Effect of Zirconia Polymorph on Vapor-Phase Ketonization of Propionic Acid

catalysts

Article Eﬀect of Zirconia Polymorph on Vapor-Phase Ketonization of Propionic Acid

Shuang Ding *, Jiankang Zhao and Qiang Yu Collaborative Innovation Center of Chemical Science and Engineering, School of Chemical Engineering and Technology, Tianjin University, Tianjin 300072, China; [email protected] (J.Z.); [email protected] (Q.Y.) * Correspondence: [email protected]

Received: 26 August 2019; Accepted: 11 September 2019; Published: 13 September 2019

Abstract: Vapor-phase ketonization of propionic acid derived from biomass was studied at 300–375 ◦C over ZrO2 with different zirconia polymorph. The tetragonal ZrO2 (t-ZrO2) are more active than monoclinic ZrO2 (m-ZrO2). The results of characterizations from X-ray diffraction (XRD) and Raman suggest m-ZrO2 and t-ZrO2 are synthesized by the solvothermal method. NH3 and CO2 temperature-programmed desorption (NH3-TPD and CO2-TPD) measurements show that there were more medium-strength Lewis acid base sites with lower coordination exposed on m-ZrO2 relative to t-ZrO2, increasing the adsorption strength of propionic acid. The in situ DRIFTS (Diffuse reflectance infrared Fourier transform spectroscopy) of adsorbed propionic acid under ketonization reaction reveal that as the most abundant surface intermediates, the monodentate propionates are more active than bidentate propionates. In comparison with m-ZrO2, the t-ZrO2 surface favors monodentate adsorption over bidentate adsorption. Additionally, the adsorption strength of monodentate propionate is weaker on t-ZrO2. These differences in adsorption configuration and adsorption strength of propionic acid are affected by the zirconia structure. The higher surface concentration and weaker adsorption strength of monodentate propionates contribute to the higher ketonization rate in the steady state.

Keywords: monoclinic ZrO2; tetragonal ZrO2; ketonization; propionic acid; in situ infrared spectroscopy

1. Introduction The depletion of fossil fuels and the increase in environmental issues are leading researchers to search for alternative energy sources [1–3]. Thus, investigations of biomass have been widely carried out. Biomass can be easily converted to bio-oil via fast pyrolysis [1,4]. Carboxylic acid constitutes an important component of bio-oil, which possesses acidity and instability. Therefore, during the upgrading of bio-oil, the efficient conversion of carboxylic acid is an important step in reducing corrosiveness and enhance stability. In comparison to other reactions of carboxylic acid [5,6], their ketonization is an attractive one due to there being no H2 consumption. This leads to two carboxylic acid molecules achieving C–C coupling, dehydration and decarboxylation, forming ketone, water and CO2. In addition, the product of ketone, possessing higher energy content, can be further converted into longer chain hydrocarbons via aldol condensation [7,8]. The ketonization reaction can be performed on two types of catalysts, metal oxide [9–14] and zeolite [15,16], in the vapor phase and aqueous phase. The amphoteric metal oxides show higher catalytic activity, such as MnO2, TiO2, ZrO2, CeO2, etc. [12,17–22]. The reaction activity of carboxylic acid ketonization is strongly influenced by the surface structure of metal oxides [23–27]. Kim et al. performed carboxylic acid reaction on the TiO2 single-crystal surface [23]. The carboxylic acid was converted to produce ketene through unimolecular dehydration on the (011) surface and to form ketone via bimolecular reaction on the (114) surface. They attributed this difference of products to the different

Catalysts 2019, 9, 768; doi:10.3390/catal9090768 www.mdpi.com/journal/catalysts Catalysts 2019, 9, 768 2 of 15 coordination of Ti4+ cations exposed on the surface. Stubenrauch et al. studied the reaction of formic acid and acetic acid on CeO2 (111) and (100) surface [24]. On both surfaces, CH2CO was produced through acetic acid dehydration near 600 K. However, acetone was detected only on the (111) surface during acetic acid decomposition at 600 K. This result is different from that obtained by Kim et al., since Ce4+ cations of (111) surface possess one vacancy relative to the Ce4+ cations in the bulk. Wang et al. carried out ketonization of carboxylic acid on anatase and rutile TiO2, monoclinic and tetragonal ZrO2 at 503–533 K [25,26]. They showed that the reaction activity of acetic acid was higher over anatase TiO2 than that over rutile TiO2, and attributed that to the more reactive monodentate acetate present on anatase TiO2, whereas unreactive bidentate acetate was present on rutile TiO2. The effect of different nanocrystals (nanocubes, nanorods and nanopolyhedra) of CeO2 on acetic acid ketonization was studied by Snell et al. at 503 K in the condensed phase and at 623 K in the vapor phase [27]. They found that the crystal of CeO2 was disrupted in the condensed phase with the formation of metal carboxylate, whereas the bulk structure was maintained during the vapor phase reaction. They suggested that the morphology of CeO2 did not have a critical influence on the reaction activity in the condensed and vapor phase. Although some work has been done on the influence of the structure, no consensus has been reached. This may be due to the different reaction conditions (condensed phase and vapor phase) via the different reaction mechanisms or different metal oxides used. As the higher activity catalyst in the ketonization of carboxylic acid, ZrO2 and the modified ZrO2 catalysts have been studied using experimental measurements and density functional theory calculations (DFT) [28–35]. Wu et al. performed acetic acid ketonizaion over carbon promoted ZrO2 catalysts and on the Zr/Mn mixed oxides at 340 ◦C in the aqueous-phase ketonization [28,29]. They found that carbon promoted ZrO2 with tetragonal phase was more active than that possessing monoclinic phase, and the different carbon species and carbon content resulted in different stability. They suggested that carbonization degree leading to the crystal phase transformation between tetragonal ZrO2 (t-ZrO2) and monoclinic ZrO2 (m-ZrO2), and the interaction strength between carbons species and ZrO2 both influenced the reactivity. During the study of acetic acid ketonization on the Zr/Mn mixed oxides, they suggested that the incorporation of Mn resulted in monoclinic phase disappearing and acidity improved accompanying the formation of t-ZrO2/MnOx solid solution, which increased ketonization activity. Lopez-Ruiz et al. reported that the catalytic activity and stability of LaxZryOz were enhanced relative to ZrO2 for acetic acid ketonization at 568 K in the aqueous condensed phase [30]. Under the condensed-phase hydrothermal reaction conditions, ZrO2 and LaxZryOz catalysts could be restructured. ZrO2 possessed a mixed phase, with monoclinic and tetragonal; whereas LaxZryOz displayed a tetragonal phase. This indicated that the improvement of stability was caused by the phase transformation, and the higher activity was due to the greater acetic acid surface concentration. The study of carboxylic acid ketonization over ZrO2 and the modified ZrO2 catalysts was also performed in the vapor phase [26,31,34,36,37]. Parida et al. studied acetic acid ketonization over alkali-metal cations promoted ZrO2 at 623–698 K [31]. They reported that Na was the most effective promoter, probably due to the crystal phase conversion from m-ZrO2 to t-ZrO2 through adding Na. Shutilov et al. reported that 10 wt% CeO2 supported on ZrO2 showed the highest activity of pentanoic acid ketonization at 355 ◦C among 5–20 wt% CeO2/ ZrO2 [36]. The modification by CeO2 led to the formation of Ce-Zr solid solution based on t-ZrO2. They proposed that all acid sites and lattice oxygen atoms played a role in the reaction. Recently, we investigated propionic acid ketonization on CexZr1–xO2 mixed oxides at 270–350 ◦C through adjusting the Ce/Zr molar ratios [37]. The results showed that the reaction rate reached maximum at a medium-strength acid-base ratio of ~1.07 on Ce0.1Zr0.9O2 with tetragonal phase, suggesting that the balanced medium-strength acid-base sites was an important factor in the ketonization. Analysis of the above studies indicates that ZrO2 (and the modified ZrO2) based on the tetragonal phase is more active than that with the monoclinic phase during carboxylic acid ketonization. However, the structure-activity relationship for ZrO2 is unclear. Hence, in this work, the effect of zirconia polymorph on propionic acid ketonization in the vapor phase has been studied. Monoclinic and Catalysts 2019, 9, x FOR PEER REVIEW 3 of 15 zirconiaCatalysts 2019 polymorph, 9, 768 on propionic acid ketonization in the vapor phase has been studied. Monoclinic3 of 15 and tetragonal zirconia were prepared by solvothermal method using water and methanol as solvents, respectively, and tested for ketonization of propionic acid at 300–375 °C. A series of tetragonal zirconia were prepared by solvothermal method using water and methanol as solvents, characterizations demonstrated that two zirconia morphologies were formed. The greater number of respectively, and tested for ketonization of propionic acid at 300–375 ◦C. A series of characterizations medium-strength Lewis acid-base sites with lower coordination exposed on the m-ZrO2 surface are demonstrated that two zirconia morphologies were formed. The greater number of medium-strength in contrast with the values obtained for t-ZrO2. The in situ DRIFTS showed that the monodentate Lewis acid-base sites with lower coordination exposed on the m-ZrO surface are in contrast with the propionates are much more active than the bidentate propionates during2 ketonization. In comparison values obtained for t-ZrO2. The in situ DRIFTS showed that the monodentate propionates are much with m-ZrO2, t-ZrO2 favored monodentate adsorption over bidentate adsorption, and weakened more active than the bidentate propionates during ketonization. In comparison with m-ZrO , t-ZrO adsorption strength of monodentate propionates, which led to the ketonization rate of propionic2 acid2 favored monodentate adsorption over bidentate adsorption, and weakened adsorption strength of enhanced on t-ZrO2. monodentate propionates, which led to the ketonization rate of propionic acid enhanced on t-ZrO2.

