Graphite oxide–iron oxide nanocomposites as a new class of catalyst for the thermal decomposition of ammonium perchlorate

Abstract

Graphite oxide (GO) is receiving increased attention due to its special surface properties and layered structure for the synthesis of GO containing nanocomposites. It is possible that integration of GO sheets and iron oxide nanoparticles may result in enhanced properties and enlarge the application range. Herein, we report the effect of Fe₂O₃–GO nanocomposite as a new class of catalyst for the decomposition of ammonium perchlorate (AP), a rocket propellant oxidizer, and study the effect of Fe₂O₃ : GO ratio on the catalytic activity. The material was characterized by X-ray diffraction and Raman spectroscopy and the formation of Fe₂O₃ and GO were confirmed. FESEM analysis showed that the Fe₂O₃ nanoparticles are highly dispersed between and on the graphene layers. With the addition of 3% of the composite with 1 : 1 Fe₂O₃–GO ratio, the decomposition temperature of AP was reduced by 45 °C, showing a high catalytic activity for the new composite. The high catalytic activity of the in situ synthesized Fe₂O₃–GO composite may be attributed to the uniform distribution of iron oxide nanoparticles which in turn provide a number of active sites on the surface due to the presence of GO.

---

1. Introduction

In the development of space vehicles and missiles composite solid propellants (CSP) are the major source of chemical energy. CSP are heterogeneous mixtures consisting of a large proportion of oxidizer, usually ammonium perchlorate (AP), a metallic fuel like aluminium powder and a fuel/binder, generally hydroxyl terminated polybutadiene. A specific feature of the thermal decomposition of AP is its extremely high sensitivity to the action of various additives, either entering the lattice (homophase) or forming their own phase (heterophase).^1,2 These additives can accelerate or decelerate the processes of low-temperature and high-temperature thermal decompositions, which is important for the storage of AP and mix compositions based on it. It has been assumed that there exists a correlation between the effect of the additives on the rate of thermal decomposition and their influence on the combustion rates of mix compositions incorporating AP as the major component. This has stimulated the attention of many research groups, especially those dealing with the development of solid propellants, to the problem.^3–5

Graphite oxide (GO) is receiving increased attention due to its special surface properties and layered structure for the synthesis of GO-containing nanocomposites. Specifically, owing to the introduction of oxygen-containing functional groups (hydroxyl, carboxyl and epoxy groups) on carbon nanosheets, GO became active and easily absorbs polar molecules and polymers. These materials often possess unusual properties as compared with their individual components, and can be used as precursors to produce conductive polymers, cathodic materials in lithium rechargeable batteries, and other functional nanomaterials like hybrid catalysts for the thermal decomposition of ammonium perchlorate, the major oxidizer used in composite solid propellants.^6–12

To broaden the horizons for the use of GO-based composites, an idea of using nanoparticles (NPs) anchored onto GO sheets was examined. It is known that GO can be exfoliated under appropriate treatment forming quasi-two-dimensional carbon nanosheets. These exfoliated GO sheets possess large surface area and thus may be potential support materials to load NPs. In addition, the aforementioned oxygenated functional groups have been utilized as nucleation centers to anchor NPs onto some support materials (e.g. carbon nanotubes), forming NP-containing composites.¹³ Therefore, it is feasible to synthesize graphite oxide-nanoparticle composite (GONP) by depositing the NPs onto GO sheets. Meanwhile, just like other materials, we hope that GONP can lead to a possible concerted effect or an integration of the properties of the two components (GO and NPs) in the new materials that will present special features for catalysis and nanotechnology.^14–16 Even though it has these special features, comparatively few investigations have been made with regard to GONP so far.¹⁷

Iron oxides and GO-based sheets have long been treated as promising materials for catalysis and energy storage such as in ultracapacitors due to their high electrical conductivity and surface area. Iron oxide (Fe₂O₃) is an important transition metal oxide that has been widely used as a catalyst in various industrial applications. Such a broad range of use has made the preparation of Fe₂O₃ nanostructures attract much attention.^18,19 However, when used Fe₂O₃ individually, it is prone to aggregation that will greatly limit their accessible surface area, which prevents the full realization of their high catalytic activity. This problem can be solved by co-assembling Fe₂O₃ to GO—nanoparticles and sheets—to form Fe₂O₃-intercalated GO stacks. It is possible that integration of GO sheets and Fe₂O₃ NPs may result in enhanced properties and enlarge the application range, especially in the field of catalysis.^20–23

