Metal–Support Interactions on Ag/Co3O4 Nanowire Monolithic Catalysts Promoting Catalytic Soot Combustion

Tuning metal–support interactions (MSIs) is an important strategy in heterogeneous catalysis to realize the desirable metal dispersion and redox ability of metal catalysts. Herein, we use pre-reduced Co3O4 nanowires (Co-NWs) in situ grown on monolithic Ni foam substrates to support Ag catalysts (Ag/Co-NW-R) for soot combustion. The macroporous structure of Ni foam with crossed Co3O4 nanowires remarkably increases the soot–catalyst contact efficiency. Our characterization results demonstrate that Ag species exist as Ag0 because of the equation Ag+  + Co2+  = Ag0 + Co3+, and the pre-reduction treatment enhances interactions between Ag and Co3O4. The number of active oxygen species on the Ag-loaded catalysts is approximately twice that on the supports, demonstrating the significant role of Ag sites in generating active oxygen species. Additionally, the strengthened MSI on Ag/Co-NW-R further improves this number by increasing metal dispersion and the intrinsic activity determined by the turnover frequency of these oxygen species for soot oxidation compared with the catalyst without pre-reduction of Co-NW (Ag/Co-NW). In addition to high activity, Ag/Co-NW-R exhibits high catalytic stability and water resistance. The strategy used in this work might be applicable in related catalytic systems.


Introduction
Presently, diesel engines are widely used in heavy-duty vehicles because of their low operating costs, excellent durability, superior fuel efficiency, and reliability under lean conditions [1][2][3]. However, soot particulates emitted from diesel engines have caused severe damage to human health and the environment [4]. Catalytic combustion technology using oxidation catalysts combined with a diesel particulate filter is a promising after-treatment strategy to trap and eliminate soot particulates in the range of 200-500 °C [5,6].
Numerous catalysts for soot combustion have been reported using noble metals [7][8][9], perovskites [10,11], spinel-type oxides [12,13], hydrotalcites [14,15], alkaline metal oxides [16,17], transition metal oxides [18], and rare earth metal oxides [19]. Ag-based catalysts are promising candidates for catalytic soot oxidation reactions because of their low price among noble metals, and particularly the Ag 0 species has a high ability to activate oxygen [20][21][22]. Interestingly, the Ag 0 species can be automatically obtained by directly supporting Ag salt on reducible metal oxides (Co 3 O 4 , CeO 2 , MnO 2 , etc.) because of the reaction between silver and variable valence cations (for example, Ag + + Co 2+ = Ag 0 + Co 3+ ) [7]. However, because of the increased cation valence on supports (such as Co 2+ → Co 3+ ), the number of oxygen vacancies on the surface of reducible supports will inevitably be reduced. Nevertheless, oxygen vacancies play important roles in anchoring metal sites, increasing metal dispersion, and tuning the metal-support interaction, which significantly influence the activity and stability of catalysts; moreover, they are beneficial to promoting oxidation reactions [23][24][25][26]. Thus, before loading metals, a possible way to solve the problem is by pre-reducing reducible supports to construct more available oxygen vacancies. To date, few reports have used this strategy to develop efficient catalysts for soot combustion [27].
Some reducible, noble metal-free metal oxides, such as CeO 2 and MnO 2 and particularly Co 3 O 4 , are active in soot oxidation [27]. Moreover, Co 3 O 4 exhibits a high ability for NO oxidation to NO 2 [22,28], which is a more effective oxidant than O 2 for soot oxidation. Thus, Co 3 O 4 becomes a good candidate for supporting noble metal catalysts in this reaction.
In addition, soot is a type of particulate with a diameter of 25-100 nm, which makes touching the inner surface of mesopores and micropores in traditional powder catalysts difficult during a reaction, resulting in low soot-catalyst contact efficiency and unsatisfactory catalytic performance [29]. To solve this problem, Zhao's research group [12,30,31] developed a series of three-dimensional ordered macroporous catalysts using the colloidal crystal template method for catalytic soot elimination, which can greatly improve the contact efficiency and catalytic soot oxidation performance. Zhang's group [32,33] focused on multiple strategies to decrease the ignition temperature of soot combustion. In addition, our group [1,5,34] has developed a series of three-dimensional monolithic catalysts by in situ growth of nanostructured active components on monolithic substrates, which can provide a sufficient open macroporous structure to increase soot-catalyst contact opportunities and remarkably lower soot elimination temperatures.
In this work, we designed and synthesized Ag catalysts supported on pre-reduced Co 3 O 4 nanowires (Co-NW) using Ni foams as the monolithic substrate for soot oxidation. The structure-activity relationship was revealed through various characterization techniques, such as scanning electron microscopy (SEM), high-resolution transmission electron microscopy (HRTEM), X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), temperature-programmed reduction by soot (soot-TPR), temperature-programmed oxidation by NO (NO-TPO), and CO temperature-programmed desorption (CO-TPD) measurements. The pre-reduction of the Co-NW support enhances metal-support interactions on Ag/Co-NW-R and then increases the metal dispersion and the number of oxygen vacancies and improves the turnover frequency (TOF) and catalytic performance for soot oxidation.

