure of the samples changes from a Cr-BCC structure to a CrN-FCC structure with the coexistence with a Cr2N hexagonal structure at lower nitrogen fluxes. The increase in nitrogen flux also promotes the increase of the grain size of the samples. • The GLAD process promotes a zigzag microstructure on the sample. Porosity is also observed. • It has been found a negative TCR coefficient for the thin films produced with N2 flux of 4, 6 and 8 sccm, which correlates with the variation of the morphological features, namely in the columnar arrangement (type, density, hardness and surface aspect) and with the transition from Cr to CrNx cubic phase. Besides, the thermoresistivity showed very stable behavior as a function of both temperature and time. Further, the grain-boundary model allows calculating the 𝛽𝐺 coefficients theoretically, which are in agreement with the experimental ones, and therefore allow to disclose the physical mechanism behind the obtained response. • Our results pave the way to explore the CrNx system with negative TCR for the development of multifunctional high-performance hard coatings with temperature sensing characteristics. On the other hand, the conclusions achieved by the characterization of the chromium oxynitride samples were: • The increase of the N2+O2 flux during the deposition of the Cr process increased the atomic percentage of N2+O2 on the samples. More, this increase also provoked an increase of O/N ratio due to the higher reactivity of the oxygen when compared with the nitrogen reactivity. • The addition of N2+O2 produces, for lower fluxes, chromium nitride coatings. On the other hand, for higher fluxes, the formation of chromium oxide is induced due to the high oxygen reactivity. • SEM images showed granular samples with high porosity with zigzag growth, this structure was more present on samples deposited with 4, 6 and 8 sccm of N2+O2 mixture flow. The deposition rate was lower in the 10 sccm deposited sample and was approximately the same to all other samples. 39 • The amount of O2 present in the composition of the coatings leads to high electrical resistivity and increase the relative sensitivity. • The films deposited with high N2+O2 flux show a higher sensor response to humidity. The sensibility of the films should be attributed not only to the presence of oxygen in the CrNXOY coatings but also to the porosity presented in the surface of the coatings, which promote the adsorption of water vapor on its lower layers, allowing the improvement of the sensitivity of the sensors. . 40 6. Future work The obtained results of this work allow us to open new lines of study and mechanisms for the future. The inclusion of nitrogen and oxygen into chromium matrices has shown to be advantageous in creating a temperature and humidity transductor. Further investigation needs optimization and calibration in this array of sensors. The implementation of a functional prototype with integrated sensor calibrated, as well with various environmental conditions is also needed. New thin films with different transition metals should also be considered and produced with inclined columns and spirals by Glancing Angle Deposition technique. The GLAD architectures (zig-zags) have brought excellent thermoresistive and humidity responses, with promising results for sensor devices applications. The possibility to sculpture the Cr-based thin films will expand the possibilities to control the material, at the nanometer scale, and hence to enhance the elasticity of the films deposited on flexible-like polymers. The obtained results also raise the possibility of upscaling these materials for use into a production scale. 