*2.4. H2-TPR Analysis*

[30].

The H2−TPR data of CuxMn3−xO<sup>4</sup> samples are exhibited in Figure 5. Four peaks were observed on the Mn2O<sup>3</sup> sample at 385, 466, 524 and 651 ◦C, respectively. The relatively weak reduction peak at low temperature is due to the existence of surface species that can be easily reduced, that is, Mn2O<sup>3</sup> is reduced to Mn3O4. The strong reduction peak at high temperature can be attributed to the reduction of Mn3O<sup>4</sup> to MnO, which is attributed to the manganese in the spinel phase. Mn3O<sup>4</sup> is generally considered to consist of Mn2+ and Mn3+. However, Mn4+ appears in the samples due to the equilibrium state of 2Mn3+ Mn4+ + Mn2+. This phenomenon shows that the valence state of the Mn cation was complex in the Mn3O<sup>4</sup> spinel, which may be of significance and be responsible for the completion of the catalytic cycle. For the Cu1Mn2O<sup>4</sup> spinel in Figure 5, there are only two well-defined reduction peaks at 298 and 351 ◦C. The first reduction peak at 298 ◦C was attributed to the reduction of Cu2+ to Cu<sup>+</sup> , and the second reduction peak at 351 ◦C corresponded to the three reduction processes: the reduction of Mn4+ <sup>→</sup> Mn3+, Mn3+ <sup>→</sup> Mn2+ and Cu<sup>+</sup> <sup>→</sup> Cu<sup>0</sup> . The changes in the reduction peak number, reduction temperature and

peak intensity showed that there is electron transfer between Cu ions and Mn ions in the spinel lattice (Mn3+ + Cu2+ Mn4+ + Cu<sup>+</sup> ), and the presence of the strong interaction between Cu and Mn could play a synergistic role in the reducibility of the catalysts, leading to the enhancement of the catalytic cycle in CO-SCR [29]. *Catalysts* **2022**, *12*, x FOR PEER REVIEW 5 of 12

**Figure 3.** SEM images of Mn2O<sup>3</sup> (**a**,**b**), Cu1Mn2O<sup>4</sup> (**c**,**d**), Cu1.5Mn1.5O<sup>4</sup> (**e**,**f**), Cu2Mn1O<sup>4</sup> (**g**,**h**) and CuO (**i**,**j**). Cu, Mn and O elemental mapping recordings from Cu1.5Mn1.5O<sup>4</sup> (**k1**–**k4**) and the EDS result (**k5**). **Figure 3.** SEM images of Mn2O<sup>3</sup> (**a**,**b**), Cu1Mn2O<sup>4</sup> (**c**,**d**), Cu1.5Mn1.5O<sup>4</sup> (**e**,**f**), Cu2Mn1O<sup>4</sup> (**g**,**h**) and CuO (**i**,**j**). Cu, Mn and O elemental mapping recordings from Cu1.5Mn1.5O<sup>4</sup> (**k1**–**k4**) and the EDS result (**k5**).

For the Cu1.5Mn1.5O<sup>4</sup> sample, the low-temperature reduction peak reaches the same temperature (298 ◦C), and the high-temperature reduction peak moves to a lower temperature (342 ◦C). This phenomenon can be explained by the reduction in lattice distortion and the strong interaction between copper and manganese. Compared with Cu2Mn1O4, the two reduction peaks of Cu2Mn1O<sup>4</sup> (at 299 and 328 ◦C) have shifted to lower values. It is noteworthy that as the Cu doping content increased, the low-temperature reduction peaks of all catalysts became stronger. These results indicate that the interaction between Cu and Mn is enhanced, and that the redox property is improved with an increasing Cu doping amount.

*2.4. H2-TPR Analysis*

**Figure 4.** TEM image (**a**) and HRTEM image (**b**) of the Cu1.5Mn1.5O4sample.

Cu and Mn is enhanced, and that the redox property is improved with an increasing Cu

**Figure 3.** SEM images of Mn2O<sup>3</sup> (**a**,**b**), Cu1Mn2O<sup>4</sup> (**c**,**d**), Cu1.5Mn1.5O<sup>4</sup> (**e**,**f**), Cu2Mn1O<sup>4</sup> (**g**,**h**) and CuO (**i**,**j**). Cu, Mn and O elemental mapping recordings from Cu1.5Mn1.5O<sup>4</sup> (**k1**–**k4**) and the EDS result (**k5**).

three reduction processes: the reduction of Mn4+ → Mn3+, Mn3+ → Mn2+and Cu<sup>+</sup> → Cu<sup>0</sup>

The changes in the reduction peak number, reduction temperature and peak intensity

, and the second reduction peak at 351 °C corresponded to the

.