Catalyst Preparation
Ag/Co 3 O 4 nanowires on Ni foam were synthesized via a simple hydrothermal and incipient wetness impregnation method [1], as illustrated in Scheme 1.
The Ni foam (thickness: ca. 1.5 mm, porosity: ≥ 98%, and mechanical strength: ≥ 1 MPa) was purchased from LANKE battery materials Co. Ltd. The Ni foam was immersed in 2 mol/L HCl in an ultrasound bath for 5 min to remove the surface oxide layers and then rinsed with deionized water Scheme 1 Schematic diagram of the synthesis of the as-prepared catalysts on monolithic Ni foam 1 3 and absolute ethanol for 5 min, respectively. In a typical synthesis, 1 mmol of Co(NO 3 ) 2 ·6H 2 O, 2 mmol of NH 4 F, and 5 mmol of CO(NH 2 ) 2 were dissolved in 50 mL of water. After intense stirring for 30 min, the solution was transferred into a 100-mL Teflon-lined stainless steel autoclave. Subsequently, a piece of the clean Ni foam was immersed in the reaction solution. The autoclave was sealed and maintained at 120 °C for 5 h and cooled naturally to room temperature. The as-synthesized precursor was ultrasonically cleaned with water for 5 min and rinsed with absolute ethanol three times. After drying at 60 °C, the precursor was calcined at 500 °C in air for 2 h and denoted as Co-NW. The clean Ni foam substrate was also calcined for 2 h at 500 °C for comparison and denoted as Ni foam.
Ag/Co-NW was prepared using the impregnation method. The Co-NW precursor was impregnated with an aqueous solution of silver nitrate, followed by drying at 60 °C, and then annealed at 500 °C in air for 2 h. This catalyst was designated as Ag/Co-NW. For comparison, Co-NW was pre-reduced at 200 °C in 8% H 2 /N 2 flow, and then Ag was impregnated on Co-NW-R with the same procedure of Ag/ Co-NW, and the obtained catalyst was denoted as Ag/Co-NW-R. The pre-reduced Co-NW was also calcined for 2 h in 500 °C for comparison and denoted as Co-NW-R. According to the inductively coupled plasma (ICP) method (Vista MPX instrument), the loading of silver in the prepared catalysts amounted to 5 wt% (the mass fraction of Ag in the catalyst without the Ni foam substrate).

Catalyst Characterizations
The samples were characterized by XRD on a Rigaku D/ MAX-2500 diffractometer (Cu Kα radiation), XPS on an ESCALAB instrument, SEM on a Hitachi S-4800 scanning electron microscope, and transmission electron microscopy (TEM) and EDS mapping on a JEOL-JEM-2100F electron microscope.
On the basis of the diameter (d) of Ag on catalysts determined by TEM, the dispersity (D Ag ) of Ag crystallites was calculated from the expression D Ag (%) = 1.34/d, assuming that they had a spherical morphology [35].
Soot temperature-programmed reduction (soot-TPR) tests were performed in a fixed-bed reactor at a heating rate of 5 °C/min in N 2 (100 mL/min) on a CO-CO 2 IR analyzer. NO temperature-programmed oxidation (NO-TPO) experiments were measured on the same reactor with 250 mg of catalyst in 600-ppm NO/10% O 2 /N 2 (100 mL/min) at a rate of 5 °C/min and recorded using a chemiluminescence NO − NO 2 − NO x analyzer (Model 42i-HL, Thermo Scientific). H 2 temperature-programmed reduction (H 2 -TPR) tests were performed on a TP-5080 instrument (50-mg sample, 8% H 2 /N 2 , 10 °C /min). CO temperature-programmed desorption (CO-TPD) experiments were performed on the same instrument equipped with a mass spectrometer (Hiden). Before CO-TPD, the catalysts were pretreated at 100 °C in He flow and then adsorbed 7% CO/N 2 for 1 h. After cooling to room temperature, the system was switched to He flow and heated to 900 °C at a rate of 10 °C/min.

