Versatile Coordination Polymer Catalyst for Acid Reactions Involving Biobased Heterocyclic Chemicals

: The chemical valorization/repurposing of biomass-derived chemicals contributes to a biobased economy. Furfural (Fur) is a recognized platform chemical produced from renewable lignocellulosic biomass, and furfuryl alcohol (FA) is its most important application. The aromatic aldehydes Fur and benzaldehyde (Bza) are commonly found in the slate of compounds produced via biomass pyrolysis. On the other hand, glycerol (Gly) is a by-product of the industrial production of biodiesel, derived from fatty acid components of biomass. This work focuses on acid catalyzed routes of Fur, Bza, Gly and FA, using a versatile crystalline lamellar coordination polymer catalyst, namely [Gd(H 4 nmp)(H 2 O) 2 ]Cl · 2H 2 O ( 1 ) [H 6 nmp=nitrilotris(methylenephosphonic acid)] synthesized via an ecofriendly, relatively fast, mild microwave-assisted approach (in water, 70 ◦ C/40 min). This is the ﬁrst among crystalline coordination polymers or metal-organic framework type materials studied for the Fur/Gly and Bza/Gly reactions, giving heterobicyclic products of the type dioxolane and dioxane, and was also effective for the FA/ethanol reaction. 1 was stable and promoted the target catalytic reactions, selectively leading to heterobicyclic dioxane and dioxolane type products in the Fur/Gly and Bza/Gly reactions (up to 91% and 95% total yields respectively, at 90 ◦ C/4 h), and, on the other hand, 2-(ethoxymethyl)furan and ethyl levulinate from heterocyclic FA.

In this work, the Fur/Gly and Bza/Gly reactions were studied in the presence of [Gd(H 4 nmp)(H 2 O) 2 ]Cl·2H 2 O (1) [H 6 nmp=nitrilotris(methylenephosphonic acid)] (Scheme 1), which is an attractive material in that it may be synthesized via a simple, ecofriendly microwave-assisted approach using water as solvent, and possesses acid properties for catalytic reactions under relatively mild conditions [116]. 1 also proved effective for removing sulfur and nitrogen pollutant compounds from diesel-type mixtures [117].
1 is a (two dimensional) crystalline lamellar coordination polymer with Gd 3+ centers ({GdO 8 } coordination polyhedra) and a flexible organic linker (H 4 nmp 2− ) ( Figure 1) [116]. The organic linker is a zwitterionic species with a protonated central nitrogen atom and peripheral phosphonate groups, each bearing a single negative charge (PO 3 H − ). The Catalysts 2021, 11,190 3 of 24 phosphonate groups act as µ 2 -O,O' bridging moieties, and P=O and P-OH surface groups are formed. The influence of the reaction conditions and mechanistic and structural features involved in these catalytic systems are discussed based on experimental studies. Scheme 1. The chemical valorization/repurposing of biomass-derived chemicals such as furfural (Fur), furfuryl alcohol (FA) and glycerol (Gly). The reactions of Fur/Gly and benzaldehyde/Gly to useful cyclic acetals, and FA/ethanol to furanic ether (EMF) and levulinate ester (EL).

