In this study, chemical-looping combustion was performed with highly viscous vacuum residue. A fuel reactor with a fuel-injection system for liquid fuels was designed and built for a chemical-looping reactor with the nominal fuel input of 10 kW(th). The gas velocities in the riser section and at the gas-distribution nozzles of this unit are comparable to those of industrial circulating fluidized-bed boilers. Reference experiments were performed with an ilmenite oxygen carrier and two different fuel blends that contained 40 wt.% and respectively 80 wt.% of vacuum residue in fuel oil 1. Fuel conversion was in line with that of experiments from an earlier campaign, where fuel oil 1 was used as fuel. The fuel contained a significant fraction of sulfur, but no SO2 was detected in the flue gas of the air reactor. More experiments were performed using an oxygen carrier based on calcium manganite. The oxygen carrier was exposed to fluidization at hot conditions (more than 600 degrees C) for about 95 h, out of which fuel was injected during a total of 9.6 h. Undiluted vacuum residue, fuel oil 1 as well as different blends of these two were used as fuel. Furthermore, the parameters fuel flow, fuel-reactor temperature and air flow in the air reactor were varied to observe trends in fuel conversion. The experiments were carried out with a fuel flow corresponding to 4.0-6.2 kW(th) and an oxygen carrier-to-fuel ratio of about 1300-2000 kg/mW(th) (fuel-reactor bed mass per thermal fuel-power). With undiluted vacuum residue as fuel and a fuel-reactor temperature of 1000 degrees C, up to 93% of all carbon leaving the fuel reactor was in the form of CO2. Carbon leakage from fuel reactor to air reactor was usually below 1% for all fuel types tested, but no SO2 was detected in the off-gas from the air reactor. The reactivity of the calcium manganite-based material decreased over the course of the experiments, which is likely due to sulfur poisoning. No defluidization or agglomeration problems were experienced over the course of the experimental campaign. (C) 2016 Elsevier B.V. All rights reserved.