Introduction

The reaction between polytetrafluoroethylene (PTFE, Teflon™) and aluminum has been widely investigated due to the importance of this system as an energetic additive of rocket fuels and pyrotechnics components [1, 2]. Nevertheless, effective ignition of aluminum is problematic due to strong oxide layer that naturally forms on the aluminum surface [36]. This issue becomes critical for the nanosized aluminum particles, where the oxide layer on aluminum particles makes a significant portion of the initial reactants [7]. The PTFE can potentially remove the oxide layer from the aluminum surface, thereby increasing the contact area between oxygen and aluminum. As a result the reaction rate increases, which expands the energy release of the system.

Several research groups have investigated the reaction behavior of alumina with PTFE [811]. However, the experiments were performed at low heating rates (<20 °C min−1). It should be noted that in nanoenergetic type of reactions heating rates are significantly higher (40,000 °C s−1) [12], which can alter the interaction mechanism between PTFE and Al2O3. It is known that self-heating within reactive systems can accelerate reactions exponentially and lead to an explosion [13]. No prior investigations have been reported in literature using high heating rates, comparable to those in systems with self-sustained reaction behavior (e.g., thermites type of reactions), for the PTFE–Al2O3 system.

In this work we demonstrate that the heating rate is crucial for the interaction behavior between PTFE and Al2O3. We found that below a critical heating rate (<150 °C min−1) endothermic decomposition of PTFE governs the interaction process, while at higher heating rates, similar to thermite reactions, the exothermic reaction between PTFE and Al2O3 dominates.

Experimental

The chemicals used for this study were PTFE (99.99 %, ~1 µm, Sigma-Aldrich), Al2O3 (99.99 %, ~150 μm, TA Instruments), and argon (99.999 %, Airgas Company). Differential scanning calorimetry (DSC) (Q-600, TA Instruments) was used to study PTFE–Al2O3 reactions with heating rates in the range of 20–200 °C min−1. The instrument has a mass resolution of 0.1 μg, which provides the ability to work with very small amounts of reactants (~20 mg). We used vented cups with an argon gas flow rate of 100 mL min−1. The experiments at higher heating rates were performed using a high-speed temperature scanner (HSTS-1). Heating rates of samples can be varied from 10 to 10,000 °C min−1 with the temperature range of scanning, from room temperature up to 1,300 °C. Experiments can be stopped automatically at a given time or temperature with subsequent fast cooling of samples (up to 200 °C s−1). The instrument can be operated in vacuum, air, and inert gas at P ≤ 1 atm (static conditions). Custom Microsoft-based software with graphical interface operates the HSTS-1 apparatus. In our experiments we used heating rates up to 780 °C min−1. Samples (20–50 mg) were placed in the middle of Pt-foil (thickness of 50 µm) and heated by passing electrical current through the foil. Thermocouples were positioned to provide non-inertial temperature measurements. The heating rate of the sample was controlled by the reaction enthalpy and applied power. Temperature history and applied power were continuously recorded during the heating process.

A Nicolet iS5 Fourier transform infrared (FTIR) spectrometer (Thermo Scientific) with an ID-5 accessory was used to perform FTIR spectroscopy on a small quantity (~5 mg) of powder that was pressed to receive a flat surface. The measurement was performed in the wavenumber range of 4,000–400 cm−1. The composition and crystal structure of the powder was determined by X-ray diffraction using a Siemens D5000 diffractometer with Cu Kα radiation (λ = 1.54056 Å). The scans were taken at room temperature over a wide range of angles, 2θ = (20–80°), at 0.05° intervals.

Results

Thermodynamic analysis

For the estimation and control of the maximum reaction temperature and product composition, knowledge of the adiabatic reaction temperature and produced condensed phases are needed [14]. Thermodynamic predictions of the adiabatic temperature and equilibrium composition of multicomponent multiphase systems requires minimization of the Gibbs free energy (G) subject to mass and energy balances [15]. For a system with N (g) gas and N (s) solid components, at constant pressure, the free energy of the system can be expressed as

$$ F \left(\left\{ {n_{\text{k}} } \right\}, \left\{ {n_{\text{s}} } \right\}\right) = \sum\limits_{k = 1}^{{N^{({\rm g})} }} {n_{\text{k}} \left( {k_{\rm B} T\ln\frac{{p_{\text{k}} }}{p} + G_{\text{k}} } \right)} + \sum\limits_{l = 1}^{{N^{({\rm s})} }} {n_{\text{l}} G_{\text{l}} } , $$
(1)

where p k is the partial pressure of the kth gas-phase component, while n l and G l are the number of moles and molar Gibbs free energy of components (l). The total energy balance determines the adiabatic combustion temperature, \(T_{\rm c}^{\rm ad}:\)

