Elaboration and Characterization of n-Type Organic Semiconductor (Fullerene C₆₀) Deposed by Ultrasonic Technique for Sustainable OTFT Fabrication

Abstract

This study focuses on the deposition of fullerene (C₆₀) as thin film on glass substrate by ultrasonic chemical bath deposition (UCBD) processing, under ambient temperature. Highly effective results were obtained from the films based on the solution of C₆₀ dissolved in toluene mixed with 2-methoxyethanol. The obtained films were characterized by X-ray diffraction (XRD), infrared spectroscopy (IR), scanning electron microscopy (SEM), and energy dispersive X-ray spectroscopy (EDS). The XRD examination of the thin films reveals the presence of the C₆₀ cubic phase compared to the powder reference. The molecular structure obtained by Rietveld refinement shows no bonding between the molecules in C₆₀ powder, while in the deposed thin film the bonding is established. The molecules are bonded between them by pentagons of the right and left molecule. Each four neighbor molecules bond between them and they are all able to geometrically tie to the neighboring molecules under a crystalline FCC structure. The Sherrer and W-H methods were used to investigate microstructural parameters. The lattice parameter and the crystallite size show the same variation tendency. The average lattice parameter for the powder and the deposed films C₆₀-3h, C₆₀-5h, and C₆₀-8h is 14.0652, 14.1901, 14.0529, and 14.1848 Å, respectively, and the crystallite size calculated by the Sherrer method is 37.51, 38.98, 34.35, and 41.54 nm, respectively, as well. The IR spectrum indicated the presence of chemical π bonds (c=c) that are very suitable for enhancing the electronic properties of the material, and SEM analysis illustrated a dense, homogeneous without pinhole structures in the film morphology. Moreover, EDS emphasizes the presence of high carbon concentration and fewer stranger atoms. As a result, despite the UCBD technique being old and not very often applied in the field of organic materials, it is still a cost effective and good alternative method for organic thin film deposition.

1. Introduction

Organic semiconductors (OSCs) have gained significant attention in recent decades due to their unique advantages in solution processing, chemical versatility, and light weight, in addition to their mechanical flexibility. Owing to these potential advantages, organic semiconductors are now employed in many electronic applications as active layers, replacing traditional inorganic materials such as organic photovoltaics (OPVs) [1,2,3], organic field-effect transistors (OFETs) [4,5], and bioelectronic devices [6,7]. Among these applications, organic thin film transistors (OTFTs), used as bioelectronic devices, have progressively and currently become an important research topic focusing on the field of organic bioelectronics.

One of the primary features of organic thin film transistors (OTFTs) is that they utilize organic compounds, which can be either polymers or small molecules, facilitating cost-effective and large-scale production. The flexible substrates, including fabric, foil, fiber, plastic, and paper, enable OTFTs to be effectively employed in environmental monitoring, biomedical applications, and chemical sensing. This versatility arises from their capability to detect a wide range of analytes, including gases, biomolecules, and ions. RFID tags for item-level tracking and inventory management can also be produced at a low cost and with flexibility. Additionally, organic thin film transistors (OTFTs) may be investigated for the development of low-power solutions suitable for Internet of Things (IoT) applications, including smart homes, industrial automation, and wearable technology [8].

Organic semiconductors (OSCs) are materials that possess the unique ability to transition between being effective conductors and efficient insulators. This transformation is primarily governed by the doping process or variations in the electric field, which dictate the material’s conductive state [9]. This typically involves adding or removing electrons to change the charge carrier density and allow the material to conduct charge, with electron-transporting polymers being classified as n-type materials, while p-type materials transport holes [10,11,12,13,14].

Currently, p-type materials dominate the transistor logic circuits due to their high performance, especially in field-effect mobility; for example, the hole mobility (${\mu }_p={10}^{-1}$cm²/V.s) for pentacene is very large compared to that its electron mobility (${\mu }_e={5\times 10}^{-3}$cm²/V.s) [15,16], although the development of electron transport materials (ETMs) has consistently lagged behind [17]. Progress in n-type materials for OTFTs is needed to enable complementary logic circuits assembled with matched p-type materials that reduce static power consumption, allowing faster circuit speeds, extra complex circuits, and improved operational stability [18]. P-type and n-type operation with well-matched electrical and thermal conductivities are also needed in OTE generators [19].