2. Results Results and and Discussion Discussion

2.1. Catalyst Catalyst Characterizations Characterizations

2.1.1. Structure Structure Identification Identiﬁcation

The XRD patterns of ZrO 2 areare shown shown in in Figure Figure 11A.A. The main peaks at 24.2, 28.2, 31.4, 40.8,40.8, 45.0,45.0, 50.0 and 55.6° 55.6◦ are assigned to the monoclinic phase of ZrO 2 (m(m-ZrO-ZrO2,, JCPDS JCPDS cards cards no. no. 37 37-1484)-1484) [38,39] [38,39].. While the four peaks at 30.2, 35.2, 50.4 and 60.060.0°◦ belong to the tetragonal phase of ZrO22 (t(t-ZrO-ZrO2,, JCPDS JCPDS cards no. no. 17 17-0923).-0923). The The crystallite crystallite size size and and lattice lattice distance distance were were both bothmeasured measured by XRD, by XRD,listed listedin Table in 1.Table The1 .calculation The calculation results results using using the Scherrer the Scherrer equation equation show show that thatthe thecrystallite crystallite size size of m of-ZrO m-ZrO2 is 29.5is nm,9.5 nm, which which is larger is larger than than that that of oft-ZrO t-ZrO2 (8.42 (8.4 nm). nm). The The lattice lattice distance distance is is 0.3164 0.3164 nm nm and and 0.2987 0.2987 nm, nm, corresponding to the (-111)(-111) plane of m-ZrOm-ZrO2 and the (111) plane of t-ZrOt-ZrO2,, respectively. respectively. Considering Considering the similar diffractograms diﬀractograms between the tetragonal and cubic zirconiazirconia phasesphases (t-ZrO(t-ZrO22 or c-ZrOc-ZrO2),), the Raman spectra of ZrO 2 werewere performed to to dist distinguishinguish their their structures, structures, as as shown shown in in Figure 11B.B. TheThe Raman spectra ofof ZrOZrO22 showsshows vibration vibration bands bands at at 105, 105, 179, 179, 188, 188, 222, 222, 307, 307, 334, 334, 344, 344, 381, 381, 480, 480, 557, 557, 617, 617, and 1 and632 cm632− cm, which−1, which are assignedare assigned to the to m-ZrOthe m-ZrO2 [392, 40[39,40]]. Meanwhile,. Meanwhile, the vibrationthe vibration bands bands at 150, at 150, 272, 272, 315, 1 346215, and462 642and cm642− cmbelong−1 belong to t-ZrO to t-2ZrO[392, 40[39,40]]. The. The XRD XRD and and Raman Raman characterizations characterizations both both conﬁrm confirm that thatm-ZrO m2-ZrOand2 t-ZrOand 2 t-haveZrO2 beenhave synthesized been synthesized using solvothermal using solvothermal method. In method. addition, In the addition, polymorph the polymorphof two catalysts of two presented catalysts no presented changes no after changes propionic after acid propionic ketonization acid ke attonization 350 ◦C for at 1 350 h (Figure °C for S1).1 h (FigureThe BET S1). surface The areaBET andsurface pore area volume and ofpore ZrO volume2 were measuredof ZrO2 were by themeasured N2 adsorption-desorption. by the N2 adsorption As- desorption.shown in Figure As shown S2, the in N Figure2 sorption S2, the isotherms N2 sorption correspond isotherms to thecorrespond type IV, suggestingto the type theIV, stackingsuggesting of 2 thecrystallites stacking results of crystallites in the formation results in of the mesoporous. formation of The mesoporous. surface area The of m-ZrO surface2 (103area mof /mg)-ZrO is smaller2 (103 2 mthan2/g) that is smaller of t-ZrO than2 (127 that m of/g), t-ZrO which2 (127 is consistent m2/g), which with is the consistent larger crystallite with the sizelarger of m-ZrOcrystallite2. size of m-ZrO2.

(A) (B) 480 617 (-111) 179 334 188 344 557 632 (111) (-121) (022) 105 222 307 381 (110) (020) (-202) (-311) m-ZrO2

m-ZrO2

Intensity (a.u.) Intensity (a.u.) Intensity

t-ZrO 2 t-ZrO2 (111) (200) (220) (311) 150 272 315 462 642 20 30 40 50 60 70 80 100 200 300 400 500 600 700 800 o -1 2 ( ) Shift (cm )

Figure 1. XX-ray-ray diffraction diﬀraction patterns of ZrO22 (A), and Raman spectra of ZrO 2 ((BB).).

Catalysts 2019, 9, x FOR PEER REVIEW 4 of 15

Catalysts 2019, 9, 768 4 of 15 Table 1. BET surface area, pore volumes and crystal structure of ZrO2.

Surface area Pore volume Lattice spacing a Crystallite size a Catalyst Table 1. BET surface area, pore volumes and crystal structure of ZrO2. 2 3 Surface(m /g) Area Pore(cm Volume/g) Lattice(nm) Spacing a Crystallite(nm) Size a Catalyst (m2/g) (cm3/g) (nm) (nm) m-ZrO2 103 0.307 0.3164 9.5 m-ZrO2 103 0.307 0.3164 9.5 t-ZrO 127 0.160 0.2957 8.4 t-ZrO2 127 0.160 0.2957 8.4 a The calculations of lattice distance and crystallite size were carried out by using peaks of (-111) for m-ZrO2 a The calculations of lattice distance and crystallite size were carried out by using peaks of (-111) for (2θ = 28.2◦, monoclinic phase), and (111) for t-ZrO2 (2θ = 30.2◦, tetragonal phase). m-ZrO2 (2θ = 28.2°, monoclinic phase), and (111) for t-ZrO2 (2θ = 30.2°, tetragonal phase). 2.1.2. Surface Properties 2.1.2. Surface Properties The surface chemistry properties were investigated by XPS analysis. Figure2 shows the XPS The surface chemistry properties were investigated by XPS analysis. Figure 2 shows the XPS spectra of Zr 3d and O 1s regions of ZrO2. The Zr 3d spectra exhibit two peaks of Zr 3d3/2 and spectra of Zr 3d and O 1s regions of ZrO2. The Zr 3d spectra exhibit two peaks of Zr 3d3/2 and 3d5/2 3d5/2 [30,41]. For t-ZrO2, the binding energy at Zr 3d3/2 and 3d5/2 are 184.18 and 181.74 eV, respectively, [30,41]. For t-ZrO2, the binding energy at Zr 3d3/2 and 3d5/2 are 184.18 and 181.74 eV, respectively, which both shift slightly to the higher binding energy by 0.14 eV compared to that of m-ZrO2. This which both shift slightly to the higher binding energy by 0.14 eV compared to that of m-ZrO2. This result indicates that the electrons become deﬁcient on t-ZrO2. The O 1s spectra were ﬁtted by two peaks.result Theindicates lower that binding the energyelectrons at ~529.9become eV deficient is ascribed on tot-ZrO the lattice2. The oxygenO 1s spectra (O-I), whereaswere fitted the by higher two bindingpeaks. The energy lower at ~531.9 binding eV energy is attributed at ~529.9 to the eV oxygen is ascribed ions withto the lower lattice coordination oxygen (O or-I), hydroxyl-like whereas the higher binding energy at ~531.9 eV is attributed to the oxygen ions with lower coordination or groups (O-II) [41,42]. The area ratios of O-II and O-I are similar, and are 0.16 on m-ZrO2 and 0.18 hydroxyl-like groups (O-II) [41,42]. The area ratios of O-II and O-I are similar, and are 0.16 on m-ZrO2 on t-ZrO2. and 0.18 on t-ZrO2. O-I (A) Zr 3d 3d5/2 (B) O 1s 181.60 529.87 3d3/2 184.04 O-II 531.89

m-ZrO2 m-ZrO2 181.74 529.90

184.18 Counts per second per Counts

531.90 t-ZrO t-ZrO2 2 188 186 184 182 180 178 176 536 534 532 530 528 526 Binding energy (eV) Binding energy (eV)

FigureFigure 2. 2.XPS XPS spectraspectra ofof ZrOZrO22.(. (AA)) ZrZr 3d3d andand ((BB)) OO 1s. 1s. DashedDashed line,line, experimentalexperimental data;data; solidsolid lines,lines, curvecurve ﬁttings. fittings.