General Considerations
The scaled-up synthesis of the hybrid material 1 (Figure 1) was successfully accomplished, as confirmed by powder X-ray diffraction, which was similar to that reported in the literature (Supplementary Figure S1) [117]. Additional characterization studies are discussed in the catalyst stability Section 2.5. Material 1 was explored as an acid catalyst for the reactions of Bza (Section 2.2) and Fur (Section 2.3) with Gly, to give heterobicyclic products. The interest in the repurposing/valorization of Fur, Bza and Gly is related to the The catalytic potentialities of 1 were further investigated for converting heterocyclic furfuryl alcohol (FA) to useful bioproducts. FA represents the highest market volume of Fur, produced from lignocellulosic biomass. Fur and FA have wide application profiles [104,[118][119][120]. The reaction of FA with (biomass-derivable) ethanol gives 2-(ethoxymethyl)furan (EMF) and ethyl levulinate (EL) (Scheme 1) [121]. The applications of EMF include bio-solvent [122], pharmaceuticals, food flavor [123][124][125][126], aging flavor for beer storage [127][128][129][130][131] or white wine [132] and fuel additive [122,[124][125][126][133][134][135][136][137] since it has a high energy density, cetane number [138] and capacity to reduce soot emissions [124,134,139,140]. On the other hand, EL is an attractive oxygenated fuel extender (its high oxygen content enhances engine efficiency, improves flow properties, etc. [141,142]) with a promising global market [141,143]. Tian et al. [139] reported that EMF and EL present superior antiknock features to Euro95 gasoline, and Lange et al. [144] highlighted the limited footprint CO 2 emissions associated with the use of EMF in fuels. To the best of our knowledge, solely two coordination polymers or MOFs were studied for the conversion of FA to EMF/EL, namely Hf-UiO-66-SO 3 H [145] and MIL-101 (Cr)-SO 3 H [146,147], both functionalized with sulfonic acid groups.
Hence, 1 is the first of crystalline coordination polymers/MOFs investigated in the Fur/Gly and Bza/Gly reactions, and the first non-sulfonic acid crystalline coordination polymer studied in the FA/ethanol reaction. A relatively fast and mild synthesis protocol of 1, using water as solvent, is reported. 1 effectively promoted the target biomass-related reactions. The influence of the reaction conditions and insights into the reaction mechanism and catalyst structural features contributing to the formation of the cyclic acetals are reported.

General Considerations
The scaled-up synthesis of the hybrid material 1 (Figure 1) was successfully accomplished, as confirmed by powder X-ray diffraction, which was similar to that reported in the literature (Supplementary Figure S1) [117]. Additional characterization studies are discussed in the catalyst stability Section 2.5. Material 1 was explored as an acid catalyst for the reactions of Bza (Section 2.2) and Fur (Section 2.3) with Gly, to give heterobicyclic products. The interest in the repurposing/valorization of Fur, Bza and Gly is related to the Catalysts 2021, 11, 190 4 of 24 fact that Fur is an established platform chemical industrially produced from lignocellulosic biomass, Gly is industrially associated (as a by-product) with biodiesel production and Fur and Bza are aromatic aldehydes commonly found in the slate of compounds produced via biomass pyrolysis [105,148,149]. 1 promoted the reactions of Fur/Gly and Bza/Gly, leading to the corresponding heterobicyclic products of the type dioxolane and dioxane (five and six membered ring acetals, respectively).
More than 80% of the global Fur produced is converted to furfuryl alcohol (FA) for several industrial sectors [150]. Broadening the applications of these heterocyclic biobased chemicals may positively contribute to a biobased economy. Paving the way in this future direction, the catalytic potential of 1 was herein studied for the acid catalyzed reaction of FA with ethanol to the useful bioproducts 2-(ethoxymethyl)furan (EMF) and ethyl levulinate (EL) (Section 2.6) [121,124,[127][128][129][130][131]141,151]. The carbon mass balances closed in at least 96% for the Fur/Gly and Bza/Gly reaction systems, and 89% for the FA/ethanol system.

