随着“双碳”政策的不断推行，以及人们环保意识的提高，CO作为典型的大气污染物已经成为工业废气排放的主要监控对象。目前钢铁行业普遍采用的氧气顶吹转炉炼钢法会产生大量含有高浓度CO的转炉煤气。由于转炉炼钢周期的波动性，一部分煤气不满足回收条件只能通过甲烷伴烧的方式对外进行排放，不仅造成了能源的浪费，还会产生大量的温室气体CO₂，和NO_x气体等二次污染物。而CO自持催化燃烧技术是解决这一问题的有效途径，此外为了减轻CO₂的捕捉成本，本文还报道了一种新的CO处理方式即化学链燃烧。这两种燃烧方式对于钢铁行业节能减排具有重要意义。本文以高浓度的转炉放散煤气为研究对象，基于理论分析和实验方法从定性定量角度深入分析CO在过渡金属催化剂上的催化性能与反应机理，最后从工业利用角度对工业级催化剂进行实验探究，研究结果可为转炉放散煤气的处理提供设计方案，也为进一步提升催化剂的合理设计提供理论基础。

（1）利用溶胶凝胶法，并且引入Ce、Sr两种元素对LaMnO₃钙钛矿催化剂进行改性处理，并考察了葡萄糖在制备过程中的加入对催化剂的活性影响。结果表明在制备过程中加入糖，并加入的Ce、Sr改性的催化剂活性得到了提升，而未加糖的改性钙钛矿催化剂活性下降。葡萄糖的加入在焙烧过程中创造了弱还原性气氛夺走催化剂中的氧原子产生更多的氧空缺，使得Ce、Sr两种原子进一步进入到晶体中改善了B位点Mn离子的价态，而不是团聚在表面的活性位点上。此外Ce、Sr两种离子可以通过自身价态与Mn离子之间的变化促进电子之间的转移，形成了双金属活化效应。通过活性试验比较出了6种催化剂的活性大小顺序为LCMO-SW > LSMO-SW > LMO-SW > LMO-W > LSMO-W > LCMO-W，结合红外热像仪采用二分法确定了3种含糖催化剂的贫燃极限，结果表明其大小与催化剂的活性顺序有关，而不同流量下的贫燃极限则表明随着流量的增大会先减少而后增加。结合XPS、H₂-TPR等各项表征实验确定LCMO-SW催化剂活性最高的主要原因是具有最高浓度的Mn⁴⁺离子和氧空位。原位红外的静态和动态实验结构表明CO难以与Mn原子发生线性吸附，而在高温下CO会在催化剂表面形成碳酸盐物种，进一步分解从而得到CO₂，该反应机理遵循Langmuir-Hinshelwood（L-H）机理。

（2）利用溶胶凝胶将不同量的CuO物种负载在不同形貌的CeO₂载体上，通过理论与实验的方法考察催化燃烧与化学链燃烧过程中，催化剂形貌结构的演变规律及催化循环性能。结果表明在CuO含量相同时，棒状催化剂的活性要高于球状催化剂，这是因为棒状催化剂暴露的晶面以（100）晶面为主，而（100）晶面具有较低的Ce-O结合能，因此其氧的流动性更强。并且适当含量的CuO物种可以促进反应，超过一定量的CuO会覆盖在载体表面上使得暴露的活性位点数量减少，活性降低。催化燃烧的热稳定性要比化学链燃烧强，化学链燃烧在多次循环后活性发生下降。通过XRD、XPS、SEM等各项表征实验发现，化学链燃烧会导致还原性下降，且催化燃烧过后的催化剂形貌完整，而化学链燃烧会过度消耗体相晶格氧导致晶格坍塌。同时对两种燃烧进行了热力学分析，催化燃烧可以产生自持燃烧现象，可以持续对外界进行放热，而尽管化学链燃烧也是一个放热过程，但无法产生自持燃烧，但由于化学链燃烧中载氧体缓慢释放晶格氧的作用，有序地调节了热能的释放，因此燃烧时的㶲损更低。