TheThe surface surface acidity acidity and and basicity basicity of of ZrO ZrO22were were characterizedcharacterized using using DRIFTS DRIFTS of of pyridine pyridine adsorption, adsorption, temperature-programmedtemperature-programmed desorption desorption of of NH NH3 3and and CO CO2 2(NH (NH33-TPD-TPD andand COCO22-TPD).-TPD). FigureFigure S3S3 displaysdisplays thatthat the the DRIFTS DRIFTS of of pyridine pyridine adsorption adsorption on on m-ZrO m-ZrO2 2is is similar similar to to that that on on t-ZrO t-ZrO22.. TheThe bandsbands atat 16041604 1575,1575, 1 4+ 1487,1487, andand 1442 1442 cm cm−−1 areare assignedassigned toto thethe pyridinepyridine adsorbedadsorbed onon the the coordinatively coordinatively unsaturated unsaturated Zr Zr4+ 1 cationscations (Lewis (Lewis acid acid sites), sites), the the band band at at 1592 1592 cm cm−−1 isis attributedattributed toto thethe pyridinepyridine adsorbedadsorbed on on thethe hydrogenhydrogen ofof hydroxylhydroxyl group group [ 43[43,44],44]..It It shouldshould bebe notednoted thatthat thethe bandsbands correspondingcorrespondingto tothe theBrønsted Brønstedacid acidsites sites areare notnot observedobserved onon the the surface surface of of ZrO ZrO2.2. Thus,Thus, thethe resultsresults ofof DRIFTSDRIFTS ofof pyridinepyridine adsorptionadsorption implyimply therethere are are only only Lewis Lewis acid acid sites sites present present on on the the surface surface of of m-ZrO m-ZrO2 2and and t-ZrO t-ZrO22.. NHNH33-TPD-TPD was performed to further further investigate investigate the the acidity acidity of of ZrO ZrO2, 2displayed, displayed in inFigure Figure 3A.3A. In Inaddition, addition, the the densities densities of of acid acid sites sites were were calculated calculated a andnd listed listed in in Table 2 2.. TheThe acidityacidity of of Lewis Lewis acid acid sitessites is is dependent dependent on on their their coordination. coordination. According According to theto the broad broad and and asymmetric asymmetric features features of NH of3 -TPDNH3- profiles,TPD profiles, the profiles the profiles were fittedwere fitted into three into three peaks. peaks. In accordance In accordance with with previous previous studies studies [39,45 [39,45]], the, centersthe center of desorptions of desorption peak peak adsorbed adsorbed on the on weak the acidweak site, acid the site, medium the medium strength strength acid site acid and site the and strong the acid site are below 200 ◦C, between 200 and 400 ◦C, and above 400 ◦C, respectively. The densities of

Catalysts 2019, 9, x FOR PEER REVIEW 5 of 15 strong acid site are below 200 °C, between 200 and 400 °C, and above 400 °C, respectively. The densities of weak acid sites are same on the two ZrO2 catalysts. There are no strong acid sites present on the two catalysts surface. The third desorption peak shifts slightly from 341 °C on m-ZrO2 to 347 °C on t-ZrO2, which corresponds to the adsorption on the lower coordination Zr4+ cations assigned to the medium strength acid site. This implies that the acid strength of the medium strength acid site increases slightly on t-ZrO2, which is consistent with the change of binding energies of Zr 3d. However, the intensity of that peak on t-ZrO2 is much lower than that on m-ZrO2, suggesting the more Zr4+ cations with lower coordination existence on m-ZrO2. CO2-TPD was used to probe the basicity of ZrO2. The profiles of CO2 desorption and densities of base sites calculated are shown in Figure 3B and Table 2. For metal oxides, the coordinatively unsaturated oxygen ions act as base sites, whose basicity is also related to the coordination [26]. The profiles were divided into three regions according to the desorption temperature. The desorption peak (<200 °C) is assigned to CO2 adsorption on the weak base site (surface -OH group) with the formation of bicarbonate, the desorption peak (200 ~ 400 °C) is attributed to CO2 adsorption on the mediumCatalysts 2019-strength, 9, 768 base site (Mx+-O2− pair) along with the formation of bidentate carbonate, and5 ofthe 15 desorption peak (>400 °C) is ascribed to CO2 adsorption on the strong base site (low-coordination O2−) accompanying the formation of unidentate or polydentate carbonate [34,45]. Few strong base sites weak acid sites are same on the two ZrO2 catalysts. There are no strong acid sites present on the two are present on the catalysts, 0.02 μmol/m2 on m-ZrO2 and 0.03 μmol/m2 on t-ZrO2. In contrast with t- catalysts surface. The third desorption peak shifts slightly from 341 ◦C on m-ZrO2 to 347 ◦C on t-ZrO2, ZrOwhich2, on corresponds the surface to of the m adsorption-ZrO2, the concentration on the lowercoordination of weak base Zr sites4+ cations is smaller, assigned but tothe the amount medium of mediumstrength- acidstrength site. base This impliessites is higher. that the This acid result strength indicates of the mediumthat a greater strength number acid siteof low increases-coordination slightly O2− ions of Mx+-O2− pairs is exposed on the m-ZrO2 surface. NH3-TPD and CO2-TPD suggest that there on t-ZrO2, which is consistent with the change of binding energies of Zr 3d. However, the intensity of are more medium-strength Lewis acid base sites with lower coordination exposed4+ on m-ZrO2 than that peak on t-ZrO2 is much lower than that on m-ZrO2, suggesting the more Zr cations with lower on t-ZrO2, which is related to the different structures of m-ZrO2 and t-ZrO2. In addition, those sites coordination existence on m-ZrO2. possess stronger acid-base properties, and thus strengthen propionic acid adsorption [26].

(A) 0.1 (B) 0.1

341 m-ZrO

m-ZrO2 2 MS(a.u.) Signal

347 t-ZrO2 t-ZrO2

100 200 300 400 500 600 100 200 300 400 500 600 o o Temperature ( C) Temperature ( C)

Figure 3. NHNH33--TPDTPD profiles profiles ( (AA)) and and CO CO22--TPDTPD profiles profiles ( (BB)) of of ZrO ZrO22.. Solid Solid line, line, experimental experimental data; data; short short dash lines, curve fittings. fittings.

Table 2. Quantiﬁcation of acid sites and base sites. Table 2. Quantification of acid sites and base sites. Acid Site a (µmol/m2) Base Site b (µmol/m2) Catalysts Acid site a (μmol/m2) Base site b (μmol/m2) Catalysts Total Weak Medium Total Weak Medium Strong

m-ZrO2 2.21Total Weak 0.79 Medium 1.42 1.55Total Weak 0.49 Medium 1.03 Strong 0.02 t-ZrO2 1.89 0.74 1.15 1.45 0.61 0.81 0.03 m-ZrO2 2.21 0.79 1.42 1.55 0.49 1.03 0.02 a b Derived from NH3-TPD. Derived from CO2-TPD. t-ZrO2 1.89 0.74 1.15 1.45 0.61 0.81 0.03

a Derived from NH3-TPD. b Derived from CO2-TPD. CO2-TPD was used to probe the basicity of ZrO2. The profiles of CO2 desorption and densities of base sites calculated are shown in Figure3B and Table2. For metal oxides, the coordinatively 2.2. Catalytic Performance unsaturated oxygen ions act as base sites, whose basicity is also related to the coordination [26]. The profiles were divided into three regions according to the desorption temperature. The desorption peak (<200 ◦C) is assigned to CO2 adsorption on the weak base site (surface -OH group) with the formation of bicarbonate, the desorption peak (200 ~ 400 ◦C) is attributed to CO2 adsorption on the x+ 2 medium-strength base site (M -O − pair) along with the formation of bidentate carbonate, and the 2 desorption peak (>400 ◦C) is ascribed to CO2 adsorption on the strong base site (low-coordination O −) accompanying the formation of unidentate or polydentate carbonate [34,45]. Few strong base sites 2 2 are present on the catalysts, 0.02 µmol/m on m-ZrO2 and 0.03 µmol/m on t-ZrO2. In contrast with t-ZrO2, on the surface of m-ZrO2, the concentration of weak base sites is smaller, but the amount of medium-strength base sites is higher. This result indicates that a greater number of low-coordination 2 x+ 2 O − ions of M -O − pairs is exposed on the m-ZrO2 surface. NH3-TPD and CO2-TPD suggest that there are more medium-strength Lewis acid base sites with lower coordination exposed on m-ZrO2 than on t-ZrO2, which is related to the different structures of m-ZrO2 and t-ZrO2. In addition, those sites possess stronger acid-base properties, and thus strengthen propionic acid adsorption [26]. Catalysts 2019, 9, 768 6 of 15

Catalysts 2019, 9, x FOR PEER REVIEW 6 of 15 Catalysts2.2. Catalytic 2019, 9, x Performance FOR PEER REVIEW 6 of 15

TheThe catalytic activity activity of of m m-ZrO-ZrO2 2andand t-ZrO t-ZrO2 for2 for propionic propionic acid acid ketonization ketonization were were studied studied at 300 at– The catalytic activity of m-ZrO2 and t-ZrO2 for propionic acid ketonization were studied at 300– 300–375375 °C and◦C andatmospheric atmospheric pressure pressure in the in the integral integral fixed fixed-bed-bed reactor reactor with with the the same same space space time time (W/F, (W/F, 375 °C and atmospheric pressure in the integral fixed-bed reactor with the−11 same space time (W/F, defineddefined asas ratioratio ofof catalystcatalyst weightweight andand propionic propionic acid acid flow flow rate, rate, g catgcatg·gfeedfeed− ·h)h) of of 0.05 0.05 h, as shown in defined as ratio of catalyst weight and propionic acid flow rate, gcat·g· feed−1·h)· of 0.05 h, as shown in FigureFigure4 4.. The The conversion conversion of of propionic propionic acid acid is is enhanced enhanced with with increasing increasing the the temperature temperature from from 300 300C °C to Figure 4. The conversion of propionic acid is enhanced with increasing the temperature from 300◦ °C 375to 375C °C (Figure (Figure4A). 4A). The The main main product product is is 3-pentanone, 3-pentanone, the the selectivity selectivity of of which which is is higher higher than than 97.5% 97.5% to 375◦ °C (Figure 4A). The main product is 3-pentanone, the selectivity of which is higher than 97.5% onon the two catalysts (Figure4 4B),B), and and thethe minorminor productsproducts includeinclude methylketene,methylketene, propionicpropionic anhydrideanhydride on the two catalysts (Figure 4B), and the minor products include methylketene, propionic anhydride andand propanal,propanal, inin accordanceaccordance withwith others’others’ reportsreports onon ZrOZrO22 [[25,26]25,26].. The surface surface-based-based intrinsic reaction and propanal, in accordance with others’ reports on ZrO2 [25,26]. The surface-based intrinsic reaction raterate ofof propionicpropionic acidacid isis higherhigher onon t-ZrOt-ZrO22 thanthan thatthat onon m-ZrOm-ZrO22 in the temperature rangerange ofof 300–375300–375 ◦°CC rate of propionic acid is higher on t-ZrO2 than that on m-ZrO2 in the temperature range of 300–375 °C (Figure(Figure5 5),), which which is is consistent consistent with with previous previous studies studies on on acetic acetic acid acid ketonization ketonization [ 28[28,31],31]. . (Figure 5), which is consistent with previous studies on acetic acid ketonization [28,31]. 60 (A) (B) 60 m-ZrO 2 (A) 100 (B) m-ZrO 50 t-ZrO2 100 50 t-ZrO 2 2 90 40 90 40 80 30 80 30 20 70 20 70 m-ZrO 10 60 2 m-ZrO2 Selectivity to 3-pentanone (%) 3-pentanone to Selectivity 60

