Effect of various additives on aluminum oxide thin films prepared by dip coating, thermal behavior, kinetics and optical properties

Abstract

Aluminum oxide thin films attract research interest due to their properties. Aluminum acetate was used as an Al source with acetic acid, oxalic acid, and nitric acid as additives. The transmittance and the thickness of the films strongly depend on the additives, with the approximate bandgap energy changing from 5 ev to 5.4 ev. The aluminum oxide film deposited by dip-coating is presented great uniform surface morphology. The knowledge of the degradation kinetics of materials is essential for investigating the thermal stability of compounds. The acetic acid thin film proved to be the most efficient additive by demonstrating interesting optoelectronic properties. The thin films deposited by dip-coating were characterized by using X-ray grazing incidence diffraction, SEM, UV-Visible spectroscopy. Gamma aluminum oxide thin films prepared by acetic acid can be a good candidate for a wide range of optical applications.

Introduction

Aluminum oxide thin films (Al₂O₃) exhibits several interesting properties [1, 2], such as high optical transparency [3], high abrasive and corrosion resistance [4, 5], high chemical and thermal stability [6], and wide bandgap [7]. Due to the mentioned properties, aluminum oxide thin films possess a wide range of applications in optoelectronics and microelectronics devices serving as insulating [8], hard protective layers [9], surface passivation e.g. for silicon solar cells [10].

The crystalline structure of Al₂O₃ can take many different forms, including: α, γ, χ, η, θ, κ, δ and ρ [11].

Among these transformations, γ-Al2O3 is a significant material that is employed in a variety of applications, including catalytic substrates in the automotive and petroleum sectors, composite materials for spacecraft, and abrasive and thermal wear coatings [12].

The method of deposition and the temperature of the substrate are frequent factors in the creation of these phases. At low temperatures, the amorphous phase occurs, the γ−alumina phase takes over at intermediate temperatures, and the k and α-Al2O3 phases grow at high temperatures [13].

There are various chemical and physical methods applied to prepare aluminum oxide thin films such as chemical vapor deposition (CVD) [14], atomic layer deposition (ALD) [15], Pulsed laser deposition (PLD) [16, 17], spray pyrolysis (SP) [18,19,20], magnetron sputtering [16, 21], anodization [22,23,24] spin coating [25], and dip coating [26,27,28]. Dip-coating was used due to simplicity, low cost, and the ability to produce high-quality films, and it is widely utilized for the deposition of oxide thin films [25].

This work aims to investigate the solvent effect on the structural, optical, and morphological properties of aluminum oxide thin film by changing different solvents. The degradation kinetics of materials is essential for investigating the thermal stability of compounds. Degradation kinetics can be studied by several methods but one of the most popular and simplest techniques widely used in the literature is the thermogravimetric analysis [29].

The morphology of the films has been investigated by grazing incidence X-ray diffraction (GIXRD), the quality of the film by scanning electron microscopy (SEM). The optical properties of aluminum oxide thin films deposited by dip-coating were characterized by UV-visible spectrophotometry. The degradation kinetics of materials was determined by thermogravimetry (TG) techniques.

Materials and methods

Preparation

The aluminum oxide thin films synthesis process is shown in (Fig.1). the starting solution was prepared by dissolving the aluminum acetate Al (OH) (C₂H₃O₂)₂ (Sigma-Aldrich, at) in NaOH (Molar, at) solution under an air atmosphere with 0.11 M ratio of OH to Al, the mixed solution was continuously stirred at 55 °C for 1 h.

Fig. 1

Synthesis of γ-Al₂O₃ thin films

To dissolve the basic aluminum acetate, various acids (nitric acid, acetic acid, and oxalic acid) were added to the basic solution to get pH = 7. The acids initialize the precipitation and stabilize the sol system [25, 26]. The acetic acid received an emphasized attention, the acetic acid plays an important role as a bidentate ligand between two Al ions furthering the formation of a 3D network [30].

IGEPAL CO-520 or poly (oxyethylene) nonylphenyl ether) (Sigma-Aldrich) was added as a non-ionic surfactant to the colloid solution to improve the connection between layer and substrate [31]. Thin films were deposited by a dip coating method on quartz substrate at room temperature with a constant speed (5 cm/min) by an automated dip-coater (PTL-MM01 Dip Coater). The substrates were ultrasonically cleaned with acetone, ethanol, and deionized water dried. After coating the samples were dried at 70 °C for 1 h and then heated at 600 °C for 3 h with a heating rate of 2 °C/min (Fig.2). the aluminum oxide thin films synthesis process is shown in (Fig.1).