Reaction of Benzaldehyde and Glycerol to Heterobicyclic Products
Compound 1 was active for the reaction of Bza with Gly, in the temperature range 50-120 • C, without adding co-solvent, leading mainly to the corresponding dioxolane (five membered ring acetal), namely 2-phenyl-1,3-dioxalan-4-yl)methanol, and dioxane (six membered ring acetal), namely 2-phenyl-1,3-dioxan-5-ol (cis/trans mixture), in high total yields of up to 97% (90 • C, 24 h). Catalyst 1 favored the formation of the dioxane, formed in a molar ratio dioxane:dioxolane of up to 3. The predominance of the dioxane product in the Bza/Gly reaction agrees with the literature [97] for various types of catalysts: MoO x -TiO 2 -ZrO 2 (molar ratio dioxane:dioxolane = 1.13) [111], MoO 3 /SiO 2 (1.48) [93], TSA-MCM-48 (1.91) [113] and SO 4 2− /SnO 2 (1.5) [33]. The temperature and molar ratio Gly:aldehyde are important parameters influencing the catalytic reaction [110]. Hence, the Bza/Gly system was firstly studied in the presence of 1, using a molar ratio Gly:Bza of 1:2 or 1:3, and a reaction temperature of 90 or 120 • C, 24 h (Table 1). Increasing the molar ratio Gly:Bza from 1:3 to 1:2 led to higher total acetals yield, especially at 90 • C (70% and 97% respectively, at 24 h; entries 1 and 2, Table 1). On the other hand, the total acetals yields were comparable at 90 and 120 • C, using Gly:Bza = 1:2 (97% and 94% yield at 24 h respectively, entries 2 and 4, Table 1). The total acetals yield did not change considerably between 4 and 24 h reaction at 90 • C, likely being thermodynamically limited (95% and 97% respectively, Gly:Bza = 1:2). These results suggested that 1 may favorably operate at reaction temperatures lower than 120 • C, 4 h reaction, using a molar ratio Gly:Bza = 1:2. fact that Fur is an established platform chemical industrially produced from lignocellulosic biomass, Gly is industrially associated (as a by-product) with biodiesel production and Fur and Bza are aromatic aldehydes commonly found in the slate of compounds produced via biomass pyrolysis [105,148,149]. 1 promoted the reactions of Fur/Gly and Bza/Gly, leading to the corresponding heterobicyclic products of the type dioxolane and dioxane (five and six membered ring acetals, respectively). More than 80% of the global Fur produced is converted to furfuryl alcohol (FA) for several industrial sectors [150]. Broadening the applications of these heterocyclic biobased chemicals may positively contribute to a biobased economy. Paving the way in this future direction, the catalytic potential of 1 was herein studied for the acid catalyzed reaction of FA with ethanol to the useful bioproducts 2-(ethoxymethyl)furan (EMF) and ethyl levulinate (EL) (Section 2.6) [121,124,[127][128][129][130][131]141,151]. The carbon mass balances closed in at least 96% for the Fur/Gly and Bza/Gly reaction systems, and 89% for the FA/ethanol system.

Reaction of Benzaldehyde and Glycerol to Heterobicyclic Products
Compound 1 was active for the reaction of Bza with Gly, in the temperature range 50-120 °C, without adding co-solvent, leading mainly to the corresponding dioxolane (five membered ring acetal), namely 2-phenyl-1,3-dioxalan-4-yl)methanol, and dioxane (six membered ring acetal), namely 2-phenyl-1,3-dioxan-5-ol (cis/trans mixture), in high total yields of up to 97% (90 °C, 24 h). Catalyst 1 favored the formation of the dioxane, formed in a molar ratio dioxane:dioxolane of up to 3. The predominance of the dioxane product in the Bza/Gly reaction agrees with the literature [97] for various types of catalysts: MoOx-TiO2-ZrO2 (molar ratio dioxane:dioxolane = 1.13) [111], MoO3/SiO2 (1.48) [93], TSA-MCM-48 (1.91) [113] and SO4 2− /SnO2 (1.5) [33]. The temperature and molar ratio Gly:aldehyde are important parameters influencing the catalytic reaction [110]. Hence, the Bza/Gly system was firstly studied in the presence of 1, using a molar ratio Gly:Bza of 1:2 or 1:3, and a reaction temperature of 90 or 120 °C, 24 h (Table 1). Increasing the molar ratio Gly:Bza from 1:3 to 1:2 led to higher total acetals yield, especially at 90 °C (70% and 97% respectively, at 24 h; entries 1 and 2, Table 1). On the other hand, the total acetals yields were comparable at 90 and 120 °C, using Gly:Bza = 1:2 (97% and 94% yield at 24 h respectively, entries 2 and 4, Table 1). The total acetals yield did not change considerably between 4 and 24 h reaction at 90 °C, likely being thermodynamically limited (95% and 97% respectively, Gly:Bza = 1:2). These results suggested that 1 may favorably operate at reaction temperatures lower than 120 °C, 4 h reaction, using a molar ratio Gly:Bza = 1:2. The influence of the catalyst mass load was thus studied using Gly:Bza = 1:2, at 50, 70 and 90 °C, until 4 h reaction (since product yields did not increase considerably after this time) (Figures 2 and 3). In the studied ranges of reaction conditions, the dioxane and dioxolane were always the main products, formed in a molar ratio dioxane:dioxolane in the range 2.0-2.8. For each reaction temperature, a 10-fold increase in the catalyst load fact that Fur is an established platform chemical industrially produced from lignocellulosic biomass, Gly is industrially associated (as a by-product) with biodiesel production and Fur and Bza are aromatic aldehydes commonly found in the slate of compounds produced via biomass pyrolysis [105,148,149]. 1 promoted the reactions of Fur/Gly and Bza/Gly, leading to the corresponding heterobicyclic products of the type dioxolane and dioxane (five and six membered ring acetals, respectively). More than 80% of the global Fur produced is converted to furfuryl alcohol (FA) for several industrial sectors [150]. Broadening the applications of these heterocyclic biobased chemicals may positively contribute to a biobased economy. Paving the way in this future direction, the catalytic potential of 1 was herein studied for the acid catalyzed reaction of FA with ethanol to the useful bioproducts 2-(ethoxymethyl)furan (EMF) and ethyl levulinate (EL) (Section 2.6) [121,124,[127][128][129][130][131]141,151]. The carbon mass balances closed in at least 96% for the Fur/Gly and Bza/Gly reaction systems, and 89% for the FA/ethanol system.