The H2−TPR data of CuxMn3−xO<sup>4</sup> samples are exhibited in Figure 5. Four peaks were observed on the Mn2O3 sample at 385, 466, 524 and 651 °C, respectively. The relatively weak reduction peak at low temperature is due to the existence of surface species that can be easily reduced, that is, Mn2O<sup>3</sup> is reduced to Mn3O4. The strong reduction peak at high temperature can be attributed to the reduction of Mn3O<sup>4</sup> to MnO, which is attributed to the manganese in the spinel phase. Mn3O<sup>4</sup> is generally considered to consist of Mn2+ and Mn3+. However, Mn4+ appears in the samples due to the equilibrium state of 2Mn3+ ⇄ Mn4+ + Mn2+. This phenomenon shows that the valence state of the Mn cation was complex in the Mn3O<sup>4</sup> spinel, which may be of significance and be responsible for the completion of the catalytic cycle. For the Cu1Mn2O<sup>4</sup> spinel in Figure 5, there are only two well-defined reduction peaks at 298 and 351 °C. The first reduction peak at 298 °C was attributed to the

**Figure 4.** TEM image (**a**) and HRTEM image (**b**) of the Cu1.5Mn1.5O<sup>4</sup> sample. doping amount.

**Figure 4.** TEM image (**a**) and HRTEM image (**b**) of the Cu1.5Mn1.5O4sample.

*Catalysts* **2022**, *12*, x FOR PEER REVIEW 6 of 12

reduction of Cu2+ to Cu<sup>+</sup>

**Figure 5.** H2−TPR curves of the as-synthesized CuxMn3-xO<sup>4</sup> samples. **Figure 5.** H2−TPR curves of the as-synthesized CuxMn3−xO<sup>4</sup> samples.

#### *2.5. XPS Analysis 2.5. XPS Analysis*

XPS was obtained on the CuxMn3−xO<sup>4</sup> sample, and the spectra of Cu 2p, Mn 2p and O 1 s scans, as well as C from the reference, are shown in Figure 6a. The Cu 2p 3/2 spectra could be divided into two characteristic peaks attributed to Cu+ (931.0 eV) and Cu2+ (934.1 eV) by performing peak-fitting deconvolutions, we can also see that accompanied by two distinct satellite peaks (marked by Sat.) at 938.2–945.9 and 959.7–964.7 eV in Figure 6b, which confirms the presence of Cu2+. The content of Cu+/0 of Cu1.5Mn1.5O<sup>4</sup> is the highest among the CuxMn3−xO4. This result validated the existence of electron transfer between Cu XPS was obtained on the CuxMn3−xO<sup>4</sup> sample, and the spectra of Cu 2p, Mn 2p and O 1 s scans, as well as C from the reference, are shown in Figure 6a. The Cu 2p 3/2 spectra could be divided into two characteristic peaks attributed to Cu<sup>+</sup> (931.0 eV) and Cu2+ (934.1 eV) by performing peak-fitting deconvolutions, we can also see that accompanied by two distinct satellite peaks (marked by Sat.) at 938.2–945.9 and 959.7–964.7 eV in Figure 6b, which confirms the presence of Cu2+. The content of Cu+/0 of Cu1.5Mn1.5O<sup>4</sup> is the highest among the CuxMn3−xO4. This result validated the existence of electron transfer between Cu ions and Mn ions (Mn3+ + Cu2+ Mn4+ + Cu<sup>+</sup> ) in the Cu1.5Mn1.5O<sup>4</sup> spinel (Table 2). The spectra recorded from the Cu1.5Mn1.5O<sup>4</sup> sample consist of a broad spin-orbit double peak, indicating the presence of more than one Mn contribution. An obvious feature of this spectrum is that the high binding energy side of the main peaks 2p3/2 and 2p1/2 are obviously the Mn 2p3/2 spectra, and could be divided into three characteristic peaks attributed to Mn2+ (640.7 eV), Mn3+ (641.8 eV), and Mn4+ (643.9 eV), respectively (Figure 6c). The results show that the Cu1.5Mn1.5O<sup>4</sup> sample contains the highest content of Mn4+ ions (54.4%), which indicates that Cu replaces the low valence Mn cations and significantly promotes the formation of high valence Mn cations. This result support the TPR results. In other words, due to the strong interaction between manganese and copper oxide (Cu), there are some electronic interactions between Mn4+ and Cu<sup>+</sup> (Cu−O−Mn bridge). To study the different O species on the surface of the CuxMn3−xO<sup>4</sup> samples, the O 1 s photoelectron spectra were obtained, as shown in Figure 6d. The deconvoluted peaks indicate that there are two different kinds of O species on the surface of the catalyst. The split peak at a lower binding energy of approximately 531.4 eV corresponds to lattice oxygen (denoted as Oα), and the other peak at approximately 529.5 eV is assigned to surface chemisorbed oxygen,

