1. Introduction
Naturally occurring phenols represent a valuable source of aryl-cored compounds with significant potential for a wide range of applications in the chemical industry. Abundant lignocellulosic biomass serves as a major source of these phenolic compounds, which can be obtained through various thermochemical, chemical and/or biotechnological processes [
1,
2,
3]. This renewable feedstock holds great promise for the synthesis of high-value building blocks, polymers, and other valuable materials [
1,
2,
4], which take advantage of the structural stiffness of aromatic core. Biorenewable plant phenolics represent key intermediates with which to develop high-performance multifunctional polymers or composite materials [
5].
However, to fully exploit their potential and integrate them into sustainable industrial practices, it is essential to develop efficient and environmentally friendly upgrading processes [
1,
5,
6,
7]. Such processes should focus on maximizing yield, minimizing waste, and using green chemistry principles to reduce energy consumption and reliance on non-renewable resources.
The presence of oxygen atoms bonded to the benzenoid nucleus in phenolic compounds imparts electron-rich characteristics, making them particularly reactive toward electrophilic aromatic substitution (Ar-S
E) reactions. Aldehydes—either naturally occurring or readily synthesized from (bio)alcohols—act as effective electrophiles and are therefore expected to react efficiently with bio-sourced arenes. The reaction between aldehydes and phenolics is well established and frequently results in the coupling of two arene units, as the initial substitution product can further alkylate a second arene molecule. On this basis, phenolic resins or benzoxazines can be easily obtained by reaction with formaldehyde [
5]. Moreover, other alkylants such as alkyl- or acyl-halides can transform phenolics into vinylic monomers by forming new ether or ester bonds with the phenolic oxygen atom, respectively. Epoxy monomers are also obtainable through interaction with epichlorohydrin, while polyurethanes can be prepared with isocyanates [
5]. Some other specific processes such as bioconversion [
8], continuous flow methods [
9], or reductive catalytic technologies [
10] have also been described.
Going back to the interaction between phenolics and formaldehyde, it is interesting to note that the outcome of the process can be profoundly changed by the presence of chloride ions. First described by Grassi-Cristaldi and Maselli in 1898 [
11] and later expanded upon by Stephen [
12], Blanc [
13], and Quelet [
14], this effect is attributed to a strong interaction between the halide and the aldehyde, leading to the formation of a chlorinated intermediate—such as a Cl,O-acetal [
15]—which acts as the true electrophilic species in the reaction with the aromatic partner.
Despite the long history of the Blanc–Quelet reaction [
16], its application to bio-based phenols has been the subject of only a few studies [
17]. In this work, we investigated the application of the Blanc–Quelet procedure for the selective chloromethylation of representative bio-based phenols, including vanillin, salicylaldehyde, piceol, eugenol, guaiacol, and
p-cresol. The resulting benzyl halides (
2a–
f,
Figure 1) serve as valuable building blocks for the production of bio-based materials and chemicals [
18,
19,
20]. Diarylmethanes, novolac-, resol-, or reosol-type oligomers and resins are some of the valuable derivatives obtainable from benzyl chlorides.
One interesting application of compounds
2a–
f, which exemplify an extension of their synthetic versatility, is their amidation to form benzyl formamides (
3a–
f,
Figure 1). These molecular targets exhibit good potential as a source of many other functionalized building blocks and materials. Formamides are highly versatile intermediates, serving as precursors to a variety of functional groups and scaffolds. They can be converted into amines [
21,
22,
23] or isocyanides [
24,
25,
26] and may also be transformed into isocyanates [
27], which in turn enable the synthesis of urethanes [
28], ureas [
29,
30], and thiocarbamates [
31]. Additionally, specialized methods have been developed for coupling formamides with various synthons to generate valuable structures such as aminonitriles [
32], tetrazoles [
33], and lactams [
34], which are often key structural motifs in low-molecular-weight active pharmaceutical ingredients.
This synthetic approach represents a novel and practical strategy for upgrading bio-based phenols into valuable chemical building blocks.
