Trends in Low‐Temperature Combustion Derived Thin Films for...

2000464 (1 of 25) © 2020 Wiley-VCH GmbH

www.advelectronicmat.de

Review

Trends in Low-Temperature Combustion Derived Thin Films for Solution-Processed Electronics

Pavan Pujar, Srinivas Gandla, Dipti Gupta,* Sunkook Kim,* and Myung-Gil Kim*

DOI: 10.1002/aelm.202000464

1. Introduction

The emergence of large-area electronics with unprecedented multifunctionalities of optical and electrical properties and ubiquitous coverage, such as transparent displays, building sensors, smart windows, car surface, smart textiles, and solar

Appreciable advancement of low-temperature combustion processing brings a step closer to the fulfillment of large-area, flexible electronics. The maximum temperature of deposition is successfully reduced below the softening temperature of the polymeric substrates. The method embodies the incorporation of fuel- and oxidizer-ligands in the precursor, which leads to an exothermic reaction resulting in low-temperature conversion and/or densification of the metal oxide thin films. A series of electrically conducting, semiconducting, and high permittivity dielectric metal oxides are deposited on varied substrates, including flexible plastic substrates, such as polyimide and aromatic polyester. The combustion processing is efficient in depositing metal, metal oxide, metal–metal oxide composite thin films at considerably low temperatures. In addition, the combustible precursors are considered compatible with diverse solution deposition methods such as spin coating, spray coating, inkjet printing, and blade coating. The combustion processed metal oxide thin films have the potential to exhibit acceptable device perfor-mance of thin film transistors, solar cells, gas sensors, organic light emitting diodes, and so on. The present review not only covers the on-going research in solution combustion-based deposition, but also discusses potential of the various devices with combustion-derived thin films of functional oxides as active/passive components.

Dr. P. Pujar, Dr. S. Gandla, Prof. S. Kim, Prof. M.-G. KimSchool of Advanced Materials Science and EngineeringSungkyunkwan UniversitySuwon, Gyeonggi-do 16419, South KoreaE-mail: [email protected]; [email protected]. D. GuptaPlastic Electronics and Energy LaboratoryDepartment of Metallurgical Engineering and Materials ScienceIndian Institute of TechnologyMumbai, Maharashtra 400076, IndiaE-mail: [email protected]

The ORCID identification number(s) for the author(s) of this article can be found under https://doi.org/10.1002/aelm.202000464.

cells over several km2, demand superior optoelectronic properties to conventional semiconductors and an ever-lower fabrica-tion cost than conventional vacuum-based fabrication. The limited performance of traditional electronics, especially those based on hydrogenated amorphous sil-icon (a-Si:H) added with the utilization of vacuum deposition techniques, impose challenges in large-area deposition on flexible substrates.[1,2] The thin film depo-sition on thermal sensitive flexible poly-meric substrates demands a processing temperature below the softening tempera-ture of respective substrate. Thus, a new class of materials, along with low-temper-ature deposition techniques, is needed to address the applications where tradi-tional electronics is severely limited. The alternate materials such as carbon nano-tubes, silver nanowires (AgNWs), organic molecules, and 2D materials have shown potential in realizing flexible and trans-parent devices. Although organic semi-conducting molecules have had successful usage in light-emitting diodes, they still need to overcome inferior electrical func-

tionality and poor environmental stability for general usage in large-area electronics.

On the other hand, metal oxides are considered to be the right candidates for both active and passive components of various electronic devices. The large bandgap of metal oxides makes them suitable for the passive-dielectric layer of thin film transistors (TFTs)[3,4] and controlling the defect chemistry of the metal oxides results in tunable electrical conductivity.[5] Because of the versatile nature of metal oxides, they have been used as transparent semiconductors, conductors, and dielec-trics.[6] Moreover, the tunable defect chemistry of metal oxides makes them electrically conductive, exhibiting conductivity greater than 104 S cm−1.[5] Thus, both large-bandgap and tun-able defect chemistry offers a unique amalgamation of optical transparency and electrical conductivity. Further, the functional metal oxides must be deposited in the form of thin films to mold them into different device architectures. The conven-tional sputtering techniques are appropriate in depositing metal oxides, but for low-cost devices, more innovative depo-sition techniques need to be employed. Solution processing is one such method, where alkoxides or metal salts dissolved in a suitable solvent is typically utilized for the deposition of metal oxides through various techniques such as spin coating,

Adv. Electron. Mater. 2020, 2000464

http://crossmark.crossref.org/dialog/?doi=10.1002%2Faelm.202000464&domain=pdf&date_stamp=2020-09-11

www.advancedsciencenews.com

© 2020 Wiley-VCH GmbH2000464 (2 of 25)


dip coating, spray coating, inkjet printing, etc.[7] The deposited wet film needs to be thermally treated at elevated temperatures to ultimately fetch away impurity-ligands and promote the for-mation of metal–oxygen bonds.[8] However, the high processing temperature requirements for impurity removal and condensa-tion reaction restrict the usage of thermally sensitive soft poly-meric substrates, which hinder the implementation of solution processing for next-generation flexible or stretchable electronic applications. The advancements in the solution processing such as sol–gel on chip,[9] ultraviolet (UV)-annealing,[10] combustion processing,[1] synthetic metal complexes,[5] oxo-cluster com-pounds,[11] etc. have been addressed to minimize the tempera-ture of deposition. In all of these, combustion processing offers low-temperature deposition with relatively simple precursors and general applicability for various materials. The combustible precursor is composed of organic fuels (reducing agents), such as urea and acetylacetone, and primarily nitrate of the respec-tive metal cation as oxidizers.

In combustion process, the simultaneous charge exchange reaction results in the intense liberation of heat energy with a self-catalytic reaction pathway.[12,13] The amount of heat energy output from the reaction is mostly dependent on the type of reactants; the resultant temperature of the reaction is termed “exothermicity.”[12] The heat energy output of the reaction allows metal salts to decompose at relatively low levels of supplied energy. Thus, the combustible precursors are capable of depos-iting functional metal oxides at low processing temperatures as energy input to initiate combustion reaction.[1,12] The com-bustion processing have been a well-established synthesis tech-nique for bulk metal oxides and other functional ceramics.[13,14] In 2011, combustion processing has been employed in depos-iting ultrathin functional metal oxides at low temperatures, which clearly showed a general roadmap for the realization of low-temperature deposition.[1] The scientific interest in such combustion derived thin films is evident from the number of publications addressing the issue. The feasibility of combustion processing in depositing metal oxides at low temperatures can fulfill the current demand for low-cost and large-scale micro-electronics. This review is dedicated to the on-going research in combustion derived thin films of various functional materials, its mechanism, and its application in TFTs, solar cells, trans-parent conducting oxides (TCOs), and so on.

2. Design of Functional Metal Oxides

2.1. Amorphous Metal Oxide Semiconductors

The discovery of a-Si:H and low-temperature polycrystalline silicon (LTPS) has advanced display technology to new hori-zons.[15,16] Currently, both technologies have secured a place in the commercial display sector. In the case of LTPS, high electron mobility (μ > 80 cm2 V−1 s−1),[17] outstanding electrical stability, and good form factor makes it a right candidate for applications such as mobile telephones.[17] On the other hand, a-Si:H based TFTs drive the largest market of televisions due to its large-scale uniformity, ease of scalability, and low-cost production.[18,19] Although a-Si:H is a popular candidate for the mainstream backplane of displays, yet its carrier μ is severely

limited (0.5–1.0 cm2 V−1 s−1).[15] Also, the limitations such as inferior current-carrying capacity, optical opacity, and moderate mechanical flexibility impose challenges for advanced large scale, flexible electronics.[15] To overcome such impediments and to enhance the electrical functionalities, various alterna-tive materials have been proposed, such as transition metal oxides, organic small molecules, 2D materials (MoS2 and gra-phene), organic polymers, and single-wall carbon nanotubes (SWCNTs).[20–22] The gist of these unconventional functional materials is depicted in Figure 1.

Recently, post-transition metal oxides have been widely accepted and successfully commercialized for display applica-tions with various range of electrical properties from insulator, semiconductor, and conductor. The recent discovery of multi-component amorphous metal oxide semiconductors, such as amorphous indium gallium zinc oxide (a-IGZO), have shown potential to act as active components of TFTs.[3,23,24] The advan-tage of the amorphous post-transition metal oxide system lies in its orbital configuration. The metal oxides with heavy-metal-cation [electronic configuration: (n − 1)d10 ns0] possess large spatial distribution of s-orbitals at conduction band minima with low effective mass[3] (Figure 2). The μ of charge carriers in amorphous metal oxides can be uniform over a large area with the absence of irregular scattering sites such as grain bounda-ries (Figure 2). Another advantage of amorphous metal oxides is that the feasibility of low-temperature deposition is compat-ible with flexible substrates.[1,5]

The practical application of a-IGZO based TFTs is dependent on the design of display panels.[17] For other applications, such as mobile telephones demanding high μ (>20 cm2 V−1 s−1) comparable to LTPS requires indium-rich a-IGZO.[25–27] On the other hand, for large-area displays, the critical aspect is to reduce the cost of fabrication. Unlike LTPS TFTs, the TFTs based on a-IGZO can be fabricated without intentional ion doping and crystallization, which minimizes process com-plexity and cost of production. Furthermore, after the discovery of a-IGZO, various firms have adopted it as an active layer of TFTs and improved the quality of the displays. Gadgets such as laptops, mobiles, televisions, etc. have already integrated a-IGZO TFTs (Table 1).[16,28]

2.2. Metal Oxides as Transparent Conductors

TCOs are another class of functional metal oxide with consider-ably high carrier concentration (>1020 cm−3) and excellent optical transparency (T550nm > 80%). The tin-doped indium oxide (ITO) with In:Sn = 9:1, as an industrial standard material, finds appli-cation as an electrode in optoelectronic devices, energy-efficient windows, and transparent heaters.[5,29] Further, the presence of high proportion of indium in ITO increases the cost. The total commercial value of ITO can be brought down signifi-cantly by either choosing an alternative TCO or by utilizing a low-cost deposition technique. Several alternate TCOs with an acceptable combination of both optical transparency and elec-trical conductivity have been reported over the years (Table 2), such as indium doped cadmium oxide (In0.05Cd0.95O),[30] alu-minum doped zinc oxide (Al:ZnO; AZO), fluorine doped tin oxide (F:SnO2; FTO). However, the alternative materials still



© 2020 Wiley-VCH GmbH2000464 (3 of 25)


suffer from several drawbacks, such as lack of adequate opto-electronic performance, environmental toxicity, and difficulty in mild etching process.

2.3. High κ Metal Oxide Dielectrics

Besides, tunable semiconducting and conducting characteris-tics, the metal oxide thin films of Al2O3, ZrO2, and TiO2, etc. exhibit high dielectric permittivity (κ) and excellent dielectric strength, which allows the low voltage operation of TFTs.[37] Apart from amorphous metal oxides, few other dielectrics satisfy the essential requirements are mainly categorized as self-assembled monolayers (SAMs), electrolyte-based high capacitance materials, fluorinated polymers, and inorganic–organic hybrid materials.[6,38] However, there are issues asso-ciated with each of the dielectrics mentioned above.[6] On the other hand, a moderately thick high κ metal oxide maintains an appreciable areal capacitance with low leakage current density. The selection of an optimum metal oxide is based on specific prerequisites that are worth considering. Amorphous metal oxide dielectric with low surface roughness is the basic morphological requirement. Also, a large breakdown field (>4 MV cm−1), high κ magnitude (Figure 3), and low leakage

current density (<1 nA cm−2 at 1 MV cm−1) are the added essentials.[6,39]

2.4. Challenges in Conventional Vacuum Processing Methods

Irrespective of the type of functionalities, metal oxide thin films have been deposited using vapor phase deposition techniques for industrial applications. These commercially accepted thin film deposition methods enjoy the luxury of being electrically robust and mechanically durable.[5,27] The vapor phase deposi-tion techniques include thermal evaporation, sputtering, pulsed laser deposition, and atomic layer deposition. The thermal deposition is widely used to deposit metalized products and the deposition of compositionally robust metal oxides requires a proper tuning of oxygen concentration.