47 146. [66] A.S. Afify, S. Ahmad, R.A. Khushnood, P. Jagdale, J.M. Tulliani, Elaboration and characterization of novel humidity sensor based on micro-carbonized bamboo particles, Sensors Actuators, B Chem. 239 (2017) 1251–1256. https://doi.org/10.1016/j.snb.2016.09.130. [67] A. Kromka, M. Davydova, B. Rezek, M. Vanecek, M. Stuchlik, P. Exnar, M. Kalbac, Gas sensing properties of nanocrystalline diamond films, Diam. Relat. Mater. 19 (2010) 196–200. https://doi.org/10.1016/j.diamond.2009.10.006. [68] L. Du, Y. Zhang, Y. Lei, H. Zhao, Synthesis of high-quality CdS nanowires and their application as humidity sensors, Mater. Lett. 129 (2014) 46–49. https://doi.org/10.1016/j.matlet.2014.05.002. [69] R. Demir, S. Okur, M. Şeker, Electrical characterization of CdS nanoparticles for humidity sensing applications, Ind. Eng. Chem. Res. 51 (2012) 3309–3313. https://doi.org/10.1021/ie201509a. [70] S. Okur, N. Üzar, N. Tekgüzel, A. Erol, M. Çetin Arkan, Synthesis and humidity sensing analysis of ZnS nanowires, Phys. E Low-Dimensional Syst. Nanostructures. 44 (2012) 1103–1107. https://doi.org/10.1016/j.physe.2010.08.015. [71] N. Üzar, S. Okur, M.Ç. Arikan, Investigation of humidity sensing properties of ZnS nanowires synthesized by vapor liquid solid (VLS) technique, Sensors Actuators, A Phys. 167 (2011) 188–193. https://doi.org/10.1016/j.sna.2010.10.005. [72] Y.P. Leung, W.C.H. Choy, T.I. Yuk, Linearly resistive humidity sensor based on quasi one-dimensional ZnSe nanostructures, Chem. Phys. Lett. 457 (2008) 198–201. https://doi.org/10.1016/j.cplett.2008.04.005. [73] G. Dražič, M. Trontelj, Preparation and properties of ceramic sensor elements based on MgCr2O4, Sensors and Actuators. 18 (1989) 407–414. https://doi.org/10.1016/0250-6874(89)87046-5. [74] Y. Yokomizo, S. Uno, M. Harata, H. Hiraki, K. Yuki, Microstructure and humidity-sensitive properties of ZnCr2O4-LiZnVO4 ceramic sensors, Sensors and Actuators. 4 (1983) 599–606. https://doi.org/10.1016/0250-6874(83)85073-2. [75] K.C. Sajjan, A.S. Roy, A. Parveen, S. Khasim, Analysis of DC and AC properties of a humidity sensor based on polyaniline-chromium oxide composites, J. Mater. Sci. Mater. Electron. 25 (2014) 1237–1243. https://doi.org/10.1007/s10854-014-1715-7. [76] T. Chandrasekhar, N. Sasidhar, B. Chethan, Y.T. Ravikiran, Effect of chromium oxide in improving 48 humidity sensing properties of polypyrrole/chromium oxide composite, AIP Conf. Proc. 2274 (2020) 020012. https://doi.org/10.1063/5.0022613. [77] S. Pokhrel, K.S. Nagaraja, Electrical and humidity sensing properties of Chromium(III) oxide-tungsten(VI) oxide composites, Sensors Actuators, B Chem. 92 (2003) 144–150. https://doi.org/10.1016/S09254005(03)00251-X. [78] W. Daoping, C. Dandan, W. Lingling, W. Yuan, Q. Li, T. Wang, High