（3）针对3CuCe-R催化剂利用化学链燃烧对催化燃烧解耦的方式进一步研究了活性晶格氧的种类及其贡献度。结果证明CO可以与催化剂表面的活性晶格氧发生反应，并且其“取出”与温度有关，随着反应温度的升高，参与的晶格氧数量越多。但反应速率随着时间明显下降，这是由于当表面晶格氧消耗完之后体相晶格氧需要克服一定量的能垒迁移至表面参与反应。当催化剂进行催化反应是主要由表面的Cu –[O] -Ce固溶体中的晶格氧（α）和表面分散性氧化铜的晶格氧（β）参与，在有气相氧参与反应时，体相晶格氧并不会参与反应。进一步通过氧同位素交换实验定量分析了两种晶格氧的量以及对催化反应的贡献度——α氧的数量尽管要小于β晶格氧，但由于其转换频率较快，因此对整个反应的贡献程度达到59.6%，占据主要地位。气相氧除了补充晶格氧反应过后的氧空位之外还能与表面的晶格氧直接发生交换反应，其总的交换速率（R_o）受氧解离速率（R_a）与掺入催化剂晶格氧速率（R_i）限制，三者均会随着温度的增大而增大，尽管R_i的增大速率明显要快，但始终R_i＜R_a，表明R_o主要受到的R_i限制。

（4）通过浸渍涂覆法制备5种不同孔径的CuCe蜂窝陶瓷催化剂，比较了它们的催化活性。结果表明随着孔径的变大活性下降。在此基础上探究了不同流量下CO对催化剂活性的影响，其临界当量比的趋势与实验室粉末催化类似。对不同孔径蜂窝陶瓷内部流场进行了模拟，发现对于孔径较小的催化剂容易在入口处形成涡旋区，导致灰颗粒不易排出造成堵塞，压力场的分布同样表明孔径较小压损较大不利于灰颗粒的流动。综合催化效果以及速度场与压力场的结果，选择孔径为5 mm的蜂窝陶瓷催化剂最为合适。中试平台的实验证明，催化不仅对CO具有良好的催化效果，同时也会对烟气中NO_x气体有还原效果，CO转换效率97%，而NO_x转换效率为67%。

With the continuous implementation of the "dual carbon" policy and the increase in environmental awareness, CO, as a typical atmospheric pollutant, has become a primary monitoring target for industrial exhaust emissions. The oxygen top-blowing converter steelmaking method commonly used in the steel industry produces a large amount of converter gas containing high concentrations of CO. Due to the variability of the steelmaking cycle, some gas does not meet the recovery conditions and can only be emitted through methane co-firing, resulting not only in energy waste but also in the generation of a large amount of greenhouse gas CO₂ and secondary pollutants such as NO_x gases. CO self-sustaining catalytic combustion technology is an effective way to address this issue. Additionally, this paper reports a new CO treatment method, namely chemical looping combustion, which has the unique advantage of low-energy CO₂ capture. These two combustion methods are of great significance for energy conservation and emission reduction in the steel industry. This paper takes the high-concentration converter off-gas as the research object and deeply analyzes the catalytic performance and reaction mechanism of CO on transition metal catalysts from both qualitative and quantitative perspectives based on theoretical analysis and experimental methods. Finally, from the perspective of industrial application, the paper explores experimental studies on industrial-grade catalysts, providing design schemes for the treatment of converter off-gas and theoretical basis for further improving the rational design of catalysts.

(1) Utilizing the sol-gel method and introducing Ce and Sr elements to modify LaMnO3 perovskite catalysts, this study investigated the impact of adding glucose during the preparation process on catalyst activity. Results showed that the activity of Ce, Sr modified catalysts with added sugar during preparation was enhanced, whereas the activity of modified perovskite catalysts without sugar decreased. The addition of glucose created a weakly reducing atmosphere during calcination, removing oxygen atoms from the catalyst to generate more oxygen vacancies, thereby further incorporating Ce and Sr atoms into the crystal and improving the valence state of Mn ions at the B-site, rather than aggregating at surface active sites. Moreover, Ce and Sr ions could promote electron transfer between themselves and Mn ions, forming a bimetallic activation effect. Through activity tests, the activity order of six catalysts was determined as LCMO-SW > LSMO-SW > LMO-SW > LMO-W > LSMO-W > LCMO-W. Combined with infrared thermal imaging, the lean burn limits of three sugar-containing catalysts were determined using the bisection method, showing that the lean burn limit is related to the activity order of the catalysts, and the lean burn limit first decreases and then increases with increasing flow rate. Characterization experiments such as XPS and H₂-TPR determined that the highest activity of LCMO-SW catalysts was mainly due to the highest concentration of Mn⁴⁺ ions and oxygen vacancies. In-situ infrared static and dynamic experiments indicated that CO could hardly adsorb linearly with Mn atoms, but at high temperatures, CO would form carbonate species on the catalyst surface, further decomposing to produce CO₂, following the Langmuir-Hinshelwood (L-H) mechanism.