Sputtering is widely used for the deposition of metal oxides. Although sputtering produces robust metal oxide thin films even at room temperature on plastic substrates, the process is costly due to specialized targets and equipment. In the case of pulsed laser deposition (PLD), the metal oxide target is irradi-ated with high-power UV pulses to vaporize the metal oxide in a plasma plume. Although PLD can grow morphologically robust thin films with well-controlled chemical compositions,

Figure 1. Gist of various functional materials.



© 2020 Wiley-VCH GmbH2000464 (4 of 25)


slow deposition rates and small area deposition make the process capital intensive. Further, the atomic layer deposition (ALD) technique is based on subjecting the substrate to the alternate pulses of gaseous reagents. The self-limiting reactions between the gaseous reagents allow the growth of atomically thin films. The ALD deposited thin films are robust and can have large area uniformity. ALD processes are generally used to grow metal oxide dielectrics. However, slow deposition rates make ALD a costly and time-consuming technique.

The vapor phase deposition methods remain ideal for the exploration of new material systems because of their ability to yield morphologically robust thin films with benchmark prop-erties.[5] However, atmospheric-pressure solution-based deposi-tion techniques are an attractive and alternative to high-vacuum assisted deposition techniques due to their scalability, ease of fabrication, and potential to lower manufacturing costs.[5] The realization of solution-based deposition had come into exist-ence, when L. Gordon and his research colleagues, during the conduction of gravimetric analysis through the precipitation method, observed transparent layers retained on the inner sur-face of the glass beaker.[40,41] The transparent layers opened up a new area of understanding the nature of these films and their implications. This accidental discovery led to a new area of thin film deposition called “solution processing.” In contrast to the inherent issues of various vacuum-based techniques, solution-based deposition technique offers relatively simple alternatives, which is classified into three main categories: spin coating, spray coating, and printing. Spin coating evenly distributes the precursor solution onto the substrate via centrifugal force. The thickness of the film depends on acceleration, the viscosity of the solution, the volatility of the solvent used, and the con-

centration of the solute. Spray coating uses a mist of atomized droplets injected from a nozzle. The method demands a volatile and low viscous solvent for efficient atomization. Spray deposi-tion is commercially used for the deposition of transparent and electrically conductive thin films. The spray coating requires optimization to achieve smooth thin films. Further, printing is an additive deposition process, which allows simultaneous dep-osition and patterning. The commonly used printing processes

Figure 2. Design concept of amorphous metal oxide semiconductors and the difference between the micrographs of polycrystalline and multicompo-nent amorphous metal oxides.

Table 1. List of commercial products based on a-IGZO TFTs.

Product Maker Display

Laptop and tablet

Radius 12 TOSHIBA 12.5″, 3840 × 2160, 352 ppi

Skylake NS850 NEC 15.6″, 3840 × 2160, 282 ppi

X3 Plux v3 Aorus 13.3″, 3200 × 1800, 276 ppi

i6 Air Cube 9.7″, 2048 × 1536, 264 ppi

iPad Pro Apple 12.9″, 2732 × 2048, 264 ppi

Monitor

PN-321Q Sharp 31.5″, 3840 × 2160, 140 ppi

Ultrasharp UP3214Q Dell 31.5″, 3840 × 2160, 140 ppi

PA322UHD-BK-SV NEC 31.5″, 3840 × 2160, 140 ppi

Phone

m1 note Meizu 5.5″, 1080 × 1920, 401 ppi

Redmi 2 Prime Xiaomi 4.7″, 720 × 1280, 312 ppi

Torque G02 Kyocera 4.7″, 720 × 1280, 312 ppi

Numbia Z5S ZTE 5″, 1080 × 1920, 441 ppi



© 2020 Wiley-VCH GmbH2000464 (5 of 25)


are ink-jet, gravure, screen printing, flexo, slot casting, and aerosol printing.[42] For example, screen printing is widely used to deposit metallic lines for solar cells. Besides, ink-jet printing has been successfully used to deposit both organic and inor-ganic-metal oxides. Figure 4 shows generally employed solu-tion-based deposition techniques.

3. Solution Processing of Functional Metal Oxides

Solution deposition of electronically functional metal oxide thin films generally falls into two main categories a) pre-formed oxide nanoparticle dispersion and b) soluble chemical precursor.[43] In the first case, the dispersion of nanoparticles of the material of interest needs the additional step of synthe-sizing nanoparticles and designing proper dispersants and surface-active agents. In addition, the challenges such as lim-ited dispersion stability, difficulty of multicomponent system, and high-temperature requirement for ligand removal, make dispersions of nanoparticles inappropriate for industrial imple-mentation in solution processed large area electronics. How-ever, in the case of soluble chemical precursor, the formation of metal oxide thin films on a substrate of interest is accom-plished by a series of steps.[44] In the first step, a precursor solu-tion is formulated using various metal reagents, solvent(s), and stabilizers. In the second step, the deposition of the formulated homogeneous precursor results in a wet film, which leads to the removal of carrier solvent. The escape of the solvent allows the densification via viscoelastic flow. Further, in the third step: the thermal treatment (i.e., annealing) of wet films promotes

the removal of compounds with low boiling points and builds oxo-connecting bridges between the cations through hydrolysis and condensation reactions.[5,44] The densification is greatly affected by intermediate phases and other particle inclusions. The subtle control of solution composition and deposition profile is crucial to achieving high quality film. Also, in gen-eral, the high-temperature treatment is essential for the com-plete densification and removal of organic residuals. Further, the high-temperature annealing facilitates the short-range dif-fusion resulting in the crystallization of the thin film (fourth step). The amorphous oxide semiconductors do not require a high-temperature crystallization process, but high film density is always desired. The quality of the thin film is dependent on the type of starting materials used in formulating the coating/printing solution. The most commonly used precursor system is sol–gel, which is classified into two categories: a) alkoxides and b) metal salts with stabilizer.[45] The sol–gel process initi-ates with the alcoholic solution of an alkoxide (M-(OR)n), where n is the valency of cation, and R is an alkyl group. The hydrol-ysis of the alkoxide results in hydroxyl groups, which on poly-condensation produces a polymeric network. These reactions take place simultaneously and produce low molecular weight by-products, such as water and alcohol.[45][46] The terminology “sol–gel” relates to the stable suspension of tiny particles called “the sol,” which further evolves into a 3D polymeric network called “the gel.” The formation of sol is the direct consequence of hydrolysis and polycondensation reactions of the starting material, which on gelation and drying produces required metal oxides (both bulk-powders and thin films). In the case of sol–gel derived thin films, the gelation step can be termed “film gelation”[5]. The process of formation of a complex poly-meric network is a time-dependent phenomenon, and the time required is called “aging.” Aging a precursor has direct impli-cations on the quality of the thin film. The bridging between metal cations and oxygen anions takes place as a result of continuous hydrolysis and condensation reactions (Figure 5). The transformation from M–OR to M–O–M and final polym-erization to form a continuous network of M–O–Ms demands high temperatures. Nevertheless, a thin film with large frac-tion of M–O–Ms network does not solely serve the purpose. The formed metal and oxygen bonds should sinter and create dense metal oxide thin film with negligible porosity. Due to the requirement of high-temperature, sol–gel processing is not

Table 2. List of alternate TCOs with their properties.

TCO σ [S cm−1] Thickness [nm] Transparency [%] Ref.

Al:ZnO 300 750 85 [31]

In0.05Cd0.95O 17 000 650 90 [30]

Mo:In2O3 1369 – 90 [32]

Ga:ZnO 1000 300 85 [33]

Nb:TiO2 1136 40 97 [34]

Ti:In2O3 800 200 90 [35]

F:SnO2 4546.8 1100 85 [36]

Figure 3. Periodic table showing criteria for the selection of metal oxide dielectrics based on the magnitude of κ. Adapted with permission.[6] Copyright 2018, Wiley-VCH.



© 2020 Wiley-VCH GmbH2000464 (6 of 25)


feasible for thermally sensitive polymeric substrates. Moreover, the high processing temperatures not only demand more input power, but also limit processing flexibility for the inclusion of thermally labile functional materials, such as quantum dot, self-assembled monolayer, and organic materials. The altera-tions of soluble chemical precursors are to decrease the tem-perature of processing. The steps involved in the deposition of dense thin films need to be carefully designed to reduce the maximum temperature of processing. The first step of formu-lating the precursor is subjected to scientific scrutiny by many researchers and developed various classes of precursors with distinct chemistries.[44] The following paragraphs describe a few such precursor systems.

3.1. Chelating Agent-Based Sol–Gel Precursors

The modification in the sol–gel precursor is brought by the incorporation of chelating ligands.[47] The intentional addition

of keto- or amine-groups alters the organic part of metal alkoxide reagents resulting in stabilization towards hydrol-ysis.[44] This method applies to diverse metal oxides with transition, post-transition, and alkaline earth metal cations. However, the high-temperature requirement makes these precursors hard to integrate with thermally sensitive plastic substrates.[48]

3.2. Polymer-Assisted Precursors

The reagents such as metal nitrates with water-soluble polymer forms a stable and homogeneous precursor.[49] The polymer capped metal cations arrest undesired chemical reactions and precipitation. Although the deposited films are of uniform thickness and controlled stoichiometry, high annealing temperatures are necessary for depolymeriza-tion, impurity removal, metal-oxide network formation, and crystallization.[50]

Figure 4. Schematic depicting various solution-based film deposition techniques. Adapted with permission.[5] Copyright 2011, Royal Society of Chemistry.

Figure 5. Hydrolysis and condensation reactions involved in “sol–gel” processing.



© 2020 Wiley-VCH GmbH2000464 (7 of 25)


3.3. Metal Organic Decomposition-Based Precursors

This is relatively less versatile compared to conventional sol–gel. The precursor is composed of short alkyl chain carboxy-lates dissolved in polar solvents.[48] The precursor ensures no secondary reactions between the reagents. The deposited thin films demand a high thermal budget for the complete elimination of organic residuals. The large outflux of resid-uals compromises the thin film homogeneity and chemical stoichiometry; also, high crystallization temperatures are essential.[51]

3.4. Liquid Exfoliation of Layered Materials

This is a relatively new technique of formulating stable disper-sions of nanometer thick layered materials.[52] The method uti-lizes oxidizing agents, ultrasonic waves, and ion intercalation strategies to break weak out-of-plane bonds of bulk oxide mate-rials, such as Nb6O17, Ti0.91O2, TaO3, Sr2RuO4, Ti3O7, etc.[52,53] The synthesized nanosheets are dispersed into a suitable sol-vent with appropriate surfactants/dispersants. Thin films can not only be grown at room temperature, but also deposition of highly textured ultrathin metal oxides is possible. However, the poor coverage control with uniform thickness, the limited material scope, and high temperature requirement for sur-factant removal should be overcome for large area electronic application.

3.5. Sol–Gel-on-Chip

The precursor used in this case is an alkoxide cluster, which on hydrolysis on the surface of the film results in the forma-tion of metal oxide at low processing temperatures. Unlike con-ventional sol–gel, this process utilizes a mediated hydrolysis in the humid atmosphere. The process was first demonstrated for amorphous ternary (indium zinc oxide: IZO) and quaternary (IGZO) metal oxides.[9]

3.6. Combustible Precursors

The metal nitrate(s) and organic fuel(s) are the ingredi-ents of combustible precursors.[13,54,55] The exothermic reaction between these species can be ignited at low tem-peratures compared to the endothermic decomposition of pure metal nitrates. Figure 6 presents a reaction coor-dinate diagram of combustion and conventional thermal decomposition pathways. The high localized temperature produced from the combustion reaction allows the forma-tion of metal oxide at minimum supplied energy. The use of combustible precursors was first introduced in 2011 by Kim et al. and successfully demonstrated a spectrum of functional metal oxide thin films at low-temperature (200 °C).[1] The following section is dedicated to the impli-cation of combustion reactions in depositing functional metal oxides.