Herein, we report a simple, efficient and scalable procedure for the synthesis of Fe₂O₃–GO sheets by in situ depositing Fe₂O₃ NPs between and on the surface of GO sheets from thermochemical oxidation of ferric nitrate. The assemblies of GO sheets are intercalated with iron oxides using solvothermal approach, and the structure is characterized by means of microscopy techniques (scanning electron microscopy). The chemical and physical properties of the assemblies are investigated by various analytical techniques. The catalytic activity of the composites was analyzed on the thermal decomposition of ammonium perchlorate by thermogravimetry analysis. More importantly, the optimum Fe₂O₃–GO ratio was investigated which may open up a whole new field for high energy application of carbon-based materials as support for catalytic systems.

2. Materials and methods

2.1 Chemicals

All chemicals used in this synthesis were of analytical grade and used without further purification. Analytical grade reagents viz., ferric nitrate trihydrate (Qualigens fine chemical, India) graphite flakes (Sigma-Aldrich, batch no. 282863), potassium permanganate, hydrochloric acid, sulphuric acid, hydrogen peroxide, barium chloride, (Merck, India) were used for the synthesis of catalysts. Ammonium perchlorate made in house with purity >99% was used for studying the catalytic activity.

2.2 Synthesis of graphite oxide

GO used in this work was prepared from purified natural graphite flakes according to modified Hummers method.^24,25 In a typical procedure, 10 g of graphite flakes was added to 230 ml of cooled (0 °C) concentrated H₂SO₄. 30 g of KMnO₄ powder was gradually added with constant stirring, keeping the temperature of the mixture below 20 °C. The mixture was then stirred at 35 °C for 30 min using a magnetic stirrer at 3000 rpm. Then 470 ml of distilled water was added slowly and the temperature was increased to 98 °C. The solution was maintained at that temperature for another 15 min. To the solution 1.4 l of distilled water was added followed by 10 ml of 30% H₂O₂ solution to terminate the reaction. The solid product obtained was separated by centrifugation, washed repeatedly with 5% HCl solution until it was free from sulphate (tested with BaCl₂). After complete removal of sulphate, the suspension was dried in a vacuum oven at 60 °C for 48 h to obtain GO.

2.3 Synthesis of Fe₂O₃–GO hybrids

A series of four Fe₂O₃–GO were prepared with Fe₂O₃ : GO weight percentage of 2 : 1, 1 : 1, 0.5 : 1, and 0.2 : 1. In a typical synthesis (Fe₂O₃ : GO, 1 : 1) 500 mg of GO and 1.25 mmol of ferric nitrate was added to 15 and 20 ml of ethanol solution and ultrasonicated for 5 and 30 minutes respectively. The solutions were then mixed and magnetically stirred for another 30 min. The resulting mixture was transferred and sealed into a 50 ml Teflon-lined stainless steel autoclave, heated to 140 °C in an electric oven for 24 h. After attaining room temperature, the autoclave was opened and product obtained was rinsed extensively with absolute alcohol. Finally, the product was dried in a vacuum oven at 60 °C for 6 h. The resulting Fe₂O₃–GO powder was then heated to 200 °C in nitrogen atmosphere and kept at that temperature for 2 minutes. The other composites were also prepared by the same procedure, by changing the concentration of the ferric nitrate. Products thus obtained with Fe₂O₃ : GO weight percentage of 2 : 1, 1 : 1, 0.5 : 1, and 0.2 : 1 are represented as FeGO1, FeGO2, FeGO3 and FeGO4 respectively. For comparison, Fe₂O₃ NPs without GO were prepared following the same procedure above without adding GO.