$$ \sum\limits_{i = 1}^{{N_{0} }} {H_{\text{i}} \left( {T_{0} } \right)} = \sum\limits_{k = 1}^{{N_{({\rm g})} }} {n_{\text{k}} H_{\text{k}} \left( {T_{\rm c}^{\rm ad} } \right)} + \sum\limits_{l = 1}^{{N_{({\rm s})} }} {n_{\text{l}} H_{\text{l}} \left( {T_{\rm c}^{\rm ad} } \right)} , $$
(2)

where the enthalpy of each component is

$$ H_{i}( T) = \Updelta H_{{\rm f},{\rm i}}^{0} + \int\nolimits_{{T_{0} }}^{T} {c_{{\rm p},{\rm i}} {\rm d}T} + \mathop \sum \nolimits \Updelta H_{{\rm s},{\rm i}}. $$
(3)

\( \Updelta H_{{\rm f},{\rm i}}^{0} \) is the heat of formation at 1 atm and reference temperature T 0, c p,i is the heat capacity, and \( \Updelta H_{{\rm s},{\rm i}} \) is the heat of sth phase transition for component i.

The calculations were made using thermodynamic software THERMO [16], which is based on the minimization of the Gibbs free energy and includes a database with the thermochemical properties of approximately 3,000 compounds. THERMO allows the equilibrium product composition to be determined for the given temperature and pressure as well as the maximum adiabatic front temperature and combustion product composition for the given pressure. Additionally, we used the thermo-chemical computer code HSC Chemistry-7, which includes a database of over 25,000 compounds, to introduce thermodynamic data for a number of organic compounds absent in THERMO, as well as to calculate the enthalpy of the reaction between PTFE and Al2O3. We studied the possibility of the reaction between PTFE and Al2O3, since Al particles are always coated with an Al2O3 layer. The PTFE interaction with the alumina layer is the first reaction in PTFE–Al system and is expressed in the following equation:

$$ 1. 5/{n}\left( {{\text{C}}_{ 2} {\text{F}}_{ 4} } \right)_{\text{n}} + {\rm Al}_{ 2} {\rm O}_{ 3} \to 2{\rm AlF}_{ 3} + 3{\rm CO}, \quad \Updelta {H} = - 1.0\,{\rm kJ}\,{\rm g^{-1}} . $$
(4)

Thermodynamic analysis revealed the exothermic character of reaction (4). By the enthalpy of reaction for PTFE in solid state [17] equal to 817.5 kJ mol−1, the adiabatic temperature was estimated to be T ad = 1,425 K. In that case, the enthalpy of the reaction was calculated to be −103 kJ mol−1 (−1 kJ g−1).

Figure 1 represents the dependence of the adiabatic temperature and the equilibrium concentration of condensed and gaseous phases on the molar ratio of PTFE/Al2O3. As the molar ratio of PTFE/Al2O3 increases, the maximum adiabatic temperature correspondingly increases reaching the maximum at the stoichiometric ratio of 1.5. At this ratio Al2O3 is completely converted to gaseous and solid AlF3, with the release of carbon monoxide (CO) gas. When the molar ratio of PTFE/Al2O3 is higher than 1.5 the adiabatic temperature slowly decreases due to excess of the PTFE decomposing into gaseous CF4 and carbon (graphite), and partial sublimation of solid AlF3 to gaseous AlF3 (Fig. 1). Thus, thermodynamic calculations confirm the possibility of an exothermic interaction between PTFE and alumina.

Fig. 1
figure 1

Dependence of the maximum adiabatic temperature and equilibrium concentration of condensed and gaseous phases on molar ratio of PTFE/Al2O3 during the reaction between PTFE and Al2O3

DSC of PTFE–Al2O3

Thermolysis of PTFE is an endothermic process associated with a complex radical decomposition mechanism [1820]. In order to reveal the impact of heating rate on the reaction mechanism between PTFE–Al2O3, thermo-calorimetric analysis was performed using four different heating rates: 20, 80, 150, and 160 °C min−1 in flowing argon (100 mL min−1). The DTA/TG curves of the PTFE–Al2O3 mixture at 20 °C min−1 heating rate are plotted in Fig. 2a. The DTA result presents a clear endothermic peak at 320 °C, which represents the melting of PTFE with 27.7 J g−1 of latent heat. The absence of mass change indicates that there is no reaction between melted PTFE and alumina at this temperature. At 500 °C the decomposition of PTFE starts with an endothermic peak utilizing 289 J g−1 with associated 59.4 % mass loss. According to reaction (4), 33.3 % mass loss should be expected if PTFE is reacting with Al2O3, and 59.5 % mass loss should be recorded for complete decomposition and elimination of PTFE from the reactive system. Thus, 59.4 % mass loss indicates complete decomposition of PTFE and removal from the reaction zone as gaseous C2F (g)4 and CF (g)4 .

