Catalytic Screening for 1,2-Diol Protection: A Saccharose-Derived Hydrothermal Carbon Showcases Enhanced Performance

Featured Application

This study unveils the outperformance of a renewable sugar-based carbocatalyst for diol protection. This strategy of immense value in academia and industry enables, in addition, the upgrading and valorization of biorefinery products, such as bio-based solvents.

Abstract

A benchmarking study is reported on a series of modified carbocatalysts to efficiently promote the acetalization of 1,2-diols under heterogeneous conditions. Among the catalysts surveyed, a hydrothermal carbon generated from saccharose, a cheap, abundant, and biobased material, showed excellent performance when tested on two representative diols. All catalysts have been thoroughly characterized, focusing on surface acidity and composition. Optimal working parameters such as temperature and catalyst loading could be established. Remarkably, sonication improved the diol protection, which proceeded faster at 25 °C. The catalyst could be easily recycled and reused several times. In addition, the protocol was successfully translated from batch to continuous flow operation using a packed-bed reactor.

1. Introduction

Protection of hydroxy functional groups, especially vicinal 1,2-diols and 1,3-diols, constitutes a key strategy in synthetic pursuits, both for research and industry-oriented targets. Ideally, the protecting group can be removed after installing other functions and/or achieving a specific derivatization. Although, in line with the green chemistry principles, the development of protecting-group-free routes is desirable to minimize additional reagents and side products [1], and sometimes, there is no avoiding functional protection as the target is too complex. Figure 1 displays a series of polyhydroxylated compounds (1–7) whose complexity prevents straightforward access without relying on protecting groups, such as bioactive saccharides or aminoglycosides [2,3]. Ingenol (6) mimics the role of diacylglycerol, the endogenous activator of protein kinase C [4]. Molnupiravir (7) is particularly attractive, as this modified nucleoside was developed and licensed in less than two years to treat SARS-CoV2 infection. It can be obtained from uridine in five steps, the key ones involving acetal protection of vicinal OH groups [5]. On a larger scale, glycerol, the major residue during biofuel preparation, can be upgraded by reaction with acetone. The resulting acetonide (2,2-dimethyl-1,3-dioxolan-4-yl)methanol is marketed as solketal (8). This bio-based green solvent has found other applications, such as a plasticizer or a safe additive for pharmaceutical formulations [6,7].

Selective protection of vicinal diols can be accomplished by numerous methods [8,9,10], albeit acetalization represents a facile and convenient protocol affording the corresponding five-membered dioxolanes, usually in high yields. Major efforts have been directed toward heterogeneous catalysis in the search for sustainable approaches that bypass the use of harsh acid conditions and allow for rapid separation and purification. Both metal-doped and metal-free catalysts have been developed [11].

In a recent study, we have shown that a sulfonated carbocatalyst derived from an activated carbon enabled efficient protection–deprotection of 1,2-diols under heterogeneous conditions [12]. Here, we report on a systematic screening of carbocatalysts tailored for vicinal diol acetalization, which can be easily generated from readily available precursors. In particular, we show that a hydrothermal carbon (HTC) obtained from saccharose led to enhanced results relative to other heterogeneous catalysts employed. This catalyst improves the previous protocol in terms of reaction times, catalyst efficiency, catalyst stability, and sustainability [12]. Hydrothermal carbonization has emerged in recent decades as a promising alternative to other pyrolytic and carbonization methods, using milder conditions and leading to materials with suitable microporosity and structuredness for a broad range of applications [13,14].

Given the acid character required for acetalization, other modified catalysts have been tested, as discussed below. Sulfated zirconia (Zr-S) exhibiting high acidity appears to be an appropriate material for this end [15,16], which has been used for alkane isomerization at low temperatures [17]. It combines both Brönsted and Lewis acidity due to the electron-withdrawing effect of sulfate groups bonded to the zirconia surface. Previous research evidenced that Zr-S had comparable or superior activity to other solid catalysts, including zeolites, ion exchange resins, and sulfated metal oxides such as titania and tin oxide, for the esterification of lauric acid with alcohols [18]. Moreover, Zr-S efficiently catalyzed the esterification of various saturated/unsaturated fatty acids [19,20] and also the simultaneous esterification and transesterification of oils [21,22].

2. Materials and Methods

2.1. Analytical and Spectroscopic Acquisition Data

Adsorption of N₂ was performed at 77 K in a Quadrasorb Evo equipment from Quantachrome Instruments (Boynton Beach, FL, USA), with a previous degasification at 120 °C for 24 h. Elemental analyses were determined on a Leco CHNS-932 analyzer (Leco, St. Joseph, MI, USA). X-ray fluorescence spectra were recorded on a Bruker S8 Tiger instrument (Bruker, Billerica, MA, USA). X-ray photoelectron spectroscopy (XPS) analyses were performed on FlexPS-ARPES-E equipment from SPECS (Seongnam, Republic of Korea) with monochromatic Kα Al radiation, 12 KV voltage, and 6 mA current. Binding energy values were internally referenced to the C1s peak at 284.8 eV. Point of Zero Charge (PZC) measurements were obtained following the method described by Nabais and Carrott [23]. For the determination of acid groups, 0.15 g of sample was mixed with 30 mL of 0.01 M NaOH, and these experiments were performed in triplicate. The mixture was stirred for 24 h at 25 °C, filtered, and the resulting filtered solution (20 mL) was titrated with 0.01 M nitric acid. For the quantification of basic groups, 0.15 g of sample was mixed with 30 mL of 0.01 M HNO₃, and the mixture was stirred for 24 h at 25 °C, filtered, and the resulting filtered solution was titrated with 0.01 M NaOH. Phenolphthalein was used as an indicator. Acid and basic sites were expressed as mEq/g. FT-IR spectra were recorded in the range of 4000−400 cm⁻¹ on a THERMO spectrophotometer (THERMO, Waltham, MA, USA) using an ATR (attenuated total reflectance) device. Scanning electron microscopy (SEM): SEM images were obtained using QUANTA 3D FEG (FEI Company, Hillsboro, OR, USA) with a secondary electron detector, 15.0 kV acceleration voltage, and at high vacuum. Melting points were determined on an Electrothermal IA 9000 apparatus (Electrothermal, London, UK) and were uncorrected. NMR spectra were recorded at 298 K on Bruker Ultrashield TM 500 Plus and Bruker Ultrashield TM 300 instruments. Samples were dissolved in CDCl₃ using tetramethylsilane (Me₄Si, TMS) (δ = 0.00 ppm) as an internal reference. Thin-layer chromatography (TLC) was performed on aluminum sheets coated with 60 F₂₅₄ silica from Merck (Rahway, NJ, USA).