Among the n-type organic semiconductors widely used as ETMs, fullerene (C₆₀) has been included in OFETs and organic electronics due to its high electron mobility [16]. Its spherical soccer ball shaped molecule and exceptional electronic properties, fullerenes constitute an important research topic in medicine and biology fields [20,21,22,23]. C₆₀-based nanomaterials have good prospects in cancer treatment [24], such as killing HeLa cells [25] and providing radioprotection to radiosensitive mammalian cells. An aqueous dispersion of C₆₀ fullerene has outstanding anti-inflammatory effect, which also acts as an antioxidant and an anti-aging medicine. Nearly all these applications are not yet available to the public [26].

In order to fabricate an OTFT, one needs first to choose the deposition technique suitable for the organic material (fullerene). Thermal sublimation of pure C₆₀ was first used in early 1900 [27] to deposit uniform fullerene films on sapphire and KBr substrates. Then, electron-beam-excited plasma deposition used C₆₀ fullerene clusters as a carbon source for depositing amorphous carbon films. This method allows the deposition of films with controlled properties, such as hardness and lubricity [28]. Additionally, vapor deposition without plasma treatment shows a crystalline structure of C₆₀ film that involves plasma and reveals an amorphous microstructure. In contrast, the laser evaporation of graphite was used by Kroto et al. to produce C₆₀ fullerenes (named Buckminsterfullerene) [29]. The C₆₀-modified electrodes, mostly used in biosensors, have been prepared using dropping, vapor deposition, electrochemical deposition, and electropolymerization processes [30]. Among the inexpensive methods for organic-based devices, spin coating is a solution-processing technique. It is also used during the fabrication of OPV and OTFT devices when C₆₀ is involved [31,32]. Ultrasonic chemical bath deposition (UCBD) can also be a good alternative technique for uniform and well-characterized thin film.

The solution approach is better suited to prepare polymer films since polymer molecular chains tend to break at high temperatures [33]. To enhance the homogeneity and quality of thin film growth, research has investigated the combination of an ultrasonic bath and thin film deposition techniques [32,33,34,35]. The uniformity of the deposed thin layer can be enhanced by ultrasonic agitation in the bath, which minimizes the effects of convection and gravity, resulting in more consistent film thickness and morphology. By improving the mass transport of precursors and reactants to the substrate, its energy can increase the processing speed and rate of deposition. Additionally, a more consistent and regulated environment for film growth can be produced by decreasing defects like pinholes, cracks, and heterogeneities and improving the film’s adhesion to the substrate, both of which will enhance the film’s quality.

The possibility to obtain a functionalized C₆₀ fullerene is high due to its unique and versatile structure, which has proven to be the ideal foundation for impressive structural diversity, driving significant advancements in the field of organic chemistry and opening new possibilities in the synthesis of advanced compounds.

Functionalization in the perspective of C₆₀ fullerene refers to the process of attaching functional groups or molecules to the surface of the fullerene [36,37,38].

Functionalization can significantly enhance the electrical properties of fullerenes. For instance, chemical functionalization of fullerenes can increase their miscibility with host polymers, even if the fullerenes can sometimes lose certain original properties. The functionalization can improve the dispersion of C₆₀ within the host matrix and enhance their interfacial adhesion, which in turn improves the electrical conductivity of the composite [39].

Once the films are deposed, different characterization methods are required to examine the crystallization of the structure, its morphology, and the chemical elements contained therein.

