Enhanced mass-transfer rates in nanofluids have been demonstrated in recent studies. A semiempirical theory for predicting mass-transfer rates in nanofluids was proposed in our earlier work, based on data in model apparatuses. In the present work, the applicability of this theory to a more complex hydrodynamic environment, that of a packed bed, is studied. Experiments that have been performed with silica nanofluids on the absorption of carbon dioxide with and without chemical reaction have been conducted in a laboratory packed bed. Enhancements are generally larger in physical absorption than in absorption with reaction, as expected by the theory. The average error in the prediction of packed bed data using the theory is on the order of 21%, which suggests that the theory can form a basis for design. A phase plot has been presented to identify regions in the parametric space for significant enhancement.