2.2. Sonochemical Reactions

Reactions activated by ultrasound were performed using a JP Selecta ultrasound bath with a nominal frequency of ~40 kHz. Vials were put inside cavitation intensifying bags (CIBs), obtained from BuBble Clean^® (Enschede, The Netherlands), then immersed into the bath and irradiated for a given time. Reactions were conducted at ambient temperature (25 ± 2 °C) as monitored by a thermometer. To avoid overheating caused by prolonged sonication, the temperature was kept constant by pouring ice into the bulk liquid.

2.3. Synthesis of Catalysts

2.3.1. Zirconia (Zr)

Zirconia, ZrO₂, was synthesized from ZrCl₄ by treatment with aqueous ammonia solution according to the experimental procedure reported by Scurrell [24].

2.3.2. Sulfated Zirconia (Zr-S)

Zirconia was subsequently sulfated using H₂SO₄, following the reported procedure [15].

2.3.3. Norit with Zirconia (N-Zr)

Activated carbon Norit (1 g) was added to a solution of ZrCl₄ (1 g in 10 mL of water). Ammonia (30%) was subsequently added dropwise under stirring until a pH value of 10 was obtained. The suspension was stirred for 1 h, filtered, and rinsed several times with water.

2.3.4. Norit with Zirconia Treated with Concentrated Sulfuric Acid (N-Zr-S1)

N-Zr (1 g) was treated with 15 mL of commercial sulfuric acid (98%) under stirring for 90 min. The solid was filtered, rinsed several times with water, and dried in an oven.

2.3.5. Norit with Zirconia Treated with Diluted Sulfuric Acid (N-Zr-S2)

N-Zr (1 g) was treated with 15 mL of 0.5 M sulfuric acid under stirring for 30 min. The solid was filtered, rinsed several times with water, and dried in an oven.

2.3.6. Norit Hydrothermally Treated with Saccharose (N-HT)

Norit (12 g) was added to 50 mL of an aqueous solution of saccharose (0.10 g L⁻¹), and the resulting mixture was placed in a PTFE vessel reinforced with a stainless-steel outer jacket. After a thermal treatment at 160 °C for 15 h, the solid was filtered, rinsed several times with water (200 mL), and dried at 110 °C for at least 8 h.

2.3.7. Norit Hydrothermally Treated with Saccharose and Sulfuric Acid (N-HT-S)

N-HT was treated with concentrated sulfuric acid (98%) at 25 °C for 90 min, using a ratio of 20 mL of acid per gram of N-HT. The solid was washed with distilled water (until no sulfates were detected in the washings) and dried at 110 °C for at least 8 h.

2.3.8. Carbon Merck^® Treated with Sulfuric Acid (M-S)

Activated carbon Merck^® (M) was treated with concentrated sulfuric acid (98%) at 25 °C for 90 min, using a ratio of 20 mL of acid per gram of M. The solid was washed with distilled water (until no sulfates were detected in the washings) and dried at 110 °C for at least 8 h.

2.3.9. Carbon Merck^® Treated with Nitric Acid (M-N)

Activated carbon Merck^® (M) was treated with concentrated nitric acid (65%) at 25 °C for 90 min, using a ratio of 20 mL of acid per gram of M. The solid was washed with distilled water and dried at 110 °C for at least 8 h.

2.3.10. Hydrothermal Carbon from Saccharose (HT)

An aqueous solution of saccharose (20 g in 50 mL of distilled water) was placed in a PTFE vessel reinforced with a stainless-steel outer jacket. After a thermal treatment at 180 °C for 20 h, the solid was filtered, rinsed several times with water (200 mL), and dried at 110 °C for at least 8 h.

2.3.11. Hydrothermal Carbon from Saccharose Treated with Sulfuric Acid (HT-S)

Hydrothermal carbon from saccharose (HT) was treated with concentrated sulfuric acid (98%) at 25 °C for 90 min, using a ratio of 20 mL of acid per gram of HT. The solid was washed with distilled water (until no sulfates were detected in the washings) and dried at 110 °C for at least 8 h.

2.4. Synthesis of 1,3-Dioxolanes in Batch Mode (Sealed Vial)

2.4.1. 4-((2-Methoxyphenoxy)methyl)-2,2-dimethyl-1,3-dioxolane (11)

A mixture of 3-(2-methoxyphenoxy)propane-1,2-diol (0.5000 g, 2.52 mmol), the catalyst HT-S (0.0251 g), and 2,2-dimethoxypropane (5 mL) was placed in a sealed vial at 80 °C under stirring. After 20 min, the catalyst was filtered through a glass microfiber filter with a nylon membrane of 0.22 μm, and the resulting filtrate was concentrated under reduced pressure to give acetonide 11 as a white crystalline solid (0.5803 g, 2.44 mmol, 97%). M.p. 50–51 °C. IR (ATR) ῡ_max 1593, 1578, 1510, 1465 (arom), 1183, 1147, 1123, 1056 (C-O-C) cm⁻¹. ¹H-NMR (500 MHz, CDCl₃) δ 6.97–6.88 (m, 4H, H-arom), 4.55–4.50 (m, 1H, CH), 4.18 (dd, J = 8.5 Hz, J = 6.5 Hz, 1H, CH₂), 4.12 (dd, J = 9.5 Hz, J = 5.0 Hz, 1H, CH₂), 3.98–3.94 (m, 2H, CH₂), 3.85 (s, 3H, OCH₃), 1.46 (s, 3H, CH₃), 1.39 (s, 3H, CH₃) ppm. ¹³C-NMR (125 MHz, CDCl₃) δ 149.9, 148.3, 122.0, 121.0, 114.6, 112.2 (C-arom), 109.7 ([CH₃]₂C), 74.1 (CH), 70.3 (CH₂), 67.3 (CH₂), 56.0 (OCH₃), 26.9 (CH₃), 25.5 (CH₃) ppm. Calculated analysis for C₁₃H₁₈O₄ (238): C, 65.53; H, 7.61. Found: C, 65.60; H, 7.50.