potentially including the chemisorbed oxygen O<sup>2</sup> <sup>2</sup><sup>−</sup> or defective O<sup>−</sup> (denoted as Oβ). The doping of Cu leads to the partial substitution of Cu atoms for Mn atoms in the −O−Mn− structure (O−Cu), which enhances the instability of O species and forms more active O species. This result is similar to the conclusion in the reported literature [30]. other peak at approximately 529.5 eV is assigned to surface chemisorbed oxygen, potentially including the chemisorbed oxygen O<sup>2</sup> <sup>2</sup><sup>−</sup> or defective O<sup>−</sup> (denoted as O*β*). The doping of Cu leads to the partial substitution of Cu atoms for Mn atoms in the −O−Mn− structure (O−Cu), which enhances the instability of O species and forms more active O species. This result is similar to the conclusion in the reported literature [30].

ions and Mn ions (Mn3+ + Cu2+ ⇄ Mn4+ + Cu+) in the Cu1.5Mn1.5O<sup>4</sup> spinel (Table 2). The spectra recorded from the Cu1.5Mn1.5O<sup>4</sup> sample consist of a broad spin-orbit double peak, indicating the presence of more than one Mn contribution. An obvious feature of this spectrum is that the high binding energy side of the main peaks 2p3/2 and 2p1/2 are obviously the Mn 2p3/2 spectra, and could be divided into three characteristic peaks attributed to Mn2+ (640.7 eV), Mn3+ (641.8 eV), and Mn4+ (643.9 eV), respectively (Figure 6c). The results show that the Cu1.5Mn1.5O4 sample contains the highest content of Mn4+ ions (54.4%), which indicates that Cu replaces the low valence Mn cations and significantly promotes the formation of high valence Mn cations. This result support the TPR results. In other words, due to the strong interaction between manganese and copper oxide (Cu), there are some electronic interactions between Mn4+ and Cu<sup>+</sup> (Cu−O−Mn bridge). To study the different O species on the surface of the CuxMn3−xO<sup>4</sup> samples, the O 1 s photoelectron spectra were obtained, as shown in Figure 6d. The deconvoluted peaks indicate that there are two different kinds of O species on the surface of the catalyst. The split peak at a lower binding energy of approximately 531.4 eV corresponds to lattice oxygen (denoted as O*α*), and the

*Catalysts* **2022**, *12*, x FOR PEER REVIEW 7 of 12

**Figure 6.** Survey spectra (**a**), Cu 2p (**b**), Mn 2p (**c**) and O 1 s, (**d**) XPS spectra of CuxMn3−xO<sup>4</sup> cata-**Figure 6.** Survey spectra (**a**), Cu 2p (**b**), Mn 2p (**c**) and O 1 s, (**d**) XPS spectra of CuxMn3−xO<sup>4</sup> catalysts.


**Table 2.** XPS results of all the catalysts.

lysts.

## **3. Catalytic Performances of the Catalysts**

*3.1. Catalytic Reduction of NO with CO*

In the temperature range 100–400 ◦C, the catalytic performance of the synthesized materials for the reduction of NO by CO is shown in Figure 7. It can be seen that pure Mn2O<sup>3</sup> has CO-SCR catalytic activity, the NO conversion rate can reach 100% at a temperature of approximately 350 ◦C, and the CO conversion rate is the worst. It can be clearly found that the catalytic activity of all Cu-doped catalysts is significantly higher than that of manganese oxide catalysts in the test temperature range. The CO-SCR activities of the Cu1.5Mn1.5O<sup>4</sup> catalyst exhibited the best NO conversion when the temperature was below

