All chemical reagents used to be of pure analytical grade. The TiO₂ with anatase structure and γ-Al₂O₃ powder were prepared by sol-gel method. Titanium (IV) isopropoxide (Ti[OCH(CH₃)₂]₄) formed a solution after 30 min of stirring absolute ethanol. Ethylacetoacetate (Eacac/Ti = 1, molar ratio) was added in the solution to control hydrolysis and condensation reaction rates. After 30 min of stirring an aqueous solution of nitric acid (HNO₃, 0,1 M) was introduced to the mix (HNO₃/Ti = 10, molar ratio), and stirring continued until a yellow gel formed. Then, the sample was dried, and the excess solvent was removed in an autoclave under the supercritical conditions, and calcined at 500˚C with ramping rate of 1˚C∙min⁻¹ under a flow rate 30 mL/min of oxygen. The same steps were used to synthesize the Al₂O₃ with Aluminum-tri-sec-butoxide (C₁₂H₂₇AlO₃) and 1-butanol, the uncolored gel formed was dried in different condition and calcined at 600˚C. The supported catalysts Ag/Al₂O₃ and Ag/TiO₂ were prepared by impregnation with an aqueous solution of silver nitrate (AgNO₃).

BET surface area, pore volume and pore diameter of all samples were obtained from N2 adsorption isotherms using an ASAP 2020 micromeritics analyzer.

X-ray diffraction (XRD) measurements of catalysts were measured by using PANalytical X’pert PRO MPD diffractometer with a CuKα1 radiation (λ = 1.54060 Ǻ) and operated at 40 kV and 40 mA. The patterns were taken in the 2θ range of 10˚ - 70˚.

The programmed H₂ temperature reduction profiles (H₂–TPR) were carried out on chemisorption analyzer Autochem 2920 equipped with a TCD detector The TPR technique is based on the determination of hydrogen consumption for the reduction of metal oxides as a function of temperature. The temperature was in the range of 40˚C to 550˚C using a 5% H₂ /Ar gas mixture with a ramp of 5˚C/min and a flow rate of 30 ml/min.

Fourier transform infrared spectroscopy (FTIR) was used to identify types of molecular bonds in samples using the Perkin Elmer spectrometer. The samples were analyzed in the frequency range 400 - 4000 cm⁻¹ with an irradiating source in the IR medium.

The source of formaldehyde used during the tests, therefore, comes from the dissolution of paraformaldehyde at 50˚C overnight with stirring. The activity tests for the catalytic oxidation of HCHO (100 ppm) over the Ag/Al₂O₃ and Ag/TiO₂ catalysts (50 mg) were performed with a fixed-bed quartz flow reactor by passing a mixture gas of 19.5% O₂, 78% N₂, and 2.5% H₂O balance at a total flow rate of 100 cm³∙min⁻¹, within the temperature range of 50˚C - 350˚C. After reaching the steady state, the effluent gas was analyzed (MKS Multigas 2030) for HCHO conversion and CO₂ selectivity. The HCHO conversion was calculated by this equation:

HCHOConversion ( % ) = [ HCHO ] i n − [ HCHO ] o u t [ HCHO ] i n * 100

where [ HCHO ] i n and [ HCHO ] o u t are HCHO concentration at the inlet and outlet, respectively.

The apparent selectivity for carbon dioxide, are calculated as presented by the following equation:

( Pi ) Selectivity ( % ) = [ Pi ] ∑ ​ [ Pj ] * 100

where [ Pi ] is the product concentration and [ Pj ] is the sum of all reaction product concentrations.

The previous selectivity values will make it possible to calculate the yields of these products as shown on the equation:

( Pi ) Yield ( % ) = HCHOConversion * S ( Pi )

The catalytic activities and the selectivity of by-products, of pure supports (TiO₂ and Al₂O₃) and the catalysts (Ag/TiO₂ and Ag/Al₂O₃) were evaluated for the HCHO oxidation as shown in Figure 1. It is evident that the most efficient catalyst for HCHO oxidation was the Ag/TiO₂, while the worst active catalyst was pure Al₂O₃ the order of activity observed was Ag/TiO₂ >> Ag/Al₂O₃ > TiO₂ > Al₂O₃. To more understand the effect of support modification, the catalytic properties of TiO₂ and Al₂O₃ without silver were studied. The results obtained showed that the pure supports have an average activity, but cannot completely convert HCHO into CO₂ and H₂O. The main by-products found it in the outlet of gases is the CH₃OH and CO with a low selectivity of dioxide of carbon (CO₂).

Zhang and all [15] proved that TiO₂ and Al₂O₃ showed no activities for HCHO oxidation in the range of temperature 35˚C - 125˚C. The study of Hang and al [23] showed that TiO₂ and hydrogenated TiO₂ have a little activity after 200˚C. According to these data, the catalytic activity of the supports synthesized by sol-gel method is significantly important. Therefore, the supports modification by impregnation of silver on the surface, results in an increase of catalysts performance as well as the selectivity of formaldehyde oxidation to CO₂. The Ag/TiO₂ catalyst has better oxidation activity, the complete HCHO conversion to CO₂ is achieved at 95˚C. Total elimination of formaldehyde over Ag/Al₂O₃ catalyst is exhibited at 125˚C. We can conclude that the addition of Ag can provide sufficient active sites for the HCHO oxidation reaction. Note that the efficiency of the Ag/TiO₂ catalyst in the conversion of HCHO at low temperature (particularly T₅₀) is better than some literature data Table 1.