2.4.2. 4-((4-Chlorophenoxy)methyl)-2,2-dimethyl-1,3-dioxolane (12)

A mixture of 3-(4-chlorophenoxy)propane-1,2-diol (0.5071 g, 2.50 mmol), the catalyst HT-S (0.0249 g), and 2,2-dimethoxypropane (5 mL) was placed in a sealed vial at 80 °C under stirring. After 20 min, the catalyst was filtered through a glass microfiber filter with a nylon membrane of 0.22 μm, and the resulting filtrate was concentrated under reduced pressure to give acetonide 12 as a white crystalline solid (0.5664 g, 2.33 mmol, 93%). M.p. 42–43 °C. IR (ATR) ῡ_max 1596, 1581, 1490, 1451 (arom), 1203, 1169, 1152, 1051 (C-O-C) cm⁻¹. ¹H-NMR (500 MHz, CDCl₃) δ 7.23 (d, J = 9.0 Hz, 2H, H-arom), 6.84 (d, J = 8.5 Hz, 2H, H-arom), 4.47 (q, J = 6.0 Hz, 1H, CH), 4.16 (dd, J = 8.5 Hz, J = 6.5 Hz, 1H, CH₂), 4.02 (dd, J = 9.5 Hz, J = 5.5 Hz, 1H, CH₂), 3.93–3.87 (m, 2H, CH₂), 1.46 (s, 3H, CH₃), 1.40 (s, 3H, CH₃) ppm. ¹³C-NMR (125 MHz, CDCl₃) δ 157.3, 129.5, 126.2, 116.0 (C-arom), 110.0 ([CH₃]₂C), 74.1 (CH), 69.2 (CH₂), 66.9 (CH₂), 26.9 (CH₃), 25.5 (CH₃) ppm. Calculated analysis for C₁₂H₁₅ClO₃ (243): C, 59,39; H, 6,23. Found: C, 59,50; H, 6,07.

2.5. Synthesis of 4-((4-Chlorophenoxy)methyl)-2,2-dimethyl-1,3-dioxolane (12) in Flow Mode

A 0.20 M solution of 3-(4-chlorophenoxy)propane-1,2-diol in 2,2-dimethoxypropane, prepared in a volumetric flask, was pumped using a syringe pump (Infusetek CFSP-I) into a packed bed reactor containing 353 mg of HT-S with a flow rate of 1.246 mL/min (30 s contact time with the catalyst bed). The system was pressurized to 7 bar using a backpressure regulator (Upchurch Scientific, Washington, DC, USA) to stabilize the system. Under steady-state conditions, the liquid stream was collected from the reactor output for 60 min. The crude mixture was evaporated under reduced pressure, resulting in 3.447 g (14.2 mmol, 95% yield) of the title compound. Spectroscopic data matched the values reported in Section 2.4.2).

2.6. Computational Details

DFT computational calculations were performed using the M06-2X [25] hybrid density functional in conjunction with the 6-311++G(d,p) basis set [26,27], without any geometrical restrictions, as implemented in the Gaussian16 package (Wallingford, CT, USA) [28]. In all cases, frequency calculations were also carried out to confirm the existence of energy minima by the absence of imaginary frequencies. Solvent effects (chloroform) were modeled through the solvation model density (SMD) method [29]. NMR chemical shifts were calculated with the GIAO method [30], using the optimized structure of tetramethylsilane as a reference.

3. Results and Discussion

3.1. Synthesis and Characterization of Catalysts

The catalysts employed in this study (including their precursors) are listed in

Table 1. Catalysts employed in this study.

Code	Description
N	Norit (commercially available)
Zr	ZrO2
Zr-S	Sulfated ZrO2
N-Zr	N with ZrO2
N-Zr-S1	N-Zr treated with concentrated sulfuric acid
N-Zr-S2	N-Zr treated with diluted sulfuric acid
N-HT	N hydrothermally treated with saccharose
N-HT-S	N-HT treated with sulfuric acid
M	Commercially available (supplier: Merck®)
M-S	M treated with sulfuric acid
M-N	M treated with nitric acid
HT	Hydrothermal carbon from saccharose
HT-S	HT treated with sulfuric acid

. Synthetic details can be found in Section 2.

Parent-activated carbons Merck (M) and Norit (N) are essentially microporous. According to the IUPAC classification, they display type IV isotherms [31], which suggests the presence of mesopores, some of which have narrow apertures (Figure 2 and Figure 3). Both solids display H4 loops, which are closed at P/P₀ = 1, consistent with the presence of slit-shaped mesopores. The porosity of these solids is not altered substantially after treatment with acid oxidants. Hydrothermal carbons derived from saccharose (HT and HT-S), however, showed very low adsorption capacity as evidenced by their poor porous development (Figure 3). Hydrothermal heating of N in the presence of saccharose to give N-HT notably increases the porosity. Further treatment of this material with sulfuric acid to form N-HT-S reduces its porosity.