Catalytic Activity Measurement
The catalytic activities for soot oxidation were evaluated by soot temperature-programmed oxidation (soot-TPO) using Printex-U soot (Degussa) as a model in a fixed-bed flow reactor. To achieve the loose contact condition between catalysts and soot particulates, 15 mg of soot was dispersed in 25 mL of ethanol by ultrasound to obtain a suspended soot-ethanol solution, which was gradually dropped on the monolithic catalyst via a precisely controlled pipette gun, and then the mixture was dried at 60 °C for 2 h. The mass ratio of soot/catalyst (excluding the weight of the Ni foam substrate) was approximately 1:10. For each reaction, the monolithic catalyst was heated from 200 to 650 °C at a heating rate of 2 °C/min in the reactant gas flow (100 mL/min) of 0-or 600-ppm NO and 10% O 2 balanced with N 2 . The products of CO 2 and CO were online analyzed by an IR analyzer. The catalytic activities were evaluated using the temperatures at 10% (T 10 ) and 50% (T 50 ) soot conversion during the soot-TPO reaction. The CO 2 selectivity ( S CO 2 ) was defined as S CO 2 = C CO 2 /(C CO + C CO 2 ), where C CO and C CO 2 were the concentrations of CO and CO 2 , respectively.

Isothermal Kinetic Measurements
The isothermal kinetic experiments (isothermal reactions and isothermal anaerobic titrations) were performed to calculate reaction rates, active oxygen amounts, and TOF values of the monolithic catalysts [36]. To ensure that all reactions occurred in the dynamic region with a low soot conversion (< 10%), the isothermal oxidation tests were conducted in a 10% O 2 /N 2 flow at 280 °C (150 mL/min). When the CO 2 concentration became steady, the 10% O 2 / N 2 flow was switched to a pure N 2 flow at the same volume flow rate, and the isothermal anaerobic titration began. The outlet gas was monitored online by an infrared gas analyzer. The reaction rate (ν) was calculated according to the following equation: where Q is the molar gas flow rate (mol/s); c is the molar fraction of CO 2 estimated by isothermal reactions; and m is the mass of the catalyst (g); n is the amount of substance of CO 2 (mol).
The number of active oxygen species (O * ) was quantified by integrating the diminishing rate of CO 2 concentration with time. The TOF was obtained as follows: where P 0 represents the atmospheric pressure (Pa);V represents the volumetric flow rate (m 3 /s);A represents the integral of the CO 2 concentration curves as a function of time during the isothermal anaerobic titration (s); R represents the ideal gas constant; T represents the room temperature (K).
In Fig. 2, the structure of the as-prepared catalysts, the average diameter of silver on the Co-NW surface, and the features of Ag-Co 3 O 4 composites are investigated using SEM and TEM. Figure S1 displays the SEM images of the Ni foam, which possesses a 3D-macroporous structure with diameters ranging from 200 to 500 µm. In Fig. 2a, d and g, the monolithic catalysts exhibit nanowire morphology, which is well-distributed on the skeleton of the Ni foam substrate. The nanowires are self-assembled into grass-like clusters, which can provide enough macroporous space for soot deposition [5]. The HRTEM images in Fig. 2c, f and i demonstrate that these nanowires are composites of Co 3 O 4 [21].
After loading Ag, the crossed nanowire morphology of Ag/Co-NW and Ag/Co-NW-R remain unchanged, except that the surface of the nanowire becomes slightly rough in Fig. 2 a, d and g. The lattice spacing of 0.236 nm belonging to Ag (111) in Fig. 2f, i confirms the presence of metallic Ag on the catalysts [21]. In Fig. S2, Ag nanoparticles (NPs) are homogeneously dispersed on Co 3 O 4 nanowires with a low content from TEM-EDS mapping. Moreover, as shown in Fig. S3j, k, the average size of Ag NPs is approximately 5.8 nm and 4.4 nm for Ag/Co-NW and Ag/Co-NW-R, respectively. We calculated the dispersion of Ag based on the above results and found that it was approximately 1.5-fold larger on Ag/Co-NW-R compared to Ag/Co-NW, as shown in Table S1.