10 t-ZrO2

Conversion of propionic acid (%) acid propionic of Conversion Selectivity to 3-pentanone (%) 3-pentanone to Selectivity t-ZrO2 Conversion of propionic acid (%) acid propionic of Conversion 0 50 0 300 320 340 360 380 50 300 320 340 360 380 300 320Temperature340 (o360C) 380 300 320Temperature340 (oC)360 380 o o Temperature ( C) Temperature ( C)

FigureFigure 4. ReactionReaction activity activity of of propionic propionic acid acid ketonization ketonization on onZrO ZrO2 as2 aas function a function of temperature. of temperature. (A) Figure 4. Reaction activity of propionic acid ketonization on ZrO2 as a function of temperature. (A) (ConversionA) Conversion of propionic of propionic acid; acid; (B) (selectivityB) selectivity to 3 to-pentanone. 3-pentanone. Reaction Reaction conditions: conditions: Pacid P =acid 3.9= kPa,3.9 kPa,Ptotal Conversion of propionic acid; (B) selectivity to 3-pentanone. Reaction conditions: Pacid = 3.9 kPa, Ptotal P=total 101.325= 101.325 kPa, W/F kPa, = W 0.05/F = h,0.05 Ar/propionic h, Ar/propionic acid acid= 25,= time25, time on stream on stream is 30 is min 30 min for foreach each temperature. temperature. = 101.325 kPa, W/F = 0.05 h, Ar/propionic acid = 25, time on stream is 30 min for each temperature.

o 2.1 300o C 2.1 300 Co ) 325 C

-2 o

) 325 Co cat -2 1.8

350o C cat m 1.8

· 350 Co m

-1 375 C

· o

-1 1.5 375 C min

· 1.5 min · 1.2 1.2 0.9 0.9 0.6 0.6 0.3

0.3 Ketonizationrate (mmol

Ketonizationrate (mmol 0.0 0.0 m-ZrO t-ZrO m-ZrO 2 t-ZrO 2 2 Catalysts 2 Catalysts FigureFigure 5. TheThe area area-based-based intrinsic intrinsic ketonization ketonization rate rate on on ZrO ZrO2. Reaction2. Reaction conditions: conditions: T = T300=–300–375375 °C, P◦totalC, Figure 5. The area-based intrinsic ketonization rate on ZrO2. Reaction conditions: T = 300–375 °C, Ptotal P=total 101.325= 101.325 kPa, P kPa,acid = P3.9acid kPa,= 3.9 Ar/Propionic kPa, Ar/Propionic acid = acid25, Time= 25, on Time stream on stream is 30 min, is 30 the min, conversion the conversion is < 18% is = 101.325 kPa, Pacid = 3.9 kPa, Ar/Propionic acid = 25, Time on stream is 30 min, the conversion is < 18%

Catalysts 2019, 9, 768 7 of 15

100.3 kJ/mol on t-ZrO2, which is consistent with a previous report of 117 kJ/mol on m-ZrO2 [18] and 103–109 kJ/mol on ZrO2 with mixed phase [21]. Please note that the variation trend of Ea is in accordance with that of propionic acid ketonization rate, where the higher Ea causes a lower Catalystsketonization 2019, 9,rate. x FOR PEER REVIEW 7 of 15

Temperatrue (oC) 380 360 340 320 300

5.6 )]

-1 4.8

cat

· -1

min 4.0 ·

3.2

Ln [r(mmol Ln m-ZrO : 124.2 kJ/mol 2.4 2 t-ZrO2: 100.3 kJ/mol

0.184 0.190 0.196 0.203 0.210 -1 1000/T (K ) Figure 6. Arrhenius plots of propionic acid ketonization on ZrO . Reaction conditions: T = 300–370 C, Figure 6. Arrhenius plots of propionic acid ketonization on ZrO22. Reaction conditions: T = 300–370 °C,◦ P = 101.325 kPa, P = 3.9 kPa, Ar/Propionic acid = 25, time on stream is 30 min, the conversion is Ptotal = 101.325 kPa, Pacidacid = 3.9 kPa, Ar/Propionic acid = 25, time on stream is 30 min, the conversion is < 18% by adjusting the space time (W/F). < 18% by adjusting the space time (W/F).

2.3. DRIFTS Study of Propionic Acid and 3-Pentanone Adsorption on the Surface of ZrO2 2.3. DRIFTS Study of Propionic Acid and 3-Pentanone Adsorption on the Surface of ZrO2 To elucidate the influence factors on the different reaction activity of propionic acid over m-ZrO2 To elucidate the influence factors on the different reaction activity of propionic acid over m-ZrO2 and t-ZrO2, the propionic acid adsorption was investigated under reaction conditions using DRIFTS. and t-ZrO2, the propionic acid adsorption was investigated under reaction conditions1 using DRIFTS. Figure7 and Figure S4 display the spectra recorded. The peaks at 1789 and 1769 cm − are assigned to Figure 7 and Figure S4 display the spectra recorded. The peaks at 1789 and 1769 cm−1 are assigned to the C=O stretching vibration (νC=O) of vapor-phase propionic acid (C2H5COOH) [37,46]. Based on the the C=O stretching vibration (νC=O) of vapor-phase propionic1 acid (C2H5COOH) [37,46]. Based on the literature [5,46,47], the bands at 2981, 2944, and 2887 cm− are ascribed to the C–H stretching vibration literature [5,46,47], the bands at 2981, 2944,1 and 2887 cm−1 are ascribed to the C–H stretching vibration (νC–H), the bands at 1374 and 1294 cm− are attributed to the C–H bending vibration (δC–H), and the (νC–H), the bands 1at 1374 and 1294 cm−1 are attributed to the C–H bending vibration (δC–H), and the band at 1080 cm− belongs to the C–H in-plane rocking vibration (ρC–H). The dissociated adsorption band at 1080 cm−1 belongs to the C–H in-plane rocking vibration (ρC–H). The dissociated adsorption configurations of propionic acid (Scheme1) can be distinguished through the νOCO (νas-νs, frequency configurations of propionic acid (Scheme 1) can be distinguished through the4 △νOCO (νas-νs, frequency separation between the asymmetric vibration (νas) and the symmetric vibration (νs) of OCO bonds). In separation between the asymmetric vibration (νas) and the symmetric vibration (νs) of OCO bonds). addition, νOCO follows the order of monodentate > free ionic > bidentate [48,49], and the ionic ν of In addition,4 △νOCO follows the1 order of monodentate > free ionic > bidentate [48,49], and the ionic4 △ν sodium propionate is 158 cm− [37]. Thus, the monodenate propionates (C2H5COO*) are detected on of sodium propionate is 158 cm−1 [37]. Thus, the1 monodenate propionates1 (C2H5COO*) are detected m-ZrO2 and t-ZrO2, locating at 1590, 1417 cm− and 1597, 1414 cm− , respectively. Besides those, two on m-ZrO2 and t-ZrO2, locating at 1590, 1417 cm−1 and 1597, 1414 cm−1, respectively. Besides those, kind bidentate propionates (*C2H5COO*) are observed on m-ZrO2 and t-ZrO2, including chelating two kind bidentate propionates (*C21H5COO*) are observed on m-ZrO2 and t-ZrO2, including1 bidentate propionates (1556, 1560 cm− ) and bridging bidentate propionates (1512, 1470 cm− ; 1511, chelating 1bidentate propionates (1556, 1560 cm−1) and bridging1 bidentate propionates (1512, 1470 cm−1; 1470 cm− )[5,16,47]. The peaks of 1597 and 1560 cm− on t-ZrO2 shift to the lower wavenumber 1511, 1470 cm−1) [5,16,47]. The peaks1 of 1597 and 1560 cm−1 on t-ZrO2 shift to the lower wavenumber on m-ZrO2 (1590 and 1556 cm− ), indicating the stronger adsorption of propionic acid on m-ZrO2. on m-ZrO2 (1590 and 1556 cm−1), indicating the stronger adsorption of propionic acid on m-ZrO2. With increasing the time of propionic acid fed, the intensity of C2H5COOH peaks increases, then With increasing the time of propionic acid fed, the intensity of C2H5COOH peaks increases, then decreases rapidly and disappears after the removal of propionic acid in the vapor phase. During decreases rapidly and disappears after1 the removal of propionic acid in the vapor phase. During propionic acid fed, a peak at 2358 cm− is detected on t-ZrO2 surface (Figure S4C), which belongs to propionic acid fed, a peak at 2358 cm−1 is detected on t-ZrO2 surface (Figure S4C), which4+ belongs to the asymmetric stretching mode of CO2 binding with coordinatively unsaturated Zr cations [50]. the asymmetric stretching mode of CO2 binding with coordinatively unsaturated Zr4+ cations [50]. However, a tiny CO2 peak is observed on the m-ZrO2 surface (Figure S4A). This result indicates that However, a tiny CO2 peak is observed on the m-ZrO2 surface (Figure S4A). This result indicates that less CO2 is produced on m-ZrO2 than on t-ZrO2, which is consistent with the lower ketonization rate less CO2 is produced on m-ZrO2 than on t-ZrO2, which is consistent with the lower ketonization rate of propionic acid on m-ZrO2 (Section 2.2). of propionic acid on m-ZrO2 (Section 2.2).