2.4 Characterization

Powder X-ray diffraction data of samples was collected on a Bruker D8-Discover powder X-ray diffractometer in a Bragg–Brentano configuration with a CuKα radiation (1.5418 Å) at a scan rate of 2.5 deg min⁻¹. Crystalline phases were identified by comparing the experimental diffraction patterns to Joint Committee on Powder Diffraction Standards (JCPDS). Raman spectra of samples were recorded using a Witec Alpha 300R confocal microscope using an excitation wavelength of 532 nm. Perkin-Elmer GX FTIR spectrometer was used for FTIR analysis. Field emission scanning electron microscopy (FESEM) observations were performed to examine the morphology of the samples using Carl Zeiss, Supra 55 model field emission scanning electron microscope equipped with an energy-dispersive X-ray spectrometer (EDX). Simultaneous thermogravimetry-differential scanning calorimetry (TG-DSC), TA Instruments Q600, was employed for thermal characterization. In all the thermal analysis experiments a sample mass of 5 ± 1 mg was used under a nitrogen atmosphere in platinum sample cups at a heating rate of 5 °C min⁻¹. For the evaluation of catalytic activity of synthesized Fe₂O₃–GO composites, fine AP (average particle, 50 μm) was homogeneously mixed with catalysts in the ratio 97 : 3 and subjected to thermal decomposition studies using TG-DSC techniques.

3. Results and discussion

GO, prepared by oxidation of graphite with strong oxidizing agents consists of distorted graphene layers bearing carbonyl, hydroxyl, and epoxy groups on the basal planes as well as carboxylic groups on the edges of the carbon sheets.^20,21 All of these factors make the material acidic as well as highly hydrophilic and explain its easy composite formation in alkaline or alcohol solutions. Mechanism of formation of GO from graphite was already reported and hence it is not worthy to mention again.²⁶ We are going to discuss the mechanism of formation of Fe₂O₃–GO system which is not reported elsewhere.

3.1 Driving force for the exfoliation of graphite and intercalation with Fe₂O₃

The intercalation process in GO is chemical as well as physical in nature.¹⁸ Even though it easily disperses in water due to the ionizable edge acid (–COOH) groups, a closer look at its structure reveals that it may actually be amphiphilic.²⁶ Intercalation in GO can occur by reactive adsorption of the guest species (iron oxide in this case), where attractive forces among the functional groups of GO and the guest species are essentially ionic.¹⁶ However, the intercalation of certain species, mainly polar, is thought to be controlled by hydrogen bonding.¹⁴ Thus, The intercalant occupies and thereby expands the interplanar spacing of GO along its c-axis. It is already reported that intercalation of solvents from liquid phase showed that common solvents such as alcohols and both chlorine and aromatic compounds readily penetrate into the GO interspacing.²⁷ Such molecules remained trapped in the GO structure after drying and caused irreversible changes in the GO interlayer distribution.

3.2 Raman spectra

Oxidations of graphite to produce GO introduces several chemical and structural changes in the parent graphite: increase in both the oxygen content and the interlayer distance and the formation of sp³-hybridized carbons which reduces the planar structure of the graphene layers. The extent of these changes depends not only on the synthetic methods but also on the graphite sources used. Therefore, our GO was first of all characterized by suitable and available techniques. Raman spectroscopy is a widely used tool for the characterization of carbon products, especially considering the fact that conjugated and double carbon–carbon bonds lead to high Raman intensities. Highly ordered graphite has only a couple of Raman-active bands visible in the spectra (Fig. 1), the in-phase vibration of the graphite lattice (G band) at 1575 cm⁻¹ as well as the (weak) disorder band caused by the graphite edges (D band) at approximately 1355 cm⁻¹.

Fig. 1 Raman spectra of graphite after oxidation and exfoliation processes. The position of D, G and 2D band in graphite and GO is represented by a vertical, dotted line.

Both the G and the D bands in graphite undergo significant changes during the transformation of graphite to GO. Specifically, the G band broadens significantly and shift to higher frequencies (blue-shift), and the D band grows in intensity. A notable fact is that the G band peak is located at a higher frequency in GO than that in graphite (1590 vs. 1575 cm⁻¹). Since in all measurements the same laser excitation frequency (532 nm) was used, we can exclude spurious shifts which are known to affect the D band of graphite. A literature survey points to several explanations for this blue shift, but the most plausible explanation is the presence of isolated double bonds produce another band, the D^I band, resonate at higher frequencies than the G band of graphite, located at 1620 cm⁻¹ which can partially merge with the G band.^28–30

From the spectra, it is evident that presence of iron oxide (confirmed by the peaks in the range 200–300 cm⁻¹) didn't produce any significant changes in the Raman spectra of the GO. D and G bands are well distinguishable and the 2D (even though it is weak) band is observable even in the presence of iron oxide. This gives an indication that the iron oxide particles were highly dispersed between and on the graphene layers (supported by SEM analysis discussed later in the paper).²⁰ As the ratio of iron oxide in the composite increases the peaks intensity in the range 200–300 cm⁻¹ increase.