The behavior is different for those carbons constituted by zirconia (Figure S1). The deposition of zirconia on Norit (N-Zr) hampers the access of N₂ into the pores, probably due to their blockage, thereby reducing the porosity of this material. Treatment with diluted sulfuric acid (N-Zr-S2) promotes the lixiviation of this mineral, resulting in a higher porosity. This effect is even more significant when concentrated sulfuric acid is used (N-Zr-S1) since a higher amount of ZrO₂ is lixiviated from the solid. Zr and Zr-S also show type IV adsorption isotherms with a lower proportion of micropores. Furthermore, the higher slope of the curves in the middle range of pressure values suggests that the volume of mesopores is larger for Zr and Zr-S compared to the carbonaceous materials grafted with zirconia.

Table 2. Specific surface values of catalysts.

Catalyst	Specific Surface (m2 g−1)
N	1249
Zr	352
Zr-S	311
N-Zr	888
N-Zr-S1	1248
N-Zr-S2	952
N-HT	1412
N-HT-S	1036
M	839
M-S	847
M-N	895
HT	27
HT-S	0.031

gathers the specific surface values of the catalysts obtained after applying the BET model to their N₂ adsorption isotherms [32]. As expected, the higher the porous development, the larger the specific surface since the physic structure of the catalyst does not change after the treatment employed. Activated carbons possess higher surfaces; zirconia displays medium-range values, and hydrothermal carbons exhibit almost negligible specific surfaces.

Elemental analysis data of the catalysts are collected in

Table 3. Elemental analyses of catalysts.

Catalyst	C (%)	H (%)	N (%)	S (%)	O (%)
N	89.1	1.4	0.5	0.5	8.6
N-Zr	77.6	3.3	0.7	0.4	18.0
N-Zr-S1	70.8	3.2	0.5	0.8	24.8
N-Zr-S2	70.7	4.4	0.5	2.2	22.3
N-HT	93.5	0.8	1.4	0.5	3.8
N-HT-S	86.8	0.2	0.2	0.9	11.9
M	91.9	0.7	0.7	0.8	6.0
M-S	83.3	1.4	0.7	1.4	13.2
M-N	85.6	1.0	1.0	0.7	11.7
HT	64.3	6.7	0.0	0.0	29.0
HT-S	59.6	3.2	0.0	3.4	33.8

(excluding mineral matter). Oxygen content was calculated by difference. As expected, carbonaceous materials are mainly composed of carbon. Treatment of N-Zr, M, N-HT, and HT with sulfuric or nitric acid raises the oxygen content, a fact consistent with the oxidation of the solids. The percentage of sulfur and nitrogen also increases depending on the acid used in this process. Hydrothermal treatment with saccharose of carbon N to give N-HT leads to a lower content of hydrogen and oxygen, whilst the ratio of other elements is enhanced, in agreement with a higher graphitization of the solid.

Hydrothermal carbon HT shows a minor amount of C and a higher content in H and O than that showed by activated carbons such as M or N since the hydrothermal process is conducted at lower temperatures than pyrolytic carbonization and, therefore, oxygenated groups are degraded in a minor extent. The amount of oxygen and sulfur is also increased when HT is treated with sulfuric acid, as stated above for other catalysts.

Zirconium-based carbons show a different behavior.

Table 4. Elemental compositions of zirconium-based catalysts.

Catalyst	C (%)	H (%)	N (%)	S (%)	O + Ash (%)	Zr (%) a
N	84.4	1.3	0.5	0.5	13.4	0.0
N-Zr	46.2	2.0	0.4	0.2	51.2	25.6
N-Zr-S1	66.0	3.0	0.4	0.8	29.8	1.1
N-Zr-S2	45.6	2.8	0.3	1.4	49.9	22.3
Zr	-	-	-	0.0	26.0	74.0
Zr-S	-	-	-	2.5	28.1	69.4

^a Determined by X-ray fluorescence.

shows global compositions (both organic and inorganic), including Zr and Zr-S, for comparative purposes. The addition of zirconia to N (N-Zr) results in a lower proportion of carbon while increasing the mineral content. Treatment with concentrated sulfuric acid (N-Zr-S1), with the aim of sulfonating the carbonaceous precursor, promotes the lixiviation of mineral content, which is consistent with a lower ratio of zirconium and a larger amount of the other elements. When N-Zr is treated with diluted sulfuric acid (N-Zr-S2), with the idea of sulfating the deposits of zirconia on carbon N, lixiviation occurs in a minor extension, and the amount of sulfur increases significantly. With respect to Zr-S, although sulfur content is important, it is mainly composed of zirconium and oxygen, as expected. Trace impurities, not included in the table, constituted by silicon and copper, were detected in the two mineral residues of these catalysts.

Elemental compositions measured by XPS are collected in

Table 5. Elemental compositions obtained by XPS.

Catalyst	C (%)	O (%)	N (%)	S (%)	Zr (%)
N	95.5	4.5	0.0	0.1	-
N-Zr	72.5	18.0	0.5	0.0	7.1
N-Zr-S1	88.6	8.5	0.6	0.3	0.2
N-Zr-S2	68.7	22.2	0.3	1.3	7.5
N-HT	95.0	4.8	0.0	0.1	-
N-HT-S	93.2	5.6	0.0	0.4	-
M	94.1	5.0	-	0.1	-
M-S	93.1	5.7	-	0.3	-
M-N	91.2	7.1	-	0.3	-
HT	78.0	22.0	-	-	-
HT-S	73.8	24.6	-	1.6	-

. In general terms, these results are in agreement with those obtained by elemental analysis (Table 3): carbon is the main component of carbonaceous materials; the oxygen content of HT is much larger than that detected for the rest of the carbonaceous catalysts; treatment with acids gives rise to the oxidation of the carbons with increased content of nitrogen or sulfur, depending on the acid employed; treatment of N-Zr with concentrated sulfuric acid leads to a substantial lixiviation of zirconia (N-Zr-S1), decreasing the amount of zirconium and oxygen; sulfur is not detected in Zr, whereas in Zr-S, it is present in a 2.5%. Moreover, the chemical treatment with acids decreases the oxygen content when compared to data obtained by elemental analysis, suggesting that oxygenated groups are degraded further when they are attached to the surface of the catalyst.