Redox Properties
The soot-TPR and CO-TPD experiments were performed to investigate the active oxygen species of the catalysts. As shown in Fig. 3, the soot-TPR curves of the catalysts consist of three temperature ranges, representing three types of oxygen species. The low-temperature soot-TPR reduction peak at 200-400 ºC can be assigned to interfacial active oxygen species in the case of Ag-O-Co, which is probably caused by the strong interaction between Ag and Co 3 O 4 . These oxygen species can be assigned to surface-adsorbed oxygen species. In addition, the peaks at 400-650 °C originate from surface lattice oxygen (O 2− ), and the peaks above 650 °C are ascribed to the bulk lattice oxygen (O 2− ) of the catalysts, which has little impact on the catalytic activity [5]. Compared with soot, CO is more easily oxidized, so CO-TPD can be a better technology than soot-TPR for revealing the activity of oxygen species on the catalyst [37]. Generally, the adsorbed CO reacts with the reactive oxygen species on the catalyst and then is oxidized to CO 2 during the temperatureprogrammed process [37,38]. Figure 4 shows the CO-TPD profiles of the samples. The desorption peaks below 200 °C are attributed to the desorption of CO 2 produced directly by the reaction of adsorbed CO with surface-adsorbed oxygen species. The broad peaks between 200 and 350 °C for all three catalysts belong to the desorption of CO 2 oxidized by surface lattice oxygen species. No obvious difference in the shape of the peaks is apparent for all catalysts below 350 °C. The peaks above 350 ºC can be ascribed to the desorption of CO 2 oxidized by bulk lattice oxygen species. According to the temperature range (200-600 °C) for soot oxidation, surface-adsorbed oxygen species and surface lattice oxygen species are the main active oxygen species over these catalysts [1]. For soot-TPR and CO-TPD, the peaks of surfaceadsorbed oxygen species and bulk lattice oxygen species  for the Ag-loaded catalysts shift to lower temperatures, particularly for Ag/Co-NW-R, which illustrates that Ag-Co 3 O 4 interactions can activate surface oxygen species and promote bulk lattice oxygen mobility. In Figs. 3, 4, the peak areas of the catalysts follow the order of Ag/Co-NW-R > Ag/ Co-NW > Co-NW, which illustrates that Ag/Co-NW-R possesses more active oxygen species than other catalysts. Figure 5 and Table 1 display the XPS results of the samples. In Fig. 5a, Ag species display Ag 3d 5/2 and Ag 3d 3/2 peaks at binding energies of 368.0, 368.3, 374.0, and 374.3 eV, respectively, with a splitting value of 6.0 eV, which demonstrates that Ag species exist in the metallic state on the catalysts [25,39]. Notably, the peaks of Ag/Co-NW-R shift to higher binding energy (BE) compared with Ag/Co-NW, demonstrating the enhanced metal-support interaction on Ag/Co-NW-R, which is also suggested by the decreased size of Ag NPs [27] in Fig. 2.