3.3 IR spectra

FTIR spectra of pristine GO and Fe₂O₃–GO (Fig. 2) was recorded within the absorption range 400 to 4000 cm⁻¹. The following functional groups were identified in pristine GO and Fe₂O₃–GO systems: O–H stretching vibrations (3420 cm⁻¹), C–O stretching vibration (1720–1740 cm⁻¹), C–C vibrations from unoxidized sp² C–C bonds (1590–1620 cm⁻¹), and C–O vibrations (1250 cm⁻¹). The FTIR spectrum of GO (Fig. 2) shows a broad peak between 3000 and 3700 cm⁻¹ in the high-frequency area corresponding to the stretching vibration of OH groups of water molecules adsorbed on graphite oxide. This shows the strong hydrophilicity of the material. The presence of absorption peak in the medium frequency area, at 1630 cm⁻¹ can be attributed to the stretching vibration of C=C.

Fig. 2 FTIR spectra of GO and Fe₂O₃–GO composite.

The stretching vibration of carboxyl groups on the edges of the layer planes or conjugated carbonyl groups is observed at 1715 cm⁻¹. Finally, the absorption peaks at 1385 cm⁻¹ and 1110 cm⁻¹ correspond to the stretching vibration of C–O of carboxylic acid and C–OH of alcohol, respectively. Even though these peaks are observed, there is an additional peak at 2930 cm⁻¹ and 2850 cm⁻¹ (weak) which correspond to the symmetric and antisymmetric stretching vibrations of unoxidized CH₂, showing that the complete oxidation of graphite is not taking place during conversion. For the IR spectra of Fe₂O₃–GO composites, along with these peaks an additional peak at 530 cm⁻¹ is observed corresponds to the stretching vibration of Fe–O bond. It is interesting to note that there is no change in the absorption frequencies of GO even in the presence of Fe₂O₃ nanoparticles. It is also to be noticed here that the peak corresponding to Fe–O stretching is grown in intensity as the concentration of Fe is increasing.

3.4 XRD

A series of characteristic peaks were observed in the XRD pattern of Fe₂O₃–GO (Fig. 3b). The positions and relative intensities of the reflection peak of all composites agree with the diffraction peaks of standard α-Fe₂O₃, according to standard data (JCPDS 33-0664).^18,19

Fig. 3 XRD of graphite and GO (a), Fe₂O₃ and Fe₂O₃–GO composites (b).

Oxidation of graphene sheets results in an increase of the interlayer distance in GOs compared to graphite. The expansion of the lattice is considered as one of the main indicators for the purity of the material and the degree of its oxidation. The high oxidation state of the GO samples in this study is confirmed by analysis of XRD patterns recorded from pristine powders at ambient conditions (Fig. 3a). As expected, GO sample exhibited a significant increase of interlayer distance compared to graphite (deduced from the position of (001) reflection), up to 7.18 Å. These results indicate that the interlayer space of graphite became mostly occupied by oxygen-containing functional groups upon oxidation. The reduction of the average size of coherent diffraction domains in GO as compared to its parent graphite indicates that severe disruption of graphite layers took place during the oxidation process.^30,31 It should be noted that graphene oxide layers in GO are packed turbostratically and there is no ordering for various functional groups in the structure. Also, GO samples absorb some water from air at ambient humidity, and complete drying of samples seems to be very difficult. Therefore, the values of d-spacings cited above might reflect different abilities of samples to become hydrated by air at ambient humidity.

3.5 Electron microscopy and EDX

Field emission scanning electron microscopy (FESEM) observations were performed to examine the morphology of the samples. Fig. 4 shows the FESEM images of the pristine Fe₂O₃ and Fe₂O₃–GO composites. These figures (Fig. 4C–F) show that the graphite oxide layers are well separated due to thermal exfoliation and Fe₂O₃ particles are occupied between and on the surface these layers.

Fig. 4 (A and B) FESEM images of Fe₂O₃ nanoparticles (C–F) high-magnification SEM images of Fe₂O₃–GO nanocomposites. (C) FeGO1 (D) FeGO2 (E) FeGO3 (F) FeGO4.