performance humidity sensors based on CeO2 nanoparticles, Sensors Actuators B. Chem. 215 (2015) 125–132. https://doi.org/10.1016/j.snb.2015.03.051. [79] H. Feng, C. Li, T. Li, F. Diao, T. Xin, B. Liu, Y. Wang, Three-dimensional hierarchical SnO2 dodecahedral nanocrystals with enhanced humidity sensing properties, Sensors Actuators B. Chem. 243 (2017) 704– 714. https://doi.org/10.1016/j.snb.2016.12.043. [80] J. Xu, K. Yu, J. Wu, D. Shang, L. Li, Y. Xu, Z. Zhu, Synthesis , field emission and humidity sensing characteristics of honeycomb-like CuO, J. Phys. D. Appl. Phys. 42 (2009) 075417. https://doi.org/10.1088/0022-3727/42/7/075417. [81] P.K. Kannan, R. Saraswathi, J.B.B. Rayappan, A highly sensitive humidity sensor based on DC reactive magnetron sputtered zinc oxide thin film, Sensors Actuators, A Phys. 164 (2010) 8–14. https://doi.org/10.1016/j.sna.2010.09.006. [82] X. Peng, J. Chu, B. Yang, P.X. Feng, Sensors and Actuators B : Chemical Mn-doped zinc oxide nanopowders for humidity sensors, Sensors Actuators B. Chem. 174 (2012) 258–262. https://doi.org/10.1016/j.snb.2012.07.011. [83] B.C. Yadav, R. Srivastava, C.D. Dwivedi, P. Pramanik, Sensors and Actuators A : Physical Synthesis of nano-sized ZnO using drop wise method and its performance as moisture sensor, Sensors Actuators A. 153 (2009) 137–141. https://doi.org/10.1016/j.sna.2009.05.010. [84] S. Rajarajan, G. Ramkumar, A comparative study of humidity sensing and photocatalytic applications of pure and nickel (Ni) -doped WO3 thin films, Appl. Phys. A. 123 (2017) 401. https://doi.org/10.1007/s00339-017-0983-5. [85] V. Jadkar, A. Pawbake, R. Waykar, A. Jadhavar, A. Date, D. Late, H. Pathan, S. Gosavi, S. Jadkar, Synthesis of g-WO3 thin films by hot wire-CVD and investigation of its humidity, Phys. Status Solidi A. 214 (2017) 1600717. https://doi.org/10.1002/pssa.201600717. 49 [86] Z. Wang, L. Shi, F. Wu, S. Yuan, Y. Zhao, M. Zhang, The sol – gel template synthesis of porous TiO2 for a high performance humidity sensor, Nanotechnology. 22 (2011) 275502. https://doi.org/10.1088/09574484/22/27/275502. [87] J. Wang, M. Su, J. Qi, L. Chang, Sensors and Actuators B : Chemical Sensitivity and complex impedance of nanometer zirconia thick film humidity sensors, Sensors Actuators B Chem. 139 (2009) 418–424. https://doi.org/10.1016/j.snb.2009.03.070. [88] Q. Qi, Y. Feng, T. Zhang, X. Zheng, G. Lu, Influence of crystallographic structure on the humidity sensing properties of KCl-doped TiO2 nanofibers, Sensors Actuators B Chem. 139 (2009) 611–617. https://doi.org/10.1016/j.snb.2009.03.041. [89] W. Lin, C. Liao, T.-C. Chang, S.-H. Chen, R.-J. Wu, Humidity sensing properties of novel graphene/TiO2 composites by sol-gel process, Sensors Actuators B. Chem. 209 (2015) 555–561. https://doi.org/10.1016/j.snb.2014.12.013. [90] Z. Li, A.A. Haidry, B. Gao, T. Wang, Z. Yao, The effect of Co-doping on the humidity sensing properties of ordered mesoporous TiO2, Appl. Surf. Sci. 412 (2017) 638–647. https://doi.org/10.1016/j.apsusc.2017.03.156. [91] Z. Li, H. Zhang, W. Zheng, W. Wang, H. Huang, C. Wang, A.G. Macdiarmid, Y. Wei, D.U. V, V. Pennsyl, Highly Sensitive and Stable Humidity Nanosensors Based on LiCl Doped TiO2 