X-ray diffraction, which shows distinctive diffracted peaks corresponding to various reflection planes, is used to examine the crystallite size, crystal structure, and phase present in the synthesized samples of nanocrystals. Also, since no crystal is perfect because of its size, if a crystal extends in all directions, it is an ideal crystal and starting from optimum crystallinity is represented by the broadening of the XRD peak. The size constriction and line cause the intrinsic strain. Size confinement, line dislocations, stacking faults, and sintering stress are the causes of the intrinsic strain [40,41,42,43]. Different synthesis factors affect the structural characteristics and microstructural parameters [41]. Peak broadening in XRD is caused by two factors: lattice strain and crystallite size. Lattice strain is the extent to which crystal defects cause the distribution of lattice constants and lattice dislocations [41,44]. Due to polycrystalline aggregates within a particle, crystallite and particle sizes are typically not comparable [45]. Some efficient methods to study microstructural parameters and structural characteristics have been published in the literature. The microstructural parameters analyzed in this study utilized a comparison of the Williamson–Hall (W-H) and Sherrer methods. The Williamson–Hall method is generally considered more accurate because it accounts for both the size and strain of the crystallites, providing a more comprehensive analysis of the structural parameters. However, the accuracy can depend on the assumptions made in the models used, such as the uniform deformation model (UDM), uniform deformation stress model (USDM), and uniform defect energy density model (UDEDM), which can yield different strain values [46]. As a counterpart, the Warren–Averbach method, now known as the Scherrer method, is currently widely used. It is a simple approach that relates the peak width only to the size of the crystallites. However, it does not account for lattice strain caused by defects like point defects, grain boundaries, and stacking faults. This approach showed that the constant Ks in the equation of this method is 1.84 for two-dimensional peaks and 0.89 for three-dimensional peaks [47]. These two calculation methods are used to determine the microstructural parameters based on XRD spectra.

In the present work, fullerene (C₆₀) films were deposed using UCBD, a solution process, and analyzed by X-ray diffraction (XRD), infrared spectroscopy (FTIR), scanning electron microscopy (SEM), and energy dispersive X-ray spectroscopy (EDS). The microstructural properties and parameters are studied based on the characterization results.

2. Materials and Deposition Method

2.1. Chemical and Thin Film Preparation

The chemicals used in this work were of analytical reagent grade. Toluene and 2-methoxyethanol were obtained from Honneywell (Charlotte, NC, USA) and Sigma-Aldrich (St. Louis, MO, USA), having a purity of 99.7 and 99.0%, respectively. The black powder of fullerene C₆₀ with a purity of 99.5 wt% was obtained from XFNANO (Shenzhen, China).

The C₆₀ thin films were synthesized under ambient temperature using the UCBD method. The elaboration starts on two steps. First, a solution of 10 mL composed of 80% toluene and 20% of 2-methoxyethanol was prepared, based on the study by Krishna et al. [48]. The 2-methoxyethanol was used as it is known to enhance adhesion to metal substrates due to the formation of chemical bonds at the metal/oxide interface. Then, the solution was added to 2.8 mg/mL fullerene, based on the study by Ruoff et al. [49]. The mixture was stirred for 2 h. During this UCBD technique, represented in Figure 1, a steel basket was not used. The deposition time is considered as an important parameter to ensure the total evaporation of the solvent and to obtain good quality fullerene thin film. The films resulting from UCBD were characterized by different spectroscopy techniques.

2.2. Sample Characterization

The XRD analysis of the C₆₀ thin film was conducted using a Rigaku miniflex 600 diffractometer (Rigaku, Tokyo, Japan). The X-ray diffraction technique was employed using CuKα (40 kV of current) radiation within a 2-theta angle range from 5 to 40° with a 0.02 step. Infrared spectroscopy (FTIR) was performed to identify the chemical bonds in Carbon C₆₀ thin film by using an Agilent Cary 600 instrument (SpectraLab Scientific Inc., Markham, ON, USA), equipped with an ATR diamond, and specifically within the wave number range of 4000–500 cm⁻¹. The high-resolution images of the microstructures of the films were acquired utilizing an environmental scanning electron microscope (E-SEM), type Prisma ThermoFisher Scientific (Waltham, MA, USA), with a tungsten filament. Finally, the basic elements of the films were analyzed by energy dispersive spectrometry (EDS). The error of the EDS detector for Carbon (C) and Oxygen (O) varies from $\pm 0.5$to $\pm 2$%.