From Fig. 4A and B it can be seen that the pristine Fe₂O₃ NPs were polydisperse and seriously aggregated. After an introduction to GO, the particles maintained their original shape with a good monodispersity. Fig. 4C shows that, in comparing with pure Fe₂O₃, the tendency for aggregation is reduced for the FeGO1 sample. When the ratio is changed to 1 : 1 (FeGO2), the tendency for agglomeration is again reduced and the number of intercalated Fe₂O₃ particles is increased (Fig. 4D). The average particle size of the pristine Fe₂O₃ NPs was about 17 nm. In the presence of GO, the particle size is reduced to 7 nm which shows the effect of GO support on the size of Fe₂O₃ NPs. It has been reported that during the thermal exfoliation, GO actually undergoes disproportionation reactions, in which some of the partially oxidised sp³ carbon atoms become fully oxidised to carbonaceous gases such as CO₂ and the rest reduced to sp².^32,33 The rapid evolution of gaseous species such as H₂O and CO₂ during the deoxygenation process fragmented the GO and results in a black plume of GO. Deoxygenation was triggered on the spot and then rapidly propagated across the entire film. The evolution of CO₂ and H₂O caused a large apparent volume expansion of ∼100–300 times, thus producing very low-bulk density.³² Fig. 4C–F shows the cross-sectional view of a GO film, showing much more loosely packed GO sheets after exfoliation. For the composite with other two combinations, since the Fe₂O₃–GO ratio is very less, a number of intercalated Fe₂O₃ particles is lesser than the other two samples. Hence, from these observations, it is concluded that composite with Fe₂O₃–GO ratio of 1 : 1 was the optimum concentration to obtain a well dispersed and intercalated Fe₂O₃ particles with lesser aggregation.

Fig. 5A shows the compositional analysis of the materials using EDX. The EDX technique can be used to determine the composition of a specimen as a whole as well as the composition of individual components. The areas under selected peaks can also be used to provide semi-quantitative elemental composition information. The compositional analysis results shown in Fig. 5A reveal that Fe, C, and O are the main elements present within the inspection field, with varying concentrations. From elemental mapping images shown in Fig. 5B it is confirmed that the Fe₂O₃ particle are well dispersed between and on the graphene layers.

Fig. 5 (A) EDX spectrum of Fe₂O₃–GO composites at different compositions. EDX spectra confirm the regular increase in Fe concentration. (B) Elemental mapping images of Fe₂O₃–GO composites. Red – oxygen, green – carbon and white – iron. The regular decrease in Fe concentration was confirmed from these mapping images.

3.6 Surface area

Surface area for all the samples was measured using BET method and is tabulated in Table 1. From Table 1, it can be seen that surface area of all samples shows a direct dependence on the amount of Fe₂O₃ used during the synthesis. Accordingly, the surface area shows a significant increase with an increase in the GO content. For pristine GO, the surface area was about 205 m² g⁻¹. On addition of 0.2% of Fe₂O₃ surface area is reduced to 145 m² g⁻¹. As the concentration of Fe₂O₃ is increased, the surface area is still reduced and finally reached a value of 90 m² g⁻¹ for the sample with Fe₂O₃–GO ratio of 2 : 1.

Table 1 Surface area of samples measured using BET method

Sample	Surface area (m^2 g^−1)
Fe2O3	45
FeGO1	90
FeGO2	100
FeGO3	114
FeGO4	145
GO	205

---

3.7 Catalytic activity on the thermal decomposition of AP

AP is the most common oxidant in composite solid propellants since its thermal decomposition characteristics directly influence the combustion behaviour of the propellant. More specifically, the activation energy, reaction rate and kinetics of thermal decomposition of AP are closely related to the performance of solid propellants. A lot has been said in the literature about the mechanism of decomposition of AP, but still it a matter of debate.^8,33 According to electron transfer mechanism of decomposition of AP, the rate determining step for the Low-Temperature Decomposition (LTD) is the transfer of an electron from ClO₄⁻ to NH₄⁺, while for High-Temperature Decomposition (HTD) rate determining step is the transformation of oxygen to superoxide. In the presence of burn rate modifiers, the rate of these LTD and HTD are accelerated and thus, the decomposition reaction is completed at low temperature. Unique properties of metal oxides like a large number of acid sites; ability to accept and release electrons makes them a good conductor of electrons thereby accelerating the decomposition reactions, especially redox reactions.