Electrospun Nanofibers, J. Am. Ceram. Soc. 130 (2008) 5036–5037. [92] K. Katayama, K. Hasegawa, Y. Takahashi, T. Akiba, H. Yanagida, Humidity Sensitivity of Nb2Os-doped TiO2 Ceramics, Sensors Actuators A. 24 (1990) 55–60. [93] H. Yin, J. Ni, W. Jiang, Z. Zhang, K. Yu, Synthesis , field emission and humidity sensing characteristics of monoclinic VO2 nanostructures, Phys. E. 43 (2011) 1720–1725. https://doi.org/10.1016/j.physe.2011.05.030. [94] M.S. Pawar, P.K. Bankar, M.A. More, D.J. Late, Ultra-thin V2O5 Nanosheets Based Humidity Sensor, Photodetector and Its Enhanced Field Emission Properties, RSC Adv. 5 (2015) 88796–88804. https://doi.org/10.1039/C5RA17253E. [95] C. Xu, K. Miyazaki, Humidity sensor with manganese oxide for room temperature, Sens. Achuztors B. 13– 14 (1993) 523–524. https://doi.org/https://doi.org/10.1016/S0925-4005(97)00330-4. [96] V. Jeseentharani, A. Dayalan, K.S. Nagaraja, Co-precipitation synthesis, humidity sensing and 50 photoluminescence properties of nanocrystalline Co2+ substituted zinc(II)molybdate (Zn1–xCoxMoO4; x = 0, 0.3, 0.5, 0.7, 1), Solid State Sci. 67 (2017) 46–58. https://doi.org/10.1016/j.solidstatesciences.2017.02.008. [97] W. Qu, J.U. Meyer, Thick-film humidity sensor based on porous MnWO4 material, Meas. Sci. Technol. 8 (1997) 593–600. https://doi.org/10.1088/0957-0233/8/6/002. [98] W. Qu, Wojtek Wlodarski, Jorg-Uwe Meyer, Comparative study on micromorphology and humidity sensitive properties of thin–film and thick–film humidity sensors based on semiconducting MnWO4, Sensors Actuators B Actuators. 64 (2000) 76–82. https://doi.org/https://doi.org/10.1016/S09254005(99)00487-6. [99] S. Upadhyay, Humidity-sensitive characteristic of Ba0.99La0.01SnO3, Phys. Status Solidi Appl. Mater. Sci. 205 (2008) 1113–1119. https://doi.org/10.1002/pssa.200723006. [100] H. Klym, A. Ingram, O. Shpotyuk, I. Hadzaman, O. Hotra, Y. Kostiv, Nanostructural Free-Volume Effects in Humidity-Sensitive MgO-Al2O3 Ceramics for Sensor Applications, J. Mater. Eng. Perform. 25 (2016) 866–873. https://doi.org/10.1007/s11665-016-1931-9. [101] K.I. Arshak, K. Twomey, Investigation into a novel humidity sensor operating at room temperature, Microelectronics J. 33 (2002) 213–220. https://doi.org/10.1016/S0026-2692(01)00150-1. [102] M. Sabarilakshmi, K. Janaki, Effect of Mg concentration on structural, optical and humidity sensing performance of SnO2 nanoparticles prepared by one step facile route, J. Mater. Sci. Mater. Electron. 28 (2017) 8101–8107. https://doi.org/10.1007/s10854-017-6516-3. [103] R. Sundaram, K.S. Nagaraja, Electrical and humidity sensing properties of lead(II) tungstate-tungsten(VI) oxide and zinc(II) tungstate-tungsten(VI) oxide composites, Mater. Res. Bull. 39 (2004) 581–590. https://doi.org/10.1016/j.materresbull.2003.12.014. [104] Y. Shimizu, H. Okada, Humidity-Sensitive Characteristics of Porous La-Ti-V-0 Glass-Ceramics, J. Am. Ceram. Soc. 72 (1987) 51–53. https://doi.org/https://doi.org/10.1111/j.1151-2916.1989.tb06149.x. [105] Q. Yuan, N. Li, W. Geng, Y. Chi, J. Tu, X. Li, C. Shao, Humidity sensing properties of mesoporous iron oxide/silica composite prepared via hydrothermal process, Sensors Actuators, B Chem. 160 (2011) 334– 340. https://doi.org/10.1016/j.snb.2011.07.057. [106] A.M. Edwin Suresh Raj, C. Mallika, K. Swaminathan, O.M. Sreedharan, K.S. Nagaraja, Zinc(II) oxide-zinc(II) molybdate composite humidity sensor, Sensors Actuators, B Chem. 81 (2002) 229–236. 