4. Combustion Reaction

The bulk combustion reaction was first proposed as a modifica-tion in sol–gel.[56] The alkoxides, carbonates, and metal nitrates are employed along with carboxylic acids, such as citric acid. The metal cations from nitrates undergo chelation with the multi carboxylic acids present in the solution.[56,57] Various such chelation events result in the formation of complexes consti-tuting an organic matrix with uniform distribution of metal cations. The method was introduced to synthesize bulk ceramic powders such as titanates and niobates;[57] furthermore, it has been successfully utilized in synthesizing complex metal oxides. Another method using modified precursor chemistry is the “glycine-nitrate” process,[57] where a solvent-free viscous gel contains a glycine–nitrate complex impregnated with metal cat-ions. The complex organic network containing glycine does not allow the metal cations to suspend, which in turn enables the precursor to have compositional homogeneity. Consequently, a series of complex metal oxide powders have been synthesized. It is important to note that the reaction between the glycine and the metal ion complexes is highly exothermic, and the glycine component acts as a reducer of metal salt, which in turn gets oxidized due to the presence of oxygen-ligands of the metal salt. This simultaneous oxidation-reduction (redox) reaction effec-tively results in the formation of metal oxide.[12] The reducing agent is called “the fuel,” and the chemical species which pro-vides oxygen for the redox reaction is termed “the oxidizer.” Generally, metal nitrates serve the purpose of oxidizer and also as the source of cations. The exothermic reaction between the fuel and the oxidizer are utilized systematically in building materials of interest. The exothermic combustion reaction ini-tiates at low-temperatures, and the energy output decides the path of the reaction. If the heat energy output is higher than the amount of energy required to initiate the reaction, then the combustion is self-propagating. The ingredients of the combus-tion reaction (fuel, oxidizer, and requisite temperature) are pic-torially presented in the form of a triangle (Figure 7) along with the various potential fuels. Over the years, the bulk combus-tion is utilized to synthesize nanoscale materials such as metal

Figure 6. a) Reaction coordinate diagram of combustion and conven-tional thermal decomposition pathways. Adapted with permission.[1] Copyright 2011, Nature Publishing Group.



© 2020 Wiley-VCH GmbH2000464 (8 of 25)


oxides, sulfides, alloys, and metals.[13] The recent advancements in the combustion reactions involve the synthesis of 2D mate-rials and the deposition of functional thin films.[13]

4.1. Combustion Derived Metal Oxide Thin Films

Combustible precursors are used in the fabrication of thin films of functional metal oxides, metals, and metal–metal oxide composites. The first study on the fabrication of In2O3, IZO, ITO, and zinc tin oxide (ZTO) was reported in 2011,[1] where a combustible precursor containing metal nitrates and fuel in an organic solvent (2-methoxyethanol: 2-ME) was spin-coated and annealed at a temperature of 200 °C to produce functional metal oxides. The deposited thin films of all metal oxides were smooth (Figure 8a). Additionally, the semiconducting In2O3 thin films deposited at 200 °C were crystalline (Figure 8b). The low-temperature deposition of functional metal oxides has widely been explored after its first realization, in various areas of elec-tronic device fabrication such as TFTs, solar cells, organic light-emitting diodes (OLEDs), and gas sensors. The list of different applications, along with performance parameters, is shown in Table 3. Further the combustion processing is also investigated to develop new oxide systems such as yttrium doped In2O3 (IYO). The ternary IYO is preferred over quaternary IGZO due

Figure 7. Triangle mentioning the ingredients of combustion reactions and different fuels.

Figure 8. a) Surface topography from atomic force microscopy of In2O3, In0.7Zn0.3O1.35, Zn0.3Sn0.7O1.7, In9.0Sn1.0O15.5 thin films deposited on n++ silicon substrate and annealed at 200 °C, b) grazing incidence angle X-ray diffraction of In2O3 thin films annealed at different temperatures, c) surface mor-phology of deposited In2O3 thin films by scanning electron microscopy via multiple coating steps and annealed at 250 °C. Adapted with permission.[1] Copyright 2011, Nature Publishing Group.



© 2020 Wiley-VCH GmbH2000464 (9 of 25)


to reduced scattering sites and effectively suppressed oxygen vacancies with minimum fraction of yttrium.[58] The IYO thin films annealed at 250 °C were found to be amorphous with microscopically smooth morphology (Figure 9a,b). The TFTs based on IYO/SiO2 have shown a μ, threshold voltage (Vth), and on:off ratio (Ion:Ioff) of 0.75 cm2 V−1 s−1, 105–106, 10 V, respec-tively. Figure 9c,d shows the transfer and output curves of IYO/SiO2 TFTs. Although initial demonstrations present the efficacy of low-temperature combustion processing in the deposition of robust functional metal oxide thin films, there were still signifi-cant issues of morphology control in thick film deposition. In general, the single step deposition of thick metal oxide results in porous thin film, which is not advisable for electrical func-tionalities. Figure 8c shows the surface morphology of depos-ited In2O3 thin films with different thicknesses. To achieve a dense or porous/pinhole-free thin film, multiple coatings with intermediate densification through thermal annealing is essen-tial. On the other hand, controlled spraying of a combustible precursor onto a preheated substrate achieves better densifica-tion.[59,60] The subsequent section deals with the advancements in the combustion processing of thin films.

4.2. Advancements in Combustion Processed Thin Films

A thin film of functional metal oxide is considered as robust if it is dense with nonporous morphology. In this regard, the low-temperature spray combustion derived films have shown extraordinary performance equivalent to those of sputtered thin films.[59] However, the gaseous products of combustion readily compromise the quality of spin-coated thin films (Figure 10). While the sole advantage of spray combustion is that the depos-ited film is of high density, but controlling the thickness and surface roughness of the thin films can be challenging. The thickness of the films needs to be as low as possible (<5–10 nm) in one step deposition because the exothermicity of the reac-tion retains for a short period.[12,59,61] Successful utilization of

the combustion peak temperature depends on the thickness of the film. Figure 10 shows the strategy of thin film deposition using both spin and spray combustion techniques. Moreover, the nanoscopic density, roughness, and thickness of thin films

Figure 9. a) X-ray diffraction patterns of IYO thin films with varying yttrium concentration annealed at 250 °C, b) AFM surface topography of 2.5 mol% IYO annealed at 250 °C, c) transfer, and d) output charac-teristics of IYO/SiO2 TFTs. Adapted with permission.[58] Copyright 2012, American Chemical Society.

Table 3. Performance parameters of thin films of various materials deposited using combustion technique. HIL: hole injection layer, OLED: organic light-emitting diode, T: annealing temperature, F–M–O: ferroelectric–magnetic–optical, AcAc: acetylacetone, 2,4-PENT: 2,4-pentanedionate, EG: eth-ylene glycol, SMCCR: Suzuki–Miyaura cross-coupling reaction, 2-ME: 2-methoxyethanol, An-E: anhydrous ethanol, g-Ac: glacial acetic acid, t: film thickness, CE: current efficiency, EQE: external quantum efficiency, PS: polarization, M: magnetization, Eg: bandgap, conc: concentration.

Application Material T [°C] Fuel Solvent t [nm] Performance parameters reported Ref.

“HIL” in OLED NiOx 150 AcAc 2-ME ≈15 CE 45.8 cd A−1 EQE 10.9% [89]

Acetone vapor sensor SnO2 350 Urea 2-ME ≈40–60 Sensitivity = 5 with conc: 10 ppm [90]

Thin film catalyst ethanol oxidation

Pd 250 Glycine H2O ≈260 Yield: 89% (one cycle) (SMCCR) [91]

“HIL” in OLED Cu: NiOx 200 AcAc 2-ME ≈30 CE 85.3 cd A−1 EQE 22.3% [92]

“HIL” in OLED NiO 220 Glycine An-E ≈85 CE 4.56 cd A−1, Lifetime: 2711 h [93]

F–M–O Thin film

BiFeO3 500 2,4-PENT H2O/HNO3/g-Ac ≈215 PS ≈ 67 μC × cm2 M: 2.50 emu × g−1 Direct Eg ≈ 2.82 eV

[94]

Gas sensor (Porous sheet)

NiO 400 EG H2O – Concmin: 5 ppm Concmax:1000 ppm

[95]

Phototransistor with (combustion derived semiconductor)

ZnSnO 350 AcAc 2-ME ≈47 OFF state: UV sensitivity ≈105 ON state: UV sensitivity ≈101

[96]



© 2020 Wiley-VCH GmbH2000464 (10 of 25)


of technologically relevant semiconducting IZO and IGZO are presented (Figure 10).[60] The effortless transformation of the precursor to metal oxide takes place when the atomized droplets come in contact with the preheated substrate. Since the mole-cular mixing ensures the compositional homogeneity, all atom-ized droplets embody the required quantity of both fuel and oxidizer for the combustion. Since the densification is strongly dependent on the exothermicity of the combustion reaction, it is essential to use stoichiometric amounts of fuel and oxidizer. Any deviation in the quantities results in the creation of pores or pinholes. In addition, due to prolonged processing time, the cost associated with multilayer spin deposition may reach conventional vacuum-based fabrication methods. In contrast, the single step spray deposition has the capability to achieve electrically robust dense thin film within a short time span. Figure 11a,b illustrates the quality of 20 nm thick single step spray deposited and multilayer (four times coating with 5 nm thickness in each step) spin deposited In2O3 along with its per-formance in TFTs[59] (Figure 11c,d).

The single step spray combustion has resulted in polycrystal-line In2O3 with pronounced crystallinity compared to multilayer spin deposition. In contrast, multicomponent systems such as IGZO and IZO are found to be amorphous with single step spray deposition. TFTs based on spin (multilayer) and spray

(single step) deposited In2O3 (at 200 °C) channel layer have clearly shown the improved performance with spray deposi-tion.[59] The static performance parameters such as μ, Vth, and Ion:Ioff have boosted from 0.83 ± 0.17 to 1.44 ± 0.22 cm2 V−1 s−1, 11.5 ± 5.1 V to 9.6 ± 5.0 V, and ≈105–106, respectively.[59] The TFTs with spray combustion deposited IGZO has presented high per-formance approaching those fabricated by magnetron sputtered IGZO (μspray: 8.4 ± 0.99 and μsputtering: 10.9 ± 0.12 cm2 V−1 s−1 with sol–gel processed ZrOx).[59] Further, the integration of spray combustion deposited dielectrics and semiconductors had been reported for the realization of low voltage operation of TFTs.[60] Figure 12 shows the surface roughness, and transfer character-istics of TFTs based on spray combustion derived dielectrics (Al2O3 and/or ZrO2) and semiconductors (IGZO and/or IZO).

Furthermore, the feasibility of spray combustion is also investigated in depositing thin films of pure metallic systems.[62] The formation of metals from the combustion reaction is an outcome of fuel-rich precursors. The presence of excess fuel, in turn, helps in reducing the metal oxide to pure metal.[63] Based on this, pure metals such as nickel and alloys of nickel–copper–iron have been synthesized using glycine as a fuel.[63] With an increase in the fuel:oxidizer ratio (φ) from 0.75 to 1.0, 1.25, and 1.75, the transition from nickel oxide (NiO) to pure nickel has been achieved. Both fuel-lean (φ < 1) and stoichiometric φ

Figure 10. Thin film deposition strategy depicting the advantages of multiple spin coatings with intermediate annealing and spray combustion tech-nique along with the AFM topography of IZO and IGZO semiconducting films deposited at different temperatures. (Note: σ is the root mean square roughness) Adapted with permission.[60] Copyright 2016, Wiley-VCH.



© 2020 Wiley-VCH GmbH2000464 (11 of 25)


processes provide excess oxygen to the reaction resulting in the formation of NiO. The X-ray diffraction (Figure 13a) of the SEM micrographs revealed the formation of metallic nickel at larger magnitudes of φ (Figure 13b).[64]

In view of this, there was an attempt to deposit pure silver and silver–copper oxide composite films on glass substrates.[62] Nitrates of copper and silver served the purpose of the oxidizer, and citric acid acted as the fuel. The presence of excess fuel helped in reducing silver oxide to pure silver. Figure 14 shows, step-by-step derivation, and comparison of combustion reac-tions involved in the formation of both silver oxide and metallic-silver. The amount of fuel required to form pure silver is higher than that of its oxide. This analogy is further implemented in depositing electrically conductive silver–copper oxide composite films.[62] The aqueous precursors of silver and copper were mixed and sprayed on a preheated substrate (at 170 °C). The maximum electrical conductivity reported was of the order of 105 S cm−1 for the composite films with 50% silver and the rest copper oxide.[62] Further, it is interesting to note that the mole-cular level mixing of the precursor did not yield an alloy of silver and copper; instead, the reaction between copper and oxygen prevailed. Therefore, it is clear that spray combustion is effi-cient in the in situ fabrication of metal–metal oxide composite films. The advantage of spray combustion is not only attributed to the realization of dense and robust metal oxide thin films, but also in the achievement of short processing periods.