200 ◦C. High-valence state spinel is the active component of the CO-SCR reaction, which is more conducive to showing better low-temperature activity, as reported in the literature. The CO conversion of the Cu-doped catalyst has a similar trend with the increase of temperature, and the CO conversion data are inconsistent with the NO conversion data above 200 ◦C, implying that CO reduced partial metal oxides in Figures 7c and S1 (consistent with the H2−TPR result). From the CO catalytic activity results, it can be seen that the Cudoped catalyst shows better catalytic activity than the pure Mn2O<sup>3</sup> sample. The Cu2Mn1O<sup>4</sup> sample shows a higher CO catalytic oxidation activity, which suggests that excessive copper doping causes the adsorption of CO to be stronger than that of NO. This also implies that the Cu−O−Mn structure in spinel is the active site of CO-SCR (corresponding to the XRD results). The reaction of CO-SCR under O2-rich conditions was performed to investigate the effect of O<sup>2</sup> on the catalytic performance. As shown in Figure S2, the NO conversion of the catalyst significantly decreased, and CO conversion increased with the increase of temperature. It can be found that the main reason affecting the NO conversion is the competitive reaction between CO and NO with O2, resulting in the decline of performance. Improving the low-temperature catalytic performance of the catalyst under oxygen conditions will be the focus of our future research. 200 °C. High-valence state spinel is the active component of the CO-SCR reaction, which is more conducive to showing better low-temperature activity, as reported in the literature. The CO conversion of the Cu-doped catalyst has a similar trend with the increase of temperature, and the CO conversion data are inconsistent with the NO conversion data above 200 °C, implying that CO reduced partial metal oxides in Figures 7c and S1 (consistent with the H2−TPR result). From the CO catalytic activity results, it can be seen that the Cu-doped catalyst shows better catalytic activity than the pure Mn2O<sup>3</sup> sample. The Cu2Mn1O<sup>4</sup> sample shows a higher CO catalytic oxidation activity, which suggests that excessive copper doping causes the adsorption of CO to be stronger than that of NO. This also implies that the Cu−O−Mn structure in spinel is the active site of CO-SCR (corresponding to the XRD results). The reaction of CO-SCR under O2-rich conditions was performed to investigate the effect of O<sup>2</sup> on the catalytic performance. As shown in Figure S2, the NO conversion of the catalyst significantly decreased, and CO conversion increased with the increase of temperature. It can be found that the main reason affecting the NO conversion is the competitive reaction between CO and NO with O2, resulting in the decline of performance. Improving the low-temperature catalytic performance of the catalyst under oxygen conditions will be the focus of our future research.

**Sample Mn4+/Mn Mn3+/Mn Mn2+/Mn Cu2+/Cu Oα/O Oβ/O** Mn2O<sup>3</sup> 2.7 50.7 46.6 - 55.4 44.6 Cu1Mn2O<sup>4</sup> 31.6 50.8 17.6 70.1 53.1 46.9 Cu1.5Mn1.5O<sup>4</sup> 54.4 36.0 9.6 66.2 32.3 67.7 Cu2Mn1.5O<sup>4</sup> 33.4 54.4 12.2 86.7 37.5 62.5 CuO - - - 100 51.1 48.9

In the temperature range 100–400 °C, the catalytic performance of the synthesized materials for the reduction of NO by CO is shown in Figure 7. It can be seen that pure Mn2O<sup>3</sup> has CO-SCR catalytic activity, the NO conversion rate can reach 100% at a temperature of approximately 350 °C, and the CO conversion rate is the worst. It can be clearly found that the catalytic activity of all Cu-doped catalysts is significantly higher than that of manganese oxide catalysts in the test temperature range. The CO-SCR activities of the Cu1.5Mn1.5O<sup>4</sup> catalyst exhibited the best NO conversion when the temperature was below

*Catalysts* **2022**, *12*, x FOR PEER REVIEW 8 of 12

**Table 2.** XPS results of all the catalysts.

**3. Catalytic Performances of the Catalysts** *3.1. Catalytic Reduction of NO with CO*

**Figure 7.** (**a**) NO conversion and (**b**) CO conversion of all the catalysts in the CO-SCR.

Therefore, Cu doping is conducive to the performance improvement of the CuxMn3−xO<sup>4</sup> catalyst because there is a strong synergistic effect between the binary metal oxides. It is accepted that the active phase of spinel is the highly reactive center in the catalytic reaction process. The active phase of Cu1.5Mn1.5O<sup>4</sup> spinel plays an important role in the CO-SCR reaction, and the catalytic performance of the spinel structure catalyst is better than that of the other catalysts. The stability of this catalyst was further confirmed by the XRD (Figure S3a) and TEM analyses (Figure S3b,c), which showed no obvious change in the structure after the reaction at 400 ◦C.