Fig. 2

Schematically graph of heat treatment of gamma aluminum oxide thin films

Kinetic approach

There are several approaches for estimating kinetic parameters from thermogravimetric data. It is therefore essential to specify the approach adopted in any kinetic exploitation of the experimental mass loss data. In this work, the kinetic parameters are determined by the Coats Redfern method. The expression to achieve these parameters is obtained from the reaction rate. Thus, in the kinetic analysis of thermal decomposition reactions, the reaction rate is written according to the following form:

$$ \frac{d\alpha}{d t}= kf\left(\alpha \right) $$

(1)

Where α is a characteristic variable of reaction progress, related to the mass of the sample m by the formula:

$$ \alpha =\frac{m_0-{m}_t}{m_0-{m}_f} $$

(2)

Where, m₀ is the initial weight of the sample, m_t is the weight of the sample at the particular temperature T, and m_∞ is the weight at the end of degradation step, f(α) represents the mode of degradation of the substance. The function f(α) does not depend on the temperature but rather on the degradation mode of the subjected matter.

The different modes proposed in the literature [32,33,34] are grouped in Table 1. In this same Table, the function $g\left(\alpha \right)=\underset{0}{\overset{\alpha }{\int }}\frac{d\alpha}{f\left(\alpha \right)}$ represents the integral form of the function f(α) [35].

Table 1 Thermal degradation modes proposed for gas-solid reactions [32,33,34]

k is the reaction rate constant. It is accepted that k following the Arrhenius law:

$$ k=A\exp \left(\frac{-E}{RT}\right) $$

(3)

Where E is the apparent activation energy in kJ/mol, R is the perfect gas constant in Joules. K^− 1. mol^− 1, A is the pre-exponential factor or frequency factor in min^− 1, T is the absolute temperature in °K,

A, E and f(α) are called the kinetic triplets of a reaction.

D₁, D₂,..are symbols given to models.

Procedure for kinetic parameters determination

In order to determine the kinetic parameters of our samples, the Coats–Redfern method [29], which is given in Eq.4, is expressed as follows:

$$ Ln\left(\frac{g\left(\alpha \right)}{T^2}\right)= Ln\frac{AR}{\beta {E}_{\alpha }}-\frac{E_{\alpha }}{RT} $$

(4)

Where α is a characteristic variable of the reaction progress of the sample and T is the absolute temperature, g(α) represents functions commonly used for the description of thermal decomposition (Table 1). A is the frequency factor, E_a is the activation energy, R is the gas constant, and β is the heating rate.

A plot of Ln (g(α)/T²) against 1/T will give a straight line of slope –E/R and an intercept of ln (AR/ βE) for an appropriate form of g(α). Thus, based on the correct form of g(α), the activation energy and the pre-exponential factor could be respectively determined from the slope and intercept terms of the regression line.

Table 1 lists the most common kinetic g(α) functions, which were used in this study for the estimation of reaction mechanisms from dynamic TG curves by using Coats–Redfern method.

Investigation methods

Scanning electron microscopy (SEM): The surface covering, and the layer thickness has been studied by a FEI Quanta 3D FEG scanning electron microscope (SEM). The SEM images were prepared by the Everhart-Thornley secondary electron detector (ETD), its ultimate resolution is 1–2 nm. Since the conductance of the particles investigated is high enough to remove the electric charge accumulated on the surface, the SEM images were performed in a high vacuum without any coverage on the specimen surface. For the best SEM visibility, the particles were deposited on a HOPG (graphite) substrate surface. SEM combined with energy disperse X-ray spectroscopy (EDX) is applied for spatially resolved chemical analysis of layers.

Grazing incidence X-ray diffraction (GIXRD) measurements were performed by a Rigaku Smartlab X-ray diffractometer equipped with a 1.2 kW copper source (radiation wavelength: CuKα; λ = 0.15418 nm). To reduce the effect of the substrate, a grazing incidence parallel-beam geometry was used with an incidence angle of ω = 1°. Scans were performed in the range 2Θ between 10°–110° with an 1D silicon strip detector (D/Tex ultra 250) by a speed 0.2°/min. The XRD data were collected over the 2θ range of 9–90° with a step size 0.005°. Identification of phases was performed by comparing the diffraction patterns with standard PDF cards.