The processing time is a crucial parameter, which decides the total cost of production. It is important to note that the as-deposited thin films of combustible precursors require prolonged annealing time (≈10 min–2 h) to eliminate organic contaminants entirely and to ensure high densification with porous-free morphology.[64] Such a postannealing step demands a continuous supply of heat energy, raising the cost of production. In this regard, modification in combus-tion chemistry is utilized to reduce processing time signifi-cantly (10–60 s).[64] The use of co-fuel system, coupled with blade coating technique, has reduced not only the time of processing, but has also proved that the efficacy of combus-tion in depositing wafer-scale functional metal oxides.[64] Figure 15a–f shows a wafer (4″) scale TFTs with IGZO-sem-iconductor and Al2O3-dielectric along with μ statistics, char-acteristic curves, and Vth shift versus time plots. The wafer-scale TFTs (100 devices) have shown μ, Vth, and Ion:Ioff of 18 ± 2.5 cm2 V−1 s−1, 0.55 ± 0.27 V, and ≈104–105 respectively with a negligible Vth shift.[64] The addition of co-fuels yields high densification and enhanced metal oxide lattice formation. The substantial improvement in the TFT performance is attrib-uted to an optimum concentration of co-fuel in the precursor (Figure 16a). In an investigation, a range of co-fuels has been employed with primary acetylacetone.[65,66] The improvement in the amount of heat energy liberated from the exothermic reaction (ΔHexo), due to the presence of co-fuels, is greater

Figure 11. a) High-resolution, bright-field TEM images with selected area energy-filtered electron diffraction patterns of In2O3 films deposited by multilayer spin coating and single step spray combustion. b) AFM surface topography of 20 nm In2O3 films. c) TFT transfer characteristics and d) μ distribution statistics for 20 nm thick In2O3 on Si/SiO2 substrates. (Note: CS: combustion synthesis, SCS: spray combustion synthesis). Adapted with permission.[59] Copyright 2015, United States National Academy of Sciences.



© 2020 Wiley-VCH GmbH2000464 (12 of 25)


than those with only acetylacetone. Also, various co-fuels obey a linear relationship between ΔHexo and TFT μ (Figure 16b).[66] Note that excess co-fuel in the system can act adversely and reduce the performance of TFTs due to carbonaceous impu-rity. Moreover, the precursors without a primary fuel also show exothermic combustion response; the presence of an organic solvent serves the purpose of fuel as well.[67] The dual role of

organic solvent, 2-ME, has shown marginal improvement in the performance of SiO2/ZTO TFTs in the absence of urea-fuel,[67] which is attributed to increased generation of defect states at the semiconductor–dielectric interface. Another inves-tigation[68] on semiconducting indium zinc tin oxide (IZTO) deposited using the dual role of 2-ME has quantified the frac-tion (0.3%) of 2-ME serving the purpose of fuel.

Figure 12. AFM topography and transfer characteristics of TFTs based on spray combustion derived various functional metal oxides processed at a) 300 and b) 350 °C. (Note: σ is the root mean square roughness). Adapted with permission. [60] Copyright 2016, Wiley-VCH.

Figure 13. a) X-ray diffractograms b) scanning electron micrographs of combustion products synthesized with different magnitudes of φ. Adapted with permission.[63] Copyright 2017, American Chemical Society.



© 2020 Wiley-VCH GmbH2000464 (13 of 25)


Most of the studies on combustion derived metal oxide thin films are based on organic solvent, mainly 2-ME. Due to the large carbonaceous load from the organic solvent, the as-depos-ited film needs high annealing temperatures. Thus, the replace-ment of organic solvent by an ecofriendly aqueous solvent is reported in depositing high κ Al2O3.[69] A comparative study on 2-ME and water-based combustible precursors of Al2O3 has revealed that the thin film deposited from as-prepared (without aging) aqueous combustible precursors are amor-phous and depict high areal capacitance. On the other hand, 2-ME based precursors require aging of 72 h (Figure 17a,b), to achieve nearly similar areal capacitance. The TFTs based on aqueous combustion processed 10 nm thick Al2O3, and solu-tion-processed gallium zinc tin oxide (GZTO)-semiconductor (Figure 17c), have depicted a saturation μ and turn-on voltage of 1.3 cm2 V−1 s−1 and 0.5 V, respectively.[69] Figure 17d shows the transfer characteristics of aqueous Al2O3/solution processed GZTO TFT. Another way of reducing carbonaceous load is to incorporate both fuel- and oxidizer-ligand in a single chemical species.[70] Figure 18 shows the Oak Ridge thermal ellipsoid plots of water-soluble indium nitrate–urea [In(urea)6(NO3)3], and zinc nitrate–urea [Zn(urea)4(H2O)2][NO3]2 complexes;

where nitrates are the oxidizers and urea is the fuel. Such complexes are directly utilized in the fabrication of IGZO TFTs and the performance parameters, such as μ, Vth, and Ion:Ioff is estimated to be 3.1 cm2 V−1 s−1, 4.3 V, and ≈107, respectively.[70]

Although combustible precursors fall into the category of soluble chemical precursors, their utilization is also dem-onstrated in the nanoparticle-based system as electrical low-temperature solder.[71] The thin films processed via nanopar-ticle precursor route show poor electrical connectivity due to retained surface capping agents, which compromises the inherent electrical characteristics of functional nanoparticles. The integration of a nanomaterial-based precursor system with low-temperature combustion improves not only the electrical connectivity between the nanostructures, but also preserves the intrinsic electronic character. The as-deposited zinc oxide (ZnO) nanorod-based TFTs has depicted an electron μ of the order 10−2–10−3 cm2 V−1 s−1.[71] However, when nanorod based thin films are impregnated/overcoated with the ZnO combus-tible precursor, the electron μ is enhanced to 0.2 cm2 V−1 s−1. Figure 19a,b shows the thickness, and surface morphology (inset) of ZnO nanorod thin films and nanorods impregnated in combustion derived ZnO matrix, respectively. Figure 19c–e

Figure 14. Derivation of stoichiometric combustion reactions to achieve silver oxide and its comparison with reaction forming metallic silver. (Note: n = number of moles of fuel required for one more of an oxidizer).



© 2020 Wiley-VCH GmbH2000464 (14 of 25)


shows the transfer and output characteristics of ZnO nanorod-based and nanorods impregnated in combustion derived ZnO matrix-based TFTs. Further, the low-temperature combustion processing is also employed in the fabrication of p-type metal oxides such as copper doped nickel oxide (Cu:NiO).[72] The low-temperature (150 °C) deposited Cu:NiO (5 at% copper doping) has been found to exhibit excellent electrical characteristics. The TFTs composed of Cu:NiO and high κ ZrO2 has shown a μ, Ion:Ioff, and Vth of 1.5 cm2 V−1 s−1, 104, 0.45 V, respectively. Figure 20 shows the transfer curves of ZrO2/Ni:CuO TFTs with varying doping concentrations of copper and histogram of TFT performance parameters (μ and Vth).

To further improve the performance of both n- and p-type TFTs, combustion processing is accompanied by photochem-ical activation. UV irradiation promotes the formation of metal–oxygen bonds through the successive breaking of alkoxy groups.[54,73,74] Park et al. introduced the photochemical activa-tion process for the conventional sol–gel precursor system of ternary-IZO and quaternary-IGZO.[73] The successful integra-tion of combustion processing and photochemical activation has resulted in outstanding TFT characteristics at low tempera-tures (<200 °C)[75] (Figure 21a). The far UV (FUV) irradiation has depicted a substantial improvement in the dielectric response of AlOx (Figure 21b) and the performance of IGZO and In2O3

Figure 15. a) Blade coating process with inset: 4″ wafer, b) optical micrograph showing the top view of IGZO/Al2O3 TFT, c) μ statistics, d–e) output and transfer characteristics IGZO/Al2O3 TFT, f) bias stress measurements (ΔVth versus time). Adapted with permission.[64] Copyright 2019, United States National Academy of Sciences.

Figure 16. a) μ of 10% and 20% co-fuel content-derived Si/SiO2/IGZO/Al TFTs, b) plot of TFT μ versus ΔHexo. (Note: Sor: sorbitol, Gly: glycine, VC: l-ascorbic acid, Ery: erythritol, Glu: glucose, Xyl: xylitol, Suc: sucrose). Adapted with permission.[66] Copyright 2019, Wiley-VCH.



© 2020 Wiley-VCH GmbH2000464 (15 of 25)


TFTs.[75] The photochemical activation and combustion pro-cessing are also employed in the deposition of multilayer oxide dielectrics (HfOx and AlOx), and the TFTs fabricated with the compound dielectric have revealed stable performance.[76] Also, p-type TFTs with lithium doped nickel oxide (Li:NiOx) active layer have been investigated using combined photochemical activation and combustion processing.[77] The UV irradiation has contributed to improving hole transport at considerably low processing temperature (150 °C). Figure 21c–e shows the transfer characteristics of TFTs with varying concentrations of lithium in Li:NiOx along with the statistical distribution of performance parameters. Further, combined UV irradiation and combustion processing have also been employed in large area AlOx/In2O3 TFTs on a flexible substrate using flexographic printing.[78]

4.3. Combustion Processed Transparent Electrodes

Although combustion processing ensures the acceptable perfor-mance of functional metal oxides at considerably low tempera-tures, the increase of annealing temperature further enhances the functionalities, which can be attributed to defect chemistry and enhancement in crystallinity. For example, commercial-grade TCO (i.e., ITO) with a composition of In:Sn::9:1 has shown an improvement in electrical conductivity with annealing temperature (200 °C: 0.35 S cm−1 and 250 °C: 130 S cm−1),[1] due to the formation of oxygen vacancies, contributing to the overall charge concentration. Similar observations have been made in complex ternary oxides such as IZTO.[79] The low-temperature crystallinity achieved in IZTO films is evidence of the fuel-acety-lacetone additive. Apart from conventional dopants (i.e., Sn and

Figure 17. Capacitance versus voltage characteristics of Si/AlOx/Al capacitors with a) aqueous, b) 2-ME processed AlOx depicting the effect of aging, c) high-resolution focused ion beam-scanning electron cross-sectional micrograph of TFT, d) transfer characteristic of Si/aqueous processed AlOx/GZTO/Al TFT. Adapted with permission.[69] Copyright 2014, American Chemical Society.

Figure 18. Oak Ridge Thermal Ellipsoid plots of metal–urea complexes. Adapted with permission.[70] Copyright 2018, Wiley-VCH.



© 2020 Wiley-VCH GmbH2000464 (16 of 25)


Zn in In2O3), doping of several elements from the group 3/4f has been demonstrated using the combustion processing. In addition, combustion processing is useful in depositing other electrically functional metal oxides, such as CaVO3,[80] which

is of enormous importance because of the utilization of low-cost materials for TCO such as calcium rather than rare-earth indium. Moreover, CaVO3 is classified as a “correlated metal,” which shows a metallic conductivity when deposited at near

Figure 19. The thickness and surface morphology (inset) of a) ZnO nanorod thin films, b) ZnO nanorods impregnated in combustion derived ZnO matrix, c) transfer, and d–e) output characteristics TFTs (Si/SiO2/ZnO/Al) based on ZnO nanorod and ZnO nanorod impregnated in combustion derived ZnO matrix (nanocomposite). Adapted with permission.[71] Copyright 2012, American Chemical Society.

Figure 20. a) The transfer characteristics of combustion derived Cu:NiO TFTs with varying concentrations of copper doping. Histograms depicting TFT performance parameters, b) μ, c) Vth for 5 at% copper doped nickel oxide. Adapted with permission.[72] Copyright 2017, Wiley-VCH.