51 https://doi.org/10.1016/S0925-4005(01)00957-1. [107] K. Hayat, F. Niaz, S. Ali, M.J. Iqbal, M. Ajmal, M. Ali, Y. Iqbal, Thermoelectric performance and humidity sensing characteristics of La2CuO4 nanofibers, Sensors Actuators, B Chem. 231 (2016) 102–109. https://doi.org/10.1016/j.snb.2016.02.127. [108] A.C. Caballero, M. Villegas, M. Viviani, M.T. Buscaglia, M. Leoni, Effect of humidity on the electrical response of porous BaTiO3 ceramics, J. Mater. Sci. Lett. 18 (1999) 1297–1299. https://doi.org/https://doi.org/10.1023/A:1006662805186. [109] J.G. Kim, Electrical properties and fabrication of porous BaTiO3-based ceramics, J. Mater. Sci. Lett. 21 (2002) 477–479. https://doi.org/10.1023/A:1015334623555. [110] J. Yuk, T. Troczynski, Sol-gel BaTiO3 thin film for humidity sensors, Sensors Actuators, B Chem. 94 (2003) 290–293. https://doi.org/10.1016/S0925-4005(03)00371-X. [111] J. Wang, B.K. Xu, S.P. Ruan, S.P. Wang, Preparation and electrical properties of humidity sensing films of BaTiO3/polystrene sulfonic sodium, Mater. Chem. Phys. 78 (2003) 746–750. https://doi.org/10.1016/S0254-0584(02)00421-2. [112] J. Wang, H. Wan, Q. Lin, Properties of a nanocrystalline barium titanate on silicon humidity sensor, Meas. Sci. Technol. 14 (2003) 172–175. https://doi.org/10.1088/0957-0233/14/2/303. [113] B. Ertug, T. Boyraz, O. Addemir, Investigation of the electrical conductivity and humidity sensitivity characteristics of BaTiO3 ceramics with PMMA additive, Arch. Metall. Mater. 57 (2012) 437–442. https://doi.org/10.2478/v10172-012-0043-2. [114] L. Wang, Y. He, J. Hu, Q. Qi, T. Zhang, DC humidity sensing properties of BaTiO3 nanofiber sensors with different electrode materials, Sensors Actuators, B Chem. 153 (2011) 460–464. https://doi.org/10.1016/j.snb.2010.11.016. [115] J. Lu, L. Cheng, Y. Zhang, J. Huang, C. Li, Effect of the seed layer on surface morphology and humidity sensing property of CoTiO3 nanocrystalline film, Ceram. Int. 43 (2017) 5823–5827. https://doi.org/10.1016/j.ceramint.2017.01.080. [116] R. Ganeshkumar, C.W. Cheah, R. Zhao, Annealing temperature and bias voltage dependency of humidity nanosensors based on electrospun KNbO3 nanofibers, Surfaces and Interfaces. 8 (2017) 60–64. https://doi.org/10.1016/j.surfin.2017.04.010. [117] D. Li, J. Zhang, L. Shen, W. Dong, C. Feng, C. Liu, S. Ruan, Humidity sensing properties of SrTiO3 52 nanospheres with high sensitivity and rapid response, RSC Adv. 5 (2015) 22879–22883. https://doi.org/10.1039/c5ra00451a. [118] Y. Tian, M. Xu, P. Huang, E. Lei, H. Hao, C. Cui, The abnormal sensitivity and its mechanism of (Ba, Pb)TiO3 semiconductor ceramics, Chinese Sci. Bull. 50 (2005) 936–939. https://doi.org/10.1360/04we0107. [119] Y.-C. Yeh, T.-Y. Tseng, D.-A. Chang, Electrical Properties Porous Titania Ceramic Humidity Sensors, J. Am. Ceram. Soc. 71 (1989) 1472–1475. https://doi.org/https://doi.org/10.1111/j.11512916.1989.tb07679.x. [120] Z.A. Ansari, T.G. Ko, J.H. Oh, Humidity sensing behavior of thick films of strontium-doped lead-zirconiumtitanate, Surf. Coatings Technol. 