We have investigated the as-synthesized Fe₂O₃–GO composite as an additive to the thermal decomposition of AP, together with its individual component (Fe₂O₃) for comparison. Thermogravimetry-differential scanning calorimetry (TG-DSC) measurements were used to examine this catalytic process. The TG, DTG and DSC curves for pure AP and AP with catalysts are shown in Fig. 6 and 7. From Fig. 7 it can be seen that three peaks centered at 242, 293, and 360 °C are observed, corresponding to the endothermic phase transition, the exothermic low-temperature decomposition (LTD) and the exothermic high-temperature decomposition (HTD), in agreement with those reported earlier.^34,35 The endothermic transformation happens from the orthorhombic phase to the cubic phase (240–250 °C).

Fig. 6 TG and DTG curves for the thermal decomposition of AP with and without catalysts. High catalytic activity of the new hybrid composites was confirmed from these diagrams.

Fig. 7 DSC curves for the thermal decomposition of AP with and without catalysts.

On addition of 3 wt% catalysts, the two stage decomposition of pure AP becomes almost a single stage with a lowering of decomposition temperature by 40–50 °C. A closer look at DSC curves (Fig. 7) revealed that the LTD of AP is not much affected by the catalyst whereas HTD is actively catalyzed, reducing its initiation temperature. Decrease in temperature is more significant by the addition of FeGO2 compared to all other catalysts which is attributed to the well dispersed and intercalated Fe₂O₃ particles as revealed by the FESEM analysis.

The decomposition kinetics also supporting the high catalytic activity of FeGO2. Kinetic parameters via activation energy (E_a) and rate constants (k) for the thermal decomposition of pure AP and AP in presence of catalysts were calculated using isoconversional Kissinger–Akahira–Sunose (KAS) method which is based on the following equation.³⁶

where β is the heating rate, T_α is the temperature in kelvin corresponding to a fixed degree of conversion, α, A is the pre-exponential factor, R is the gas constant, E_α is the activation energy at a given degree of conversion and g(α) is the integral form of kinetic model function. E_α for a given degree of conversion is obtained from the slope of the linear fit of the plot versus 1/T_α. The average of E_α value is reported as activation energy (E_a) of the decomposition reactions. The rate constants (k) for the reactions were calculated using Arrhenius equation as shown in eqn 2.

TG analysis of pure AP, AP in presence of Fe₂O₃ and Fe₂O₃–GO (FeGO2) catalysts were carried out at heating rates of 2, 3 and 5 °C min⁻¹. And the derived kinetic parameters are tabulated in Table 2. From the table it can be see that pure AP has E_a and k value of 90 kJ mol⁻¹ and 5.5 × 10³ min⁻¹ respectively. Addition of Fe₂O₃ and FeGO2 reduce the E_a values to 60 and 55 kJ mol⁻¹ while k values increase to 8.2 and 9.8 × 10³ min⁻¹ respectively. This reflects that FeGO2 catalyst has superior activity than Fe₂O₃.

Table 2 Activation energy and rate constant for the thermal decomposition of pure AP, AP in presence of Fe₂O₃ and FeGO2 catalysts

Sample	Kinetic parameter
	Activation energy, Ea (kJ mol^−1)	Rate constant k, (min^−1)
Pure AP	90	5.5 × 10^3
AP + Fe2O3	60	8.2 × 10^3
AP + FeGO2	55	9.8 × 10^3

---

It can be seen from the above experimental results that Fe₂O₃–GO hybrids, especially FeGO2, significantly promote both the low-temperature and high-temperature exothermic process of AP (Fig. 6 and 7). It has been well addressed that GO exhibits a good conductivity and electrons in GO move a lot faster than they do among metal atoms. Due to its exceptional electronic characteristics, electrons can travel long distances in GO without being scattered.³⁷ Thus, it can be inferred that Fe₂O₃–GO hybrids used here could offer accelerated electron flow from ClO₄⁻ to NH₄⁺ and transforming to NH₃ and HClO₄. The O₂ formed from HClO₄ is transformed to superoxide ion (O₂⁻) at a faster speed due to the presence of GO as a perfect bed for the flow of electrons. The superoxide ions together with other products of HClO₄ could then help the decomposition of NH₃ and finally complete the decomposition of AP.

