Three inorganic ion exchange sorbents were evaluated for Cs-137 removal from the Idaho Nuclear Technology and Engineering Center (INTEC), acidic tank waste. A commercially available, Russian manufactured, potassium copper hexacyanoferrate (FS-2), crystalline silicotitanate (IONSIV IE-911), manufactured by UOP, and ammonium molybdophosphate-polyacrilonitrile composite sorbent (AMP-PAN), produced at the Czech Technical University, Prague, were tested. Approximately 800 mLs of actual radioactive tank waste were used to evaluate the FS-2 and IONSIV IE-911 sorbents. Feed flowrates of 6-7 mLs/hr through a 1 cm(3) column were used for both tests. Initial Cs-137 breakthrough with IONSIV IE-911 was observed at approximately 100 mLs and 50% breakthrough was observed at 660 mLs. A Cs-137 breakthrough of 20% was observed after 775 mLs of waste was processed through the column containing FS-2. Tests were also performed with both of these sorbents using simulated tank waste. Results for the IONSIV IE-911 compare well for both tests; however, there are some slight discrepancies with the FS-2 results. These discrepancies are believed to be the result of mercury sorption arising from different mercury speciation between the actual and simulated waste feeds. Over 80 wt% of the cesium loaded onto the FS-2 sorbent was eluted with 8 M HNO3, while no attempt was made to elute cesium from the IONSIV IE-911 sorbent. The AMP-PAN test used a 1.5 cm(3) column, 1600 mLs of actual radioactive waste, and a flowrate of 40 mLs/hr (26 bed volumes/hr). Initial Cs-137 breakthrough was observed after processing 600 mLs, and 0.15% breakthrough was observed after processing 1550 mLs of waste. Over 83% of the cesium was eluted using 44 mLs of 5 M NH4NO3. Two loading and elution cycles were performed with simulated waste and AMP-PAN. Approximately 4800 mLs (3200 BV) of simulant were processed before 50% breakthrough was observed. Over 70% of the cesium was eluted in 85 mL (57 BV). Only 4050 mLs (2700 BV) of simulant were processed before 50% breakthrough was observed in the second loading cycle and 50% of the cesium was eluted in 117 mL (78 BV). The results indicate that AMP-PAN has the highest capacity for cesium and sorption appears to not be affected by the presence of mercury in the waste, giving it clear advantages over IONSIV IE-911 and the FS-2 sorbent. The AMP may also be easily dissolved from the PAN binder using sodium hydroxide.