179 (2004) 182–187. https://doi.org/10.1016/S02578972(03)00820-X. [121] M. Ahmadipour, M.F. Ain, Z.A. Ahmad, Effect of thickness on surface morphology, optical and humidity sensing properties of RF magnetron sputtered CCTO thin films, Appl. Surf. Sci. 385 (2016) 182–190. https://doi.org/10.1016/j.apsusc.2016.05.098. [122] J. Wang, B. Xu, G. Liu, J. Zhang, T. Zhang, Improvement of nanocrystalline BaTiO3 humidity sensing properties, Sensors Actuators, B Chem. 66 (2000) 159–160. https://doi.org/10.1016/S09254005(00)00308-7. [123] A. Kazemzadeh, F.A. Hessary, N. Jafari, New solid state sensor for detection of humidity, based on Ni, Co, and Mn oxide nano composite doped with Lithium, Sensors Transducers J. 94 (2008) 161–169. [124] L. Jingbo, L. Wenchao, Z. Yanxi, W. Zhimin, Preparation and characterization of Li+-modified CaxPb1-xTiO3 film for humidity sensor, Sensors Actuators, B Chem. 75 (2001) 11–17. https://doi.org/10.1016/S0925-4005(00)00686-9. [125] A. Bonavita, A. Caddemi, N. Donato, P. Accordino, S. Galvagno, G. Neri, U. Messina, C. Sperone, S. Agata, U. Messina, C. Sperone, S. Agata, Electrical characterization and modeling of thin-film humidity sensors, in: Proc. IEEE Conf. Electron. Circuits Syst. ICECS 2001, Sept. 2–5, Malta, 2001: pp. 673–676. https://doi.org/10.1109/ICECS.2001.957565. [126] D. Depla, S. Mahieu, J.E. Greene, Sputter Deposition Processes, in: P.M. Martin (Ed.), Handb. Depos. Technol. Film. Coatings, 3rd ed., Elsevier Inc, Oxford, 2010: pp. 253–296. https://doi.org/https://doi.org/10.1016/B978-0-8155-2031-3.00005-3. 53 [127] M.T. Johnson, C.B. Carter, J. Michael, SEM Analysis of Oxide Thin Films and Reactions, J. Am. Ceram. Soc. 82 (1999) 1644–1646. http://dx.doi.org/10.1111/j.1151-2916.1999.tb01981.x. [128] S. Nasrazadani, S. Hassani, Modern analytical techniques in failure analysis of aerospace, chemical, and oil and gas industries, in: A.S.H. Makhlouf, M. Aliofkhazraei (Eds.), Handb. Mater. Fail. Anal. with Case Stud. from Oil Gas Ind., 1st ed., Butterworth Heinemann, Oxford, 2016: pp. 39–54. https://doi.org/10.1016/B978-0-08-100117-2.00010-8. [129] D. Bell, A. Garratt-Reed, Energy Dispersive X-ray Analysis in the Electron Microscope, 1st ed., BIOS Scientific Publishers Limited, Oxford, 2003. https://doi.org/10.4324/9780203483428. [130] N.P. Barradas, C. Jeynes, S.M. Jackson, NOM RBS / Simulated annealing analysis of buried SiCO , layers formed by implantation of 0 into cubic silicon carbide, Nucl. Instruments Methods Phys. Res. B. 136– 138 (1998) 1168–1171. https://doi.org/https://doi.org/10.1016/S0168-583X(97)00686-1. [131] M.H. Herman, Applications of Rutherford backscattering spectrometry to refractory metal silicide characterization, J. Vac. Sci. Technol. B 2. 748 (1984). https://doi.org/10.1116/1.582873. [132] N.P. Barradas, C. Pascual-Izarra, Double scattering in RBS analysis of PtSi thin films on Si, Nucl. Instruments Methods Phys. Res. B. 228 (2005) 378–382. https://doi.org/10.1016/j.nimb.2004.10.074. [133] N.P. Barradas, M.A. Reis, Accurate calculation of pileup effects in PIXE spectra from first principles, XRAY Spectrosc. 