Deconvoluted C 1s data of carbonaceous catalysts are collected in Table S1. Most of the catalysts display three components: one is detected at 284.8 eV, corresponding to the reference and assigned to the most reduced form of carbon (C-C and C-H bonds); a second component is observed at 286 eV, attributable to carbon atoms partially oxidized (i.e., ether, alcohol, aldehyde); and a third component close to 290 eV, due to highly oxidized forms of carbon like carboxylic or ester groups [12].

Formation of zirconia causes a minor change in the composition of parent carbon N since the reduced fraction decreases, whereas the most oxidized forms increase. Such an effect is even more significant after the treatment with sulfuric acid, acting as an oxidant as well. Hydrothermal treatment of N to give N-HT barely modifies the chemical composition of the surface, but treatment with sulfuric acid drives the oxidation of the material (Figure 4).

A similar trend is observed when M is treated with acids. As expected, the oxidation is larger with nitric acid (M-N). In this case, four components are distinguished: the peak at 285.7 eV can be assigned to C–O bonds; the peak at 287.6 eV is assigned to carboxyl groups mainly, and the peak at 290.7 eV is attributable to carboxyl or derivatives.

Hydrothermal carbon derived from saccharose showed a different spectrum since it possesses the highest oxygen content and only two components are present: the reduced form is minor, whilst the major component is highly oxidized (component 3, 289 eV). However, after the treatment with sulfuric acid, the latter component disappears, showing instead the intermediate component at 286.5 eV (Figure 4). This suggests that the acid reacts with the most oxidized functional groups, removing them from the catalyst surface.

The interpretation of O 1s XPS spectra in the range of materials studied, namely, activated or hydrothermal carbon, along with other carbonized substrates (Table S2), is complex indeed as one functional group gives rise to more than one peak and, in addition, different functional groups often lead to close signals. Therefore, no clear-cut interpretations can be formulated accurately. However, it is worth pointing out that the signal observed near 531 eV is consistent with the presence of both oxygen and metal in the N-Zr, N-Zr-S1, N-Zr-S2, Zr, and Zr-S samples.

Deconvoluted N 1s data (Table S3) show one main component coming from the pristine carbon (N or M). These nitrogen atoms are not completely oxidized (like nitro, nitroxy, or nitrate groups), so a peak above 405 eV should be detected.

S 2p deconvoluted components (Table S4) can be divided into two groups. Reduced forms of sulfur, like organic and inorganic sulfides, are assigned to the peak at 164 eV. On the other hand, the peak at 168 eV is due to oxidized forms such as sulfates, sulfones, or sulfonic acids.

The treatment of N-Zr with sulfuric acid increases the amount of sulfur in an oxidized form due to the presence of sulfonic groups (N-Zr-S1) or sulfates (N-Zr-S2), where the presence of four oxygen atoms around the sulfur agrees with the shift of the peak to higher values of binding energies.

The hydrothermal treatment of N with saccharose modifies the ratio of the two types of sulfur atoms. This change is more favorable toward oxidized forms of sulfur after the treatment with sulfuric acid. The same behavior is observed for HT-S, although, in this case, the precursor does not contain sulfur, thereby proving that the sulfur present in this catalyst comes from sulfuric acid. Interestingly, a small amount of sulfur is reduced in this reaction as a peak at about 164 eV is also detected. The duplicity of some peaks (163.4 and 164.5 eV; 167.4 and 165.8 eV in HT-S) is due to the existence of 2p_3/2 y 2p_1/2 configurations. In most cases, however, the low relation signal/noise of the spectrum hampers the visibility of both multiplicities.

The S 2s signal of Zr-S contains two peaks, 226 and 234 eV. The first peak can be associated with the presence of adventitious carbon. The second peak can be attributed to the formation of sulfates on the surface of ZrO₂. Another peak is observed at 169.7 eV in the S 2p spectrum, corresponding to sulfates groups, not detected in Zr.

Deconvoluted Zr 3d spectrum gives two peaks at 183.0 y 185.5 eV, typical for ZrO₂ and assigned to 3d_5/2 y 3d_3/2 multiplicities, respectively. Likewise, the Zr 3p spectrum shows two peaks at 334 and 338 eV in Zr and Zr-S.

PZC values and quantification of acid and basic sites of catalysts are collected in

Table 6. PZC values, acid, and basic sites of catalysts.

Catalyst	PZC a	Acid Groups b	Basic Groups b
N	9.1	0.32	0.56
Zr	7.8	1.07	0.76
Zr-S	3.9	1.57	ND c
N-Zr	7.2	0.68	0.81
N-Zr-S1	4.9	0.93	0.09
N-Zr-S2	4.7	1.34	0.13
N-HT	9.2	0.43	0.47
N-HT-S	4.4	0.74	0.12
M	10.0	0.53	0.25
M-S	3.8	0.76	0.03
M-N	7.4	0.82	0.12
HT	2.8	1.96	ND c
HT-S	1.9	ND c	ND c

^a Point of zero charge. ^b Measured as mEq g⁻¹. ^c Not determined due to the high acidity of these solids.

. Carbon N is slightly basic, as observed for M and Zr. N-HT also possesses a basic character, suggesting that the hydrothermal treatment with saccharose barely alters the PZC measurement. On the other hand, the carbonaceous material prepared by hydrothermal treatment of saccharose shows high acidity. Chemical treatment with acids leads to a reduction in PZC values, including that of HT. The low PZC value of HT can be explained by a lower degradation experimented with saccharose when subjected to the hydrothermal process, which agrees with the abundance of hydroxyl and carboxyl groups.

Treatment with nitric acid enhances the acidity of the solids due to the formation of oxygenated functional groups with acid behavior, especially carboxylic acids. Likewise, treatment with sulfuric acid drives even more acid materials, a fact that is consistent with the creation of highly acid sulfonic groups. The treatment with acids raises the number of acid groups, whereas basic sites are reduced.