Figure 2 displays the XRD patterns of TiO₂ and Ag/TiO₂ catalysts. Diffraction peaks at 25.44˚, 38.0˚, 48.3˚, 54.43˚ and 63.4˚ correspond to (101), (004), (200), (105) and (204) and were well matched of tetragonal anatase TiO₂ (JCPDS 01-071-1166). The XRD pattern of Ag/TiO₂ showed an increase in peaks intensities with no additional peak related to the silver. These results showed that Ag

With GHSV: gas hourly space velocity.

species were widely dispersed on TiO₂ support that may have been caused by the strong interaction metal-support [25] . XRD patterns of Al₂O₃ and Ag/Al₂O₃ catalysts are shown in Figure 3. Only the phase γ-Al₂O₃ was detected Al₂O₃ [26] . As for the Ag/Al₂O₃ the diffraction peaks of Ag₂O at 38.2˚ and 64.5˚ were observed, which assigned as (111) and (220) crystal planes. In addition, a peak of Ag was detected at 44.3˚ corresponded to the (200) lattice plane, which could be indexed to face-centered cubic phase [27] . The crystallite sizes of the synthesized catalysts were calculated using the Debye-Scherrer equation and are presented in Table 2.

^aDetermined by BET, ^bDetermined by XRD.

The N₂ adsorption-desorption isotherms of Al₂O₃, Ag/Al₂O₃, TiO₂ and Ag/TiO₂ samples are presented in Figure 4(a) and Figure 4(b) show the isotherms of Al₂O₃ support and Ag/Al₂O₃ catalyst. Both solids belonged to type IV with a type H3 hysteresis, typically indicates the presence of closely packed slit pores [28] . As shown, in Figure 4(b) the isotherm of TiO₂ and Ag/TiO₂ samples resembled the type IV curves with H2 hysteresis loop, corresponding to mesoporous textures and in-bottle shaped. The Barrett-Joyner-Halenda (BJH) pore size distribution (Figure 5(a), Figure 5(b)) Al₂O₃, Ag/Al₂O₃, TiO₂ and Ag/TiO₂ catalysts indicating that all samples characterized by a monomodal size distribution. The Brunauer-Emmett-Teller (BET) surface area, pore volume and average pore diameter obtained from N₂ adsorption-desorption measurement were showed in Table 2. The surface area of both TiO₂ and γ-Al₂O₃ supports gradually decreased after silver impregnation.

The optical analysis of FT-IR was used to determine the functional groups and vibration bonds in the samples in the range of 4000 - 400 cm⁻¹. Figure 6 shows the FT-IR of Al₂O₃ support and Ag/ Al₂O₃ catalyst. The absorption peak at 3480 cm⁻¹ is related to the vibration bonding of the hydroxyl groups OH. The peak of water molecules is formed at 1630 cm⁻¹. The vibrations peaks generated in the range of 700 - 500 cm⁻¹ correspond to Al-O-Al, which proves the existence of γ-form [29] . Figure 7 presents the FT-IR spectra of TiO₂ and Ag/TiO₂. The adsorption band in the range of 3800 - 3000 cm⁻¹ are assigned to the stretching vibrations of OH. The peak absorption at 1620 cm⁻¹ is ascribed to the adsorbed water molecules. The observed peaks at the range of 900 - 450 cm⁻¹ is associated with the vibration of Ti-O-Ti [30] [31] .

Transmission electron micrographs (TEM) of Ag/Al₂O₃ and Ag/TiO₂ catalysts are shown in Figure 8 and Figure 9. The image of Ag/Al₂O₃ showed a distribution of Ag particles with nonuniform shapes, the silver present in the form of huge agglomerates composed of large particle sizes (Table 1). In contrast, in Ag/TiO₂ it can be observed that the Ag particle sizes are much smaller, and the distribution was relatively uniform. The addition of silver in the surface of TiO₂ support caused some black spots with a hexagonal structure. The results of TEM indicate that the impregnation of silver is very affected by the support used.

H₂-TPR experiments were conducted to investigate the redox ability of Ag/Al₂O₃ and Ag/TiO₂ catalysts and the results are shown in Figure 10. No reduction peaks observed in the profiles reported for the pure supports TiO₂ and γ-Al₂O₃. Two peaks were detected for Ag/Al₂O₃, peak around 100˚C caused by large Ag₂O clusters and the peak around 350˚C is assigned to small Ag₂O clusters [32] . For Ag/TiO₂ the peak below 100˚C was ascribed to reduction of oxygen species absorbed on the dispersed Ag surface, and the second peak was due to the reduction of large Ag₂O groups [27] .