35 (2006) 232–237. https://doi.org/10.1002/xrs.903. [134] Y. Waseda, E. Matsubara, K. Shinoda, X-ray Diffraction Crystallography: Introduction, Examples and Solved Problems, 1st ed., Springer, Berlin, 2011. https://doi.org/10.1007/978-3-642-16635-8. [135] M.K. Gupta, N. Sinha, B. Kumar, Synthesis and Comparative Study of ZnO Nanorods for Structural , Optical and Dielectric Behaviour, Integr. Ferroelectr. 118 (2010) 61–66. https://doi.org/10.1080/10584587.2010.489478. [136] P.R.N. Childs, Practical Temperature Measurement, 1st ed., Butterworth-Heinemann, Oxford, 2001. https://doi.org/https://doi.org/10.1016/B978-0-7506-5080-9.X5000-X. [137] L.C. Fernandes, D.M. Correia, N. Pereira, C.R. Tubio, S. Lanceros-Méndez, Highly Sensitive Humidity Sensor Based on Ionic Liquid–Polymer Composites, ACS Appl. Polym. Mater. 1 (2019) 2723–2730. https://doi.org/10.1021/acsapm.9b00675. [138] A.D. Smith, K. Elgammal, F. Niklaus, A. Delin, A.C. Fischer, S. Vaziri, F. Forsberg, M. Råsander, H. 54 Hugosson, L. Bergqvist, S. Schröder, S. Kataria, M. Östling, M.C. Lemme, Resistive graphene humidity sensors with rapid and direct electrical readout, Nanoscale. 7 (2015) 19099–19109. https://doi.org/10.1039/c5nr06038a. [139] T. Elangovan, P. Kuppusami, R. Thirumurugesan, V. Ganesan, E. Mohandas, D. Mangalaraj, Nanostructured CrN thin films prepared by reactive pulsed DC magnetron sputtering, Mater. Sci. Eng. B Solid-State Mater. Adv. Technol. 167 (2010) 17–25. https://doi.org/10.1016/j.mseb.2010.01.021. [140] K. Ibrahim, M.M. Rahman, X. Zhao, J.P. Veder, Z. feng Zhou, E. Mohammadpour, R.H. Majeed, A.N. Nikoloski, Z.T. Jiang, Annealing effects on microstructural, optical, and mechanical properties of sputtered CrN thin film coatings: Experimental studies and finite element modeling, J. Alloys Compd. 750 (2018) 451–464. https://doi.org/10.1016/j.jallcom.2018.04.012. [141] V. Uvarov, I. Popov, Metrological characterization of X-ray diffraction methods at different acquisition geometries for determination of crystallite size in nano-scale materials, Mater. Charact. 85 (2013) 111– 123. https://doi.org/10.1016/j.matchar.2013.09.002. [142] X. Pang, K. Gao, F. Luo, H. Yang, L. Qiao, Y. Wang, A.A. Volinsky, Annealing effects on microstructure and mechanical properties of chromium oxide coatings, Thin Solid Films. 516 (2008) 4685–4689. https://doi.org/10.1016/j.tsf.2007.08.083. [143] C.K. Chung, A. Nautiyal, T.S. Chen, Y.L. Chang, Grain boundary scattering for temperature coefficient of resistance (TCR) behaviour of Ta-Si-N thin films, J. Phys. D. Appl. Phys. 41 (2008) 185404 (5pp). https://doi.org/10.1088/0022-3727/41/18/185404. [144] G.S. Chung, W.J. Lee, J.S. Song, Characteristics of chromium nitride thin-film strain gauges, Sensors Mater. 13 (2001) 013–023. [145] J.H. Tyan, J.T. Lue, Grain boundary scattering in the normal state resistivity of superconducting NbN thin films, J. Appl. Phys. 75 (1994) 325–331. https://doi.org/10.1063/1.355853. [146] M.A. Park, K. Savran, Y.J. Kim, Weak localization and the Mooij rule in disordered metals, Phys. Status Solidi Basic Res. 237 (2003) 500–512. https://doi.org/10.1002/pssb.200301654. [147] Y.M. Blanter, V.M. Vinokur, L.I. Glazman, Weak localization in metallic granular media, Phys. Rev. B - Condens. Matter Mater. Phys. 73 (2006) 165322. https://doi.org/10.1103/PhysRevB.73.165322. [148] S. Srinivas, S.N. Kaul, S.N. Kane, Enhanced electron-electron interaction, weak localization and electronmagnon scattering contributions to electrical resistivity in Feand Co-based metallic glass wires, J. Non. 55 Cryst. Solids. 