FT-IR spectra (Figures S2–S5) show characteristic OH absorption bands at about 3400 cm⁻¹. Vibrational bands in the range 1500–1600 cm⁻¹ are assigned to C=C bonds typically detected in graphitic carbonaceous materials. Hydrothermal carbons HT and HT-S also display bands at 1700 cm⁻¹, consistent with the presence of carboxylic groups. In general, treatment with sulfuric acid often reveals the detection of absorption bands at ~1100 and ~617 cm⁻¹, attributable to sulfonic groups [12,33].

Figure 5a,b shows the typical spheric morphology of hydrothermal carbons prepared from carbohydrates, with different sphere sizes for both catalysts [34]. Moreover, chemical treatment with sulfuric acid does not alter the characteristic morphology of this material. Nor does it induce any significant change in particle size. Nevertheless, catalysts N-HT and N-HT-S show both an irregular morphological pattern with a lack of spheres (Figure 5c,d), thereby pointing to a modification of the surface of the catalyst without producing an activated carbon/hydrothermal carbon composite.

3.2. Screening of Heterogeneous Catalysts for Diol Protection

Firstly, we investigated the acetalization of two model diols (9 and 10) against a wide variety of heterogeneous catalysts (Scheme 1). These reactions were performed under magnetic stirring at 80 °C in a sealed vial to minimize the evaporation of the solvent and to avoid the presence of moisture in the reaction media. The catalysts employed in this screening are listed above (Table 1). Catalyst load and acetalization reagent were optimized in a previous study [12]. Reactions were monitored by TLC analysis until the complete consumption of starting materials. Purification by column chromatography was only required when analysis of the crude reaction mixtures by ¹H-NMR showed the presence of impurities.

Table 7. Preliminary screening of heterogeneous catalysts for protection of diols 9 and 10.

Entry	Diol	Catalyst (5%) a	Time	Yield
1	9	Zr	1 d	NC d
2	10	Zr	5 d	NC d
3	9	Zr-S	2 h	98% b
4	10	Zr-S	3 h	88% b
5	9	N-Zr	2 d	NC d
6	10	N-Zr	2 d	NC d
7	9	N-Zr-S1	1 d	91%c
8	10	N-Zr-S1	1 d	92%c
9	9	N-Zr-S2	1 d	84%c
10	10	N-Zr-S2	1 d	93%c
11	9	M	8 d	NC d
12	10	M	8 d	NC d
13	9	M-S	1 d	91% c
14	10	M-S	1 d	93% c
15	9	M-N	1 d	93% c
16	10	M-N	1 d	96% c
17	9	N-HT	5 d	81% c
18	10	N-HT	5 d	96% c
19	9	N-HT-S	1 h	93% c
20	10	N-HT-S	3 h	92% c
21	9	HT	9 d	NC d
22	10	HT	9 d	91% c
23	9	HT-S	20 min	97% c
24	10	HT-S	20 min	93% c

^a Percentage of catalyst with respect to the starting weight of diol. ^b Isolated yield after purification by column chromatography. ^c Isolated yield without purification by column chromatography. ^d Reaction not completed.

summarizes the results of this catalytic screening.

The acetalization reaction was not completed for catalysts Zr (entries 1 and 2) and N-Zr (entries 5 and 6). Likewise, catalyst M did not complete the reaction either, even after 8 days, probably due to its basic character. Interestingly, the reaction of diol 9 (entry 21) could not be completed after 9 days with hydrothermal carbon HT, which is strongly acidic, whereas protection of diol 10 was achieved with excellent yield (91%, entry 22) after the same reaction time. Nevertheless, and as will be demonstrated below, these findings show that the acidity of the catalysts is required but not exclusive to reach optimum catalytic activity.

Protection of both model diols was feasible with zirconium-based catalyst Zr-S, providing excellent yields, although purification by column chromatography was required to isolate pure acetonides 11 and 12.

On the other hand, catalysts N-Zr-S1 and N-Zr-S2 (entries 7–10) allowed for the formation of acetonides with excellent yields, but after reaction times of one day. Similarly, the use of catalysts M-S (entries 13 and 14) and M-N (entries 15 and 16) drove excellent yields for 11 and 12. The slightly basic catalyst N-HT also allowed for the isolation of ketals in good yields (entries 17 and 18), even though reaction times were longer, requiring 5 days for completion.

Finally, sulfonated catalysts N-HT-S and HT-S accelerated the acetalization reactions significantly. Protection of diol 9 was completed in one hour in the presence of N-HT-S (entry 19) and remarkably in only 20 min when using catalyst HT-S (entry 23), which was by far the most active catalyst found in this study. Likewise, the acetalization of 10 required 3 h with N-HT-S (entry 20), whereas the reaction catalyzed by HT-S took place again in 20 min with an excellent yield (entry 24).

Therefore, one can conclude that the best catalysts identified in this preliminary study were acid carbons N-HT-S and HT-S since reactions were completed in less than 3 h with excellent yields. Moreover, acetonides 11 and 12 were isolated with no need for purification by column chromatography.

The structure of isopropylidene ketals for 11 and 12 was corroborated by analytical and spectroscopic techniques (see Section 2) and was compared to reported data [12]. The high purity of these compounds was confirmed by their correct elemental analyses and by the fact that both acetonides crystallized spontaneously after evaporation of the solvent, even though both compounds have been previously described as oils [35,36].

Likewise, FT-IR and NMR spectra were consistent with the 1,3-dioxolane structure of 11 and 12 (Figures S8–S12). A complete assignment of NMR signals for methylene hydrogen and carbon atoms of 11 and 12 was achieved through NOESY and HMQC correlation spectra (Figures S13–S16). In addition, DFT calculations were performed with the optimized structures of ketals 11 and 12 (see Methods for computational details). Once the structures of 11 and 12 were optimized without restrictions at the M06-2X/6-311++G(d,p) level of theory [25,26,27,28], including the solvent effect with the SMD model [29], chemical shifts were calculated with the GIAO method [30], using the optimized structure of tetramethylsilane as reference.