Types of Active Species and Mechanistic Insights
The catalytic performance of 1 was compared to that of its synthesis precursors, namely the free organic ligand H 6 nmp and the lanthanide oxide Gd 2 O 3 , used in equivalent molar amounts to those added with 1 in a normal catalytic run (using 3.8 g cat L −1 ). The free ligand H 6 nmp was completely soluble in the reaction media and led the 1,3-dioxolane and 1,3-dioxane products in a total yield of 52% (dioxolane:dioxane = 1.5) at 50 • C/4 h. These results are significant considering that 1 led to 73% total yield under similar conditions. Nevertheless, H 6 nmp acted as a homogeneous organocatalyst requiring more demanding catalyst separation processes, in relation to 1. On the other hand, the lanthanide precursor Gd 2 O 3 led to a sluggish reaction (10% conversion at 50 • C/4 h).
The above results suggested that the catalytic activity of 1 seemed to be associated with Brønsted acidity. The attenuated total reflectance (ATR) FT-IR spectra of the fresh and used 1 were similar, and different from those of the catalyst synthesis precursors, H 6 nmp and Gd 2 O 3 ( Figure 6). The fresh and used solids exhibited characteristic bands of the coordinated organic linker and two very weak bands at ca. 1133 and 919 cm −1 assignable to ν as (PO 2 ) and ν as (POH) of PO 3 H − groups (which may confer Brønsted acidity) [152].   The solids also exhibited a very broad band in the range 3600-3150 cm −1 assigned to the ν(O-H) stretching modes of coordinated water (the solvent used in the catalyst synthesis), a band at ca. 1620 cm −1 due to in-plane deformation δ(H 2 O) bands of water molecules and bands in the range 3100-2800 cm −1 due to (as)symmetric ν(C-H) stretching vibrational modes of the linker (Supplementary Figure S2) [153,154].
The catalytic reaction may be triggered by interactions between the oxygen atom of the carbonyl group of the aldehyde reactant and Brønsted acid sites, according to the mechanistic proposal given in Scheme 2 (exemplified for Fur/Gly) [19,113,155]. Specifically, the formation of a hemi-acetal in the reaction of the aldehyde with Gly may be followed by the elimination of a water molecule and formation of a carbocation [79], which, in turn, suffers an attack involving the inner or terminal hydroxyl group of the Gly molecule, finally giving the cyclic acetals 1,3-dioxolane and 1,3-dioxane, respectively.