248 (1999) 211–223. https://doi.org/10.1016/S0022-3093(99)00144-1. [149] A.F. Mayadas, M. Shatzkes, Electrical-resistivity model for polycrystalline films: The case of arbitrary reflection at external surfaces, Phys. Rev. B. 1 (1970) 1382–1389. https://doi.org/10.1103/PhysRevB.1.1382. [150] R. Sanjińs, M. Benkahoul, C.S. Sandu, P.E. Schmid, F. Ĺvy, Relationship between the physical and structural properties of NbzSiyNx thin films deposited by dc reactive magnetron sputtering, J. Appl. Phys. 98 (2005) 123511. https://doi.org/10.1063/1.2149488. [151] K.S. Karimov, K.Y. Cheong, M. Saleem, I. Murtaza, M. Farooq, A.F. Mohd Noor, Ag/PEPC/NiPc/ZnO/Ag thin film capacitive and resistive humidity sensors, J. Semicond. 31 (2010) 054002. https://doi.org/10.1088/1674-4926/31/5/054002. [152] S. Wang, H. Lin, H. Bor, Y. Tsai, C. Wei, Characterization of chromium thin films by sputter deposition, J. Alloys Compd. 509 (2011) 10110–10114. https://doi.org/10.1016/j.jallcom.2011.08.052. [153] R. Mientus, R. Grötschel, K. Ellmer, Optical and electronic properties of CrOxNy films, deposited by reactive DC magnetron sputtering in Ar/N2/O2(N2O) atmospheres, Surf. Coatings Technol. 200 (2005) 341–345. https://doi.org/10.1016/j.surfcoat.2005.02.181. [154] R. Arvinte, J. Borges, R.E. Sousa, D. Munteanu, N.P. Barradas, E. Alves, F. Vaz, L. Marques, Preparation and characterization of CrNxOy thin films: The effect of composition and structural features on the electrical behavior, Appl. Surf. Sci. 257 (2011) 9120–9124. https://doi.org/10.1016/j.apsusc.2011.05.109. [155] J. Lin, Z.L. Wu, X.H. Zhang, B. Mishra, J.J. Moore, W.D. Sproul, A comparative study of CrNx coatings Synthesized by dc and pulsed dc magnetron sputtering, Thin Solid Films. 517 (2009) 1887–1894. https://doi.org/10.1016/j.tsf.2008.09.093. [156] P. Carvalho, F. Vaz, L. Rebouta, L. Cunha, C.J. Tavares, C. Moura, E. Alves, A. Cavaleiro, P. Goudeau, E. Le Bourhis, J.P. Rivìre, J.F. Pierson, O. Banakh, Structural, electrical, optical, and mechanical characterizations of decorative ZrOxNy thin films, J. Appl. Phys. 98 (2005) 023715. https://doi.org/10.1063/1.1990261. [157] J. Inoue, H. Asami, M. Hirai, T. Suzuki, H. Suematsu, W. Jiang, K. Yatsui, Oxidation properties of Cr(N,O) thin films synthesized by pulsed laser deposition, in: S. Miyake (Ed.), Nov. Mater. Process. by Adv. Electromagn. Energy Sources, 1st ed., Elsevier Science, Osaka, 2005: pp. 203–207. https://doi.org/10.1016/B978-008044504-5/50041-6. 56 [158] T. Suzuki, H. Saito, M. Hirai, H. Suematsu, W. Jiang, K. Yatsui, Preparation of Cr(Nx,Oy ) thin films by pulsed laser deposition, Thin Solid Films. 407 (2002) 118–121. https://doi.org/10.1016/S00406090(02)00023-8. [159] R. Wang, L.L.H. King, A.W. Sleight, Highly conducting transparent thin films based on zinc oxide, J. Mater. Res. 11 (1996) 1659–1664. https://doi.org/10.1557/JMR.1996.0208.