The Effect of pH Solution in the Sol–Gel Process on the Process of Formation of Fractal Structures in Thin SnO₂ Films

Abstract

In this paper, we investigated fractal cluster structures of colloidal particles in tin dioxide films obtained from lyophilic film-forming systems SnCl₄/EtOH/NH₄OH with different pH levels. It was revealed that at the ratio Sn > Cl₂ > O₂, N₂ = 0, and pH = 1.42, the growth of cross-shaped and flower-shaped structures of various sizes from several μm to tens of μm is observed. At the ratio Cl₂ > Sn > O₂ > N₂ and pH = 1.44, triangular and hexagonal structures are observed, the sizes of which are on the order of several tens of micrometers. The growth of hexagonal structures is probably affected by the presence of nitrogen in the film, according to the elemental analysis data. At the ratio Sn > Cl₂ > O₂ > N₂ and solution pH of 1.49, the growth of hexagonal and cross-shaped structures is observed, whereas flower-shaped structures are not observed. Hierarchical flower-like and cross-shaped structures are fractal. The shape of microstructures is directly related to the shape of the elementary cells of SnO₂ and NH₄Cl. A direct dependence of the formation of hierarchical structures on the volume of ammonium hydroxide additive was found. This allows for controlling the shape and size of the synthesized structures when changing the ratio of the initial precursors and influencing the final physicochemical characteristics of the obtained samples.

1. Introduction

Among the large number of physical and chemical methods for obtaining materials with various functional characteristics (gas sensitivity, conductivity, thermal conductivity, photosensitivity, etc.), one of the most interesting and simple methods is sol–gel technology [1,2,3,4,5]. Sol–gel technology includes the synthesis of materials from initial solutions (for example, film-forming), wherein at the first stage, the solution is modified during “maturation” into a sol, and at the second stage, the sol is transformed into a gel. The range of applications of sol–gel technology is very wide. For example, sol–gel technology is used to obtain optochemical sensors [6], composite photocatalysts [7], to obtain materials used for catalytic cleaning of soot particles from diesel engines [8], for the synthesis of fluorescent nanoperovskites, which are used in fingerprinting [9]. It should be noted that the sol–gel method has established itself as the basis for the latest nanotechnologies for the synthesis of nanoparticles for potential use in dosimetry [10]. Using the sol–gel method, thin films for perovskite solar cells are obtained [11]. Moreover, the sol–gel method is used to obtain nanostructures used to remove crystal violet dye from aqueous media. This dye poses a significant risk to human health, including carcinogenic and mutagenic effects, as well as a hazard to the environment due to its persistence and toxicity in aquatic ecosystems [12]. Finally, using sol–gel technology, thin nanostructured films based on tin dioxide are obtained [13]. Thin nanostructured films, nanodispersed powders, and composite systems made of SnO₂ are functional materials with many applications. Tin dioxide-based coatings are used as a three-dimensional macroporous anode in lithium-ion batteries, as active layers or a base in gas sensors, as a protective layer against corrosion, etc., refs. [14,15]. It should be noted that one of the main properties of nanostructured thin SnO₂ films is a change in electrical conductivity upon gas adsorption. In this regard, they are used in sensors of various types of gases, including those with toxic and explosive effects [16,17,18,19].

Layers with a porous surface structure are very relevant for applications as gas-sensitive materials. In this case, such a structure must be controllable. In addition, fractal cluster structures of colloidal particles forming such a surface structure are of great interest. There are various models describing the processes of formation of fractal structures. They may differ in the clustering mechanism. Such models include the diffusion-limiting aggregation model [20]. In this model, the growth of a random cluster occurs due to the adhesion of particles/clusters to the initial fractal-forming cluster. There is also a model of cluster–cluster aggregation [21]. In this model, the appearance of small clusters is achieved by combining the initial ones. It should be noted that as a result of the dissolution of the initial reagent (in this case, tin tetrachloride) in ethyl alcohol, a sol is formed. This sol is a highly dispersed colloidal system. In such systems, no precipitation of dispersed particles is observed, and the particles are maintained in a suspended state due to Brownian motion. However, Brownian motion in colloidal solutions differs from ordinary Brownian motion. Under normal conditions, dispersed particles do not collide here since they have the same charge. The gel is formed due to the coagulation of sol particles. To accelerate this process, it is necessary to introduce catalysts into the system that will produce deformation of the electric layer of the dispersed phase. It should be noted that as a result of the hydrolysis of tin tetrachloride, stannic hydroxide Sn(OH)₄ is formed. It has a gel-like structure but is unstable. Due to this, it decomposes into water and tin dioxide.