2. Materials and Methods
2.1. General Information
Reagents and solvents were reagent-grade products and were used without further purification. Vanillin (1a, Carlo Erba, Cornaredo (MI), Italy), salicylaldehyde (1b, Merk Life Science S.r.l., Milano, Italy), eugenol (BLD Pharmatech GmbH, Reinbek, Germany), guaiacol (Merk Life Science S.r.l., Milano, Italy), piceol (BLDpharm), p-cresol (TCI Europe N.V.), paraformaldehyde (PF, Merk Life Science S.r.l., Milano, Italy) and triethylamine (TEA, Merck Life Science S.r.l., Milano, Italy) were used as purchased.
Elemental analyses were performed with a FLASH 2000 CHNS/O elemental analyzer (Thermo Fisher Scientific Inc., Rodano (MI), Italy). 1H NMR and 13C NMR spectra were recorded on an Avance 600 spectrometer (Bruker, Billerica, MA, USA). Melting points were recorded on a MP30 melting point system (Mettler-Toledo S.p.A., Milano, Italy) with a 3 °C/min heating ramp. Most of the chloromethylated compounds decomposed upon heating before reaching their melting point. This behavior is attributed to their self-curing phenol-formaldehyde monomeric nature.
The compounds
2a [
35],
2b [
36],
2c [
37], and
2f [
38] are known and already characterized.
2.2. Chloromethylation
General procedure. The phenolic substrate (10 mmol) was mixed with PF (from 12 to 42 mmol, depending on the number of chloromethyl groups incorporated into the product) and aqueous 37% HCl (from 6.5 to 25.0 mL). Concentrated H2SO4 (96%, up to 1.0 mL) was also added in the case of electron-poor phenolics. In some cases, dichloromethane (DCM, up to 2.0 mL) was added as a co-solvent. The reaction mixture was stirred at a specific temperature (from 25 °C to 100 °C) for several hours until the disappearance of the substrate, which was evaluated through TLC. Filtration on a Gooch glass filter (P3) followed by aqueous 10% NaHCO3 (10 mL) and water washing (2 × 10 mL) allowed for the isolation of the product. Alternatively, extraction with DCM (3 × 10 mL) and washing of the organic phase with aq. 10% NaHCO3 (2 × 10 mL) and brine (2 × 10 mL) allowed for the isolation of the product after vacuum-drying the organic phase.
3-(Chloromethyl)-4-hydroxy-5-methoxybenzaldehyde (2a). Following the general procedure, vanillin (1a, 10 mmol) was stirred with PF (12 mmol), conc. HCl (37%, 6.5 mL), and conc. H2SO4 (0.5 mL) at 60 °C for 6 h. Filtration on a P3 glass filter allowed the isolation of the product as a brown solid in a 70% yield. 1H NMR (400 MHz, CDCl3) δ(ppm): 9.84 (s, 1H, Ar-C(=O) H), 7.53 (d, 1H, Ar–H), 7.40 (d, 1H, Ar–H), 6.43 (s, 1H, Ar–OH), 4.71 (s, 2H, Ar–CH2-Cl), 3.99 (s, 3H, O-CH3). 13C NMR (400 MHz, CDCl3) δ(ppm): 197.1 (C=O), 151.2 (C-Ar), 148.0 (C-Ar), 125.3 (C-Ar), 123.3 (C-Ar), 119.6 (C-Ar), 113.2 (C-Ar), 56.0 (OCH3), 40.2 (-CH2-). EA found: C 53.71%, H 4.69%; calcd for (C9H9O3Cl): C 53.88%, H 4.52%. Mp 116–119 °C (Lit.: 120–122 °C).