Catalyst Stability and Structural Characterization
The stability of 1 was studied using the catalyst for three 4 h batch runs of the Fur/Gly reaction at 50 • C. The catalyst regeneration simply involved washing and drying between runs (details in the Experimental Section 3.2). The conversion and products distribution remained steady in consecutive runs, suggesting that 1 was stable (Figure 7). A contact test (CT) was performed to confirm that the catalytic reaction was heterogeneous. Specifically, Gly was added to the liquid phase of the CT of Fur plus 1 and then left to react (details in the Experimental Section 3.2), which led to 37% conversion at 50 • C/4 h. This result was comparable to that obtained without catalyst (30% at 50 • C/4 h) and considerably inferior to that for a normal catalytic test in the presence of 1 (62-74% conversion, using 3.8-38 g cat L −1 ), suggesting that the catalytic reaction for 1 is heterogeneous. The molar ratios Gd:P of the fresh and used catalyst were similar (ca. 0.4), consistent with the results of the contact test. The elemental mappings of the fresh and used catalysts indicated uniform distributions of gadolinium and linker (based on phosphorous, Figure 8). No considerable changes in morphology of the fresh and used catalyst occurred, albeit the particle size seemed to have decreased somewhat (Supplementary Figure S3 and Figure 8).
The powder X-ray diffraction (PXRD) patterns of the fresh and consecutively used solids were similar, suggesting that the crystalline structure of 1 was preserved (Figure 9). The PXRD patterns of the solids recovered from the Fur/Gly reaction at different temperatures (50, 70 and 90 °C) were also similar, suggesting that 1 was thermally stable (Supplementary Figure S4). The ATR FT-IR spectra of the fresh and used solids were comparable, suggesting that the chemical features were essentially preserved during the catalytic reaction ( Figure 6 and Supplementary Figure S2).  particle size seemed to have decreased somewhat (Supplementary Figure S3 and Figure  8).
The powder X-ray diffraction (PXRD) patterns of the fresh and consecutively used solids were similar, suggesting that the crystalline structure of 1 was preserved (Figure 9). The PXRD patterns of the solids recovered from the Fur/Gly reaction at different temperatures (50, 70 and 90 °C) were also similar, suggesting that 1 was thermally stable (Supplementary Figure S4). The ATR FT-IR spectra of the fresh and used solids were comparable, suggesting that the chemical features were essentially preserved during the catalytic reaction ( Figure 6 and Supplementary Figure S2).  The powder X-ray diffraction (PXRD) patterns of the fresh and consecutively used solids were similar, suggesting that the crystalline structure of 1 was preserved (Figure 9). The PXRD patterns of the solids recovered from the Fur/Gly reaction at different temperatures (50, 70 and 90 • C) were also similar, suggesting that 1 was thermally stable (Supplementary Figure S4). The ATR FT-IR spectra of the fresh and used solids were comparable, suggesting that the chemical features were essentially preserved during the catalytic reaction ( Figure 6 and Supplementary Figure S2).
Catalysts 2021, 11, x FOR PEER REVIEW 13 of 24 Figure 8. SEM and respective elemental mappings (Gd (green); P (red)) for the fresh (a) and used (b) catalyst 1. Figure 9. Powder X-ray diffraction patterns of the consecutively used catalyst in the Fur/Gly reaction at 50 °C.

Other Biobased Systems Involving Heterocyclic Compounds
The catalytic potential of 1 was further explored for the acid catalyzed reaction of heterocyclic furfuryl alcohol (FA) with ethanol (these two reactants are industrially produced from vegetable biomass). The main products were 2-(ethoxymethyl)furan (EMF) and ethyl levulinate (EL), formed in 93% and 95% total yield at 90 and 140 °C respectively, 24 h. FA may be converted to EL via the intermediate formation of the heterocyclic product EMF, which involves furan ring opening promoted by Brønsted acidity [156][157][158]. For the FA reaction at 140 °C, the EMF yield decreased from 37% at 1 h to 3% at 24 h, whereas EL yield increased from 49% to 92% ( Figure 10). Other reaction products included angelica lactone formed in <5% yield. The formation of angelica lactone in the FA/EtOH reaction was reported in the literature for zeolite ZSM-5 as a catalyst [126,159].
The FA reaction at 90 °C in the presence of 1 gave EMF as the main product at 5 h (52% yield) and EL was formed in 25% yield; after 24 h, EMF and EL yields were 35% and 58%, respectively. These results together with those for FA/ethanol at 140 °C indicate that the products' distribution is strongly reaction temperature-and time-dependent. Figure 9. Powder X-ray diffraction patterns of the consecutively used catalyst in the Fur/Gly reaction at 50 • C.