This article discusses the process of formation of cluster fractal structures due to the so-called “competition” of coagulation and disintegration processes [22]. And as a consequence, a fractal structure of film clusters is formed, which is caused by nonlinear evolution in stochastic processes during sol–gel synthesis.

One of the important characteristics of the formation of synthesized thin films of tin dioxide is the pH level of the initial film-forming solutions. Often, experiments work with pH solutions in the range of 7 to 11 [23]. The present work is aimed at the range from 1.40 to 1.53 for pH. This is not accidental since, in this range, the characteristics of light absorption and surface resistance change. And the surface tin dioxide passes into the bulk. In addition, systems with such a pH level are lyophilic. In contrast, often in studies, mainly lyophobic systems are considered, which is definitely an advantage of this study. Also, the pH level is regulated in a small range of pH values. A small change in pH in such systems significantly affects the formation of hierarchical fractal structures of different shapes.

The formation of structures of various types affects the uniqueness of the properties of nanomaterials. It is noted that in hierarchical structures, the formation of useful functions occurs both in the nanometer and in other structural areas. Hierarchical structures formed in thin films of tin dioxide are distinguished by a large surface area, as well as high surface permeability and low density. A significant advantage of obtaining such films is the low cost of production and the environmental friendliness of the method.

In this paper, the formation of fractal cluster structures of colloidal particles in tin dioxide films obtained from lyophilic film-forming systems SnCl₄/EtOH/NH₄OH was studied. It has been shown that the use of a film-forming system with a pH level of 1.40–1.53 allows the creation of micro-nanostructures of various shapes with adjustable sizes depending on the pH of the solution.

2. Materials and Methods

The following reagents were used to conduct the experiment on the synthesis of film-forming systems:

(1)

Tin chloride pentahydrate powder SnCl₄ 5H₂O (>98% pure grade, corresponding to State Standard 6-09-3084-87 [24], “Labkhimprom” LLP);

(2)

Ethyl alcohol C₂H₅OH (corresponding to State Standard 5962-13 [25], rectified alcohol);

(3)

Concentrated aqueous solution of ammonia NH₄OH (>98% pure grade, corresponding to NH₄OH State Standard 24147-80 [26], “Labkhimprom” LLP).

A total of 5 film-forming solutions were prepared using the sol–gel method. For the preparation, the following were used:

(1)

m (SnCl₄ 5H₂O) = 3.9072 g;

(2)

V (C₂H₅OH) = 100 mL;

(3)

V (NH₄OH) = 0; 0.2; 0.4; 0.8; 1.6 mL.

SnCl₄ 5H₂O powder was poured into a flask of 100 mL and dissolved in ethyl alcohol of 50 mL. We poured 25 mL of ethyl alcohol into a separate flask and added the required amount of concentrated aqueous ammonia solution. Thus, the concentrated ammonium hydroxide was significantly diluted. The resulting solution was added dropwise to the original flask. Then, the remaining volume of ethyl alcohol was poured in. We mixed it thoroughly to avoid the formation of agglomerates (without using a stirrer, manually rotating the flask) and sent it to a dark place for a day to “mature” the solution.

Table 1. PH level of film-forming solutions depending on the volume of added ammonium hydroxide, the content of tin and ammonium ions in 100 mL of solution.