© 2020 Wiley-VCH GmbH2000464 (17 of 25)


room temperature.[81] The maximum electrical conductivity of CaVO3 at 140 °C was found to be about 3 S cm−1[80] (Table 4), which is comparatively low and requires a further improvement by increasing combustion efficiency by adopting recent advance-ments in thin film combustion. The solution combustion tech-nique was successfully adopted in depositing transparent and conducting titanium doped indium oxide (In14TiO23) films, and

it was reported that the films deposited at a temperature of 350 °C (Table 4) have resulted in excellent optical transparency (≈100%) in the visible spectrum.[82] Furthermore, the composite films comprising ITO and AgNWs have also been demon-strated using solution combustion[83] (Table 4). Metal oxides are well-known for n-type behavior due to the large concentration of anion vacancies, but few p-type metal oxides have also been

Figure 21. a) Improvement in the TFT performance at low processing temperature due to UV aid, b) capacitance versus frequency characteristic of FUV irradiated AlOx at a low processing temperature of 180 °C. The performance of Si/ZrO2/Li:NiOx/Ni TFTs: c) transfer characteristics and d–e) histograms presenting the statistical variation of performance parameters (μ and Vth). Adapted with permission.[75,77] Copyright 2016, American Chemical Society and Copyright 2018, Royal Society of Chemistry.

Table 4. Solution combustion derived TCOs and respective deposition conditions. t: film thickness, T: annealing temperature, Rs: sheet resistance, σ: electrical conductivity, Ce-AcAct: cerium acetylacetonate, 2-ME: 2-methoxyethanol, Tr: optical transparency, AcAc: acetylacetone, AgNWs: silver nanowires, FTO: fluorine doped tin oxide, IZTO: indium zinc tin oxide, ITO: indium tin oxide.

Material t [nm] T [°C] Fuel Solvent σ [S cm−1] or Rs [Ω sq−1) Tr [%] Ref.

In14TiO23 ≈166 350 AcAc 2-ME 20 S cm−1 (at 450 °C) ≈100 [82]

CuCrO2 (p-type) ≈75 180 Glycine 2-ME 0.14 S cm−1 ≈90 [84]

ITO – 250 AcAc 2-ME ≈81 Ω sq−1 – [97]

IZTO – 200 AcAc 2-ME ≈14 S cm−1 ≈83 [79]

CaVO3 – 140 AcAc 2-ME ≈3 S cm−1 – [80]

AgNWS-ITO (composite)

≈120 250 AcAc 2-ME ≈18 Ω sq−1 ≈80 [83]

Ce: In2O3 ≈100 500 Ce-AcAct 2-ME ≈60.7 Ω sq−1 ≈90 [98]

ITO ≈160 250 AcAc 2-ME ≈23 S cm−1 ≈90 [99]

FTO – 420 Urea 2-ME ≈6.6 Ω sq−1 ≈90 [100]



© 2020 Wiley-VCH GmbH2000464 (18 of 25)


reported. The origin of p-type conduction can be traced to the formation of cation vacancies such as VCu′′ in CuO, VSn′′ in SnO. One such p-type (CuCrO2) TCO has been reported to display an electrical conductivity of 0.14 S cm−1, deposited using a combus-tible precursor involving metal nitrates and glycine as oxidizers and fuel, respectively[84] (Table 4).

Additionally, combustion derived thin films have been employed as various components of solar cells, such as electron transport layers, hole transport layers, and contact electrodes. Series of metal oxides have been investigated for their electrical/optical features in solar cells. Table 5 shows the configuration of solar cells, the components with combustion assisted deposition followed by the cell performance parameters. A large number of metal oxides have been deposited using solution combustion with acetylacetone/urea-fuel. Various other potential fuels still need to be investigated for their combustion efficiency in the formation of thin films. Furthermore, along the side of com-bustion derived metal oxides in TFTs (Table 6), there are a few reports that describe attempts to fabricate high κ metal oxide dielectrics, such as hafnium oxide (HfOx),[85] and sodium beta alumina (SBA)[86] with dielectric permittivity of 14.2 and 21, respectively. In particular, the low-temperature (about 200 °C) formation of the beta phase in SBA (NaAl11O17) via spray depos-ited aqueous combustible precursor is the evidence of low-temperature phase stabilization.[12] In addition to the applica-tions mentioned above, the solution combustion method is also extended to deposit electrochemically active porous films.[87,88]

Overall, the functional thin films derived from low-temper-ature combustion processing are of high technological impor-tance due to the fact that the process is compatible with flex-ible and large area substrates. After the first realization of combustion reactions at nanoscale thin films, there have been numerous advancements such as spraying atomized drop-lets of combustible precursors to achieve single step dense and porous/pinhole morphology, use of ecofriendly water as a medium of dissolution, significant reduction in annealing time by incorporating optimized quantities of co-fuels and fetching

the dual role of organic solvent. Notably, the performance of combustion derived thin films is comparable with those depos-ited with conventional sputtering techniques. In addition, the versatility of combustion processing in depositing not only metal oxides but also metals/metal–metal oxide composite films have paved the way towards commercialization of com-bustion processing for solution processed low-cost large area electronics on flexible substrates.

Over the years, a continuous effort has been made to develop low-temperature processing and post-treatment methods for functional metal oxides thin films. Apart from combustion processing, the other methods such as alkoxide precursors, microwave annealing, high pressure annealing, aqueous pre-cursors, photochemical activation, and O2/O3 annealing have been developed.[37] A comparative plot comprising of processing temperature and device performance parameters is presented in Figure 22. Despite significant advancements in combustion processed nanoscale thin films, there are few shortcomings, which are worth noting. The proper selection of fuel, co-fuel, and solvent(s) requires a good knowledge of reaction chemistry. Also, inadequate thermal analysis tools to predict the nanometer thick precursor films’ decomposition behavior impose difficulty in selecting the right annealing temperatures. The selection of fuel, in bulk combustion processing, is well defined, and sev-eral fuels have been investigated for their combustion efficiency. But, in the case of thin film combustion, most of the studies are based on acetylacetone and urea. In this regard, a large number of organic fuels are yet to be investigated. Further, the deposition method in the case combustion-derived thin films plays a crucial role in deciding the devices’ performance. The spin coating demands multiple layers of deposit to achieve the desired thickness; such multistep deposition increases the cost of production. Nevertheless, a proper understanding of the reac-tion thermodynamics and decomposition strategy of nanometer thick as-deposited precursors plays a vital role in achieving high-performance devices. The subsequent section deals with the mechanistic understanding of combustion derived thin films.

Table 5. Performance of solar cells with one of the components (in bold letters) deposited using combustion reaction. t: thickness of thin film, T: annealing temperature, Rs: sheet resistance, Tr: optical transparency, VOC: power conversion efficiency, JSC: short circuit density, FF: fill factor, PCE: power conversion efficiency, 2-ME: 2-methoxyethanol, AcAc: acetylacetone, Zn-AcAct: zinc actylacetonate, Poly-TPD: poly (4-butyl-phenyl-diphenyl-amine).

Solar cell configuration Reaction parameters t [nm] Tr [%] Rs [Ω sq−1]

Performance parameters Ref.

Fuel Solvent T [°C] VOC [V] JSC [mA cm−2] FF PCE [%]

ITO/Cu:NiOx/CH3NH3PbI3/C60/Bis-C60/Ag AcAc 2-ME 150 ≈50 >95 1.8 1.05 22.23 0.76 17.74 [101]

Glass/Mo/CIGS/[AZO-AgNWs-AZO] Zn-AcAct 2-ME 200 ≈150 93.4 11.3 0.54 36.43 0.56 11.03 [102]

ITO/SnO2/CH3NH3PbI3/spiro/Ag Urea-AcAc 2-ME 140 ≈20 – – 1.04 19.20 0.52 10.36 [103]

AgNWS/ZnO/PTB7-Th:PC70BM/MoO3 Zn-AcAct 2-ME 180 ≈50 ≈87 30 0.80 16.48 0.64 8.48 [104]

FTO/NiO/MAPbI3/PCBM/Ag AcAc Alcohol 330 ≈40 – – 1.03 17.42 0.71 12.7 [105]

ITO/MoO3/P3HT:PC61BM/Al AcAc H2O 150 ≈25 – – 0.60 8.88 0.63 3.56 [106]

ITO/In2O3/PCBM/MA1−yFAyPbI3−xClx/Spiro/Ag AcAc 2-ME 150 – ≈85 – 1.04 20.85 0.64 13.75 [107]

ITO/NiOx/ PC71BM/LiF/Al AcAc 2-ME 175 ≈10 ≈90 5.87 0.87 10.50 0.70 6.42 [108]

ITO/PEDOT/NiO/Poly-TPD/PC70BM/Al Urea An-E 225 ≈65 ≈95 – 0.52 5.28 0.62 1.72 [109]

FTO/Zn:Nb2O5/CH3NH3PbI3/Spiro/Ag AcAc 2-ME 200 – ≈80 – 1.12 22.55 0.70 17.70 [110]

FTO/NiOx/MA1−yFAyPbI3xClx/PCBM/BCP/Ag AcAc 2-ME 250 ≈30 ≈98 – 1.12 23.7 0.76 20.2 [111]

ITO/ZnO/ CH3NH3PbI3/Spiro/Au AcAc 2-ME 150 ≈30 – – 1.0 23.03 0.7 16.47 [112]



© 2020 Wiley-VCH GmbH2000464 (19 of 25)


4.4. Mechanism of Combustion in Nanoscale Films

After the first article[1] on combustion derived functional metal oxide thin films, there have been a consecutive report about a systematic understanding of combustion reactions in

nanoscale thin films.[71] Owing to the large surface-to-volume ratio of thin films facilitate both heat and mass transfer with greater ease.[71] The factors such as predecomposition, heating profile, and sample size play a vital role in realizing thin film combustion.[71,135] The presence of fuel-ligand indeed helps to

Table 6. Performance of TFTs with components (in bold letters) deposited using combustion reaction. t: thickness of thin film, T: annealing tem-perature, μ: mobility, Vth: threshold voltage, SS: subthreshold swing, Ion:Ioff: on:off ratio, 2-ME: 2-methoxyethanol, AcAc: acetylacetone, Zn-AcAct: zinc actylacetonate, TFPD: 1,1,1-trifluoro-2,4-pentanedione, NAcAc: nitroacetylacetone, Zr-SAND: zirconia-based self-assembled nanodielectric, ZATO: zinc aluminum tin oxide, GZTO: gallium zinc tin oxide, SBA: sodium beta alumina, ILZO: indium lanthanum zinc oxide, PVP: polyvinylpyrrolidone, IGO: indium gallium oxide, g-Ac: glacial acetic acid, ZTO: zinc tin oxide, IZO: indium zinc oxide.

Architecture of TFTs Reaction parameters t [nm] Performance parameters

Fuel Solvent T [°C] μ [cm2 V−1 s−1] Ion:Ioff Vth [V] SS [V dec−1] Ref.