Catalysts 2019, 9, x FOR PEER REVIEW 8 of 15 Catalysts 2019, 9, 768 8 of 15 Catalysts 2019, 9, x FOR PEER REVIEW 8 of 15

1556 15561470 (A) 1590 1512 1417 (B) 1512 1417 15561470 159015561470 (A)0.1 1769 1590 1512 1417 (B)0.1 1512 1417 1789 1470 1080 1590 0.1 1769 0.1 1080 1789 1080 120 min 1080 120 min 15 min 60 min 15 min 3060 minmin 9 min 30 min

Absorbance(a.u.) 9 min

5 min 15 min Absorbance(a.u.) 5 min 15 min 2 min 5 min 2 min 5 min 2000 1800 1600 1400 1200 2000 1800 1600 1400 1200 -1 -1 2000 1800Wavenumber1600 1400 (cm 1200) 2000 1800Wavenumber1600 1400 (cm )1200 Wavenumber (cm-1) Wavenumber (cm-1) 1560 1470 (C) 1470 (D) 1560 1414 159715601511 1597 15111470 (C)0.1 14701414 (D)0.1 1560 1769 1080 1414 1597 1597 1511 1080 0.11789 1511 1414 0.1 1769 1080 1080 1789 120 min 120 min 15 min 60 min 15 min 60 min 9 min 30 min 9 min 30 min

5 min 15 min Absorbance(a.u.) 5 min 15 min Absorbance(a.u.) 2 min 5 min 2 min 5 min 2000 1800 1600 1400 1200 2000 1800 1600 1400 1200 -1 -1 2000 1800Wavenumber1600 1400 (cm 1200) 2000 1800Wavenumber1600 1400 (cm )1200 -1 -1 Wavenumber (cm ) Wavenumber (cm ) Figure 7. DRIFT spectra of propionic acid adsorption on m-ZrO2 (A,B) and t-ZrO2 (C,D) during Figure 7. DRIFT spectra of propionic acid adsorption on m-ZrO (A,B) and t-ZrO (C,D) during ketonizationFigure 7. DRIFT at 300 spectra °C: (A,C of) propionicunder propionic acid adsorption acid feed; ( onB,D m) -afterZrO2 2the (A,B removal) and tof-ZrO propionic22 (C,D) acid during in ketonization at 300 C: (A,C) under propionic acid feed; (B,D) after the removal of propionic acid in vaporketonization phase. at 300 ◦°C: (A,C) under propionic acid feed; (B,D) after the removal of propionic acid in vapor phase. vapor phase.

Scheme 1. Adsorption conﬁgurations of carboxylic acid on ZrO2.R = CH3CH2. Scheme 1. Adsorption configurations of carboxylic acid on ZrO2. R = CH3CH2.

Scheme 1. Adsorption configurations of carboxylic acid on ZrO2. R = CH3CH2. To clearly distinguish the variations of C2H5COO* and *C2H5COO* on the two catalysts, the To clearly distinguish the variations of C2H5COO* and *C2H5COO* on the two catalysts, the spectra were fitted using Lorentzian and/or Gaussian function sums. Figure S5 displays the examples spectraTo were clearly fitted distinguish using Lorentzian the variations and/or ofGaussian C2H5COO* function and sums. *C2H5 COO*Figure onS5 displays the two the catalysts, examples the of fitted spectra of adsorbed propionic acid, which correspond to 15 min spectra of Figure7A,C. ofspectra fitted were spectra fitted of usingadsorbed Lorentzian propionic and/or acid, Gaussian which correspond function sums. to 15 Figure min spect S5 displaysra of Figure the examples 7A, 7C. Figure8 shows the fit results. The peak area of 1080 cm 1, 1590 cm 1, and the sum of 1560 and −1 − −1 − Figureof fitted 8 showsspectra1 the of adsorbedfit results. propionic The peak acid,area ofwhich 1080 correspond cm , 1590 cmto 15, andmin thespect sumra of Figure1560 and 7A, 1510 7C. 1510 cm− were used to represent the total concentration of adsorbed propionic acid, the density of cmFigure−1 were 8 shows used the to fit represent results. theThe totalpeak concentrationarea of 1080 cm of−1 , adsorbed 1590 cm−1 propionic, and the sum acid, of the 1560 density and 1510 of C2H5COO* and *C2H5COO* on the catalysts surface, respectively [47]. As shown in Figure8A, the Ccm2H−15 COO*were and used *C to2H represent5COO* on thethe totalcatalysts concentration surface, respectively of adsorbed [47] propionic. As shown acid, in Figure the density 8A, the of total concentration of adsorbed propionic acid increases until near saturation coverages over the both totalC2H 5concentrationCOO* and *C 2ofH adsorbed5COO* on propionic the catalysts acid surface, increases respectively until near saturation [47]. As shown coverages in Figure over the 8A, both the catalysts with increasing adsorption time. After the removal of propionic acid in the vapor phase, the catalyststotal concentration with increasing of adsorbed adsorption propionic time. acidAfter increases the removal until of near propionic saturation acid coverages in the vapor over phase, the both the amount of total adsorbed propionic acid decreases with the extension of time. In accordance with amountcatalysts of with total increasing adsorbed adsorptionpropionic acid time. decreases After the with removal the extension of propionic of time. acid Inin accordancethe vapor phase, with thethe the report by Foraita et al., the total concentration of adsorbed propionic acid on m-ZrO2 is more reportamount by of Foraita total adsorbed et al., the propionic total concentration acid decreases of adsorbed with the extensionpropionic of acid time. on In m accordance-ZrO2 is more with than the than that on t-ZrO2 [47], due to the more acid-base sites existence. The change trends of C2H5COO* thatreport on by t- ZrOForaita2 [47] et, al., due the to total the moreconcentration acid-base of sitadsorbedes existence. propionic The changeacid on trendsm-ZrO of2 is Cmore2H5COO* than concentration are obvious and similar with that of total adsorbed propionic acid (Figure8B). At the concentrationthat on t-ZrO are2 [47] obvious, due toand the similar more with acid that-base of sit totales existence.adsorbed propionic The change acid trends (Figure of 8B). C2H At5COO* the concentration are obvious and similar with that of total adsorbed propionic acid (Figure 8B). At the

Catalysts 2019, 9, x FOR PEER REVIEW 9 of 15 lower coverage of propionic acid, there are more C2H5COO* on m-ZrO2 than on t-ZrO2, which may be due to the stronger adsorption of propionic acid and the lower activity of m-ZrO2. However, the C2H5COO* becomes more on t-ZrO2 relative to on m-ZrO2 when the ketonization reaches a steady state at 15 min. The amount of C2H5COO* on m-ZrO2 and t-ZrO2 decreases gradually with time after stopping propionic acid fed because of reaction and desorption. More *C2H5COO* is adsorbed onto the m-ZrO2 surface than onto t-ZrO2 (Figure 8B). The slight changes of *C2H5COO* are observed on the two catalysts throughout the whole process. Furthermore, the *C2H5COO* decreases slowly and Catalysts 2019, 9, 768 9 of 15 there is less contrast with C2H5COO* after the removal of propionic acid, owing to the stronger binding with catalysts surface. The *C2H5COO* may even be converted to C2H5COO*, directly proceedinglower coverage ketonizati of propionicon, or be acid, desorbed there are [19, more25,26,32 C2H]. 5COO* on m-ZrO2 than on t-ZrO2, which may 2 5 2 5 be dueBased to the on strongerthe above adsorption analysis of of DRIFTS, propionic the acidC H andCOO* the is lower much activitymore active of m-ZrO than the2. However, *C H COO*, the asC2 hasH5COO* been demonstrated becomes more by on Wang t-ZrO et2 relativeal. using to DFT on m-ZrOand experiments2 when the [25,26] ketonization. Additionally, reaches the a steady ratio 2 2 ofstate monodentate at 15 min. Theto bidentate amount of pro C2pionateH5COO* on on t- m-ZrOZrO is2 andsmaller t-ZrO than2 decreases that on graduallym-ZrO . The with adsorption time after 2 2 strengthstopping of propionic propionic acid acid fed on because m-ZrO of is reaction stronger and than desorption. that on t-ZrO More, *Cwhich2H5COO* is caused is adsorbed by the more onto 2 mediumthe m-ZrO-strength2 surface acid than-base onto sites t-ZrO with2 (Figure lower8B). coordination The slight changes exposing of *Con 2 theH5 COO* m-ZrO are surface. observed The on sthetronger two catalysts adsorption throughout of propionic the whole acid isprocess. not of Furthermore, benefit to improving the *C2H 5 ketonizationCOO* decreases activity. slowly These and differencesthere is less contrast in adsorption with C 2 strengthH5COO* and after adsorption the removal configurations of propionic acid, of owing propionic to the acid stronger on the binding two 2 2 catalystswith catalysts are related surface. to The the *C different2H5COO* structures may even of be m converted-ZrO and to t- CZrO2H5.COO*, Thus, directly the higher proceeding surface concentrationketonization, orand be weaker desorbed adsorption [19,25,26 ,32of ].C2H5COO* on t-ZrO2 contribute to the higher ketonization rate of propionic acid on t-ZrO2 versus on m-ZrO2 during the steady-state ketonization.

m-ZrO Monodentate on m-ZrO2 ) (A) 2 (B) -2 2.5 Monodentate on t-ZrO

t-ZrO2 12 2 cat Bidentate on m-ZrO2

Bidentate on t-ZrO2 2.0 10

8 1.5

6 1.0 Stop propionic acid injection Normalized absorbance (m absorbance Normalized 4 Stop propionic acid injection 0.5 0 10 5 15 25 35 45 55 0 10 5 15 25 35 45 55 Time (min) Time (min) 1 Figure 8. The total amount of adsorbed propionic acid (peak area of 10801080 cmcm−−1))( (AA),), the the amount of 1 monodentate propionate (peak area of 1590–15971590–1597 cm−−1)) and and bidentate bidentate propionate propionate (peak (peak area area of of 1560 1560 and 1510 cm 1)(B) on ZrO normalized by surface area. and 1510 cm−1−) (B) on ZrO2 2normalized by surface area.