V (NH4OH), mL	pH of Film-Forming Solutions	Tin Ion Content in 100 mL (In Moles)	Ammonium Ion Content in 100 mL (In Moles)	Ratio of Ammonium Ions to Tin
0	1.40	0.011	0	0
0.2	1.42	0.011	0.0025	0.227
0.4	1.44	0.011	0.005	0.455
0.8	1.46	0.011	0.01	0.909
1.6	1.49	0.011	0.02	1.818

shows the volume of ammonium hydroxide additive, the pH of film-forming systems, the content of tin and ammonium ions in 100 mL of solution, as well as their ratio.

The following chemical reactions occur in the systems:

SnCl₄ + 4C₂H₅OH → Sn(OH)₄ + 4C₂H₅Cl

(1)

SnCl₄ + 4H₂O → Sn(OH)₄ + 4HCl

(2)

T, °C
Sn(OH)₄ → SnO₂ + 2H₂O

(3)

When interacting with a concentrated aqueous solution of ammonia, the following reactions occur:

SnCl₄ + NH₄OH → Sn(OH)₄ + 4NH₄Cl

(4)

NH₄OH + HCl → NH₄Cl + H₂O

(5)

The synthesized samples were applied to a glass substrate heated to 100 °C using the spray pyrolysis method and annealed for 15 min.

The pH of the synthesized film-forming solutions was measured using a pH meter “pH—150M”. For a better understanding of the film structure and distribution of chemical elements (Sn, N, Cl) in them, SEM images of the surface of the obtained samples were made, and mapping and elemental analysis were carried out. These results were obtained on a scanning electron microscope CC-66 (China). The Raman spectrum of the samples was recorded at room temperature using an exciting wavelength of 632.8 nm on a Jobin-Yvon LabRaman HR800, Horiba (Kyoto, Japan). The transparency of the samples was measured on a UNICO Spectro Quest 2800 spectrophotometer (UNICO, Caledonia, WI, USA). The surface morphology was investigated using an Atomic Force Microscope (AFM, Nanoscience Instruments, Phoenix, AZ, USA) Nt-mdt solver pro. XPS analysis of the samples was performed on the X-ray Photoelectron Spectroscopy (XPS, NEXSA Thermo Scietific, Waltham, MA, USA) setup.

3. Results and Discussion

During the experiment, it was found that when adding ammonium hydroxide to the initial solution and then applying one layer of such a solution to the substrate at room temperature, hierarchical flower-like and cross-shaped microstructures are formed (Figure 1a,b). When analyzing the obtained microstructures, the question of the nature of their origin arose. If we turn to the shape of the unit cell of tin dioxide, it is formed by six tin atoms, which in turn are bonded to nine oxygen atoms. Conventionally, such a cell resembles a flower with six petals (Figure 1c). The shape of the unit cell of ammonium chloride is similar to a cross. Since it is formed by four hydrogen atoms bonded to a nitrogen atom (Figure 1d), these microstructures can be fractal in nature. That is, the structures of the smallest size form the same microstructures of the largest size.

To explain the fractality of the microstructures under study, a Julien fractal aggregate (Figure 1e) and a two-dimensional disordered aggregate (Figure 1f) are shown.

In the fractal Julien aggregate, a six-petal structural element forms a six-petal object of a larger size [27]. Such a fractal aggregate is formed through a sequential connection of identical particles. The initial particle in it is located at the origin of a certain rectangular coordinate system. Six other particles are attached to it, the movement of which occurs along the positive and negative basis vectors of the lattice. An ensemble of seven particles is obtained in the form of a “flower”. And each time an absolutely identical flower is attached to such a single flower. The procedure can be repeated infinitely. Since the unit cell of tin dioxide is similar to a six-petal flower, therefore, in our case, it is the basis of a fractal structure of the same shape but already on a microscale.

We assume that cross-shaped microstructures are built on a similar principle, but the basis is already a “cross” consisting of four particles. As an example, in Figure 1f, we presented a two-dimensional disordered aggregate in which the cross morphology is formed from a cross-shaped structural element. In our case, the basis of such an aggregate can be the unit cell of NH₄Cl, the shape of which resembles a cross.