#### *3.2. Structure Activity Relationship and Catalystic Reaction Mechanism*

According to reports, the active phase of CuxMn3−xO<sup>4</sup> in the redox reaction is the Mn4+ concentration on the catalyst surface [30]. On Cu-Mn spinels, the number of surface-active sites and bulk concentration of Mn4+/Mn are critical to the reaction. At the same time, Cu2+ is transformed into Cu<sup>+</sup> , and Mn3+ is transformed into Mn4+. Mn4+ is considered to be a manganese species that has a passivation effect on the redox reaction. With the doping of copper ions in Mn2O3, the spinel structure with rich lattice defects and oxygen vacancies increases the concentration of Mn4+, which can adsorb reactant molecules and improve its redox performance, enhance the mobility of active oxygen species and enhance its catalytic activity. Therefore, compared with CuO and Mn2O3, the spinel-type copper-manganese composite oxide rich in Cu<sup>+</sup> and Mn4+ will have a significantly improved activity. In other words, due to the strong synergy between the binary metal oxides, copper doping is beneficial to the stability and catalytic performance of the CuxMn3−xO<sup>4</sup> catalyst. other words, due to the strong synergy between the binary metal oxides, copper doping is beneficial to the stability and catalytic performance of the CuxMn3−xO<sup>4</sup> catalyst.

manganese species that has a passivation effect on the redox reaction. With the doping of copper ions in Mn2O3, the spinel structure with rich lattice defects and oxygen vacancies increases the concentration of Mn4+, which can adsorb reactant molecules and improve its redox performance, enhance the mobility of active oxygen species and enhance its catalytic activity. Therefore, compared with CuO and Mn2O3, the spinel-type copper-manganese composite oxide rich in Cu<sup>+</sup> and Mn4+ will have a significantly improved activity. In

Therefore, Cu doping is conducive to the performance improvement of the CuxMn3−xO<sup>4</sup> catalyst because there is a strong synergistic effect between the binary metal oxides. It is accepted that the active phase of spinel is the highly reactive center in the catalytic reaction process. The active phase of Cu1.5Mn1.5O4 spinel plays an important role in the CO-SCR reaction, and the catalytic performance of the spinel structure catalyst is better than that of the other catalysts. The stability of this catalyst was further confirmed by the XRD (Figure S3a) and TEM analyses (Figure S3b,c), which showed no obvious

According to reports, the active phase of CuxMn3−xO<sup>4</sup> in the redox reaction is the Mn4+ concentration on the catalyst surface [30]. On Cu-Mn spinels, the number of surface-active sites and bulk concentration of Mn4+/Mn are critical to the reaction. At the same time, Cu2+