The ion current versus temperature curves of ion fragments during the thermal decomposition of AP, AP with pure Fe₂O₃ and AP with FeGO2 are shown in Fig. 8a–c. It was reported that during the LTD of AP, one of the decomposition product, ammonia, is not completely oxidised due to the prevailing low temperature. This ammonia will accumulate/adsorbed on the reactive surface of residual AP preventing it from further decomposition. However, as the temperature increases, adsorbed ammonia get oxidised, initiating the HTD. Most of the transition metal oxides catalyze the decomposition of perchlorate ion and oxidation of ammonia.^33,34 The TG-MS studies of decomposition of AP with and without catalysts showed that in the LTD of pure AP the concentration of oxidized products like H₂O and HCl are very low compared to that in HTD. However, the concentration of N₂O and O₂ in LTD and HTD are almost comparable indicating incomplete oxidation during LTD.

Fig. 8 (a) Ion current versus temperature curves of ion fragments during the thermal decomposition of AP. (b) Ion current versus temperature curves of ion fragments during the thermal decomposition of AP with pure Fe₂O₃. (c) Ion current versus temperature curves of ion fragments during the thermal decomposition of AP with FeGO2.

It can be seen that the decrease in decomposition temperature is more significant by the addition of FeGO2 compared to other composites and pure Fe₂O₃. For a heterogeneous catalytic reaction like AP decomposition, where the rate determining step is the oxidation of gaseous ammonia, a catalyst (FeGO2) with high surface area and active sites will perform better than one with the lower surface area by providing a large platform for catalytic oxidation. In addition, the presence of GO, a good conductor of electrons in the system will enhance the redox reaction taking place during the decomposition of AP by acting as better carrier/conductor of electrons.¹³ Evolution of more amounts of oxidized products viz. HCl, N₂O and H₂O during the early stages of decomposition confirms the high exothermicity of the reaction in the presence of FeGO2 as a catalyst.

4. Conclusion

Fe₂O₃–GO composites with different Fe₂O₃ : GO ratio and excellent catalytic properties have been synthesized successfully by solvothermal synthesis. Fe₂O₃ nanoparticles with a size of about 17 nm are homogeneously dispersed between and on the graphene layers that could effectively prevent them from aggregating. Raman spectroscopy and XRD studies confirm the formation of GO from graphite. FESEM analysis shows that the Fe₂O₃ nanoparticles are highly dispersed between and on the graphene layers. With the addition of 3% Fe₂O₃–GO composites the low-temperature decomposition and the high-temperature decomposition temperatures of AP were reduced. Catalytic activity was high for the FeGO2 composite with a Fe₂O₃ : GO ratio of 1 : 1. The high catalytic activity of the in situ synthesized FeGO2 composite may be attributed to uniform distribution of iron oxide nanoparticles which in turn provide a number of active sites on the surface due to the presence of GO.

Acknowledgements

The authors thank Director, VSSC and colleagues in Analytical and Spectroscopy Division, VSSC for their support.

