166Ho is a promising radionuclide for targeted radionuclide therapy, because of its beneficial decay characteristics and the possibility to track its dose delivery in a patient. However, a problem for the use of 166Ho in these treatments is its limited maximally produced specific activity in neutron irradiation of a pure 165Ho target, since the produced 166Ho is chemically inseparable from the 165Ho target. The Szilard-Chalmers method is known to be able to increase the achieved specific activity. It does so by irradiating the target isotope while it is incorporated in a compound. As the target isotope undergoes (n, γ) neutron capture it forms the product isotope and escapes the compound due to recoil from prompt gamma emissions. Then by separating the product isotopes from the compound, which still contains the target isotopes, a high specific activity can be achieved. However, batch irradiations using the Szilard-Chalmers method for production of high specific activity gives low production yields. In order to increase these production yields, a continuous flow Szilard-Chalmers system is proposed. The objective of this research is to investigate the main processes at play in a continuous flow system using the Szilard-Chalmers method on 165HoDOTAfor the production of high specific activity 166Ho. Investigations on the labelling of Ho3+ to DOTA4- have shown that 96±3 % labelling efficiency is achievable. Experiments on relabelling and isotope-exchange have shown that less than 8 % of 166Ho3+ undergoes relabelling with DOTA4- or isotope-exchange with 165HoDOTA-. Also, a cation exchange separation method for separation of Ho3+ from DOTA4- and HoDOTA- was developed. Moreover, investigations on Szilard-Chalmers separation of a dried 165HoDOTAsample have shown that with a 4-minute irradiation it was possible to achieve a specific activity of 2.8±0.2 MBq mg-1 . This corresponds to an enrichment factor of 1.62±0.03 compared to an irradiation without Szilard-Chalmers separation. Decreased radiolysis by shielding and slowed neutrons, and increased initial labelling are expected to be able to increase the achieved enrichment factor even more. The continuous flow Szilard-Chalmers system could in potential prove very useful in high specific activity radionuclide production, but further improvements on its subprocesses are necessary before practical implication is worthwhile.