Therefore, the Julien aggregate shows the formation of a six-petal object due to a structural element of the same shape. It should be noted that fractal aggregates have the property of self-similarity. The two-dimensional disordered aggregate has a cross morphology, which is built from a structural element of a cross-shaped form.

In order to evaluate the functional characteristics of the synthesized samples, their microstructure was studied. The obtained data on the samples annealed on a tile at 100 °C are presented in Figure 2.

In Figure 2a, rounded microstructures are observed. The spread of microstructures over the film surface is uneven, which may be due to the uneven deposition of solution droplets on the substrate during spray pyrolysis. When 0.2 mL of ammonium hydroxide is added (Figure 2b) to the solution, cross-shaped and flower-shaped microstructures of different sizes from several µm to tens of µm are formed in the film. Figure 2c shows triangular and hexagonal microstructures with sizes of several tens of micrometers. The formation of such microstructures may indicate the formation of an intermediate compound, which was formed as a result of insufficient annealing temperature. The tendency of the microstructures to form conglomerates is noted. Figure 2d demonstrates the presence of both hexagonal and cruciform objects and a flower-like one in the surface structure. This means that the composition contains both the SnO₂ (flower) and NH₄Cl (cross) microstructures themselves, as well as an intermediate formation. In addition, the size of the microstructures is significantly larger than in the previous samples. The sizes of the microstructures reach several hundred micrometers. But there are also microstructures of the order of several microns, which can also be associated with the rate of droplet settling and the time of their fixation. Figure 2e shows the presence of hexagonal and cruciform microstructures, while flower-like microstructures are not observed. This gives reason to assume the absence of the SnO₂ compound in the composition of the films. It should be noted that this assumption is based on the fact that the unit cell of tin dioxide has the shape of a six-petal flower, while in the compound NH₄Cl, it resembles a cross.

Thus, the size of the synthesized microstructures increases significantly with the increase in pH of film-forming solutions up to pH = 1.46 and is from several micrometers at pH = 1.42 to tens of µm at pH = 1.44 and to hundreds of µm at pH = 1.46. Then, the size of the synthesized microstructures begins to decrease with the increase in pH.

Schematically, this description can be presented as follows (Figure 3).

To better understand the distribution of tin, nitrogen, and chlorine in the obtained samples, their mapping was carried out. The results are presented in Figure 4.

Figure 4a shows a uniform distribution of Sn, N, and Cl over the entire surface of the film without ammonium hydroxide additive. When 0.2 mL of ammonium hydroxide is added to the solution (Figure 4b), cross-shaped and flower-shaped microstructures begin to form. However, the distribution of Sn, N, and Cl over the entire surface of the film is uneven. Nitrogen is present throughout the film, while Sn and Cl are present only in the microstructures themselves. A similar situation is observed when 0.4 mL of ammonium hydroxide is added to the solution (Figure 4c). However, the microstructures are hexagonal. This suggests that the microstructures also contain Sn and Cl, but probably in a different ratio than in the sample shown in Figure 4b.

Figure 4d shows approximately the same picture as the previous samples (Figure 4b,c). However, the microstructures in this case are much larger. And as SEM showed, with a given amount of ammonium hydroxide in the solution, microstructures of three shapes are observed here: hexagonal, flower-shaped, and cruciform. Consequently, the formation of microstructures is associated with both the ratio of Sn, N, and Cl in their composition, and with the rate of application and the time of fixation of the film to the surface. Figure 4d shows a situation similar to Figure 4c. However, the hexagonal microstructures are much larger. The content of Cl is, however, greater than Sn.

To further study the composition of the obtained microstructures, elemental analysis was carried out. Below is the elemental analysis of samples annealed at 100 °C.

According to

Table 2. Elemental analysis of the obtained samples.