Representation of calculated ¹H and ¹³C shifts versus experimental values led to linear plots with excellent correlation coefficients, which agrees with the assignment of methylene hydrogens and carbons of acetonides 11 and 12 from NMR spectra (Figure 6 and Figure 7).

Finally, to rule out the possible catalytic effect caused by lixiviation of the catalyst HT-S, a test reaction was performed at 80 °C, in the presence of 5% of the catalyst, filtering such a catalyst after a reaction time of 5 min. The resulting solution was subsequently stirred without any catalyst at 80 °C for 24 h. NMR monitoring demonstrated that the acetalization did not evolve once the catalyst was removed from the reaction mixture (Figure 8), therefore concluding that the catalytic action was due to the solid HT-S.

3.3. Influence of Temperature on Catalytic Activity

Once we identified N-HT-S and HT-S as the best catalysts in terms of reaction times and isolated yields, the next objective was to study the influence of the temperature on these transformations, which were performed under the same conditions, with varying temperatures.

Table 8. Influence of the temperature on the formation of ketals 11 and 12.

Entry	Reaction	Catalyst	Time	Temperature	Yield
1	9→11	N-HT-S	8.0 h	60 °C	93%
2	10→12	N-HT-S	1 d	60 °C	91%
3	9→11	HT-S	1.0 h	60 °C	93%
4	10→12	HT-S	1.5 h	60 °C	92%
5	9→11	N-HT-S	3 d	40 °C	91%
6	10→12	N-HT-S	8 d	40 °C	88%
7	9→11	HT-S	4.0 h	40 °C	94%
8	10→12	HT-S	3.5 h	40 °C	93%
9	9→11	N-HT-S	7 d	25 °C	98%
10	10→12	N-HT-S	8 d	25 °C	91%
11	9→11	HT-S	1 d	25 °C	92%
12	10→12	HT-S	1 d	25 °C	91%

shows the results of this study.

As expected, the reaction slows down as the temperature decreases. Moreover, diol 9 appears to be more reactive than diol 10. Remarkably, in all experiments, excellent yields were obtained without an additional purification step.

Catalyst HT-S significantly reduced reaction times for all the temperatures studied, and even at 25 °C, the reaction was complete after one day for both diols. As shown in Table 8, reaction times increased up to one week when catalyst N-HT-S was employed instead. Therefore, the sulfonated hydrothermal carbon derived from saccharose, HT-S, is clearly the most efficient among all the catalysts included in this work, in agreement with the fact that such a catalyst showed the highest acidity, with a PZC value of 1.9.

We also performed a kinetic NMR study for the reaction of diol 10 catalyzed by HT-S at 40 °C (Scheme 2, Figure 9). The reaction progresses rapidly as diol 10 and intermediate I_10-12 are consumed, thereby forming acetonide 12 quantitatively in less than 4 h. First-order rate constants of 1.004 h⁻¹ and 0.709 h⁻¹ were determined for the first and second step, respectively, with half-life times of 0.7 and 1.0 h for diol and intermediate (Figure S17).

3.4. Catalyst Recyclability

The catalytic potential of the best catalysts was also explored by testing the reusability of N-HT-S and HT-S for the formation of ketal 11 at 80 °C. Reactions were monitored again by TLC analysis, and after consumption of the starting diol 9, the catalyst was filtered and dried at 120 °C overnight. Figure 10 and Figure 11 and Tables S5 and S6 show the results of this study.

These experiments evidence that the reaction catalyzed by N-HT-S proceeds with excellent yields after five runs, with identical reaction times in the first three runs. Remarkably, hydrothermal catalyst HT-S could be reused up to 10 times, maintaining excellent yields in all runs (>94%). Furthermore, reaction times were significantly shorter. These results show the potential of this catalyst for flow chemistry, as we shall describe later.

3.5. Activation by Ultrasound

A limiting factor for heterogeneous reactions, either solid–liquid or liquid–liquid, is how to enhance mass transfer efficiently, thus favoring mixing. A priori, this mechanical effect can be achieved under vigorous stirring and/or heating. A parallel activation that often works under milder conditions is provided by ultrasonic waves, even using a conventional cleaning bath [37]. By virtue of cavitational implosion in liquids, high local energies are attained within the bubble and at the bubble interface, which globally increases surface area and exposes a fresh layer of the catalyst to the reacting partners. Accordingly, we have also investigated the use of ultrasonic radiation to activate the formation of ketal 11 from diol 9 using the most efficient catalyst, HT-S (

Table 9. Formation of 11 through activation by ultrasonication.

Entry	Time	Yield
1	1.8 h	92%
2 a	1.8 h	93%
3 b	1.0 h	95%

^a Using a BuBble^® bag. ^b Using a BuBble^® bag filled with water.

). The reaction was carried out in a sealed vial under the same conditions, at 25 °C and under sonication. Since the acoustic radiation is not uniformly distributed and its energy is attenuated with distance from the emitter, we turned to a new class of cheap sonoreactors, which involve microstructured plastic bags (commercialized as BuBble^® bag, Figure 12). It has been reported that such bags enhance cavitation and, therefore, considerably accelerate reactions involving the use of heterogeneous catalysts [38].

Results show a significant reduction in the reaction time by applying sonication to the reaction mixture in the vial (entry 1), maintaining excellent yields. This means that the transformation is robust enough and sensitive to the mechanical action of sound waves. As shown above, this reaction required one day when it was magnetically stirred at 25 °C. On the other hand, placing the vial inside a BuBble bag drove to a similar result (entry 2). However, when the bag was filled with water, the reaction was accelerated, and acetonide 11 was formed in only one hour with excellent yield. This result is comparable to that observed in the reaction of the same diol at 60 °C (Table 8, entry 3). Overall, the cavitation intensifying bag appears to behave similarly to membrane emulsification, i.e., capable of making small droplets or smaller particles in the present case at low energy densities [39].