3. Results and Discussion

3.1. X-Ray Diffraction

Fullerene C₆₀ crystallizes in a face-centered cubic (CFC) structure, corresponding to the space group Pa-3, as shown on Figure 2. This figure shows the X-ray diffraction pattern of the pristine fullerene (as reference) and C₆₀ thin films synthesized by UCBD. According to JCPDS data [96-901-1581], the observed diffraction peaks in XRD patterns are well matched with cubic CFC structure. The configuration of the peaks suggests the presence of high crystallinity structure. The primary diffraction peaks for C₆₀ are observed at 2θ positions such as ((111), 10.97°), ((220), 17.83°), ((311), 20.91°), ((222), 21.87°), ((331), 27.55°), ((420), 28.27°), ((422), 31.03°), and ((511), 32.97°).

Indeed, as long as the deposition takes place, the peaks of the thin film deposed are closer to those of the pristine material (C₆₀ powder). In addition, the peaks that designate the functionalizations have disappeared.

The preferential orientation peak with high intensity for pristine is (111), located at 10.97°, while in the thin film (C₆₀-3h, C₆₀-5h, and C₆₀-8h) it is located at 2θ = 21.43°, representing the (222) reflection plane in the C₆₀ cubic structure. Further, emergence of small peaks is noticed at 10.84°, 17.74°, and 20.80°, corresponding to the (111), (220), and (311) planes of the cubic phase, respectively.

The peaks marked with asterisk are additional diffraction peaks, appearing for the C₆₀-3h, C₆₀-5h, and C₆₀-8h samples. These peaks agree with the vinyl stearate structure (C₂₀H₃₈O₂) according to the information provided in the JCPDS data [00-024-1980]. These peaks prove the presence of functionalization, which is mostly the outcome of interaction involving the two solvents (toluene and 2-methoxyethanol) and the fullerene under ambient temperature. This functionalization can help the adhesion between the deposed film and the substrate. It is also depicted that the peak intensity significantly increased in the C₆₀-8h sample compared to C₆₀-3h and C₆₀-5h samples; this is attributed to the increased thickness of the C₆₀ thin film.

The proposed molecular structures of the C₆₀ powder and the films were obtained by Vesta (a 3D visualization program) after Rietveld refinement and assisted by Full-proof Software 5.20, as presented in Figure 3.

From Figure 3, one can see that there is no bonding between the molecules in the proposed molecular structure of the C₆₀ powder, while in the thin film C₆₀-8h, the bonding is formed. The two molecules are bonded between the pentagon of the right and left molecules. Each of the four neighbor molecules bond between them, and they are all able to geometrically bond to the neighboring molecules under a crystalline FCC structure.

The C₆₀ dimer, which is the distance between two C₆₀ covalently bonded molecules, is calculated. For instance, between molecule C32 from the (021) plane and the molecules from the (011) plane (C31, C34, and C21), the distances are 1.7202 (Å), 1.7979 (Å), and 1.8803 (Å), respectively, as presented in Figure 4.

Physical and instrumental broadening are the two main components of the broadening in the Scherrer-method X-ray diffraction peak that occurs in nanocrystals because of the intrinsic strain effect and the crystallite size effect [50,51]. The Scherrer equation (Equation (1)) can be used to obtain the average particle size using the adjusted physical broadening [43,51]:

$$D_{Sh}={\displaystyle \frac{k\times \mathsf{\lambda }}{\beta \times cos\theta }}$$

(1)

where $D_{Sh}$ is the crystallite size, $\beta$ is the full width at half maximum of the peak, 2$\theta$ is the peak position, k is the shape factor (typically around 0.89), and λ is the incident X-ray wavelength (0.15418 nm). The dislocation density is calculated by the inverse of the square of crystallite size:

$$\delta =1/{D_{Sh}}^2$$

(2)

The calculated parameters are regrouped in

Table 1. Microstructural values for powder C₆₀-3h, C₆₀-5h, and C₆₀-8h using the Scherrer and W-H approaches.