, and Mn3+ is transformed into Mn4+. Mn4+ is considered to be a

*Catalysts* **2022**, *12*, x FOR PEER REVIEW 9 of 12

change in the structure after the reaction at 400 °C

is transformed into Cu<sup>+</sup>

*3.2. Structure Activity Relationship and Catalystic Reaction Mechanism*

Based on the above analysis, important information on the catalytic route was obtained, and a reasonable mechanism of the CO-SCR reaction on a CuxMn3−xO<sup>4</sup> catalyst was initially proposed. A proposed mechanism of the two processes is shown in Scheme 1: (i) CO and NO molecules are the first adsorbed oxygen vacancies, Mn4+ and Cu+, on the catalyst surface. In this process, the reactant molecules CO and NO are adsorbed as CO (ads) and NO (ads). Subsequently, CO (ads) reacts with the active oxygen on the catalyst to produce CO2. (ii) NO molecules are adsorbed on the catalyst surface oxygen vacancy, the oxygen O of NO reacts with the oxygen vacancy, and nitrogen gas is generated. Herein, the redox cycle occurs between bimetallic oxide components (Cu2+ + Mn3+ Cu<sup>+</sup> + Mn4+) in the CuxMn3−xO<sup>4</sup> spinels, and the Cu<sup>+</sup> and Mn4+ formed by this interaction distorts the spinel structure and promotes the generation of more surface vacancies; that is, it is conducive to the activation of reactants CO and NO and forms more active species and improves the catalytic performance for CO-SCR of the catalysts. Based on the above analysis, important information on the catalytic route was obtained, and a reasonable mechanism of the CO-SCR reaction on a CuxMn3−xO<sup>4</sup> catalyst was initially proposed. A proposed mechanism of the two processes is shown in Scheme 1: (i) CO and NO molecules are the first adsorbed oxygen vacancies, Mn4+ and Cu+, on the catalyst surface. In this process, the reactant molecules CO and NO are adsorbed as CO (ads) and NO (ads). Subsequently, CO (ads) reacts with the active oxygen on the catalyst to produce CO2. (ii) NO molecules are adsorbed on the catalyst surface oxygen vacancy, the oxygen O of NO reacts with the oxygen vacancy, and nitrogen gas is generated. Herein, the redox cycle occurs between bimetallic oxide components (Cu2+ + Mn3+ ⇄ Cu<sup>+</sup> + Mn4+) in the CuxMn3−xO<sup>4</sup> spinels, and the Cu<sup>+</sup> and Mn4+ formed by this interaction distorts the spinel structure and promotes the generation of more surface vacancies; that is, it is conducive to the activation of reactants CO and NO and forms more active species and improves the catalytic performance for CO-SCR of the catalysts.

**Scheme 1.** Schematic illustration of the proposed mechanism for the catalytic CO-SCR over CuxMn3−xO<sup>4</sup> catalysts. **Scheme 1.** Schematic illustration of the proposed mechanism for the catalytic CO-SCR over CuxMn3−xO<sup>4</sup> catalysts.

#### **4. Experimental 4. Experimental**

#### *4.1. Material Synthesis 4.1. Material Synthesis*

Specifically, CuxMn3−xO<sup>4</sup> (x = 0, 1, 1.5, 2, 3) spinels were prepared by a citratebased modified pechini method [33–35]. Cu(NO3)2·3H2O (Sinopharm Chemical Reagent Co., Ltd., Beijing, China, ≥99.0%) and Mn(NO3)<sup>2</sup> solution (Macklin, 50% in H2O) were dissolved in deionized water. In the calculated amount of copper nitrate trihydrate and 50% manganesenitrate solution (Table 3), citric acid monohydrate (Xilong Chemical Co., Ltd., Guangzhou, China, ≥99.5%) was added at a molar ratio of 1:1 (Cu+Mn/citric acid). The solution was stirred for 2 h at room temperature to obtain a homogeneous mixture and then evaporated to obtain a sticky gel. The gel was dried in a 120 ◦C oven for 6 h, forming a foam metal citrate complex. Finally, the samples were calcined in 600 ◦C air for 8 h to form spinel oxides.

**Table 3.** The chemicals and their amounts used for preparing samples.


#### *4.2. Characterization*

The powder samples were characterized by XRD with the use of a PANalytica X'Pert PRO MPD diffractometer using Cu Kα radiation (λ = 0.154 nm, 40 kV, 40 mA). The crystallite sizes of all samples were calculated using the Debye–Scherrer equation. The morphology of the particles was analyzed with a JSM-7001F field-emission SEM with energy-dispersive spectroscopy (EDS) (INCA X-MAX, JEOL, Oxford, UK) and TEM (JEM-2010F, JEOL, Tokyo, Japan). The reducibility of the catalysts was examined by H2-TPR using a Quantachrome automated chemisorption analyzer (Chem BET pulsar TPR/TPD). Briefly, 50 mg of sample was loaded into a quartz U-tube and heated from room temperature to 150 ◦C at 10 ◦C min−<sup>1</sup> under helium flow to remove moisture and impurities. Then, the sample was cooled to room temperature, followed by heating to 800 ◦C at a heating rate of 10 ◦C min−<sup>1</sup> under a binary gas mixture (10 vol.% H2/Ar) with a gas flow rate of 30 mL min−<sup>1</sup> . H<sup>2</sup> consumption was detected continuously as a function of increasing temperature using a thermal conductivity detector (TCD). The BETs were determined using N<sup>2</sup> physisorption at −196 ◦C using Quantachrome NOVA 3200e equipment. Prior to N<sup>2</sup> adsorption, each catalyst was degassed for 2 h under vacuum at 200 ◦C. The surface chemical composition was determined by XPS (Model VG ESCALAB 250 spectrometer, Thermo Electron, London, UK) using non-monochromatized Al Kα X-ray radiation (hν = 1486.6 eV).