3.6. Diol Protection in Continuous Flow

The efficiency of heterogeneous catalysts can be enhanced when packed-bed reactors are implemented in combination with continuous flow operation [40]. In a packed-bed reactor, the catalyst is loaded into a pressure-resistant column. Then, the reaction mixture is flowed through the bed using a pumping system. As the liquid phase moves through the catalyst particles, a significant increase in the mass transfer properties of the system is achieved, as the catalyst bed acts as a passive mixer. Moreover, the liquid fraction in contact with the catalyst at a given time encounters a very high catalyst-to-substrate ratio, resulting in very fast transformations [41]. Contact times of a few minutes or even seconds are common in continuous-flow reactors. An additional advantage of packed-bed reactors in continuous flow is that the catalytic reaction and the filtration step are performed simultaneously. Thus, an operation step (filtration) can be removed from the process. This advantage, with the good scalability of packed beds, makes this system particularly suitable for industrial implementation [42].

To demonstrate that heterogeneous catalyst HT-S is suitable for continuous processing in a packed-bed reactor, it was loaded into a stainless steel column (3 mm i.d., 12 cm length). The column was plugged with glass wool on both ends to avoid leaching of the catalyst during flow operation. The column, with an inner volume of 0.85 mL, could carry ca. 350 mg of catalyst. Once loaded with the catalyst, the void volume of the packed bed was 0.623 mL. This volume was used to calculate the residence time of the reaction mixture in the reactor. In a typical experiment, a 0.2 M solution of 10 was pumped through the catalyst bed using a syringe pump (Figure 13). The system was pressurized using a 7-bar backpressure regulator. The temperature of the packed-bed reactor was monitored and controlled in a hot plate (see Supplementary Materials for details, Figure S6).

Once the flow system was assembled, the reaction conditions were optimized. Thus, the reactor temperature and residence time were screened. During the optimization process, the solution of 10 in 2,2-dimethoxypropane was continuously pumped through the reactor. Once the temperature and flow rate had been set and were stable, two reactor volumes were discarded (to ensure steady-state conditions), and then an aliquot of the crude reaction mixture was collected in a round bottom flask, evaporated under reduced pressure, and analyzed by ¹H NMR. Gratifyingly, when the reactor temperature was slightly increased from 25 °C to 40 °C, excellent conversion of the substrate was obtained within a residence time of 5 min (

Table 10. Optimization of the packed bed temperature and residence time for the continuous flow protection of diol 10 with HT-S.

Entry	T (°C)	Flow Rate (mL/min)	Residence Time (min)	Yield (%) a
1	25	0.125	5	16
2	40	0.125	5	98
3	60	0.125	5	98
4	80	0.125	5	96 b
5	40	0.208	3	94
6	40	0.312	2	97
7	40	0.623	1	88
8	60	0.208	3	98
9	60	0.312	2	98
10	60	0.623	1	98
11	60	1.246	0.5	99

^a Determined by ¹H NMR. ^b Small amounts of unknown impurities were observed at 80 °C by ¹H NMR monitoring.

, entries 1 and 2). A further increase in the temperature to 60 °C did not have negative effects on the reaction outcome (entry 3), while at 80 °C, small amounts of unknown side products were observed in flow mode (entry 4). Subsequently, the residence time was optimized at 40 °C and 60 °C by increasing the pump flow rate. At 40 °C, the residence time could be decreased to 2 min while still obtaining a very good 97% conversion (entry 6). A further decrease to 1 min resulted in a significant amount of unreacted starting material (88% conversion, entry 7). Notably, at 60 °C, the flow rate could be gradually increased until the residence time of the reaction mixture in the system was only 30 s while maintaining excellent conversion of the starting material (Table 10, entries 8–11). Indeed, under a residence time of 30 s, a slight increase in conversion from 98% to 99% (no traces of starting material observed) was achieved, most likely due to an increase in mass transfer under a higher flow rate. Indeed, ¹H NMR analysis of the crude reaction mixture collected from the reactor output (Figure S7) revealed that the mixture only contained product 12 and solvents. Using the optimal reaction conditions (Table 10, entry 11), the reaction mixture was collected for 60 min in a round bottom flask. Simple evaporation of the crude solution under reduced pressure resulted in 3.447 g (95% yield) of 12.

Then, the stability of the catalyst bed was evaluated in continuous flow mode. For this purpose, the reaction mixture was pumped through the reactor under optimal conditions (60 °C, 1.246 mL/min, 30 s residence time). Aliquots of the crude reaction mixture collected from the reactor output were analyzed at 20 min intervals (Figure 14). Importantly, very good to excellent conversion was obtained throughout this experiment. Thus, although the reaction conversion decreased at 100–120 min, most likely due to a temperature fluctuation in the system, the catalyst performance was still excellent (99% conversion) after 140 min of continuous processing.

4. Conclusions

We have shown that 1,2-diol protection, an indispensable synthetic protocol during the construction of molecular targets containing reactive functional groups, can be easily achieved under heterogeneous conditions using biomass-derived carbocatalysts. An extensive benchmarking of different modified catalysts unveiled the superior performance of a sulfonated hydrothermal carbon obtained from saccharose (HT-S). All catalysts and products have been thoroughly characterized by analytical and spectroscopic methods. It is worth mentioning that the catalyst HT-S featured a higher oxygen content compared to the other catalysts, and sulfur doping did not alter its surface morphology significantly. Remarkably, the resulting acetonides could be readily isolated without requiring chromatographic purification. Two enabling technologies were also employed to improve mass transfer, thus accelerating the heterogeneous reaction. Firstly, ultrasonic activation, using cavitation intensifying bags, led to completion within one hour at 25 °C, while conventional reactions took longer at higher temperatures. Then, the protocol was performed under continuous flow conditions by loading the HT-S catalyst into a packed-bed reactor. In the flow reactor, optimal reaction conditions were set at 60 °C with a residence time of only 30 s, affording the acetonide a nearly quantitative yield. This result holds promise for scaling up and industrial applicability of diol protection under heterogeneous acid-catalysis. In addition, the catalyst could be recycled without losing activity and had excellent stability in flow mode as well.