Samples	Miller Indices (hkl)	2θ (°)	$\mathit{\beta }$ ($\mathbf{°}$)	Average Crystallite Size (nm)	Dislocation Density ($\mathit{\delta }$) (${\mathbf{n}\mathbf{m}}^{-2}$)	Average Crystallite Size (nm)	Dislocation Density ($\mathit{\delta }$) (${\mathbf{n}\mathbf{m}}^{-2}$)
				By Scherrer Method		By WSH Method
Powder	(111)	10.97	0.184	37.51	7.1 × 10−4	55	3.3 × 10−4
	(220)	17.83	0.227
	(311)	20.90	0.224
	(222)	21.88	0.208
	(331)	27.56	0.293
	(420)	28.27	0.201
	(422)	31.03	0.264
	(511)	32.975	0.243
C60-3h	(111)	10.84	0.1976	38.98	6.5 × 10−4	28.406	12.3 × 10−4
	(220)	17.72	0.258
	(311)	20.80	0.213
	(222)	21.45	0.168
C60-5h	(111)	11.02	0.236	34.35	8.4 × 10−4	24.315	1.6 × 10−4
	(220)	17.92	0.355
	(311)	21.00	0.266
	(222)	21.67	0.181
C60-8h	(111)	10.85	0.230	41.54	5.7 × 10−4	33.157	9 × 10−4
	(220)	17.77	0.246
	(311)	20.82	0.188
	(222)	21.45	0.166

.

However, Scherrer’s method takes only into account the crystallite size effect on XRD peak broadening, but it says nothing about the lattice microstructures such as intrinsic stress, which occurs in nanocrystals due to point defects, grain boundaries, triple junctions, and stacking faults [40,52]. These effects induce XRD peak broadening and are taken into account in the Williamson–Hall (W-H) method. This approach can be used for the calculation of intrinsic stress and crystallite size ($D_{W-H}$). The Williamson–Hall method is comparably easy and simple [53,54]; the uniform deformation model is given by the following equation:

$$\beta \times cos\theta =\epsilon (4sin\theta )+{\displaystyle \frac{k\times \mathsf{\lambda }}{D_{W-H}}}$$

(3)

Equation (3) plots as a line; $\epsilon$is the slope of the line, which specifies the intrinsic strain value, while the intercept gives the average crystalline size of C₆₀ nanocrystals. The origin of the lattice strain is attributed mainly to the lattice expansion or contraction in nanocrystals due to size confinement, as the atomic arrangement can be altered due to size confinement. The extracted W-H values are presented in Table 1 and shown in Figure 5.

From Table 1, the Scherrer method presents acceptable results as the crystallite sizes of the deposed films are adjacent to that of the powder, while by the WSH method there is a large rise in the value of this parameter from that of the powder to those of the deposed layers.

As Figure 5 shows, the strain (ε) can be negative, indicating the lattice contraction (compression), or positive, meaning dilatation. As the average crystallite size in the C₆₀-8h thin film is larger (41.54 nm), the strain is smaller (−0.001), showing an acceptable resulting film.

The lattice parameter (a) of the unit cell is evaluated according to the following formula [55,56]:

$$d_{hkl}={\displaystyle \frac{a}{\sqrt{\left(h^2+k^2+l^2\right)}}}$$

(4)

where $d_{hkl}$ is the interplanar spacing calculated from Bragg’s law ($2d_{hkl}\sin \theta =n\lambda$) [44]. The calculated lattice parameter values are given in

Table 2. Lattice parameter values of the unit cell for powder, C₆₀-3h, C₆₀-5h, and C₆₀-8h.

2θ (°)	a (Å)/(C60 Powder)	2θ (°)	a (Å)/C60-3h	2θ (°)	a (Å)/C60-5h)	2θ (°)	a (Å)/C60-8h
10.97	13.9455	10.84	14.1287	11.02	13.9618	10.85	14.1548
17.83	14.0514
20.90	14.0722	17.72	14.1494	17.92	14.0083	17.77	14.1257
21.88	14.0604
27.56	14.0863	20.78	14.1696	21.0	14.0493	20.82	14.1293
28.27	14.1015
31.03	14.1033	21.49	14.3162	21.67	14.1922	21.45	14.3294
32.975	14.1012
average a (Å)	14.0652		14.1910		14.0529		14.1848
			Expansion		Slight Compression		Expansion

. The trivial change in the lattice parameter (a (Å)) values depicted in Table 2 can be due to the ultrasonic agitation. A slight compression occurred in C₆₀-5h, which resulted from the average crystallite size becoming smaller (34.35 nm) compared to the powder, where $D_{Sh}$ is 37.51 nm. To represent the variation in the crystallite size and the lattice parameters of the resulting films compared to the powder, Figure 6 shows the agreement between these two parameters.