Fig. 2
figure 2

DTA–TG curves for the PTFE–Al2O3 system using heating rates of a 20, b 80, c 150, and d 160 °C min−1 under an argon atmosphere. The additional lines are showing the changes in heat flow and mass percentage values

There is no reaction between Al2O3 and PTFE at the heating rate of 20 °C min−1. This situation is similar to the case of pure PTFE decomposition [19] or when the reactive additives do not change the kinetics of decomposition [20]. The slight exothermic trend found initially on the DSC curve is due to cross-linking, depolymerization, and cyclization processes which are quickly covered by the next deep endotherm of PTFE decomposition. We did not observe any significant mass loss prior to decomposition of PTFE. The DSC data show that decomposition of PTFE occurs at about 500–600 °C for heating rate of 20 °C min−1, which is the same as reported in previous studies [11].

DTA/TG curves are shown in Fig. 2b for the 80 °C min−1 heating rate. At 320 °C PTFE is melting. Mass loss of 53 % in the TG curve demonstrates that a small amount of PTFE residues remain in the system and react with Al2O3. The DTA curve also shows a broad exothermic peak corresponding to PTFE decomposition and reaction with alumina. The primary decomposition endotherm observed in Fig. 2b dominates the exothermic behavior of the reaction. At a heating rate of 150 °C min−1 the DTA curve, Fig. 2c, shows a well-developed exothermic peak with a maximum of 658 °C. PTFE melting is unnoticeable at heating rate of 150 °C min−1. The mass loss of the system was about 44 %, which indicates that more PTFE residues react with Al2O3 in comparison with lower heating rates. The broad exothermic peak almost covers the PTFE decomposition endotherm (a tiny curve at the top of the peak).

At the higher heating rates, 160 °C min−1 (Fig. 2d), the PTFE reacts with Al2O3 generating a single exothermic peak with 1,620 J g−1 of heat released. The total mass loss was about 38 %, indicating that a significant portion of PTFE remains in the system and reacts with Al2O3. Therefore, it is evident that the heating rate affects the interaction behavior in PTFE–Al2O3 mixtures converting interactions from endothermic (decomposition of PTFE) to exothermic (PTFE–Al2O3 reaction) modes.

Activation energy of PTFE–Al2O3

We estimated the activation energy of the PTFE–Al2O3 reaction from DSC data using the isoconversional method suggested by Starink [21, 22]. Reference [23] suggests the isoconversional method provides a more accurate activation energy value than the Kissinger and Ozawa methods. The Starink method determines the activation energy by the following equation:

$$ ln\left\{ {\frac{{T^{1.8} }}{\beta }} \right\} = \left( {1.007 - 1.2 \times 10^{ - 5} E_{\rm a} } \right)\frac{{E_{\rm a} }}{RT} + {\rm const}, $$
(5)

where E a is the apparent activation energy (in kJ mol−1), β is the heating rate used in thermal analysis, TA (in K min−1), T is the peak temperature of the exothermic curve (in K), and R is the universal gas constant. E a is estimated from the slope of the graph of ln(T 1.8/β) versus 1/T.

The activation energy of PTFE/Al2O3 reactions were estimated by performing two set of DSC experiments using heating rates of 20, 80, 100, and 150 °C min−1 and 160, 170, 180, and 200 °C min−1. At the lower heating rates the dominating reaction in the system of PTFE–Al2O3 was the decomposition of PTFE, while at higher heating rates the dominating reaction was the exothermic interaction between PTFE and Al2O3. The activation energy for the PTFE–Al2O3 system was estimated to be 265 kJ mol−1 at low heating rates, which is more accurately interpreted as the decomposition energy of PTFE (Fig. 3a). This value is in good agreement with the value reported in [17] of 275 kJ mol−1 for pure PTFE at lower heating rates. At higher heating rates 160, 170, 180, and 200 °C min−1, the activation energy was estimated to be 21 kJ mol−1 (Fig. 3b). This indicates that there is a critical heating rate regime, where the interaction mechanism changes from PTFE decomposition to reactions between Al2O3 and PTFE. It is possible, that Al2O3 behaves similar to silicon containing reducers, which act as a catalyst to accelerate the decomposition of PTFE [20].

