Recovery of Sulfur from Flue Gases Using a Copper Oxide Absorbent

Performance of silica-supported copper oxide sorbents for SOx/NOx-removal from flue gas I. Sulphur dioxide absorption and regeneration kinetics

“…In Fig 2, the weight gain during sulphatlon under standard test con&tlons 1s plotted versus the reaction time, for SIX of the sorbents tested If curves of slmllarly prepared sorbents are compared, a sorbent mth a larger copper content shows a larger weight gain after a certam reaction time, 1 e it has a higher sulphatlon activity This 1s m contrast with the results of McCrea et al [4], who reported a decrease of the sulphatlon activity for copper contents larger than 4-6 wt -% The dfferent results are probably explamed by hfferences m internal pore structure, but unfortunately McCrea et al &d not specify the pore structure of then sorbents Differences m sulphatlon actlvlty per unit copper content are represented by the slopes of the curves plotted m Fig 3 Agam, sorbents prepared by the same technique but mth a &fferent copper content can be compared. For each type, whether prepared by ion-exchange (np ), homogeneous deposltlon-preclpltatlon (pr) or vacuum Impregnation (lm), it shows that the sulphatlon actlvlty per unit copper content 1s lower If the copper content 1s larger This IS probably due to a larger size of the CuO deposits for the sorbents with a larger copper content An Increase of the deposit size with an mcreasmg copper content was confirmed by Van der Graft et al [ 131 for the pr-sorbents By usmg transmlsslon electron microscopy they determined a copper deposit size m the reduced sorbent of 2 3-2 7 nm and 3 5-5 0 nm for the pr (4 7) and pr (17 4) sorbent, respectively As expected, the lm-sorbents with large CuO deposits show a relatively low sulphatlon activity per unit copper content, while this actlvlty 1s relatively high for the pr-sorbents with the CuO dispersed very well on a large internal surface area…”

“…At higher temperatures sulphatlon remains mcomplete due to the reverse reaction and to partial decomposition of the sulphate formed to CuO and SO, McCrea et al [4] reported thus to start at approximately 450" C for CuO supported on y-alumina Above 7OO"C, no copper sulphate was formed at all They also measured that the y-alumina support itself was sulphated to approximately l-2 wt -% Sulphation of the y-alumma support was also mentioned by Prmclotta et al [ 181 Several papers report on kinetic studies of the sulphation reaction Yeh et al [ 191 as well as Vogel et aI [ 201 found the mtrmsrc reaction rate of the sulphatlon of CuO supported on y-alumina to be &rectly proportional to the SOz concentratron m the gas phase and to the fraction of unreacted CuO In case of flue gas comhtlons the sulphatron rate is generally measured to be mdependent of the oxygen concentration because rn flue gas there 1s a large excess of oxygen when compared to sulphur droxlde Laguerle and Barreteau [21] studed SO, absorption on an alumma-supported CuO sorbent in a counterflow multrstage fluuhsed bed reactor They found no influence of the NO, concentration (between 0 and 150 ppm) while the presence of ll-17% carbon &oxide led to a shght decrease m SO, absorption degree compared to expenments without any carbon &oxide m the aimulated flue gas Vanatlon of the water vapour concentratron between 2 and 20% was determined to have a considerable effect on the SOz absorptron degree A maximum in the SO:! absorption degree occurred between 6 and 10% for SOz inlet concentratrons of 2000-4000 ppm and a bed temperature of approxrmately 300" C For an SO2 inlet concentratron of 1600 ppm the maximum was absent, the SOz absorption degree decreased contmuously with mcreaslng water vapour content of the simulated flue gas Data on the Influence of temperature on the sulphatron krnetms are rather scarce Deberry and Sladek [ 161 reported an actrvatlon energy of 112 kJ/mol for pure CuO, determined from microbalance experiments.…”

“…In previous stuckes on CuO processes and related kinetics, predominantly impregnated CuO on y-alumina sorbents were tested with a Cu weight fraction of 510% [4,14,15] The question of the optimum copper content was discussed by McCrea et al [ 41. They reported for a y-alumina support that more than 4-6 wt -% of copper led to a decrease of the SO, absorption rate probably due to a decreased porosity of the sorbent.…”

“…A CuO process with a fluuksed bed absorber has been developed at the Pittsburgh Energy Technology Center (PETC ) to enable contmuous operation mstead of the swmg operation of the SFGT process [ 2,4,5] However, the pressure drop for the flue gas is higher when compared to the parallel passage reactor and the allowable superficial flue gas velocity is considerably lower [6,7] We are studying the apphcation of a relatively new contactor, the gas-solid trickle flow reactor [ 81. In this reactor a ddute flow of solid particles is contacted counter-currently with the gas phase over a regularly stacked packing Favourable properties of the gas-solid trickle flow reactor are* (1) a low pressure drop, (u ) hmited axial dispersion m the gas and solids phase, (m) excellent heat and mass transfer between both phases, and (iv) counter-current operation Therefore, a gas-sohd tnckle flow reactor IS expected to be an efficlent absorber m a CuO process for the simultaneous removal of SO, and NO, from flue gases, m whmh the advantage of the contmuous operation of the flunhsed-bed process is combined with the low pressure drop and the high flue gas velocity of the SFGT process…”

Adsorption and desorption of sulfur dioxide on novel adsorbents for flue gas desulfurization. Final report, September 1, 1993--August 31, 1994

“…The operating cost of adsorption FGD process can be substantially reduced if the attrition loss of the adsorbent is mmimmd and sulphation and regeneration are carried out at same temperature 1300-450 OC) ( McCrea et al, 1970).…”