RESEARCH COMMUNICATIONS CURRENT SCIENCE, VOL. 112, NO. 7, 10 APRIL 2017 1561 *For correspondence. (e-mail: viren@aero.iitb.ac.in) Ignition delay study of aluminium oxide liquid nano-fuel in a shock tube D. K. Tripathi 1 , G. Garg 2 , U. Agrawal 2 , V. Menezes 1, *, U. V. Bhandarkar 2 and B. P. Puranik 2 1 Department of Aerospace Engineering, and 2 Department of Mechanical Engineering, Indian Institute of Technology Bombay, Powai, Mumbai 400 076, India The ignition delay of aluminium oxide (Al 2 O 3 ) liquid nano-fuel was compared with that of base-fuel to study the feasibility of its use for high-speed aerospace applications. The base-fuel was aviation turbine fuel that was mixed with Al 2 O 3 nanoparticles to produce a nano-fuel which could be used for regenerative cool- ing of the combustor walls before injection. The experiments were carried out in a shock tube. The fuel was introduced into the shock tube in the form of a wall droplet. The ignition delay time of the nano-fuel was observed to increase slightly, by about 11% (maximum) in comparison with the baseline, at an equivalence ratio of unity. Keywords: Aluminium oxide, ignition delay, liquid nano-fuel, shock tube. IGNITION delay sets the residence time of a fuel in the combustion chamber of a high-speed engine, and is a use- ful parameter in determining the capability of ignition and flame sustainability in a flow where a packet of air– fuel mixture resides for a few milliseconds 1 . Estimation of ignition delay of a fuel produced for high-speed opera- tions is essential to avoid fluctuations in energy release and specific impulse of the engine. Aviation turbine fuel (ATF) is a preferred hydrocarbon fuel owing to its high energy density and stable thermo- dynamic properties. We produced a fuel by mixing ATF with alumina (Al 2 O 3 ) nanoparticles for the dual purpose of regenerative cooling of combustor walls and preheat- ing the fuel before injection. Heat transfer studies have indicated an increase between 20% and 30% in the heat transfer coefficient with 0.3% volume fraction loading of Al 2 O 3 nanoparticles 2 . The nano-fuel has a probable appli- cation in high-speed engines, such as pulse detonation engines. The objective of the present study was to con- firm that the ignition delay of the admixture was within the acceptable limits. The experiments were carried out in a shock tube of variable inner diameter (i.d.). The inner diameter and length of the driver and driven sections of the tube were 51 mm, 37 mm and 2.5 m, 3 m respectively. Figure 1 shows a schematic of the experimental set-up. The test gas was air, which was maintained at an appropriate pres- sure for the test in the driven section of the tube. The driver and driven sections were separated by an alumin- ium diaphragm (1.2 mm thickness) that was exploded by pressurization of the driver section by purified nitrogen gas. The explosion launched an incident shock wave into the driven section that propagated through the test gas and reflected from the driven-end, enhancing the shock- ing effect. The propagation of the planar shock waves finally left behind a reservoir of high temperature and pressure in the driven section, which could simulate a combustor for the study of ignition characteristics of fuels. The liquid fuel to be tested was preloaded into the end of the driven section as a wall droplet, which was atomized on collision with the propagating shock waves. The propagating shock waves accomplished the function of a fuel injector in the present case 3 . The Mach number of the incident shock wave was modulated to obtain the desired test conditions. The shock tube consisted of pressure transducers/ gauges and a photodiode to measure the pressure and photonic output during the ignition tests. The pressure transducers were flush-mounted in the tube wall at the end of the driven section, whereas the photodiode was mounted on the external surface of the driven-end flange that had an acrylic optical window of 35 mm diameter to transmit light during ignition. The pressure transducers (PCB-Piezotronics, USA) had sensitivities and peak load- ing limits of 1 and 5 mV/psi, and 5000 and 1000 psi (models 102B and 102B04) respectively. The photodiode (model PDA36A-EC, THORLABS, USA) had an operat- ing wavelength bandwidth of 350–1100 nm and an adjustable gain of 0–70 dB. The signals from the sensors were acquired on a PC-based data acquisition system with NI PCI-6115 S data cards (National Instruments, USA), at a sampling rate of 1 MS/s. The images of fuel ignition were obtained using Phantom V 710 high-speed video camera, with sampling rate and spatial resolution of 64 kfps and 320  240 pixels respectively. Figure 2 presents the output of the tube-end pressure transducer (S 2 ) and photodiode for dry runs (without fuel). The first jump in the pressure marks the arrival of the incident shock wave at S 2 , and the next jump indicates the arrival of the reflected shock wave from the tube end. The output of the photodiode did not show any rise dur- ing the dry run. The Mach number of the incident shock wave (M S ) can be calculated using eq. (1), with the pres- sure jump across the incident shock wave (P 1 to P 2 ) and  (ratio of the specific heat capacities of test gas, i.e. air) 4 2 S 1 1 1 1. 2 P M P             (1) The temperature across the moving shock wave can be calculated using eq. (2), where T a and T b are the tempera- ture ahead and behind the shock wave, and P a and P b are