3.2. Infrared Spectroscopy

The IR spectra of these thin films (C₆₀-3h, C₆₀-5h and C₆₀-8h), presented in Figure 7, also confirm the presence of carbon-to-carbon bonds, essentially the π bonds that are essential for the moving process of the charge carriers.

The C₆₀ thin film IR spectra revealed several pertinent absorption peaks at 525, 575, 721, 800, 1180, 1373, and 1464, ascribed to C-C vibrational modes. The peak at 1736.89 is assigned to the C=O vibration mode [57], which may result from environmental exposure, as revealed by the EDS data shown in Figure 8, and those at 2842.11, 2914.72, and 2956.43 cm⁻¹ are attributed to C−H stretching bands [58]. All of these peaks were found to be in agreement with previous studies [57,59].

The C=C bonds, as indicated in the samples (Figure 7), are responsible for π-orbitals, which at their turn are delocalized over the entire molecule, leading to the formation of a continuous π-electron system. Forming molecular orbitals that are occupied by π-electrons enables electrons to move freely within the molecule, influencing its electronic properties.

3.3. E-SEM and EDS Analysis

An environmental scanning electron microscope (E-SEM) was used to examine the microscopic morphology of the materials. SEM and EDS characterization analysis are shown in Figure 8.

Figure 8. SEM and EDS characterization of fullerene thin film deposed by UCBD during (a) 3 h, (b) 5 h, and (c) 8 h.

Figure 8. SEM and EDS characterization of fullerene thin film deposed by UCBD during (a) 3 h, (b) 5 h, and (c) 8 h.

As shown by the E-SEM results, the structure illustrates that the morphology of the film improves as the deposition period increases. The EDS results confirm that the content of carbon increased with the deposition duration; however, oxidation may also increase. Note that the specific gravity of C₆₀ powder elements (Carbon and Oxygen) are 98.5 and 1.5 (in PP At.%) and 98.87 and 1.13 (in At%), respectively.

Indeed, this last figure indicates that the thin film expels the chemical elements not proper to C₆₀ material and is restored as the solvent evaporates. However, as the deposition time increases, oxidation takes place. Additionally, note that C₆₀ films prepared by UCBD are dense and homogenous, and without the presence of pinholes.

4. Conclusions

Solution processing has been used to successfully prepared fullerene (C₆₀)-based samples, which are then examined via X-ray diffraction analysis, IR spectroscopy, SEM, and EDS. Although the UCBD approach is not widely used to deposit fullerene (C₆₀) thin films, it produced good results, as demonstrated by XRD, which helped extensively the study of structural and crystallographic morphology of the obtained films by peak analysis. As the period of deposition was considered a variable, the crystallinity improves with longer duration. Also, the XRD demonstrated that by increasing the period of deposition, more peaks would vanish due to the presence of functionalization. In addition, from the viewpoint of molecular structure, no bonding existed between molecules in C₆₀ powder sample; whereas, in the thin film C₆₀-8h, five bonds appeared, forming a pentagon bond. In addition, the preferential plane changed from (111) for the C₆₀ powder to that of (222) for the samples deposed by UCBD. Moreover, the observed diffraction peaks in the XRD patterns are well matched with cubic CFC structure. The XRD spectrum was also used to ascertain the microstructural values using Scherrer and W-H computational approaches and the lattice parameter for powder, C₆₀-5h, and C₆₀-8h by Bragg’s law. The IR revealed the presence of C=C bonds in the samples, which can improve conductivity, thus allowing electrons to move. The E-SEM and EDS showed the enhancement of morphology of the deposed films and the disappearance or reduction in elements other than Carbon.