UV–Visible spectroscopy: The transmittance of thin films was determined by UV–visible spectroscopy using Dynamica spectrophotometer with UV Detective program at room temperature, in the range of 200–1000 nm.

Thermogravimetric (TG) of bulk aluminum oxide were made on a simultaneous thermal analyzer of the "LabsysTMEvo (1F)" type and SETARAM brand. This device consists of a TG microbalance associated with DTA sensor with a single rod, a metal resistor furnace up to 1600 °C, and multitasking software controlling the various modules. The tests are carried out from ambient temperature to 800 °C with a temperature rise rate of 7 °C min-1 under argon with a flow rate of 10 cm³.min-1. The initial mass of the sample is about 10 mg.

Results and discussion

To determine the thermal behavior of the thin films derived from the aluminum acetate, thermogravimetry (TG) was used. The TG curve for AlOOH sol obtained directly by the acetic acidic treatment is shown in Fig. 3. The small weight loss that occurred at 148 °C is mostly caused by desorption of the adsorbed moisture and solvent evaporation, while a large weight loss happens within the temperature range from 182 °C to 540 °C is due to the decomposition of the organic groups by the release of CO₂ and H₂O. No visible weight loss is detected at temperatures above 540 °C, according to the results acquired, it is suitable to calcine the aluminum oxide thin films at 600 °C. When compared to bulk aluminum oxide, the aluminum oxide colloidal sol showed a significant weight loss due to solvent evaporation at 148 °C.

Fig. 3

Thermogravimetry (TG) of bulk aluminum oxide and colloidal sol

Figure 4 shows GIXRD patterns of the films annealed at 600 °C with various additives. The crystallinity degree can be influenced by many factors such as precursor concentration, pH reaction, and solvents. The films show Bragg diffraction peaks ascribed to more or less ordered γ-Al₂O₃ crystalline phase. The quality of acids does not really affect crystallinity. These patterns show slight shifts with different acids.

Fig. 4

Grazing incidence X-ray diffraction patterns of the films annealed at 600 °C with various acids (nitric acid, acetic acid, and oxalic acid)

Scanning electron microscope (SEM) was used to investigate the surface morphology and cross-section of the films as shown in Fig. 5. According to SEM images, the films morphology can be described by a uniform, with no cracks and a non-defective surface. Moreover, the smooth surface morphology decreased from acetic to nitric acid and oxalic acid. The complex formed between the acetate acid and aluminum ions helps to develop a homogenous. The cross-sectional SEM images present layer thicknesses of approximately 90,140 and 115 nm respectively.

Fig. 5

Scanning electron microscopy of γ-Al₂O₃ thin films top view (a, c, e) respectively nitric acid, acetic acid, and oxalic acid), cross-section (b, d, f)

The chemical composition of the thin film was determined by using EDX (Fig. 6). EDX analysis from five different areas indicates the presence of Al and oxygen-rich, beside silicon which is derived from quartz substrate of layer. The EDX confirms the efficient deposition of Al₂O₃ thin film.

Fig. 6

EDX (Energy Dispersive X-ray) spectrum of Al₂O₃ thin films

To get information on the optical properties of the films, transmittance spectra of γ-Al₂O₃ films on quartz substrates were measured in the wavelength range from 200 to 1000 nm (Fig. 7). It can be noticed that the transmittance of the films is over 70% in the wavelength range from 400 to 1000 nm. The transmittance of the films decreased from acetic, nitric, and oxalic acid respectively which confirms that thickness has an inverse relation with transmittance [37], the high absorbance of gamma aluminum oxide prepared by oxalic acid can also be explained by light trapping due to their textured morphology [38]. The optical band gap relation with the absorption coefficient is given by the relation:

$$ {\left(\alpha \mathrm{h}\upsilon \right)}^2=\mathrm{C}\left(\mathrm{h}\upsilon -\mathrm{Eg}\right) $$

(5)

Fig. 7

Transmittance spectra of γ-Al₂O₃ films on quartz substrates with various acids (acetic acid, nitric acid and, oxalic acid)

Where is α absorption coefficient, C is a constant, h is Plank constant and Eg is the optical band gap. The band gap Eg can be determined by extrapolation from the linear portion of the curves till they intercept the photon energy axis (Table 2). Figure 8. The band gap values are given as 4.99, 5.21, and 5.34 eV for nitric acid, oxalic acid and acetic acid respectively. Band gap value can be change by some defect levels [39].