(2) By using the sol-gel method to load different amounts of CuO species on CeO₂ supports with different morphologies, this study investigated the evolution of catalyst morphology structure and catalytic cycling performance during catalytic combustion and chemical looping combustion. Results indicated that rod-shaped catalysts exhibited higher activity than spherical ones at the same CuO content, due to the exposed crystal planes of rod-shaped catalysts being primarily (100) planes, which have a lower Ce-O bond energy, thus enhancing oxygen mobility. An appropriate amount of CuO species could promote the reaction, but an excessive amount of CuO covering the support surface would reduce the number of exposed active sites, decreasing activity. Thermal stability of catalytic combustion was stronger than that of chemical looping combustion, with activity decreasing after multiple cycles. Characterization experiments such as XRD, XPS, and SEM found that chemical looping combustion led to a reduction in reducibility, and the catalyst morphology remained intact after catalytic combustion, whereas chemical looping combustion caused excessive consumption of lattice oxygen, resulting in lattice collapse. Thermochemical analysis of both types of combustion showed that catalytic combustion could produce self-sustaining phenomena, continuously releasing heat to the surroundings, while chemical looping combustion, despite being an exothermic process, could not sustain itself. However, due to the role of the oxygen carrier releasing lattice oxygen, chemical looping combustion regulated the release of thermal energy more orderly, resulting in lower exergy loss.

(3) Further studies on the 3CuCe-R catalyst using chemical looping combustion to decouple catalytic combustion explored the types of active lattice oxygen and their contribution. Results proved that CO could react with active lattice oxygen on the catalyst surface, and its "extraction" was temperature-dependent, with more lattice oxygen involved as the reaction temperature increased. However, the reaction rate significantly decreased over time, as lattice oxygen in the bulk had to overcome a certain energy barrier to migrate to the surface for reaction. When the catalyst underwent catalytic reactions, mainly lattice oxygen (α) in the surface Cu-[O]-Ce solid solution and lattice oxygen (β) in dispersed copper oxide participated, with bulk lattice oxygen not participating in reactions involving gaseous oxygen. Further quantitative analysis of the two types of lattice oxygen and their contributions to the catalytic reaction through oxygen isotope exchange experiments showed that although the quantity of α-oxygen was smaller than β-lattice oxygen, its faster turnover rate meant it contributed significantly to the reaction, accounting for 59.6% of the overall reaction. Gaseous oxygen, besides replenishing oxygen vacancies after the lattice oxygen reaction, could also directly exchange with surface lattice oxygen, with the overall exchange rate (R_o) limited by the oxygen dissociation rate (R_a) and the rate of incorporating oxygen into the catalyst lattice (R_i), all of which increased with temperature. Although Ri increased at a faster rate, it was always less than R_a, indicating that R_o was mainly limited by R_i.

(4) By using the impregnation coating method to prepare five different pore size CuCe honeycomb ceramic catalysts, their catalytic activities were compared. Results showed that activity decreased with increasing pore size. On this basis, the effect of different CO flow rates on catalyst activity was explored, with the trend of the critical equivalence ratio similar to that of laboratory powder catalysts. Simulation of the internal flow field of different pore size honeycomb ceramics found that smaller pore size catalysts were prone to vortex formation at the inlet, causing ash particles to be difficult to discharge and leading to blockage. The pressure field distribution also indicated that smaller pore sizes had greater pressure loss, which was unfavorable for ash particle flow. Considering the overall catalytic effect and the results of velocity and pressure fields, a honeycomb ceramic catalyst with a pore size of 5 mm was deemed most suitable. Pilot platform experiments demonstrated that the catalyst not only had a good catalytic effect on CO but also reduced NO_x gases in flue gas, with a CO conversion efficiency of 97% and NO_x conversion efficiency of 67%.