Based on the above analysis of DRIFTS, the C2H5COO* is much more active than the *C2H5COO*, Scheme 2 displays the possible surface reaction of propionic acid ketonization over ZrO2 based as has been demonstrated by Wang et al. using DFT and experiments [25,26]. Additionally, the ratio on the in situ DRIFTS results and the previous reports [19,25,26]. Propionic acid dissociative adsorbs of monodentate to bidentate propionate on t-ZrO is smaller than that on m-ZrO . The adsorption on the acid-base site as monodentate and bidentate2 adsorption modes, then forms enolate2 through α- strength of propionic acid on m-ZrO is stronger than that on t-ZrO , which is caused by the more H abstraction by the vicinal base site. 2After that, enolate and neighboring2 propionic acid proceed C– medium-strength acid-base sites with lower coordination exposing on the m-ZrO2 surface. The stronger C coupling, dehydration and decarboxylation forming 3-pentanone, H2O and CO2. adsorption of propionic acid is not of benefit to improving ketonization activity. These differences in adsorption strength and adsorption configurations of propionic acid on the two catalysts are related to the different structures of m-ZrO2 and t-ZrO2. Thus, the higher surface concentration and weaker adsorption of C2H5COO* on t-ZrO2 contribute to the higher ketonization rate of propionic acid on t-ZrO2 versus on m-ZrO2 during the steady-state ketonization. Scheme2 displays the possible surface reaction of propionic acid ketonization over ZrO 2 based on the in situ DRIFTS results and the previous reports [19,25,26]. Propionic acid dissociative adsorbs on the acid-base site as monodentate and bidentate adsorption modes, then forms enolate through α-H abstraction by the vicinal base site. After that, enolate and neighboring propionic acid proceed C–C coupling, dehydration and decarboxylation forming 3-pentanone, H2O and CO2.

Catalysts 2019 9 CatalystsCatalysts 2019 2019, ,9 9, ,x 768x FOR FOR PEER PEER REVIEW REVIEW 101010 of of of 15 15

Scheme 2. Proposed surface reaction of propionic acid ketonization over ZrO2. SchemeScheme 2. 2. Proposed Proposed surface surface reaction reaction of of propionic propionic acid acid ketonization ketonization over over ZrO ZrO2.2 . The adsorption of 3-pentanone as the main product was also studied by DRIFTS. Figure9 shows The adsorption of 3-pentanone as the main product was also studied by DRIFTS. Figure 9 shows the resultThe adsorption of 3-pentanone of 3-pentanone adsorption. as the At 50main◦C, product the carbonyl was also stretching studied vibration by DRIFTS. (νC Figure=O) is located 9 shows at the result1 of 3-pentanone adsorption. At 50 °C, the carbonyl stretching vibration (νC=OC=O) is located at the1694 result cm− ofon 3- bothpentanone catalysts, adsorption. suggesting At that 50 the°C, oxygenthe carbonyl atom ofstretching carbonyl vibration binds with (ν the) coordinatively is located at 16941694unsaturated cm cm−1−1 onon metal both both cations catalysts, catalysts, of metal suggesting suggesting oxide in that an thatη 1 theconﬁguration the oxygen oxygen atom atom [37,51 of].of The carbonyl carbonyl intensity binds binds of the with with peak the the on coordinatively unsaturated metal cations of metal oxide in an η1 1 configuration [37,51]. The intensity coordinativelym-ZrO2 is higher unsaturated than that onmetal t-ZrO cations2, which of metal indicates oxide that in more an η 3-pentanone configuration is adsorbed[37,51]. The onto intensity m-ZrO2 of the peak on m-ZrO2 2 is higher than that on t-ZrO2 2, which indicates that more 3-pentanone is ofowing the peak to the on greater m-ZrO number is higher of acid than sites that present. on t- WhenZrO , increasing which indicates the temperature that more from 3-pentanone 50 to 250 ◦ isC, adsorbed onto m-ZrO212 owing to the greater number of acid sites present. When increasing the adsorbedthe peaks onto at 1694 m-ZrO cm− decreaseowing to and the disappear.greater number Meanwhile, of acid they sites shift present. to a lower When wavenumber increasing withthe temperaturetemperaturedecreasing peaks.from from 50 50 In to to the250 250 DRIFTS°C, °C, the the peaks ofpeaks adsorbed at at 1694 1694 propioniccm cm−1−1 decrease decrease acid, and and the disappear. disappear. carbonyl stretchingMeanwhile, Meanwhile, vibration they they shift shift of toto3-petanone a a lower lower wavenumber was wavenumber not observed with with maybe decreasing decreasing because peaks. peaks. of the In In rapid the the desorption DRIFTS DRIFTS of of and adsorbed adsorbed low coverage propionic propionic of 3-petanone acid, acid, the the carboncarbonon theylyl catalysts stretching stretching surface vibration vibration [13]. of of 3 3-petanone-petanone was was not not observed observed maybe maybe because because of of the the rapid rapid desorption desorption andand low low coverage coverage of of 3 3-petanone-petanone on on the the catalysts catalysts surface surface [13] [13]. . (A)(A) (B)(B) 0.0250.025 0.0250.025

16941694 16941694 oo 250250oCoC 250250CC

oo oo 150150CC

Absorbance(a.u.) 150 C

oo 5050oCoC 5050CC

18001800 16001600 14001400 12001200 18001800 16001600 14001400 12001200 Wavenumber (cm-1-1) Wavenumber (cm-1-1) Wavenumber (cm ) Wavenumber (cm )

FigureFigure 9. 9.9. TPD TPD-DRIFTTPD-DRIFT-DRIFT spectra spectra of of 1.0 1.0 kPa kPa 3 3-pentanone3-pentanone-pentanone adsorption adsorption on onon m m-ZrOm-ZrO-ZrO2 22( A(A) )and and t - t-ZrOtZrO-ZrO2 2( B((BB).).).

Catalysts 2019, 9, 768 11 of 15

3. Experiments and Methods

3.1. Catalyst Preparation

The monoclinic zirconia (m-ZrO2) and tetragonal zirconia (t-ZrO2) were prepared by the solvothermal method [52] using water and methanol (Tianjin Kemiou Chemical, Tianjin, China) as solvent, respectively. 25.76 g of Zr(NO ) 5H O (Admas Reagent, Shanghai, China) was added to 3 4· 2 100 mL water or methanol to prepare the water or methanolic solution (0.6 mol/L). Then, the amount of urea (Tianjin Kemiou Chemical) was mixed with water or methanolic solution with a urea/Zr4+ molar ratio of 5 under stirring at room temperature. After that, the mixtures were added to the stainless-steel autoclave with a Teﬂon liner at 160 ◦C for 21 h. The resulting precipitates were washed with water or methanol several times, and dried for 12 h at 110 ◦C. Finally, the powders were calcined at 400 ◦C for 4 h with a heating rate of 2 ◦C/min.