Si/SiO2/ZATO/Al AcAc 2-ME 300 ≈35 5.23 ≈106 −3.74 0.91 [113]

Si/AlOx/GZTO/Al Urea H2O 350 ≈44 1.3 ≈105 0.5 0.3 [69]

Si/SiO2/IGZO/Al Urea H2O 350 ≈9.7 3.1 ≈107 4.3 – [70]

Si/HfOx/IGZO/Al Urea 2-ME 150 ≈18 31.2 ≈105 −0.04 0.08 [76]

Si/SiO2/IZTO/Al AcAc 2-ME 300 ≈45 1.44 ≈106 1.23 0.41 [114]

PI/ITO/Al2O3/IGZO/ITO AcAc 2-ME 200 ≈20 0.85 ≈108 29.21 0.49 [115]

Si/SiO2/IZO/Al AcAc 2-ME 400 ≈12 17.54 ≈107 1.43 0.35 [116]

Si/SiO2/IZO/Mo Zn-AcAct 2-ME 400 ≈4.9 1.5 ≈108 5.0 ≈0.3 [117]

Si/SBA/IZTO/Al 2-ME 2-ME 200 ≈45 4.21 ≈102 0.47 – [68]

Si/SiO2/In2O2/Al 2-ME 2-ME 200 – 3.98 ≈108 1.83 0.4 [118]

Si/SiO2/NiO(p)/[Ni/Au] AcAc 2-ME 350 ≈30 0.015 – – – [119]

Si/SiO2/AZO/Al AcAc 2-ME 250 – 1.2 ≈106 17 4.5 [120]

PI/Mo/SiO2/In2O3/Al AcAc 2-ME 250 – 2.24 ≈108 3.05 0.45 [121]

PET/ITO/ZrOx/Li:NiOx (p)/Ni AcAc 2-ME 150 – 1.69 ≈106 5.47 0.21 [77]

Si/Al2O3/IGZO/Al AcAc 2-ME 260 ≈45 2.26 ≈104 0 0.186 [122]

Si/Al2O3/IGZO/Al AcAc 2-ME 300 Al2O3 ≈ 70 IGZO ≈20

7.3 ≈105 −0.3 0.16 [123]

Si/HfOx/SAND/IGZO/Al AcAc 2-ME 300 ≈2.7 20 ≈107 0 0.19 [124]

ITO/AlO2/ZnO/Al Urea 2-ME 250 ≈250 2.4 × 10−3 ≈105 7.11 – [125]

Si/SiO2/ZTO/Al 2-ME 2-ME 350 ≈18 3.8 ≈106 –3.98 0.89 [67]

ITO/ZrO2/Cu:NiO/Ni AcAc 2-ME 150 ≈35 1.87 ≈102 1.2 0.540 [72]

Si/SiO2/In2O3/IGO/Al AcAc 2-ME 250 In2O3 ≈3 IGO ≈10

2.56 ≈108 6.3 – [126]

Si/SiO2/IGZO/Al AcAc-Glycine 2-ME 300 ≈10 6.95 ≈107 2.5 0.9–2.0 [66]

Si/SiO2/IGZO/Al AcAc-TFPD 2-ME 300 ≈11.5 1.62 ≈106 3.34 – [64]

Si/SiO2/IGO/Al AcAc 2-ME 300 ≈10–15 13.29 ≈106–7 –1.50 2.80 [127]

Si/SiO2/IGZO/Al NAcAc 2-ME 300 ≈7–8 5.69 ≈107 2.77 0.57 [128]

Si/SiO2/IGZO/Al AcAc-Sugar 2-ME and H2O 300 ≈10–11 3.87 ≈106 3.5 – [65]

Si/SiO2/In2O3/Al AcAc 2-ME 225 ≈20 1.87 ≈105–6 9.5 3.0 [59]

Si/SiO2/ILZO/Al AcAc 2-ME 250 ≈12 0.65 ≈108 23 – [129]

Si/Zr-SAND[132]/IYO/Al AcAc 2-ME 250 – 5.0 ≈104 1.01 – [58]

Si/SiO2/IZO/Al Zn-AcAct 2-ME 350 ≈8 8.70 ≈108 12.47 7.2 [8]

Si/SiO2/[In2O3+PVP]/Al AcAc 2-ME 250 ≈18–23 0.92 ≈105–6 6.4 3.7 [131]

Si/ZrO2/In2O3/Al Urea H2O 300 ≈30 2.76 ≈106 – 0.10 [132]

Si/Al2O3/ZTO/Al Urea 2-ME 500 ≈48 17.36 ≈106 1.72 0.14 [133]

PI/Al:Nd/AlOx:Nd/ZTO/Al g-Ac g-Ac 300 – 3.9 ≈105 0 0.35 [134]



© 2020 Wiley-VCH GmbH2000464 (20 of 25)


lower the metal oxide formation temperature, but a progressive reaction demands a certain degree of predecomposition of the precursor. The vacuum dried (at 70 °C) In2O3 combustible pre-cursor with acetylacetone-fuel requires a temperature of 150 °C for the complete formation of In2O3 under isothermal heating

(Figure 23a). Similar observations have also been made in ITO (Figure 23b) and ZnO systems[71] (Figure 23b). The delayed period was attributed to the generation of sufficient amount of self-catalytic species to induce combustion reaction. The exponential decrease of delay time with the rise in temperature

Figure 23. Thermal analysis of bulk precursor a) In2O3, b) ITO, and c) ZnO. d) Grazing incident angle X-ray diffraction of In2O3 thin films. Adapted with permission.[71] Copyright 2012, American Chemical Society.

Figure 22. A comparative plot of device performance of various functional metal oxides processed using different low-temperature techniques. Adapted with permission.[37] Copyright 2018, American Chemical Society.



© 2020 Wiley-VCH GmbH2000464 (21 of 25)


could explain the reaction kinetics of self-catalytic species gen-eration.[71] The complete decomposition of the combustible precursor depends on not only the heating profile, but also the sample size. At 150 °C, thicker films have resulted in the effi-cient conversion of precursor into metal oxide by generating sufficient internal heat due to combustion. On the other hand, the thin films and films that resulted from multiple thin film depositions are amorphous. Structural analysis by X-ray dif-fraction has revealed the crystalline nature of In2O3 films with thickness ≈60 and ≈105 nm[71] (Figure 23d). Besides, the film with identical thickness has resulted from multiple coats is amorphous, with a significant quantity of indium hydroxide. The absence of combustion in thin films is attributed to the loss of combustible autocatalytic species, via evaporative out-flux, before the combustion condition is reached, leading to a nonstoichiometric φ. Although the thicker films have pro-vided evidence of combustion reaction, the resultant films are of poor morphology due to the outgassing of the reaction products, which compromises electrical functionality. Thus, there exists a trade-off between the film morphology and its thickness in achieving robust functional metal oxides via low-temperature combustion reactions. Overall the combustion in thin-film could be ignited with preheating up to the desired temperature. In this case, the self-heating is bit different from conventional combustion. In traditional bulk systems, the exothermic reaction could induce sufficient heating of near materials and thus combustion reaction. In a small heat system (i.e., thin films), the self-generated heat is not enough to induce combustion; hence it requires sufficient external heat. The sum of external heat and self-heating has resulted in metal oxide formation, as revealed by the thin-film analysis. For the acetylacetone and nitrate-based system, slow heating could result in undesired fuel and/or oxidizer depletion before reaching ignition conditions.[71,136]

Although the thin film embodies required combustible elements, the kinetics of the reaction gets severely affected by parameters such as the surface-to-volume ratio and the thermal conductivity of the substrate/sample. A detailed study on the combustion mechanism of molybdenum acety-lacetone–ammonium nitrate has depicted the process con-ditions involving an optimized surface-to-volume ratio and high heating rates to promote oxidative decomposition, encouraging the occurrence of combustion reactions in thin films.[137] Although the slow heating rate of 10 °C min−1 with typical commercial thermogravimetric analysis resulted prede-composition of fuel and induced significant imbalance in fuel to oxidizer ratio, the extremely fast heating with direct place-ment of a precursor film on hot plate resulted clear combus-tion reaction behavior. For the MoO3 combustion processing, the extremely fast heating to target temperature of 250–300 °C within few second resulted organic residue free films with enhanced crystallinity.

The reports describing the mechanism and parameters that affect the decomposition pathway in thin films with fuel additives and nitrates have strictly emphasized the geometry of the spin-coated wet film. As-spun thin films, with a thickness of less than 100 nm on a substrate of size 2 × 2 cm2, will have a surface-to-volume ratio of the order of 106–107 m−1. However, in the case of spray combustion, the

situation is entirely different because the atomized droplets of several micrometers’ diameter (for example, 100 μm) can have a surface-to-volume ratio of the order of 103–104. More-over, the atomized droplets, with a high surface-to-volume ratio, on being exposed to the preheated substrate, acquire heat spontaneously. This is attributed to high heating rates, which is stated as one of the requirements for combustion. Further, it can be noted that the droplets are minuscule and directed from the nozzle to the substrate individually in order to undergo combustion. Consequently, spray deposition is expected to exhibit much easier transportation of the gaseous products of combustion. It is well-known that spray-deposited films are generally of higher thickness than spin-coated films, with multiple spray depositions eventually increasing the volume of the deposit. Given all the observations previ-ously mentioned, combustion in sprayed precursors is most likely to occur. As an example, the spray deposited ≈1.2 μm thick silver film from an aqueous precursor containing silver nitrate/citric acid in the ratio 1:2/6 at 200 °C, reveals the existence of combustion front (Figure 24). The propagation of the combustion front provides evidence of combustion in spray deposited films (supporting video). This approach of depositing metallic films at low-temperature can be used in the deposition of electrodes of various electronic devices. Fur-ther development of characterization techniques, which can track the thermal explosion in thin films with uniform abrupt heating, could further reveal the precise nature of thin film combustion of metal oxide.

5. Outlook

The review is focused on the recent advancements of combus-tion derived functional metal oxide thin films for various elec-tronic device applications. The combustion processing was initially proposed as a modification in conventional sol–gel processing to synthesize bulk materials, and the first realiza-tion of thin film combustion was brought to limelight in the year 2011. The initial results describe the efficacy of low-tem-perature deposition of semiconductors (In2O3, a-IZO, a-ZTO) and a transparent conductor (ITO) at a low-temperature of 200 °C. The variations in combustion processing, in terms of precursors and deposition methods, have been proposed for betterment in the performance of the electronic devices. The aqueous combustible precursors and the dual role of organic solvents have improved the performance of the devices, espe-cially TFTs. In addition, the spray combustion technique has resulted in an outstanding performance of TFTs matching with their sputtered counterparts. The spray deposition has also been extended in depositing electrically conductive metallic and metal–metal oxide composite thin films. Along with spray and spin coating techniques, the combustible precursors are compatible with inkjet printing and blade coating methods. The blade coated combustion derived thin films have depicted the large area compatible fabrication as well as compatibility with flexible substrates. Further, a substantial improvement in the performance of the TFTs has been reported using an inte-grated approach of combustion processing and photochemical activation.



© 2020 Wiley-VCH GmbH2000464 (22 of 25)


AcknowledgementsThe authors are thankful for the support from the National Research Foundation of Korea (2018R1D1A1B07048232, 2018R1A4A1022647).

Conflict of InterestThe authors declare no conflict of interest.

Keywordscombustion, fuels, low-temperature processing, metal oxides, oxidizers, thin films

Received: May 1, 2020Revised: June 30, 2020

Published online:

[1] M.-G. Kim, M. G. Kanatzidis, A. Facchetti, T. J. Marks, Nat. Mater. 2011, 10, 382.

[2] Y. Sun, J. A. Rogers, Adv. Mater. 2007, 19, 1897.[3] K. Nomura, H. Ohta, A. Takagi, T. Kamiya, M. Hirano, H. Hosono,

Nature 2004, 432, 488.[4] L. Wang, M.-H. Yoon, G. Lu, Y.u Yang, A. Facchetti, T. J. Marks,

Nat. Mater. 2006, 5, 893.[5] R. M. Pasquarelli, D. S. Ginley, R. O’hayre, Chem. Soc. Rev. 2011,

40, 5406.[6] A. Liu, H. Zhu, H. Sun, Y. Xu, Y.-Y. Noh, Adv. Mater. 2018, 30,

1706364.[7] M. Eslamian, Prog. Org. Coat. 2017, 113, 60.[8] Y. H. Kang, S. Jeong, J. M. Ko, J. i.-Y. Lee, Y. Choi, C. Lee, S. Y. Cho,

J. Mater. Chem. C 2014, 2, 4247.

[9] K. K. Banger, Y. Yamashita, K. Mori, R. L. Peterson, T. Leedham, J. Rickard, H. Sirringhaus, Nat. Mater. 2011, 10, 45.

[10] E. Yarali, A. Seitkhan, O. M. Bakr, A. El-Labban, U. Schwingenschlögl, V. Tung, I. McCulloch, F. Laquai, T. Liu, W. Gao, Y. Wang, T. Yang, R. Ma, G. Zhang, C. Zhong, W. Ma, H. Yan, C. Yang, Adv. Funct. Mater. 2019, 29, 1902155.

[11] J.-W. Jo, Y.-H. Kim, J. Park, J. S. Heo, S. Hwang, W.-J. Lee, M.-H. Yoon, M.-G. Kim, S. K. Park, ACS Appl. Mater. Interfaces 2017, 9, 35114.

[12] K. C. Patil, M. S. Hegde, T. Rattan, S. T. Aruna, Chemistry of Nanocrystalline Oxide Materials—Combustion Synthesis, Properties and Applications, World Scientific, NJ 2008.