Fig. 3
figure 3

Arrhenius plot for a endothermic reaction modes at heating rates <150 °C min−1 and b exothermic peaks of the DSC curves for the system PTFE–Al2O3, at heating rates >150 °C min−1

Study at high heating rates

Figure 4 shows the temperature–time history during the reaction between PTFE and Al2O3 at a heating rate of 780 °C min−1, performed using a HSTS-1 instrument. There is an endotherm in PTFE curve at approximately 360 °C, which corresponds to the melting of PTFE. PTFE decomposition takes place at about 640 °C. However, for the PTFE–Al2O3 reaction, only a single exothermic reaction between Al2O3 and PTFE is observed at around 800 °C. There is no endotherm indicating the decomposition of PTFE. As it can be seen from Fig. 4, the interaction between Al2O3 and PTFE at a heating rate of 780 °C min−1 is highly exothermic with a significant temperature rise on the temperature curve. Thus, the single exothermic peak was typical at heating rates higher than 160 °C min−1.

Fig. 4
figure 4

The HSTS curves for PTFE–Al2O3, and PTFE at a heating rate of 780 °C min−1

FTIR and X-ray analysis

FTIR analyses of the initial PTFE/Al2O3 reaction mixture and the reaction products received at the heating rate of 150 °C min−1 are presented in Fig. 5. For the initial mixture two characteristic peaks for PTFE were identified at wavenumbers 1,204 and 1,143 cm−1. No peaks were observed at 2,500–3,500 cm−1, where the absorption of hydroxyl groups are expected. For the final product we identified peaks for AlF3, which is consistent with thermodynamic predictions. FTIR analyses confirm that no hydroxyl groups are present on the alumina surface and, therefore, no substitution of hydroxyl groups by fluorination reactions can be expected. Therefore, the only interaction is between PTFE and alumina.

Fig. 5
figure 5

FTIR spectra for (a) initial mixture of PTFE–Al2O3 and (b) reaction products

X-ray analysis of the PTFE–Al2O3 reaction product at 780 °C min−1 heating rate is shown in Fig. 6. Characteristic peaks of AlF3 and Al2O3 are identified along with traces of carbon. Low angle XRD patterns of the powder contained an “amorphous hump” indicating that some amorphous AlF3 and pure carbon are also formed. The presence of Al2O3 peaks shows that for the stoichiometric mixtures of PTFE–Al2O3 some PTFE escaped from the reaction zone without reacting with Al2O3 particles. This is consistent with DSC analysis, where some PTFE residues evaporate from the reaction chamber without interacting with Al2O3.

Fig. 6
figure 6

XRD analysis for the PTFE–Al2O3 system at a heating rate of 780 °C min−1

Discussion

Aluminum oxide–PTFE thermal interactions are sensitive to heating rates. Although thermodynamic calculations show that the reaction has negative enthalpy and should be able to proceed in a self-sustaining mode, the exothermicity is not sufficient to support a self-sustaining reaction. Experimental evidence indicates that this is due to heat loss and rapid evaporation of PTFE decomposition products from the reaction zone. Prior to the thermal decomposition of PTFE, the solid–solid or liquid–solid interaction between PTFE and Al2O3 was not observed. Thus, at the low heating rates (<20° min−1) the gaseous products of PTFE decomposition evaporate from reaction zone and do not react with alumina.

At higher heating rates, a greater amount of PTFE decomposition products remain in the reaction zone and interact with the alumina. DSC also shows that at higher heating rates the exothermic effect of the reaction increases. At a heating rate of 80 °C min−1 a slight exothermic interaction was observed, while for real nanoenergetic systems the heating rates are much higher. In fact, HSTS shows that at greater heating rates a highly exothermic reaction is observed in this system. Thus, this study indicates that PTFE can exothermically react with the outer Al2O3 layer on aluminum particles, which increases the reaction rate and improves the energy and gas discharge in PTFE–Al nanoenergetic systems.

Conclusions

Thermodynamic analysis of the reaction 1.5/n(C2F4) n  + Al2O3 → 2AlF3 + 3CO demonstrated exothermic behavior with an adiabatic temperature of T ad = 1,425 K. This temperature is not sufficient to generate a self-sustaining reaction in the PTFE–Al2O3 system under standard conditions. However, the reaction rate increases between the oxidizer and Al in nanoenergetic reactions, since Al2O3 on the surface of Al particles becomes chemically active. Our study confirms that the reaction behavior is sensitive to the heating rate. The endothermic reaction dominates at low heating rates with activation energy of 265 kJ mol−1, when PTFE decomposition is observed. An exothermic reaction in the PTFE–Al2O3 system is observed at heating rates higher than 150 °C min−1 with an activation energy of 21 kJ mol−1. In nanoenergetic reactions where the heating rates are much higher, an exothermic reaction between PTFE and Al2O3 is expected. The study indicates that under high heating rates, PTFE can potentially remove the oxide layer from aluminum particles and increase the direct contact area between oxygen and aluminum, which might increase the reaction velocity and improves the energy and gas discharge in the PTFE–Al2O3 nanoenergetic materials.