#### *4.3. Measurement*

The evaluation of the catalyst was carried out with a typical fixed-bed reactor with a quartz tube (8 mm inner diameter). Two grams of the catalysts (particle size was 20–40 mesh) were used in quartz tubes between glass wool. The catalytic activity was measured using feed gas compositions of 1000 ppm NO, 2000 ppm CO and N<sup>2</sup> (the balance) at different temperatures at a rate of 30,000 h−<sup>1</sup> . First, the catalysts were treated using a CO/N<sup>2</sup> gas flow at 200 ◦C for 1 h before each test. After the catalysts were cooled to room temperature under a N<sup>2</sup> flow, they were allowed to react with the mixed gas. The CO, NO and NO<sup>2</sup> concentrations were monitored using a Testo 350 flue gas analyzer. The catalytic activity was calculated using the following formula:

$$\text{NO conversion } (\%) = \frac{\text{NO}\_{\text{in}} - \text{NO}\_{\text{out}}}{\text{NO}\_{\text{in}}} \times 100\% \tag{1}$$

$$\text{CO conversion } (\%) = \frac{\text{CO}\_{\text{in}} - \text{CO}\_{\text{out}}}{\text{CO}\_{\text{in}}} \times 100\% \tag{2}$$

where the "in" and "out" subscripts indicate the inlet and outlet concentrations of NO and CO in the steady state, respectively. The selectivity of N<sup>2</sup> was not calculated here due to no NO<sup>2</sup> being detected at the outlet.

#### **5. Conclusions**

In this work, a series of CuxMn3−xO<sup>4</sup> spinels were synthesized by the citrate-based modified pechini method. The results show that controlling the doping amount of Cu can improve the low-temperature activity of the Mn2O<sup>3</sup> catalyst. Doping Cu species could shift the redox balance in the catalyst system (Cu2+ + Mn3+ Mn4+ + Cu<sup>+</sup> ), improve the redox performance and catalytic activity of manganese oxide catalyst, and promote the grain formation and growth of the Cu1.5Mn1.5O<sup>4</sup> spinel structure instead of manganese oxides to increase the surface area and particle size. The surface of Cu1.5Mn1.5O<sup>4</sup> spinels retained a high ratio of Mn4+/Mn, more reactive oxygen species were formed than pure Mn2O<sup>3</sup> on the surface to promote the adsorption of oxygen molecules, and it enhanced the adsorption capacity of CO and NO. In general, the doping of low valence state Cu significantly enhanced the CO−SCR activity of CuxMn3−xO<sup>4</sup> spinels at low temperature, which could be an effective way to design and synthesize highly active Mn−based CO-SCR catalysts.

**Supplementary Materials:** The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/catal12060591/s1, Figure S1: CO conversion of Cu1.5Mn1.5O<sup>4</sup> catalyst in the CO-SCR (Reaction conditions: [CO] = 2000 ppm and N<sup>2</sup> as balance gas, GHSV = 30,000 h−<sup>1</sup> ); Figure S2: (a) NO conversion; (b) N<sup>2</sup> selectivity in CO–SCR reaction (Reaction conditions: [NO] = 1000 ppm, [CO] = 2000 ppm, [O<sup>2</sup> ] = 0 or 1%, and N<sup>2</sup> as balance gas, GHSV = 30,000 h−<sup>1</sup> ); Figure S3: (a) XRD patterns, and (b,c) TEM images of the catalyst of Cu1.5Mn1.5O<sup>4</sup> after reaction.

**Author Contributions:** Conceptualization, F.F.; data curation, F.F. and L.W. (Lingjuan Wang); formal analysis, F.F., L.W. (Lingjuan Wang) and L.W. (Lei Wang); methodology, F.F. and M.W.; project administration, J.L.; writing—original draft, F.F. and M.W.; writing—review and editing, F.F., M.W. and J.L.; funding acquisition, F.F. and J.L. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the Chengde Science and Technology Research and Development Project (202006A117), School-level Youth Fund Project (QN2021001), the Scientific Research Project of the Higher Education Institutions of Hebei Province (ZD2021413) and the School-level teaching reform project (JG-202021571).

**Data Availability Statement:** Data are available from the corresponding author on request.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