Other Biobased Systems Involving Heterocyclic Compounds
The catalytic potential of 1 was further explored for the acid catalyzed reaction of heterocyclic furfuryl alcohol (FA) with ethanol (these two reactants are industrially produced from vegetable biomass). The main products were 2-(ethoxymethyl)furan (EMF) and ethyl levulinate (EL), formed in 93% and 95% total yield at 90 and 140 • C respectively, 24 h. FA may be converted to EL via the intermediate formation of the heterocyclic product EMF, which involves furan ring opening promoted by Brønsted acidity [156][157][158]. For the FA reaction at 140 • C, the EMF yield decreased from 37% at 1 h to 3% at 24 h, whereas EL yield increased from 49% to 92% ( Figure 10). Other reaction products included angelica lactone formed in <5% yield. The formation of angelica lactone in the FA/EtOH reaction was reported in the literature for zeolite ZSM-5 as a catalyst [126,159].
The FA reaction at 90 • C in the presence of 1 gave EMF as the main product at 5 h (52% yield) and EL was formed in 25% yield; after 24 h, EMF and EL yields were 35% and 58%, respectively. These results together with those for FA/ethanol at 140 • C indicate that the products' distribution is strongly reaction temperature-and time-dependent.
To the best of our knowledge, only two crystalline hybrid materials were reported in the literature for FA conversion to EMF/EL. Specifically, two MOFs based on hafnium and chromium functionalized with sulfonic (Brønsted) acid groups, namely Hf-UiO-66-SO 3 H [145] and MIL-101(Cr)-SO 3 H [146,147]. Hf-UiO-66-SO 3 H led to a moderate EL yield of 62% at 120 • C/2 h [145], and MIL-101(Cr)-SO 3 H [146,147] led to a somewhat comparable EL yield to that for 1, but under different conditions. Specifically, at 100% FA conversion and 140 • C, the chromium MOF led to 76-79% EL yield at 2 h, whereas 1 led to 72% EL yield at 5 h using a lower catalyst/FA mass ratio (0.3 for 1 versus 0.44-1.02 for the chromium catalyst) and using a more concentrated FA mixture (0.33 M for 1 versus 0.28 M for the chromium catalyst). The total selectivity (EMF plus EL) was higher for 1 (90%) than for MIL-101(Cr)-SO 3 H (79%) [146,147]. A detailed comparison of the catalytic results for 1 to literature data for other types of solid acid catalysts is given in the Supplementary Material Section 3 [121,126,134,136,137,[145][146][147]157,. Overall, the catalytic performance of 1 seemed to stand on an intermediate footing among the various types of solid acid catalysts reported in the literature for the FA/ethanol system (e.g., commercial zeolites and acid resins, and modified metal oxides). For example, the organocatalyst HCC-ML-SO 3 H led to a high EL yield of 85% at 90 • C/8 h [161], albeit it was used in a catalyst/substrate mass ratio three times greater than that for 1 (58% EL yield at 90 • C/24 h). Nano ZSM-5 led to 91% yield at 140 • C/1 h [173], while 1 required 24 h to reach a similar EL yield of 92%, using a similar FA concentration (0.33-0.39 M) but requiring a much lower catalyst/FA mass ratio (0.30 for 1 versus 1.02 for ZSM-5). Superior results were reported for HPW-MesoZSM-5 [193], Ti-HPA [183], Fe-USY [194] and magnetic sulfated zirconia (MSZ) [167], which led to 91-97% EL yield at 120-130 • C and 0.5-5 h (for 1: 72% EL at 140 • C/5 h). A higher FA initial concentration was used for HPW-MesoZSM-5 and MSZ (1.04-3.12 M versus 0.33 M for 1) [193], and a lower catalyst/FA mass ratio was used for HPW-MesoZSM5 [193], Ti-HPA [183], Fe-USY [194] and MSZ (0.02-0.07 versus 0.33 for 1) [167]. However, MSZ requires catalyst regeneration at 500 • C to keep the catalytic performance steady, whereas 1 does not require thermal regeneration. To the best of our knowledge, only two crystalline hybrid materials were reported in the literature for FA conversion to EMF/EL. Specifically, two MOFs based on hafnium and chromium functionalized with sulfonic (Brønsted) acid groups, namely Hf-UiO-66-SO3H [145] and MIL-101(Cr)-SO3H [146,147]. Hf-UiO-66-SO3H led to a moderate EL yield of 62% at 120 °C/2 h [145], and MIL-101(Cr)-SO3H [146,147] led to a somewhat comparable EL yield to that for 1, but under different conditions. Specifically, at 100% FA conversion and 140 °C, the chromium MOF led to 76-79% EL yield at 2 h, whereas 1 led to 72% EL yield at 5 h using a lower catalyst/FA mass ratio (0.3 for 1 versus 0.44-1.02 for the chromium catalyst) and using a more concentrated FA mixture (0.33 M for 1 versus 0.28 M for the chromium catalyst). The total selectivity (EMF plus EL) was higher for 1 (90%) than for MIL-101(Cr)-SO3H (79%) [146,147]. A detailed comparison of the catalytic results for 1 to literature data for other types of solid acid catalysts is given in the Supplementary Material Section 3 [121,126,134,136,137,[145][146][147]157,. Overall, the catalytic performance of 1 seemed to stand on an intermediate footing among the various types of solid acid catalysts reported in the literature for the FA/ethanol system (e.g., commercial zeolites and acid resins, and modified metal oxides). For example, the organocatalyst HCC-ML-SO3H led to a high EL yield of 85% at 90 °C/8 h [161], albeit it was used in a catalyst/substrate mass ratio three times greater than that for 1 (58% EL yield at 90 °C/24 h). Nano ZSM-5 led to 91% yield at 140 °C/1 h [173], while 1 required 24 h to reach a similar EL yield of 92%, using a similar FA concentration (0.33-0.39 M) but requiring a much lower catalyst/FA mass ratio (0.30 for 1 versus 1.02 for ZSM-5). Superior results were reported for HPW-MesoZSM-5 [193], Ti-HPA [183], Fe-USY [194] and magnetic sulfated zir-