[13] A. Varma, A. S. Mukasyan, A. S. Rogachev, K. V. Manukyan, Chem. Rev. 2016, 116, 14493.

[14] W. Wen, J.-M. Wu, RSC Adv. 2014, 4, 58090.[15] X. Yu, T. J. Marks, A. Facchetti, Nat. Mater. 2016, 15, 383.[16] J. F. Wager, Inf. Disp. 2016, 32, 16.[17] J. Y. Kwon, J. K. Jeong, Semicond. Sci. Technol. 2015, 30, 024002.[18] H. Lee, J.-S. Yu, C. D. Kim, I. B. Kang, IRE Trans. Electron. Devices

2007, 55, 329.[19] A. Indluru, S. M. Venugopal, D. R. Allee, T. L. Alford, J. Disp.

Technol. 2010, 7, 306.[20] S. R. J. N. Forrest, Nature 2004, 428, 911.[21] Z. Li, Z. Liu, H. Sun, C. Gao, Chem. Rev. 2015, 115, 7046.[22] H. Schmidt, F. Giustiniano, G. Eda, Chem. Soc. Rev. 2015, 44, 7715.[23] H. Hosono, J. Non-Cryst. Solids 2006, 352, 851.[24] H. Hosono, K. Nomura, Y. Ogo, T. Uruga, T. Kamiya, J. Non-Cryst.

Solids 2008, 354, 2796.[25] J. K. Jeong, J. H. Jeong, H. W. Yang, J.-S. Park, Appl. Phys. Lett.

2007, 91, 113505.[26] A. Olziersky, P. Barquinha, A. Vila, C. Magaña, E. Fortunato,

J. R. Moranto, R. Martins, Mater. Chem. Phys. 2011, 131, 512.[27] E. Fortunato, P. Barquinha, R. Martins, Adv. Mater. 2012, 24, 2945.[28] J. F. Wager, Inf. Disp. 2014, 30, 26.[29] A. Facchetti, T. Marks, Transparent electronics: from synthesis to

applications, John Wiley & Sons, Hoboken, NJ 2010.

Figure 24. Deposition of silver film on a glass substrate by spray combustion and propagation of combustion front.



© 2020 Wiley-VCH GmbH2000464 (23 of 25)


[30] A. Wang, J. R. Babcock, N. L. Edleman, A. W. Metz, M. A. Lane, R. Asahi, V. P. Dravid, C. R. Kannewurf, A. J. Freeman, T. J. Marks, Proc. Natl. Acad. Sci. USA 2001, 98, 7113.

[31] J. Hu, R. G. Gordon, J. Appl. Phys. 1992, 71, 880.[32] D. J. Seo, S. H. Park, Phys. B 2005, 357, 420.[33] B. H. Choi, H. B. Im, J. S. Song, K. H. Yoon, Thin Solid Films 1990,

193-194, 712.[34] Y. Furubayashi, T. Hitosugi, Y. Yamamoto, K. Inaba, G.o Kinoda,

Y. Hirose, T. Shimada, T. Hasegawa, Appl. Phys. Lett. 2005, 86, 252101.[35] J.-H. Heo, K. Y. Jung, D. J. Kwak, D. K. Lee, Y. M. Sung, IEEE Trans.

Plasma Sci. 2009, 37, 1586.[36] A. E. Rakhshani, Y. Makdisi, H. A. Ramazaniyan, J. Appl. Phys.

1998, 83, 1049.[37] W. Xu, H. Li, J.-.B. Xu, L. Wang, ACL Appl. Mater. Interfaces 2018,

10, 25878.[38] M. Kishor Kumar, J. T. Kalathi, Langmuir 2019, 35, 13923.[39] J. F. Wager, D. A. Keszler, R. E. Presley, Transparent Electronics,

Springer, New York 2008.[40] L. Gordon, Anal. Chem. 1952, 24, 459.[41] L. Gordon, M. L. Salutsky, H. H. Willard, S. Rice, Phys. Today 1959,

12, 60.[42] J. W. Park, B. H. Kang, K. J. Kim, Adv. Funct. Mater. 2020, 30, 1904632.[43] X. Dai, Y. Deng, X. Peng, Y. Jin, Adv. Mater. 2017, 29, 1607022.[44] I. Bretos, R. Jiménez, J. Ricote, M. L. Calzada, Chem. Soc. Rev.

2018, 47, 291.[45] P. Jittiarporn, S. Badilescu, M. N. Al Swafa, L. Sikong, V.-V. Truong,

J. Sci. Adv. Mater. Devices 2017, 2, 286.[46] J. Livage, D. Ganguli, Sol. Energy Mater. Sol. Cells 2001, 68, 365.[47] T. Schneller, R. Waser, M. Kosec, D. Payne, Chemical solution Depo-

sition of Functional Oxide Thin Films, Springer, Berlin 2013.[48] R. W. Schwartz, T. Schneller, R. Waser, C. R. Chim 2004, 7, 433.[49] Q. Jia, T. M. McClesky, A. K. Burrell, Y. Lin, G. E. Collis, H. Wang,

A. D. Q. Li, S. R. Foltyn, Nat. Mater. 2004, 3, 529.[50] G. Zou, J. Zhao, H. M. Luo, T. M. McCleskey, Chem. Soc. Rev. 2013,

42, 439.[51] M. L. Calzada, The Sol-Gel Handbook, Wiley-VCH, Weinheim, Ger-

many 2015, pp. 841.[52] V. Nicolosi, M. Chhowalla, M. G. Kanatzidis, M. S. Strano,

J. N. Coleman, Science 2013, 340, 1226419.[53] H. Yuan, R. Lubbers, R. Besselink, M. Nijland, J. E. t. Elshof, Appl.

Mater. Interfaces 2014, 6, 8567.[54] E. Carlos, R. Martins, E. Fortunato, R. Branquinho, Chem. - Eur. J. 2020,

26, 1.[55] K. Patil, M. S. Hedge, T. Rattan, S. T. Aruna, Chemistry of Nanocrys-

talline Oxide Materials-Combustion Synthesis, Properties and Applica-tions, World Scientific, NJ 2008.

[56] M. P. Pechini, US3330697, 1967.[57] M. Rahman, Ceramic Processing and Sintering, Materials Engi-

neering Series, CRC Press, Boca Raton, FL 2003.[58] J. W. Hennek, M.-G. Kim, M. G. Kanatzidis, A. Facchetti,

T. J. Marks, J. Am. Chem. Soc. 2012, 134, 9593.[59] X. Yu, J. Smith, N. Zhou, L.i Zeng, P. Guo, Y.u Xia, A. Alvarez,

S. Aghion, H. Lin, J. Yu, R. P. H. Chang, M. J. Bedzyk, R. Ferragut, T. J. Marks, A. Facchetti, Proc. Natl. Acad. Sci. USA 2015, 112, 3217.

[60] B. Wang, X. Yu, P. Guo, W. Huang, L. Zheng, N. Zhou, L. Chi, M. J. Bedzyk, M. C. Hersam, T. J. Marks, A. Facchetti, Adv. Electron. Mater. 2016, 2, 1500427.

[61] A. Facchetti, T. J. Marks, M. G. Kanatzidid, M.-G. Kim, W. C. Sheets, H. Yan, Y. Xia, US8940578B2, 2015.

[62] A. Salian, P. Pujar, S. Mandal, J. Mater. Sci.: Mater. Electron. 2019, 30, 2888.

[63] K. V. Manukyan, A. Cross, S. Roslyakov, S. Rouvimov, A. S. Rogachev, E. E. Wolf, A. S. Mukasyan, J. Phys. Chem. C 2013, 117, 24417.

[64] B. Wang, P. Guo, L. i. Zeng, X. Yu, A. Sil, W. Huang, M. J. Leonardi, X. Zhang, G. Wang, S. Lu, Z. Chen, M. J. Bedzyk, R. D. Schaller, T. J. Marks, A. Facchetti, Proc. Natl. Acad. Sci. USA 2019, 116, 9230.

[65] B. Wang, L. i. Zeng, W. Huang, F. S. Melkonyan, W. C. Sheets, L. Chi, M. J. Bedzyk, T. J. Marks, A. Facchetti, J. Am. Chem. Soc. 2016, 138, 7067.

[66] B. Wang, M. J. Leonardi, W. Huang, Y. Chen, B. J. Eckstein, T. J. Marks, A. Facchetti, Adv. Electron. Mater. 2019, 5, 1900540.

[67] D. Salgueiro, A. Kiazadeh, R. Branquinho, L. Santos, P. Barquinha, R. Martins, E. Fortunato, J. Phys. D: Appl. Phys. 2017, 50, 065106.

[68] P. Pujar, D. Gupta, S. Mandal, J. Mater. Sci.: Mater. Electron. 2019, 30, 18647.

[69] R. Branquinho, D. Salgueiro, L. Santos, P. Barquinha, L. Pereira, R. Martins, E. Fortunato, ACS Appl. Mater. Interfaces 2014, 6, 19592.

[70] S. Sanctis, R. C. Hoffmann, N. Koslowski, S. Foro, M. Bruns, J. J. Schneider, Chem.-Asian J. 2018, 13, 3912.

[71] M.-G. Kim, J. W. Hennek, H. S. Kim, M. G. Kanatzidis, A. Facchetti, T. J. Marks, J. Am. Chem. Soc. 2012, 134, 11583.

[72] A. Liu, H. Zhu, Z. Guo, Y. Meng, G. Liu, E. Fortunato, R. Martins, F. Shan, Adv. Mater. 2017, 29, 1701599.

[73] Y.-H. Kim, J.-S. Heo, T.-H. Kim, S. Park, M.-H. Yoon, J. Kim, M. S. Oh, G. i. -R. a. Yi, Y.-Y. Noh, S. K. Park, Nature 2012, 489, 128.

[74] R. E. van de Leest, Appl. Surf. Sci. 1995, 86, 278.[75] E. Carlos, R. Branquinho, A. Kiazadeh, P. Barquinha, R. Martins,

E. Fortunato, ACS Appl. Mater. Interfaces 2016, 8, 31100.[76] E. Carlos, R. Branquinho, A. Kiazadeh, J. Martins, P. Barquinha,

R. Martins, E. Fortunato, ACS Appl. Mater. Interfaces 2017, 9, 40428.[77] J. Yang, B. Wang, Y. Zhang, X. Ding, J. Zhang, J. Mater. Chem. C

2018, 6, 12584.[78] E. Carlos, J. Leppäniemi, A. Sneck, A. Alastalo, J. Deuermeier,

R. Branquinho, R. Martins, E. Fortunato, Adv. Electron. Mater. 2020, 6, 1901071.

[79] P. Pujar, S. Gandla, M. Singh, B. Gupta, K. Tarafder, D. Gupta, Y.-Y. Noh, S. Mandal, RSC Adv. 2017, 7, 48253.

[80] G. Manjunath, R. V. Vardhan, A. Salian, R. Jagannatha, M. Kedia, S. Mandal, Bull. Mater. Sci. 2018, 41, 126.

[81] L. Zhang, Y. Zhou, L. Guo, W. Zhao, A. Barnes, H.-T. Zhang, C. Eaton, Y. Zheng, M. Brahlek, H. F. Haneef, N. J. Podraza, M. H. W. Chan, V. Gopalan, K. M. Rabe, E. Engel-Herbert, Nat. Mater. 2016, 15, 204.

[82] P. Pujar, R. V. Vardhan, D. Gupta, S. Mandal, Thin Solid Films 2018, 660, 267.

[83] A. Kim, H. Lee, H.-C. Kwon, H. S. Jung, N.-G. Park, S. Jeong, J. Moon, Nanoscale 2016, 8, 6308.

[84] J. Wang, T. B. Daunis, L. Cheng, B.o Zhang, J. Kim, J. W. P. Hsu, ACS Appl. Mater. Interfaces 2018, 10, 3732.

[85] J. Weng, W. Chen, W. Xia, J. Zhang, Y. Jiang, G. Zhu, J. Sol-Gel Sci. Technol. 2017, 81, 662.

[86] B. Gupta, P. Pujar, S. S. Mal, D. Gupta, S. Mandal, Ceram. Int. 2018, 44, 1500.