Surface Chemical States
In Fig. 5b, the Co 2p XPS spectra of the catalysts show spin-orbit splitting into Co 2p 1/2 and Co 2p 3/2 , which can be deconvoluted into Co 3+ , Co 2+ , and two weak satellite peaks at relatively high binding energies (800-810 and 786-791 eV, respectively) [1]. As listed in Table 1, Co species exist as Co 3+ (779.5, 794.3 eV) and Co 2+ (781.4, 795.6 eV) [40]. Additionally, the spin-orbit splitting energy between the two peaks is 15.2 eV, which indicates that the surface cobalt species exist as Co 3 O 4 [41]. The values of surface Co 2+ /(Co 2+ + Co 3+ ) for all samples were calculated according to the corresponding peak areas in the XPS spectra. Notably, Co-NW-R possesses a higher Co 2+ concentration than Co-NW, indicating that the reduction pretreatment indeed created many Co 2+ species on Co-NW-R, as we proposed. Loading Ag on the supports consumed Co 2+ cations according to Ag + + Co 2+ = Ag 0 + Co 3+ . However, we interestingly observed an increased value of Co 2+ / (Co 2+ + Co 3+ ) on Ag/Co-NW and particularly on Ag/Co-NW-R. This result indicates that the enhanced Ag-Co 3 O 4 interaction can generate more Co 2+ species on the catalyst surface, which commonly accompanies the generation of oxygen vacancies to generate active oxygen species [7].
In Fig. 5c, the O 1 s XPS spectra are deconvoluted into three peaks. The peak at approximately 529.4 eV corresponds to surface lattice oxygen (O lat ), and the peaks at approximately 531.0 and 532.8 eV agree with oxygen species such as O 2 − and O 2 2− adsorbed on oxygen vacancies (O ads ) [42]. In Table 1, compared with Co-NW, Co-NW-R exhibits a higher ratio of O ads /(O ads + O lat ), demonstrating that more oxygen vacancies were constructed on the surface of Co-NW-R by H 2 reduction. Notably, loading Ag greatly increases the ratio of O ads /(O ads + O lat ) on the catalyst surface, as shown in Table 1, particularly for Ag/Co-NW-R. Additionally, the O 1 s peaks of the Ag-loaded catalysts shift to higher BEs, i.e., in the direction of the O ads peaks. All of these results demonstrate that the enhanced metal-support interaction benefits the creation of surface-adsorbed oxygen species. It also coincides with the results of Co 2p XPS, soot-TPR, and CO-TPD. Figure 6 and Table 2 display the soot-TPO activities and the corresponding CO 2 concentration profiles of the as-prepared monolithic catalysts in the absence or presence of 600-ppm NO. The blank experiment without any catalysts shows that T 10 , T 50 , and the CO 2 selectivity are 490 °C, 565 °C, and 55%, respectively. For all other samples, the only product of soot oxidation is CO 2 , as listed in Table 2. Compared with the blank experiment, Ni foam can decrease the ignition temperature by 62 °C. Because of the excellent redox ability of Co 3 O 4 , Co-NW and Co-NW-R exhibit much higher catalytic soot oxidation activity than Ni foam alone. Evidently, after loading Ag NPs on them, T 50 decreases remarkably, which indicates that Ag loading greatly promotes soot oxidation. The activities of the catalysts follow the order of Ag/Co-NW-R > Ag/Co-NW > Co-NW-R > Co-NW > Ni foam > Blank with and without NO. Introducing 600-ppm NO significantly lowers T 10 and T 50 for all as-prepared catalysts but particularly for Ag/Co-NW-R. Notably, Co-NW-R exhibits only slightly higher catalytic activity than Co-NW, but the activity of Ag/Co-NW-R is much higher than that of Ag/Co-NW with and without NO. This result can be attributed to the enhanced metal-support interaction, high dispersion of Ag, and presence of more active oxygen species on Ag/Co-NW-R.

Catalytic Soot Combustion Performance
As a more effective oxidant than O 2 , NO 2 can remarkably lower the oxidation temperature of soot particulates and plays a crucial role in the soot combustion process [43]. Thus, the ability to oxidize NO to NO 2 is a key factor influencing soot oxidation. Figure 7 shows the NO-TPO profiles of the catalysts. Apparently, loading Ag significantly lowers the NO oxidation temperatures and produces more NO 2 than the support alone, particularly for Ag/Co-NW-R.

Kinetic Study
To further understand the intrinsic activity of the catalysts, the isothermal anaerobic titration processes were carried out at 280 °C. Figure 8 shows the CO 2 concentration as a function of time before and after removing O 2 from the reactant flow. Table 3 and Fig. 8 display the corresponding kinetic results for the reaction rate, amount of O*, and TOF. Obviously, these values follow the order of Ag/Co-NW-R > Ag/ Co-NW > Co-NW-R > Co-NW, which is consistent with the catalytic performance. Meanwhile, the parameters of Co-NW-R are only slightly larger than those of Co-NW, suggesting that the positive effect of the pre-reduced treatment of Co-NW on soot oxidation is limited. According to the Ag wt% measured by ICP, the reaction rate per gram of Ag on Ag/Co-NW-R is nearly 1.4-fold that on Ag/Co-NW, as shown in Table 3, demonstrating the higher utilization efficiency of Ag on Ag/Co-NW-R. Additionally, as shown in Table 3, compared with the supports, the Ag-loaded catalysts not only generate more active O* species but also achieve higher intrinsic activity (TOF), demonstrating the important effect of Ag-Co 3 O 4 interactions on soot oxidation. Here compared with Ag/Co-NW, Ag/Co-NW-R has an only slightly higher TOF but an approximately 1.4-fold higher reaction rate. This result indicates that the primary function of the Ag-Co 3 O 4 interaction is to improve the dispersion of Ag sites to increase the catalyst's apparent activity.   to further investigate the impact of water vapor on soot combustion under the loose contact conditions. In Fig. 9b, introducing water vapor in the reactant flow improves the catalytic performance of Ag/Co-NW-R, which agrees with some previous reports [44,45].