Materials and Methods
All reagents and solvents were obtained from commercial sources and used as received without further purification, excluding furfural which was distilled prior to use.

Catalyst Synthesis and Characterization
The aqueous phase synthesis of [Gd(H 4 nmp)(H 2 O) 2 ]Cl·2H 2 O (1) was based on that previously reported [116], albeit the synthesis was scaled up to give ca. 0.6 g per batch in a relatively fast fashion, specifically, within 40 min under microwave irradiation at 70 • C (75 W power) with stirring (details in the Supplementary Material Section 1).
Powder X-ray diffraction (PXRD) data were collected at ambient temperature on an Empyrean PANalytical diffractometer (Panalytical, Almelo, The Netherlands) (Cu K a1,2 X-radiation, λ 1 = 1.540598 Å; λ 2 = 1.544426 Å), equipped with an PIXcel 1D detector and a flat-plate sample holder in a Bragg-Brentano para-focusing optics configuration (45 kV, 40 mA). Intensity data were collected by the step-counting method (step 0.01 • ), in continuous mode, in the range ca. 3.5 ≤ 2θ ≤ 50 • . EDS (energy dispersive X-ray spectroscopy) data was recorded on a high-resolution Hitachi SU-70 (Monocomp, Madrid, Spain) working at 15 kV, employing a Sprit 1.9 EDS microanalysis system, and samples were prepared by deposition on aluminum sample holders followed by carbon coating using an Emitech K950X carbon evaporator. Attenuated total reflectance (ATR) FT-IR spectra were measured on a Mattson-7000 infrared spectrophotometer (Dias de Sousa S.A., Maia, Portugal) equipped with a Specac Golden Gate Mk II ATR accessory with a diamond top plate and KRS-5 focusing lenses.