(a)	Element	Mass.%
	Sn	69.29
	N	0.00
	Cl	12.03
	O	18.68
	Total	100
(b)	Element	Mass.%
	Sn	55.08
	N	0.00
	Cl	30.07
	O	14.85
	Total	100
(c)	Element	Mass.%
	Sn	39.12
	N	0.49
	Cl	48.43
	O	11.95
	Total	100
(d)	Element	Mass.%
	Sn	63.03
	N	0.00
	Cl	19.98
	O	16.99
	Total	100
(e)	Element	Mass.%
	Sn	41.91
	N	1.77
	Cl	39.98
	O	16.34
	Total	100

, the film without ammonium hydroxide additive (a) contains about 69% tin, about 18% oxygen, and about 12% chlorine, while elemental analysis showed no nitrogen content at all. The film containing 0.2 mL ammonium hydroxide (Table 2, (b)) also does not contain nitrogen. The Sn content in it is about 55%, chlorine ~30%, and oxygen ~14%. In the film with the addition of 0.4 mL ammonium hydroxide (Table 2, (c)), the Sn content is ~39%, which is much less than in samples (a) and (b). It should be noted that the microstructures here also have a different shape than in sample (b). However, the chlorine content is ~48%, which is greater than in samples (a) and (b). The oxygen content here is ~12%. In addition, the nitrogen content is shown to be ~0,5%. Whereas in the previous samples, the analysis did not show it at all. According to the sample (Table 2, (d)), the elemental analysis showed the following results: Sn~63%, Cl~20%, and O~17%. The nitrogen content was not detected. With this ratio of elements, three types of microstructures described earlier are formed in the film. In addition, the microstructures are much larger. And finally, when studying the last sample (Table 2, (e)), the content of elements is presented in the following percentage ratio: Sn~42%, Cl~40%, O~16%, N~1.8%. As in sample (c), the presence of nitrogen in the film is noted here. And the microstructures are also presented in a hexagonal shape. Therefore, it can be concluded that the formation of hexagonal microstructures is affected by the presence of nitrogen in the film. Moreover, the ratio of Sn and Cl is approximately the same. That is, it can be said that they are almost in equal proportions. Schematically, this can be represented as follows (Figure 5).

Taking samples on an AFM showed the following results (Figure 6). Figure 6a,b show AFM images of tin dioxide film without the addition of ammonium hydroxide.

The images obtained for the film without the addition of ammonium hydroxide show individual protuberances scattered over the surface, the height of which exceeds 150 nm. At the same time, the linear dimensions reach 500 × 500 nm. These protuberances can be considered individual particles on the surface. The protuberances have a rounded shape with a developed apex. The surface between the protuberances is not developed and can be considered conditionally smooth.

Figure 7 shows AFM images of a film obtained from a solution containing 0.4 mL of ammonium hydroxide.

When examining the surface of the film obtained from a solution containing 0.4 mL of ammonium hydroxide, it is evident that the surface has become more developed and can no longer be considered conditionally smooth. A greater number of both individual large particles and their clusters have appeared. The particles themselves also have a developed surface, and a certain slope of the surface associated with the general hilliness of the surface is also traced. This indicates the growth of hierarchical microstructures.

Figure 8 shows the XPS analysis of the thin films. The XPS analysis of the thin films confirmed the presence of Sn⁴⁺ with Sn 3d_5/2 and Sn 3d_3/2 peaks at 487.05 eV and 495.48 eV corresponding to SnO₂ [28]. The Cl 2p spectrum shows peaks at 200.61 eV and 198.91 eV, indicating residual chloride compounds [29]. The N 1s peaks at 403.65 eV and 402.12 eV indicate the presence of ammonium and nitrite residues due to incomplete removal of NH₄OH. The spectra were corrected for C 1s (284.76 eV) since a peak shift was observed due to surface charging of the sample.

For a more detailed analysis of the microstructure of the obtained samples, Raman spectra were recorded. However, only films with hexagonal microstructures were subject to analysis (Figure 9).