Discussion
Soot combustion is a redox reaction in nature. The active oxygen species play an important role in catalytic soot oxidation. In our work, the XPS results demonstrate that Ag species exist on the catalyst surface as Ag 0 . For the catalytic soot oxidation in a N 2 atmosphere, as considered in Fig. 3, soot can only be oxidized by active oxygen species on the catalysts. In the range of 200-300 °C, little CO 2 is produced by oxidizing soot closely contacted with active sites. Compared with Co-NW, the CO 2 peak of the Ag-loaded catalysts shifts to lower temperatures, indicating that the oxidation ability of Ag sites is higher than that of Co-NW at low temperatures, particularly for Ag/Co-NW-R. With increasing temperature (400-600 °C), a large amount of soot is oxidized to CO 2 by surface lattice oxygen species, and the difference among the as-prepared catalysts is insignificant. However, when the temperature increases to 600 °C, soot is oxidized to CO 2 by bulk lattice oxygen species of as-prepared catalysts. The Ag-loaded catalysts exhibit a larger CO 2 peak area due to the higher mobility of bulk lattice oxygen promoted by the Ag-Co 3 O 4 interaction compared with their supports. Conversely, after introducing gaseous oxygen, O 2 can be continuously activated and converted into reactive oxygen species at active sites. Compared with the reaction in the N 2 atmosphere, considered in Fig. 3, the soot oxidation performance of the catalysts improves substantially, and the shape of the CO 2 curves sharpens over the reaction temperature range (200-600 °C) in Fig. 6a, indicating a faster reaction rate in an O 2 atmosphere. Moreover, the CO 2 concentrations of the supports are quite low below 350 °C, while that of the Ag-loaded catalysts increases remarkably, demonstrating that the major function of Ag 0 sites is to activate gaseous oxygen molecules to oxidize soot. According to the results of soot-TPR, CO-TPD, O 1 s XPS, and TOF, Ag/Co-NW-R possesses the most surface oxygen species and the highest TOF value due to the enhanced metal-support interaction and thus shows the lowest ignition temperature and the highest apparent activity.
After being introduced to the reactant gas, NO is readily oxidized to NO 2 , which is a better oxidant than O 2 . Additionally, T 10 and T 50 of the as-prepared catalysts for catalytic soot oxidation continue to decrease in Fig. 6c, particularly for Ag/ Co-NW-R. According to the NO-TPO results in Fig. 7, Ag/ Co-NW-R shows the highest ability for oxidizing NO to NO 2 due to the promoted generation of active oxygen species by an enhanced metal-support interaction.
The above results indicate that the pre-reduction treatment on reducible supports effectively anchors metal sites, improves the metal dispersion, and enhances metal-support interactions, thereby increasing the number of active oxygen species and improving the intrinsic activity of catalysts for oxidation reactions.

Conclusions
In summary, in this work, we facilely synthesized Ag catalysts supported on Co 3 O 4 nanowires in situ grown on monolithic Ni foam substrates. The designed catalyst exhibits high activity, stability, and water resistance for soot oxidation. Here, the Ag catalyst exists as Ag 0 due to the automatic reduction of Ag + by Co 2+ . Interactions between Ag and Co 3 O 4 improve the generation of active oxygen species to approximately twice that on the support alone. Through the reduction pretreatment of Co 3 O 4 to generate more oxygen vacancies, Ag NPs with smaller sizes were highly dispersed and anchored thereon. In particular, this pretreatment not only increases the number of active oxygen species but also improves the intrinsic activity for soot oxidation. In addition, the macropores of the monolithic catalysts with crossed Table 3 Active oxygen (O*) amounts, reaction rates (ν, ν Ag ), and the TOF of the catalysts calculated from the results in Fig. 8 Catalysts ν ν Ag O* amount TOF (10 -7 mol/ (s·g cat )) (10 -6 mol/ (s·g Ag )) (10 -4 mol/g cat ) (10 -3 s −1 )  . 9 a Stability of Ag/Co-NW-R in 600-ppm NO/10% O 2 /N 2 and b water resistance of Ag/Co-NW-R in 600-ppm NO/10% O 2 /N 2

3
Co 3 O 4 nanowires can provide more soot-catalyst contact sites and increase the soot capturing efficiency. These advantages make the catalyst a possible candidate for industrial applications.

Supporting information
SEM images of the monolithic Ni foam (Fig. S1), the TEM-EDS mapping images of the Ag/Co-NW-R catalyst (Fig. S2), the corresponding size distribution of Ag NPs determined by HRTEM images (Fig. S3), and the average diameter (d Ag ) and dispersion (D Ag ) values of Ag NPs on monolithic catalysts (Table S1).