Table 2 Optical band gaps of different thin films extracted from the Tauc plot

Fig. 8

The optical band gap of γ-Al₂O₃ films with various acids (nitric acid, acetic acid, and oxalic acid)

Thermal analysis 

Thermogravimetric analysis (TGA-DTG)

Determination of the thermal behaviour of γ-Al₂O₃ thin films was carried out from the aluminum acetate, thermogravimetry (TG) was used. The TG curves for colloidal alumina sol is shown in Fig. 3. The small weight loss that occurred at 148 °C is mostly caused by the desorption of the adsorbed moisture and solvent evaporation, While, a large weight loss happens within the temperature range from 182 °C to ~ 310 °C is due to the decomposition of the organic groups (mainly acetate ions) by the release of CO₂ and H₂O molecules. Another escape of H₂O molecules up to 540 °C can be assigned to the transformation of AlOOH to Al₂O₃. No detectable weight loss is above 540 °C.

Kinetic parameters estimation

To calculate and understand the nature of the decomposition, kinetic exploitation is made on a dynamic chemical regime assuming that the decomposition is a global reaction where physical limitation is neglected. The complete thermogram was divided into distinct sections according to their degradation steps. Curves indicating the solid-state mechanisms of alumina degradation under an inert atmosphere are shown in Fig. 9. The values of activation energy Ea, pre-exponential factor A and correlation factors R² for the first and second degradation steps are listed in Table 3. Moreover, the parameters A and E are moving in the same direction and their values depend on the mode of degradation. The relationship between A and Ea, called the “apparent compensation effect” is often mentioned in the literature. Figure 10 shows traces of the values of LnA as a function of Ea_. The effect of compensation is another way to further discrimination between degradation modes [40]. Thus, for the first region (second weight loss step in TGA thermogram), it was observed from Table 3 that the best correlation coefficients were obtained for F0, F1, F2, F3, R2, R3, P2, P3, D3, A2, and A3, with energy values running from 35.1 to 135.8 kJ/mol. Regarding, second degradation step (third weight loss step), degradation mechanisms that give the best mathematical fit for both samples were F1, F2, F3, R2, R3, D3 and A2, with values of activation energy from 7.154 to 121,25 kJ/mol. Likewise, results of the two regions show that the highest activation energies were found in the first thermal degradation regions where the main pyrolysis reaction took place and the largest weight loss occurred. The values of the pre-exponential factor (Table 3) indicate that it depends on the degradation mode.

Fig. 9

Curves indicating the solid-state mechanisms of bulk aluminum oxide degradation under inert atmosphere

Table 3 Thermal kinetic results for bulk aluminum oxide pyrolysis

Fig. 10

Compensation effect Ln A = f (E_a) for first and second regions of thermal degradation

Conclusion

In this work, gamma aluminum oxide thin films were prepared by various solvents. The X-ray diffraction patterns of the films confirm the presence of γ- Al₂O₃ can be crystallized at a temperature of 600 °C. The optical transparency of the films are over 70% in the wavelength range from 400 to 1000 nm with an average band gap value of 5.18 ev, therefore the γ- Al₂O₃ layer prepared by acetic acid shows the highest transmittance with the thinnest and smoothest surface morphology which can be relevant for many applications. The scanning electron microscopy presents a uniform no cracks and non-defective surface morphology. TGA curves indicate that active pyrolysis range is from 182 °C to 540 °C, where the major decomposition of organic groups occur up to ~ 310 °C and from ~ 310 °C to 540 °C AlOOH turns into Al₂O₃. From about 400 °C, γ-crystalline phase begins to form. The ideal temperature of final treatment is 600 °C. The results of kinetics study shows that the highest activation energies were found in the first thermal degradation region where the main pyrolysis reaction took place and the largest weight loss occurred, with values of activation energies running from 35.1 to 135.8 KJ/mol.