References

1. V. V. Boldyrev, Thermochim. Acta, 2006, 443, 1–36.
2. P. W. M. Jacobs and M. R. Whitehead, Chem. Rev., 1969, 69, 551–590.
3. K. Kishore and M. R. Sunitha, AIAA J., 1979, 17(10), 1118–1125.
4. A. P. Sanoop, R. Rajeev and K. G. Benny, Thermochim. Acta, 2015, 606, 34–40.
5. G. P. Sutton, Rocket Propulsion Elements, Wiley, 8th edn, 2010.
6. C. Hu, T. Lu, F. Chen and R. Zhang, J. Chin. Adv. Mater. Soc., 2013, 1, 21–39.
7. S. Stankovich, D. A. Dikin, G. H. B. Dommett, K. M. Kohlhaas, E. J. Zimney, E. A. Stach, R. D. Piner, S. T. Nguyen and R. S. Ruoff, Nature, 2006, 442, 282–286.
8. M. Seredych and T. J. Bandosz, Carbon, 2007, 45, 2130–2132.
9. Z. H. Liu, Z. M. Wang, X. J. Yang and K. Ooi, Langmuir, 2002, 18, 4926–4932.
10. S. Stankovich, R. D. Piner, X. Chen, N. Wu, S. T. Nguyen and R. S. Ruoff, J. Mater. Chem., 2006, 16, 155–158.
11. X. Xu, I. H. Halidou, W. Hui and L. Feng, Chem. Mat.er, 2008, 20(3), 1173–1182.
12. X. Xu, B. Lu and L. Feng, AIChE J., 2010, 56(11), 2934–2945.
13. M. J. Allen, V. C. Tung and R. B. Kaner, Chem. Rev., 2010, 110, 132–145.
14. O. C. Compton and S. T. Nguyen, Small, 2010, 6, 711–723.
15. G. Eda and M. Chhowalla, Adv. Mater., 2010, 22, 2392–2415.
16. C. Burda, X. Chen, R. Narayanan and M. A. El-Sayed, Chem. Rev., 2005, 105, 1025–1102.
17. X. Huang, Z. Y. Yin, S. X. Wu, X. Y. Qi, Q. Y. He, Q. C. Zhang, Q. Y. Yan, F. Boey and H. Zhang, Small, 2011, 7, 1876–1902.
18. M. Mohapatra and S. Anand, J. Eng. Sci. Technol., 2010, 2(8), 127–146.
19. B. Lv, Y. Xu, D. Wu and Y. Sun, Particuology, 2008, 6, 334–339.
20. A. Javier, A. Orozco and T. J. Bandosz, J. Mater. Chem., 2015, 3, 220.
21. V. K. Singh, M. K. Patra, M. Manoth, G. S. Gowd, S. R. Vadera and N. Kumar, New Carbon Mater., 2009, 24(2), 147–152.
22. A. Dey, J. Athar, P. Varma, H. Prasanta and A. K. Sikder, RSC Adv., 2014, 5, 3.
23. H. Fei, Z. Peng, L. Li, Y. Yang, W. Lu, E. L. G. Samuel, X. Fan and J. M. Tour, Nano Res., 2014, 7(4), 502–510.
24. C. Zhang, W. Lv, X. Xie, D. Tang, C. Liu and Q. Yang, Carbon, 2013, 62, 11–24.
25. L. Shahriary and A. A. Athawale, J. Energy Environ. Eng., 2014, 21, 2348–2357.
26. D. Wang, R. Kou, D. Choi, Z. Yang, Z. Nie, J. Li, L. V. Saraf, D. Hu, J. Zhang, G. L. Graff, J. Liu, M. A. Pope and I. A. Aksay, ACS Nano, 2010, 4(3), 1587–1595.
27. K. Gotoh, T. Kinumoto, E. Fujii, A. Yamamoto, H. Hashimoto, T. Ohkubo, A. Itadani, Y. Kuroda and H. Ishida, Carbon, 2011, 49, 1118–1125.
28. D. C. Marcano, D. V. Kosynkin, J. M. Berlin, A. Sinitskii, Z. Sun, A. Slesarev, L. B. Alemany, W. Lu and J. M. Tour, ACS Nano, 2010, 4, 8.
29. K. N. Kudin, B. Ozbas, H. C. Schniep, R. K. Prudhomme, I. A. Aksay and R. Car, Nano Lett., 2008, 8, 1.
30. W. Gao, L. B. Alemany, L. Ci and P. M. Ajayan, Nat. Chem., 2009, 1(5), 403–408.
31. C. Xu, X. Wang, J. Zhu, X. Yang and L. Lu, J. Mater. Chem., 2008, 18, 5625–5629.
32. D. Krishnan, F. Kim, J. Luo, R. Cruz-Silva, L. J. Cote, J. H. Dong and J. Huang, Nano Today, 2012, 7, 137–152.
33. N. Li, Z. Geng, M. Cao, L. Ren, X. Zhao, B. Liu, Y. Tian and C. Hu, Carbon, 2013, 54, 124–132.
34. D. L. Reid, A. E. Russo, R. V. Carro, M. A. Stephens, A. R. LePage and T. C. Spalding, Nano Lett., 2007, 7, 2157–2161.
35. W. Rosser and S. H. Inami, Combust. Flame, 1968, 12, 427–435.
36. H. Tanaka, Thermochim. Acta, 1999, 267, 29–44.
37. M. D. Ayrat and M. T. James, ACS Nano, 2014, 8(3), 3060–3068.

---