3.2. Catalyst Characterization

The Brunauer–Emmet–Teller (BET) surface area of ZrO2 was performed using a Micrometrics Tristar 3000 with a liquid nitrogen bath (Norcross, GA, USA). All the catalysts were pretreated at 300 ◦C for 3 h under the atmospheric pressure prior to analysis. The phase structure of ZrO2 was investigated by powder X-ray diffraction (XRD) on a Rigaku D/Max-2500 V/Pc diffractometer (Tokyo, Japan) under ambient conditions using a filtered Cu Kα radiation source (λ = 1.54056 Å) operated at 40 kV and 20 mA. Data was collected in the angle range of 20–100◦ with a scanning rate of 4◦/min. The Raman spectra were collected using a Renishaw Raman spectrometer (Wotton-under-Edge, UK) with the Ar+ laser (532 nm) being the exciting light source. The focusing spot size was about 1 1 µm. The spectra were collected at a resolution of 2 cm− with an acquisition time of 10 s/scan at room temperature. X-ray photoelectron spectroscopy (XPS) data were recorded on a Physical Electronics PHI 1600 (Chanhassen, MN, USA) with monochromatic Al Kα X-rays (1486.6 eV) operated at 250W and 15 kV in a chamber pumped down to a pressure of approximately 1.6 10 8 Pa. The correction of the binding × − energy (BE) employed the C 1s peak of adventitious carbon at 284.6 eV. Temperature-programmed desorption of NH3 and CO2 (NH3-TPD, CO2-TPD) was measured using a Cirrus 200 mass spectrometer (MKS, Austin, TX, USA) [37]. The ZrO2 (200 mg, 40–60 mesh) was loaded into a quartz tube reactor, pretreated at 350 ◦C in He (30 mL/min) for 60 min, then cooling to 50 ◦C (30 ◦C). After pretreatment, ZrO2 was saturated at 50 ◦C with 2% NH3-He (at 30 ◦C with 5% CO2-He) for 30 min (50 mL/min), and then subsequently purged with He (30 mL/min) for 60 min to remove all the physisorption NH3 (CO2). The desorption of chemisorbed NH3 (CO2) was carried out in the He flow (30 mL/min) with ramping temperature from 50 ◦C (30 ◦C) to 600 ◦C at a rate of 10 ◦C /min. The measurement of NH3 (CO2) evolution was studied through a TCD detector, and the desorption peak of NH3 (CO2) was quantified by calibrating the area using a 2% NH3-He (5% CO2-He) with a six-port valve. Diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) was performed using the PerkinElmer Frontier spectrometer (Waltham, MA, USA) with the DTGS detector. 40 mg samples were pressed into self-supporting wafers, then mounted into the in situ cell. Before the test, the samples were pretreated at 350 ◦C in He (30 mL/min) for 60 min, then cooled to the adsorption temperature. The DRIFTS of the pretreated samples were used as background. For the adsorption of pyridine, 0.5 kPa pyridine was injected at 30 ◦C for 30 min under He (30 mL/min), then purged for 60 min. Following that, the DRIFTS of pyridine adsorption were recorded. The adsorption of propionic acid was measured at 300 ◦C. 3.9 kPa propionic acid was injected with He (30 mL/min) as carrier gas. The system reached a steady state after 15 min. Then, propionic acid was stopped and purged with He (30 mL/min) for 120 min. The TPD of adsorbed 3-pentanone was carried out. The adsorption process was similar with that of propionic acid. After reaching saturated adsorption of 3-pentanone at 50 ◦C, Catalysts 2019, 9, 768 12 of 15

the in situ cell was purged under He (30 mL/min) for 60 min, then increased from 50 to 250 ◦C at a 1 heating rate of 5 ◦C/min. All the DRIFTS were recorded at a resolution of 4 cm− , with 32 scans in the 1 range of 4000–1000 cm− .

3.3. Catalytic Activity

Vapor ketonization of propionic acid was measured in a fixed-bed quartz tube reactor at 300–375 ◦C and atmospheric pressure [37]. Catalysts with 40–60 mesh were loaded in the center of reactor between the two layers of quartz wool, and pretreated at 350 ◦C for 30 min in Ar as carrier gas, controlled by the mass flow; then, the temperature was decreased to the reaction temperature. A K-type thermocouple was placed in the reactor for the measurement of the catalysts bed temperature. Propionic acid was introduced by a syringe pump (KDS100, kd scientific, Holliston, MA, USA). The molar ratio of Ar and propionic acid was 25. All lines were heated at 220 ◦C to avoid any condensation of reactants and products. The products were quantified online using a gas chromatograph (GC, 7890B, Agilent, Santa Clara, CA, USA). Finally, the products were collected by the methanol in the ice water bath. The conversion and selectivity are reported in Mol of carbon %.

4. Conclusions The ketonization rate of propionic acid in the vapor phase was aﬀected by the zirconia polymorph. The t-ZrO2 catalysts show higher activity on propionic acid ketonization than m-ZrO2 at 300–375 ◦C. There are more medium-strength acid base sites with lower coordination exposed on the m-ZrO2 surface in comparison to t-ZrO2, strengthening propionic acid adsorption, although they are not of beneﬁt in terms of increasing the reaction rate. The in situ DRIFTS of adsorbed propionic acid during reaction conditions demonstrate that monodentate and bidentate propionates are the most abundant surface intermediates. In addition, the monodentate propionates are much more active over bidentate propionates. Relative to m-ZrO2, t-ZrO2 favors monodentate adsorption over bidentate adsorption, and diminish the adsorption strength of monodentate propionates, which are related to the zirconia polymorph. The higher ketonization rate obtained on t-ZrO2 is due to the higher surface concentration and weaker adsorption of monodentate propionates contrast with that over m-ZrO2.

Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4344/9/9/768/s1, Figure S1: X-ray diffraction patterns of ZrO2 used at 350 ◦C in the ketonization of propionic acid, Figure S2: N2 sorption isotherm curves of ZrO2 catalysts, Figure S3: DRIFT spectra of adsorbed pyridine at 30 ◦C on ZrO2 catalysts, Figure S4: DRIFT spectra of propionic acid adsorption on m-ZrO2 (A,B) and t-ZrO2 (C, D) during ketonization at 300 ◦C, Figure S5: Fitted DRIFT spectra of propionic acid adsorption for 15 min on ZrO2 catalysts at 300 ◦C. Author Contributions: S.D. contributed to catalysts synthesis and characterizations, results analysis and paper writing; J.Z. contributed to results discussion and writing of the paper; Q.Y. contributed to catalysts synthesis. Funding: This research was funded by the National Natural Science Foundation of China, grant number 21676194 and 21576204. Conflicts of Interest: The authors declare no conflict of interest.

References

1. George, S.I.; Huber, W.; Corma, A. Synthesis of transportation fuels from biomass: Chemistry, catalysts, and engineering. Chem. Rev. 2006, 106, 4044–4098. 2. Goyal, H.B.; Seal, D.; Saxena, R.C. Bio-fuels from thermochemical conversion of renewable resources: A review. Renew. Sustain. Energ. Rev. 2008, 12, 504–517. [CrossRef] 3. Corma, A.; Iborra, S.; Velty, A. Chemical routes for the transformation of biomass into chemicals. Chem. Rev. 2007, 107, 2411–2502. [CrossRef][PubMed] 4. Perez, R.F.; Soares, O.S.G.P.; de Farias, A.M.D.; Pereira, M.F.R.; Fraga, M.A. Conversion of hemicellulose-derived pentoses over noble metal supported on 1D multiwalled carbon nanotubes. Appl. Catal. B Environ. 2018, 232, 101–107. [CrossRef] Catalysts 2019, 9, 768 13 of 15

5. Chen, L.; Zhu, Y.; Zheng, H.; Zhang, C.; Li, Y. Aqueous-phase hydrodeoxygenation of propanoic acid over

the Ru/ZrO2 and Ru–Mo/ZrO2 catalysts. Appl. Catal. A Gen. 2012, 411–412, 95–104. [CrossRef] 6. Rachmady, W.; Vannice, M.A. Acetic acid reduction by H2 over supported Pt catalysts: A DRIFTS and TPD/TPR study. J. Catal. 2002, 207, 317–330. [CrossRef]

7. Al-Auda, Z.; Al-Atabi, H.; Hohn, K.L. Metals on ZrO2: Catalysts for the aldol condensation of methyl ethyl ketone (MEK) to C8 ketones. Catalyst 2018, 8, 622. [CrossRef] 8. Faba, L.; Díaz, E.; Ordóñez, S. Gas phase acetone self-condensation over unsupported and supported Mg–Zr mixed-oxides catalysts. Appl. Catal. B Environ. 2013, 142–143, 387–395. [CrossRef] 9. Mekhemer, G.; Halawy, S.; Mohamed, M.; Zaki, M. Ketonization of acetic acid vapour over polycrystalline magnesia: In situ Fourier transform infrared spectroscopy and kinetic studies. J. Catal. 2005, 230, 109–122. [CrossRef] 10. Bennett, J.A.; Parlett, C.M.A.; Isaacs, M.A.; Durndell, L.J.; Olivi, L.; Lee, A.F.; Wilson, K. Aetic acid ketonization

over Fe2O3/SiO2 for pyrolysis bio-oil upgrading. ChemCatChem 2017, 9, 1648–1654. [CrossRef] 11. Snell, R.W.; Shanks, B.H. CeMOx-promoted ketonization of biomass-derived carboxylic acids in the condensed phase. ACS Catal. 2014, 4, 512–518. [CrossRef]

12. Nagashima, O.; Sato, S.; Takahashi, R.; Sodesawa, T. Ketonization of carboxylic acids over CeO2-based composite oxides. J. Mol. Catal. A Chem. 2005, 227, 231–239. [CrossRef] 13. Pham, T.N.; Shi, D.; Sooknoi, T.; Resasco, D.E. Reaction kinetics and mechanism of ketonization of aliphatic

carboxylic acids with diﬀerent carbon chain lengths over Ru/TiO2 catalyst. J. Catal. 2014, 314, 149–158. [CrossRef] 14. Lu, F.; Jiang, B.; Wang, J.; Huang, Z.; Liao, Z.; Yang, Y. Insights into the improvement eﬀect of Fe doping into

the CeO2 catalyst for vapor phase ketonization of carboxylic acids. Mol. Catal. 2018, 444, 22–33. [CrossRef] 15. Gumidyala, A.; Sooknoi, T.; Crossley, S. Selective ketonization of acetic acid over HZSM-5: The importance of acyl species and the inﬂuence of water. J. Catal. 2016, 340, 76–84. [CrossRef] 16. Wang, X.; Ding, S.; Wang, H.; Liu, X.; Han, J.; Ge, Q.; Zhu, X. Conversion of propionic acid and 3-pentanone to hydrocarbons on ZSM-5 catalysts: Reaction pathway and active site. Appl. Catal. A Gen. 2017, 545, 79–89. [CrossRef] 17. Martinez, R. Ketonization of acetic acid on titania-functionalized silica monoliths. J. Catal. 2004, 222, 404–409. [CrossRef] 18. Ignatchenko, A.V.; Kozliak, E.I. Distinguishing Enolic and Carbonyl Components in the Mechanism of Carboxylic Acid Ketonization on Monoclinic Zirconia. ACS Catal. 2012, 2, 1555–1562. [CrossRef] 19. Cai, Q.; Lopez-Ruiz, J.A.; Cooper, A.R.; Wang, J.; Albrecht, K.O.; Mei, D. Aqueous-phase acetic acid ketonization over monoclinic zirconia. ACS Catal. 2017, 8, 488–502. [CrossRef] 20. Snell, R.W.; Shanks, B.H. Insights into the ceria-catalyzed ketonization reaction for biofuels applications. ACS Catal. 2013, 3, 783–789. [CrossRef] 21. Almutairi, S.T.; Kozhevnikova, E.F.; Kozhevnikov, I.V. Ketonisation of acetic acid on metal oxides: Catalyst activity, stability and mechanistic insights. Appl. Catal. A Gen. 2018, 565, 135–145. [CrossRef] 22. Liu, C.; Karim, A.M.; Lebarbier, V.M.; Mei, D.; Wang, Y. Vapor phase ketonization of acetic acid on ceria based metal oxides. Top. Catal. 2013, 56, 1782–1789. [CrossRef] 23. Kim, K.S.; Barteau, M.A. Structure and composition requirements for deoxygenation, dehydration, and

ketonization reactions of carboxylic acids on TiO2 (001) single-crystal surfaces. J. Catal. 1990, 125, 353–375. [CrossRef]