[87] E. B. Secor, T. Z. Gao, M. H. Dos Santos, S. G. Wallace, K. W. Putz, M. C. Hersam, ACS Appl. Mater. Interfaces 2017, 9, 29418.

[88] M. Murayama, N. Kobayashi, Y. Matsushima, H. Unuma, J. Ceram. Soc. Jpn. 2019, 127, 741.

[89] Y. Sun, W. Chen, Y. Wu, Z. He, S. Zhang, S. Chen, Nanoscale 2019, 11, 1021.

[90] A. Fasoli, K. Lionti, L. Sundberg, R. Miller, F. Ceccarelli, L. Bozano, ISOCS/IEEE Int. Symp. on Olfaction and Electronic Nose (ISOEN), IEEE, Piscataway, NJ 2017.

[91] A. A. Voskanyan, C.-Y. V. Li, K.-Y.u Chan, ACS Appl. Mater. Interfaces 2017, 9, 33298.

[92] Y.-H. Li, X. Lu, R. Wang, Y.-Y. Ma, S. Duhm, M.-K. Fung, J. Mater. Chem. C 2017, 5, 11751.

[93] Y. Zhang, J. Lumin. 2017,192, 1015.



© 2020 Wiley-VCH GmbH2000464 (24 of 25)


[94] C. Gutiérrez-Lázaro, J. Am. Ceram. Soc. 2013, 96, 3061.[95] C. Dong, X. Xiao, G. Chen, H. Guan, Y. Wang, I. Djerdj, RSC Adv.

2015, 5, 4880.[96] L. Feng, G. Yu, X. Li, J. Zhang, Z. Ye, J. Lu, IEEE Trans. Electron

Devices 2016, 64, 206.[97] P. T. Tue, S. Inoue, Y. Takamura, T. Shimoda, Appl. Phys. A: Solids

Surf. 2016, 122, 623.[98] B. H. Kim, C. M. Staller, S. H. Cho, S. Heo, C. E. Garrison, J. Kim,

D. J. Milliron, ACS Nano 2018, 12, 3200.[99] S.-G. Ban, K.-T. Kim, B. D. Choi, J.-W. Jo, Y.-H. Kim, A. Facchetti,

M.-G. Kim, S. K. Park, ACS Appl. Mater. Interfaces 2017, 9, 26191.[100] B.-R. Koo, J.-W. Bae, H.-J. Ahn, Ceram. Int. 2019, 45, 10260.[101] J. W. Jung, C.-C. Chueh, A. K.-Y. Jen, Adv. Mater. 2015, 27, 7874.[102] A. Kim, Y. Won, K. Woo, S. Jeong, J. Moon, Adv. Funct. Mater. 2014,

24, 2462.[103] X. Liu, K.-W. Tsai, Z. Zhu, Y.e Sun, C.-C. Chueh, A. K.-Y. Jen, Adv.

Mater. Interfaces 2016, 3, 1600122.[104] M. Park, S.-H. Lee, D. Kim, J. Kang, J.-Y. Lee, S. M. Han, ACS Appl.

Mater. Interfaces 2018, 10, 7214.[105] Y. Qin, J. Song, Q. Qiu, Y. Liu, Y. Zhao, L. Zhu, Y. Qiang, J. Alloys

Compd. 2019, 810, 151970.[106] C. Yao, X. Xu, J. Wang, L. Shi, L. Li, ACS Appl. Mater. Interfaces

2013, 5, 1100.[107] X. Guo, Z. Lin, J. Ma, Z. Hu, J. Su, C. Zhang, J. Zhang, J. Chang,

Y. Hao, J. Power Sources 2019, 438, 226981.[108] S. Bai, M. Cao, Y. Jin, X. Dai, X. Liang, Z. Ye, M. Li, J. Cheng,

X. Xiao, Z. Wu, Z. Xia, B. Sun, E. Wang, Y. Mo, F. Gao, F. Zhang, Adv. Energy Mater. 2014, 4, 1301460.

[109] Y. Zhang, Z. Li, Ceram. Int. 2016, 42, 6360.[110] X. Ye, H. Ling, R. Zhang, Z. Wen, X. Ye, S. Hu, T. Akasaka, J. Xia,

X. Lu, J. Power Sources 2020, 448, 227419.[111] Z. Liu, J. Chang, Z. Lin, L. Zhou, Z. Yang, D. Chen, C. Zhang,

S. F. Liu, Y. Hao, Adv. Energy Mater. 2018, 8, 1703432.[112] D. Zheng, G. Wang, W. Huang, B. Wang, W. Ke,

J. L. Logsdon, H. Wang, Z. Wang, W. Zhu, J. Yu, M. R. Wasielewski, M. G. Kanatzidis, T. J. Marks, A. Facchetti, Adv. Funct. Mater. 2019, 29, 1900265.

[113] Q. Jiang, L. Feng, C. Wu, R. Sun, X. Li, B. Lu, Z. Ye, J. Lu, Appl. Phys. Lett. 2015, 106, 053503.

[114] Q. Jiang, J. Lu, J. Cheng, X. Li, R. Sun, L. Feng, W. Dai, W. Yan, Z. Ye, Appl. Phys. Lett. 2014, 105, 132105.

[115] Y. S. Rim, H. Chen, Y. Liu, S.-H. Bae, H. J. Kim, Y. Yang, ACS Nano 2014, 8, 9680.

[116] X. Liang, C. Wang, J. Liang, C. Liu, Y. Pei, Semicond. Sci. Technol. 2017, 32, 095010.

[117] D. E. Kim, S. W. Cho, Y. B. Kim, K. S. Kim, Ceram. Int. 2017, 43, 8956.

[118] J. S. Lee, Y.-J. Kwack, W.-S. Choi, ACS Appl. Mater. Interfaces 2013, 5, 11578.

[119] Y. Li, C. Liu, G. Wang, Y. Pei, Semicond. Sci. Technol. 2017, 32, 085004.

[120] V. Sarangi, S. Gandla, J. Mod. Mater. 2017, 3, 14.[121] Y. S. Rim, H. S. Lim, H. J. Kim, ACS Appl. Mater. Interfaces 2013,

5, 3565.[122] H. Wang, T. Sun, W. Xu, F. Xie, L. Ye, Y. Xiao, Y.u Wang, J. Chen,

J. Xu, RSC Adv. 2014, 4, 54729.[123] H. Wang, W. Xu, S. Zhou, F. Xie, Y. Xiao, L. Ye, J. Chen, J. Xu, J.

Appl. Phys. 2015, 117, 035703.[124] K. Everaerts, L.i Zeng, J. W. Hennek, D. I. Camacho, D. Jariwala,

M. J. Bedzyk, M. C. Hersam, T. J. Marks, ACS Appl. Mater. Inter-faces 2013, 5, 11884.

[125] E. J. Bae, Y. H. Kang, M. Han, C. Lee, S. Y. Cho, J. Mater. Chem. C 2014, 2, 5695.

[126] X. Yu, N. Zhou, J. Smith, H. Lin, K. Stallings, J. Yu, T. J. Marks, A. Facchetti, ACS Appl. Mater. Interfaces 2013, 5, 7983.

[127] S. L. Moffitt, K. L. Stallings, A. F. Falduto, W. Lee, D. B. Buchholz, B. Wang, Q. Ma, R. P. H. Chang, T. J. Marks, M. J. Bedzyk, ACS Appl. Electron. Mater. 2019, 1, 548.

[128] Y. Chen, B. Wang, W. Huang, X. Zhang, G. Wang, M. J. Leonardi, Y. Huang, Z. Lu, T. J. Marks, A. Facchetti, Chem. Mater. 2018, 30, 3323.

[129] J. W. Hennek, J. Smith, A. Yan, M.-G. Kim, W. Zhao, V. P. Dravid, A. Facchetti, T. J. Marks, J. Am. Chem. Soc. 2013, 135, 10729.

[130] Y.-G. Ha, J. D. Emery, M. J. Bedzyk, H. Usta, A. Facchetti, T. J. Marks, J. Am. Chem. Soc. 2011, 133, 10239.

[131] X. Yu, L.i Zeng, N. Zhou, P. Guo, F. Shi, D. B. Buchholz, Q. Ma, J. Yu, V. P. Dravid, R. P. H. Chang, M. Bedzyk, T. J. Marks, A. Facchetti, Adv. Mater. 2015, 27, 2390.

[132] J.-W. Choi, S.-Y. Han, M.-C. Nguyen, A. H.-T. Nguyen, J. Y. Kim, S. Choi, J. Cheon, H. Ji, R. Choi, IEEE Electron Device Lett. 2017, 38, 1259.

[133] W.-S. Choi, Y. Kim, Mol. Cryst. Liq. Cryst. 2019, 679, 111.[134] P. Gao, L. Lan, P. Xiao, Z. Lin, S. Sun, Y. Li, W. Song, E. Song,

P. Zhang, D. Luo, M. Xu, J. Peng, J. Soc. Inf. Disp. 2016, 24, 211.[135] P. Erri, P. Pranda, A. Varma, Ind. Eng. Chem. Res. 2004, 43, 3092.[136] D. Sanchez-Rodriguez, J. Farjas, P. Roura, S. Ricart, N. Mestres,

X. Obradors, T. Puig, J. Phys. Chem. C 2013, 117, 20133.[137] W. Marchal, C. De Dobbelaere, J. Kesters, G. Bonneux,

J. Vandenbergh, H. Damm, T. Junkers, W. Maes, J. D’haen, M. K. Van Bael, A. Hardy, RSC Adv. 2015, 5, 91349.

Pavan Pujar received his Ph.D. from the National Institute of Technology Karnataka Surathkal, India. During Ph.D., he jointly worked at the Indian Institute of Technology Bombay, Mumbai, India. Currently, he is working as a postdoctoral researcher at the School of Advanced Materials Science and Engineering, Sungkyunkwan University, South Korea. His research interests include the solution processing of functional metal oxides and metallic thin films, 2D materials, and their applications in transistors and integrated circuits.



© 2020 Wiley-VCH GmbH2000464 (25 of 25)


Srinivas Gandla received the M.Sc. degree in applied electronics from the Department of Physics, Osmania University, Hyderabad, India, in 2009, and the Ph.D. degree in material sci-ence engineering from the Department of Metallurgical Engineering and Materials Science, Indian Institute of Technology Bombay, Mumbai, India, in 2017. He is currently a postdoc-toral researcher with the School of Advanced Materials Science and Engineering Department, Sungkyunkwan University, Suwon, South Korea. His research interests include thin-film transis-tors, flexible and stretchable electronic devices, and sensors.

Dipti Gupta is working as an associate professor at Metallurgical Engineering and Materials Science (MEMS) Department, Indian Institute of Technology Bombay (IITB), India. She received her bachelor’s, master’s, and Ph.D. degree in Materials and Metallurgical Engineering from IIT Kanpur in 2000, 2002, and 2007, respectively. Later she worked at Korea Advanced Institute of Science and Technology (KAIST), Korea, Imperial College, London, and Seoul National University. Her research interests are in the area of flexible and stretchable electronics, wearable sensors for application in healthcare, and energy applications.

Sunkook Kim received the Ph.D. degree in Electrical and Computer Engineering from the Department of Electrical and Computer Engineering, Purdue University in Indiana, Indianapolis, IN, USA, in 2009. Since 2017, he has been an associate professor in Advanced Materials Science and Engineering with the School of Advanced Materials Science & Engineering, Sungkunkwan University, Suwon, South Korea. His research interests include augmented human electronics, next-generation high-mobility devices, and nanomaterials.

Myung-Gil Kim is an associate professor in the School of Advanced Materials Science and Engineering at Sungkyunkwan University (SKKU). He received his B.S. in Chemistry in 2006 from Korea Advanced Institute of Science and Technology (KAIST), and he obtained his Ph.D. in Chemistry from Northwestern University in 2012. His current research interests include the development of solution-based synthesis of inorganic materials and hybrid materials for thin-film transistors, sensor, secondary battery, and environmental remediation applications.


Trends in Low‐Temperature Combustion Derived Thin Films for...

Documents

Transcript of Trends in Low‐Temperature Combustion Derived Thin Films for...