Catalytic Tests
The reactions of furfural (Fur) and benzaldehyde (Bza) with glycerol (Gly) were carried out in tubular glass pear-shaped reactors equipped with a Teflon TM -coated magnetic stirring bar and a valve for purging. The batch reactors were loaded with the reactants in a molar ratio Gly: aldehyde of 1:2 or 1:3, and 1 or 10 wt.% catalyst (relative to Gly, which was the limiting reactant). The reaction of furfuryl alcohol (FA, 0.33 M) with ethanol in the presence of 1 (10 g cat L −1 ) was carried out using the same type of reactors, at 90 or 140 • C. For comparative purposes, the free linker and gadolinium precursors, namely H 6 nmp and Gd 2 O 3 respectively, were tested as catalysts, added in an equivalent molar amount to that added together with 1 wt.% (based on Gly) of 1. The reactor containing the reaction mixture was immersed in a thermostatically controlled oil bath preheated at the desired temperature and stirred at 1000 rpm to avoid diffusional limitations. Blank tests without catalysts were performed for each substrate.
Prior to sampling, after each assay, the reactor was cooled to ambient temperature and the solid catalyst was separated by centrifugation at 10,000 rpm. The analyses were always carried out for fresh samples. The evolution of the catalytic reactions was monitored by gas chromatography (GC) using an Agilent 7820A GC (Soquímica, Porto, Portugal) equipped with a capillary column (HP-5, 30 m × 0.32 µm × 0.25 mm) and a flame ionization detector, and product quantification was based on the external calibration. Individual experiments were performed for a given reaction time and the presented results are the mean values of at least two replicates (experimental error < 10%). The product (Prod) yield was calculated relative to the initial amount of the limiting reactant, Gly, using the formula: 100 × ((molar concentration of Prod at time t)/(initial molar concentration of Gly)). Conversion was based on the limiting reactant Gly. For product identification, a Shimadzu QP2010 ultra-GC-MS (Izasa Scientific, Lisbon, Portugal) equipped with a Phenomenex capillary Zebron ZB5-MS column (ZB-5, 30 m × 0.25 µm × 0.25 mm) and He as carrier gas was used, as well as commercial databases Wiley229 and NIST14.
The used catalyst was separated from the reaction mixture by centrifugation at 10,000 rpm, washed with acetone, dried at 85 • C overnight and reused for up to three 4 h batch runs of the Fur/Gly reaction at 50 • C. The contact test (CT) was carried out by heating a stirred suspension of 1 (3.8 g cat L −1 ) in Fur at 50 • C for 4 h, under similar conditions to those used for a normal catalytic test, but without Gly. The solid catalyst was separated from the liquid phase by centrifugation at 10,000 rpm and the supernatant liquid phase was passed through a 220 nm pore size PTFE membrane. Gly was added to the solution (filtrate) to give an initial Gly concentration of 4.2 M (as for a normal catalytic test), and this solution was left to react for 4 h at 50 • C, with stirring, and finally analyzed by GC.

Conclusions
The hybrid catalyst [Gd(H 4 nmp)(H 2 O) 2 ]Cl·2H 2 O (1) was synthesized in aqueous phase, in a relatively fast, mild fashion (40 min, 70 • C), in gram-scale. This material proved to be an effective solid acid catalyst for synthesizing heterobicyclic products of the type 1,3-dioxane and 1,3-dioxolane via reactions of biomass-derived furfural (Fur) and glycerol (Gly), and of benzaldehyde (fossil fuel or biomass derived) and Gly, without requiring co-solvents. The catalyst was stable upon reuse via simple washing/drying procedures, and led to up to 91% total acetals yield in the reaction of Fur/Gly (90 • C, 4 h; molar ratio (2-(furan-2-yl)-1,3-dioxolan-4-yl)methanol)/(2-(furan-2-yl)-1,3-dioxan-5-ol) ∼ = 2.4), and 95% total acetals yield in the reaction of Bza/Gly (90 • C, 4 h; molar ratio of (2-phenyl-1,3-dioxan-5-ol)/(2-phenyl-1,3-dioxalan-4-yl)methanol) ∼ = 2.8); at 24 h/90 • C, the total yields increased to 95% and 97%, respectively. To the best of our knowledge, 1 is the first among crystalline coordination polymers or metal-organic framework (MOF)-type materials studied in the literature for these reaction systems. Studies of the influence of the reaction conditions indicated that a lower Gly:Bza molar ratio and increasing the temperature and catalyst load enhanced the total cyclic acetals yields. Insights into the reaction mechanism and catalyst structural features contributing to the formation of cyclic acetals in these acid catalyzed reactions were put forward. Hybrid 1 seemed to perform relatively well based on comparative studies to literature data for various types of solid acid catalysts. Hybrid 1 was further explored for the catalytic conversion of biobased heterocyclic furfuryl alcohol (FA), industrially produced from Fur, which selectively gave ethyl levulinate (up to 92% yield, at 140 • C, 24 h). Decreasing the reaction temperature and time favored the formation of the intermediate 2-(ethoxymethyl)furan (52% yield at 5 h/90 • C). The performance of 1 compared favorably to that reported in the literature for the only two MOFs previously studied for the FA/ethanol conversion (the two MOFs were based on hafnium and chromium functionalized with sulfonic acid groups).
This work included a detailed comparative study of the results for 1 to literature data for different types of catalysts studied for the target reactions. One may envisage that the desired product distributions and superior catalytic performances may be met by exploring the highly versatile (tuneable physicochemical properties) families of crystalline coordination polymers or MOF-type catalysts for these reaction systems.

Data Availability Statement:
The compound used for the catalysis and raw characterization data studies herein reported are available from the authors upon request.