24. Stubenrauch, J.; Brosha, E.; Vohs, J.M. Reaction of carboxylic acids on CeO2 (111) and CeO2 (100). Catal. Today 1996, 28, 431–441. [CrossRef] 25. Wang, S.; Iglesia, E. Experimental and theoretical assessment of the mechanism and site requirements for ketonization of carboxylic acids on oxides. J. Catal. 2017, 345, 183–206. [CrossRef] 26. Wang, S.; Iglesia, E. Experimental and theoretical evidence for the reactivity of bound intermediates in ketonization of carboxylic acids and consequences of acid–base properties of oxide catalysts. J. Phys. Chem. C 2017, 121, 18030–18046. [CrossRef] 27. Snell, R.W.; Hakim, S.H.; Dumesic, J.A.; Shanks, B.H. Catalysis with ceria nanocrystals: Bio-oil model compound ketonization, Appl. Catal. A Gen. 2013, 464–465, 288–295. [CrossRef]

28. Wu, K.; Yang, M.; Pu, W.; Wu, Y.; Shi, Y.; Hu, H. Carbon promoted ZrO2 catalysts for aqueous-phase ketonization of acetic acid. ACS Sustain. Chem. Eng. 2017, 5, 3509–3516. [CrossRef] Catalysts 2019, 9, 768 14 of 15

29. Wu, K.; Yang, M.; Chen, Y.; Pu, W.; Hu, H. Aqueous-phase ketonization of acetic acid over Zr/Mn mixed oxides. AICHE J. 2017, 63, 2958–2967. [CrossRef] 30. Lopez-Ruiz, J.A.; Cooper, A.R.; Li, G.; Albrecht, K.O. Enhanced hydrothermal stability and catalytic activity of LaxZryOz mixed oxides for the ketonization of acetic acid in the aqueous condensed phase. ACS Catal. 2017, 7, 6400–6412. [CrossRef] 31. Parida, K.; Mishra, H.K. Catalytic ketonisation of acetic acid over modiﬁed zirconia 1 eﬀect of alkali-metal cations as promoter. J. Mol. Catal. A Chem 1999, 139, 73–80. [CrossRef] 32. Pulido, A.; Oliver-Tomas, B.; Renz, M.; Corma, A. Ketonic decarboxylation reaction mechanism: A combined experimental and DFT study. ChemSusChem 2013, 6, 141–151. [CrossRef][PubMed] 33. Ignatchenko, A.V.; DeRaddo, J.S.; Marino, V.J.; Mercado, A. Cross-selectivity in the catalytic ketonization of carboxylic acids. Appl. Catal. A Gen. 2015, 498, 10–24. [CrossRef] 34. Gaertner, C.A.; Serrano-Ruiz, J.C.; Braden, D.J.; Dumesic, J.A. Catalytic coupling of carboxylic acids by ketonization as a processing step in biomass conversion. J. Catal. 2009, 266, 71–78. [CrossRef] 35. Tosoni, S.; Pacchioni, G. Acetic acid ketonization on tetragonal zirconia: Role of surface reduction. J. Catal. 2016, 344, 465–473. [CrossRef] 36. Shutilov, A.A.; Simonov, M.N.; Zaytseva, Y.A.; Zenkovets, G.A.; Simakova, I.L. Phase composition and

catalytic properties of ZrO2 and CeO2-ZrO2 in the ketonization of pentanoic acid to 5-nonanone. Kinet. Catal. 2013, 54, 184–192. [CrossRef]

37. Ding, S.; Wang, H.; Han, J.; Zhu, X.; Ge, Q. Ketonization of propionic acid to 3-pentanone over CexZr1–xO2 catalysts: The importance of acid–base balance. Ind. Eng. Chem. Res. 2018, 57, 17086–17096. [CrossRef] 38. Baylon, R.A.L.; Sun, J.; Kovarik, L.; Engelhard, M.; Li, H.; Winkelman, A.D.; Wang, Y. Structural identiﬁcation of ZnxZryOz catalysts for cascade aldolization and self-deoxygenation reactions. Appl. Catal. B Environ. 2018, 234, 337–346. [CrossRef] 39. Albuquerque, E.M.; Borges, L.E.P.; Fraga, M.A.; Sievers, C. Realtionship between acid-base properties and

the activity of ZrO2-based catalysts for the cannizzaro reaction of pyruvaldehyde to lactic acid. ChemCatChem 2017, 9, 2675–2683. [CrossRef]

40. Rhodes, M.D.; Bell, A.T. The eﬀects of zirconia morphology on methanol synthesis from CO and H2 over Cu/ZrO2 catalysts Part I. steady-state studies. J. Catal. 2005, 233, 198–209. [CrossRef] 41. Nelson, A.E.; Schulz, K.H. Surface chemistry and microstructural analysis of CexZr1 xO2 y model catalyst − − surfaces. Appl. Surf. Sci. 2003, 210, 206–221. [CrossRef]

42. Silva, L.P.C.; Terra, L.E.; Coutinho, A.C.S.L.S.; Passos, F.B. Sour water-gas shift reaction over Pt/CeZrO2 catalysts. J. Catal. 2016, 341, 1–12. [CrossRef] 43. Zaki, M.I.; Hasan, M.A.; Al-Sagheer, F.A.; Pasupulety, L. In situ FTIR spectra of pyridine adsorbed on

SiO2-Al2O3, TiO2, ZrO2 and CeO2: General considerations for the identiﬁcation of acid sites on surfaces of ﬁnely divided metal oxides. Colloids Surf. A Physicochem. Eng. Asp. 2001, 190, 261–274. [CrossRef] 44. Sun, J.; Zhu, K.; Gao, F.; Wang, C.; Liu, J.; Peden, C.H.F.; Wang, Y. Direct conversion of bio-ethanol to isobutene on nanosized ZnxZryOz mixed oxides with balanced acid-base sites. J. Am. Chem. Soc. 2011, 133, 11096–11099. [CrossRef][PubMed] 45. Prymak, I.; Kalevaru, V.N.; Wohlrab, S.; Martin, A. Continuous synthesis of diethyl carbonate from ethanol

and CO2 over Ce-Zr-O catalysts. Catal. Sci. Technol. 2015, 5, 2322–2331. [CrossRef] 46. Hasan, M.A.; Zaki, M.I.; Pasupulety, L. Oxide-catalyzed conversion of acetic acid into acetone: An FTIR spectroscopic investigation. Appl. Catal. A Gen. 2003, 243, 81–92. [CrossRef] 47. Foraita, S.; Fulton, J.L.; Chase, Z.A.; Vjunov, A.; Xu, P.; Barath, E.; Camaioni, D.M.; Zhao, C.; Lercher, J.A. Impact of the oxygen defects and the hydrogen concentration on the surface of tetragonal and monoclinic

ZrO2 on the reduction rates of stearic acid on Ni/ZrO2. Chem. Eur. J. 2015, 21, 2423–2434. [CrossRef] [PubMed] 48. Li, C.; Domen, K.; Maruya, K.I.; Onishi, T. Spectroscopic identiﬁcation of adsorbed species derived from adsorption and decomposition of formic acid, methanol, and formaldehyde on cerium oxide. J. Catal. 1990, 125, 445–455. [CrossRef] 49. Pei, Z.-F.; Ponec, V. On the intermediates of the acetic acid reactions on oxides: An IR study. Appl. Surf. Sci. 1996, 103, 171–182. [CrossRef]

50. Pokrovski, K.; Jung, K.T.; Bell, A.T. Investigation of CO and CO2 adsorption on tetragonal and monoclinic Zirconia. Langmuir 2001, 17, 4297–4303. [CrossRef] Catalysts 2019, 9, 768 15 of 15

51. Zaki, M.I.; Hasan, M.A.; Pasupulety, L. Surface Reactions of acetone on Al2O3, TiO2, ZrO2 and CeO2: IR Spectroscopic assessment of impacts of the surface acid-base properties. Langmuir 2001, 17, 768–774. [CrossRef] 52. Li, W.; Huang, H.; Li, H.; Zhang, W.; Liu, H. Facile synthesis of monoclinic and tetragonal zirconia nanoparticles and their phase eﬀects on the behavior of supported molybdena catalysts for methanol-selective oxidation. Langmuir 2008, 24, 8358–8366. [CrossRef][PubMed]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).