The Raman spectrum of the sample was recorded at room temperature using an excitation wavelength of 632.8 nm on a Jobin-Yvon LabRaman HR800, Horiba. The spectra were recorded in the range from 100 cm⁻¹ to 900 cm⁻¹. Figure 9 shows the Raman spectrum obtained on hexagonal microstructures. The observed peaks are located at ~111 cm⁻¹, 174 cm⁻¹, 239 cm⁻¹, and 320 cm⁻¹. When identifying the obtained vibrational modes, a literature analysis was carried out on possible compounds formed during the chemical interaction of the initial reagents. For the NH₄Cl compound, the Raman modes are located in the higher frequency side of the spectrum from 3000 cm⁻¹ and more [30], which does not coincide with the region of the sample spectrum. It has been theoretically and practically confirmed that for all tin oxide phases and mixed tin hydroxochloride complexes, the Raman active modes are located in the region from 50 cm⁻¹ to higher wavenumber values [31,32]. However, for the above-mentioned phases, many vibrational modes are in the corresponding region for our sample. Four peaks from 100 cm⁻¹ to 350 cm⁻¹ are observed in the spectrum of the sample. No peaks are observed in the region of higher frequencies. The compound NH₃(CH₂)₅NH₃SnCl₆ [33] is the closest in peak intensity and location to the spectrum of our sample. Thus, an intermediate phase may form on the hexagons. The obtained data coincide with the SEM results, where Sn, N₂, and Cl are observed on the hexagons.

To further study the influence of the microstructures formed in the films on the functional properties of the synthesized films, a spectral analysis of the obtained samples was carried out. The results are shown in Figure 10. For comparison, the spectrum of the glass substrate annealed at 100 °C was also recorded since it has good transparency (Figure 10, curve 1).

As can be seen from Figure 10, the film without the addition of ammonium hydroxide (blue curve 2) has the highest transparency T~90%. It is almost as transparent as the glass substrate (red curve 1). When adding 0.2 mL of ammonium hydroxide (Figure 10, green curve 3), the transparency of the sample decreases to T~84%. When adding 0.4 mL of ammonium hydroxide to the composition of the initial film-forming solution (Figure 10, black curve 4), the transparency decreases to T~78%. It should be noted that this is the lowest transparency of all the analyzed samples. Further, when the addition of ammonium hydroxide in the initial solution is increased to 0.8 mL (Figure 10, gray curve 5), the transparency improves in comparison with the previous sample to T~81%. This result is also observed when adding 1.6 mL of ammonium hydroxide to the initial solution (Figure 10, blue curve 6). Thus, it can be concluded that the addition of ammonium hydroxide, which is the basis for the growth of hierarchical microstructures in the synthesized films, reduces the transparency of the samples, but insignificantly.

4. Conclusions

The formation of fractal cluster structures of colloidal particles in tin dioxide films obtained from lyophilic film-forming systems SnCl₄/EtOH/NH₄OH was studied. The use of a film-forming system with a pH level of 1.40–1.53 allows one to create micro-nanostructures of various shapes with adjustable sizes depending on the pH of the solution. At that, a small change in the pH level leads to a significant change in the shape and size of the resulting fractal structures.

A significant growth of the microstructures is observed upon the addition of ammonium hydroxide to the initial solution. It was noted that at the ratio Sn > Cl₂ > O₂, N₂ = 0, and pH = 1.42, the growth of cross-shaped and flower-shaped microstructures of different sizes from several µm to tens of µm is observed. At the ratio Cl₂ > Sn > O₂ > N₂ and pH = 1.44, triangular and hexagonal microstructures are observed, the sizes of which are on the order of several tens of micrometers. The growth of hexagonal microstructures is probably affected by the presence of nitrogen in the film, according to elemental analysis. At the ratio Sn > Cl₂ > O₂ > N₂ and pH of the solution of 1.49, the growth of hexagonal and cross-shaped microstructures is observed, whereas flower-like microstructures are not observed.