3,5-Bis(chloromethyl)-2-hydroxybenzaldehyde (2b). Following the general procedure, salicylaldehyde (1b, 10 mmol) was stirred with PF (25 mmol), conc. HCl (37%, 17 mL), and conc. H2SO4 (1.0 mL) at 70 °C for 20 h. After extraction with DCM, the product was obtained as a pale pink solid, in 99% yield. 1H NMR (400 MHz, CDCl3) δ(ppm): 11.49 (s, 1H, Ar-OH), 9.91 (s, 1H, Ar-C(=O)H), 7.68 (d, 1H, Ar–H), 7.59 (d, 1H, Ar–H), 4.68 (s, 2H, Ar-oCH2-Cl), 4.59 (s, 2H, Ar-pCH2-Cl). 13C NMR (600 MHz, CDCl3) δ(ppm): 196.3 (C=O), 159.4 (C-Ar), 137.8 (C-Ar), 134.1 (C-Ar), 129.3 (C-Ar), 126.8 (C-Ar), 120.4 (C-Ar), 45.1 (-CH2-), 39.6 (-CH2-). EA found: C 49.28%, H 3.71%; calcd for (C9H8O2Cl2): C 49.34%, H 3.68%. Mp 90–92 °C
1-(3-(Chloromethyl)-4-hydroxyphenyl)ethan-1-one (2c). Following the general procedure, piceol (1c, 10 mmol) was stirred with PF (12 mmol), conc. HCl (37%, 6.5 mL), and conc. H2SO4 (0.25 mL) at 40 °C for 6 h. Filtration on a P3 glass filter allowed the isolation of the product as a pink solid in 68% yield. 1H NMR (600 MHz, CDCl3) δ(ppm): 7.95 (d, 1H, Ar-H), 7.88 (dd, 8.4 Hz, 1H, Ar-H), 6.91 (d, 1H, 8.4 Hz, Ar-H), 5.81 (br s, 1H, Ar-OH), 4.70 (s, 2H, Ar-CH2-Cl), 2.57 (s, 3H, Ar-C(O)CH3). 13C NMR (400 MHz, CDCl3) δ(ppm): 196.6 (C=O), 158.5 (C-Ar), 131.6 (C-Ar), 131.4 (C-Ar), 130.8 (C-Ar), 123.9 (C-Ar), 116.3 (C-Ar), 41.8 (-CH2-), 26.6 (-CH3). EA found: C 58.60%, H 4.87%; calcd for (C9H9O2Cl): C 58.55%, H 4.91%. Mp n.d. (decomposed).
1-(3,5-Bis(chloromethyl)-4-hydroxyphenyl)ethan-1-one (2c2). Following the general procedure, piceol (1c, 10 mmol) was stirred with PF (25 mmol), conc. HCl (37%, 13 mL), and conc. H2SO4 (0.5 mL) at 100 °C for 6 h. Filtration on a P3 glass filter allowed the isolation of the product as a pink solid in 60% yield. 1H NMR (600 MHz, CDCl3) δ(ppm): 7.93 (s, 2H, Ar-H), 6.28 (s, 1H, Ar-OH), 4.72 (s, 4H, Ar-CH2-Cl), 2.58 (s, 3H, Ar-C(=O)-CH3). 13C NMR (600 MHz, CDCl3) δ(ppm): 196.2 (C=O), 157.5 (C-Ar), 131.8 (C-Ar), 130.5 (C-Ar), 124.7 (C-Ar), 42.0 (-CH2-), 26.5 (CH3). EA found: C 51.67%, H 4.15%; calcd for (C10H10O2Cl2): C 51.53%, H 4.32%. Mp n.d. (decomposed).
4-Allyl-2,3-bis(chloromethyl)-6-methoxyphenol (
2d). Following the general procedure, eugenol (10 mmol) was stirred with PF (32 mmol), conc. HCl (37%, 20 mL), and DCM (2.0 mL) at 25 °C for 4 h. Following extraction with DCM, a highly impure material was obtained, contaminated by unreacted eugenol and diverse polycondensation byproducts. Several purification methods were attempted to obtain a purer product, without complete success (see
Figure S4). The product was finally obtained as a viscous black liquid in 50% yield.
1H NMR (600 MHz, CDCl
3)
δ(ppm): 6.68 (s, 1H, Ar-H), 5.98 (m, 1H, -CH=C), 5.30 (s, 1H, Ar-OH), 5.10 (m, 1H, -C=CH
2), 5.02 (m, 1H, -C=CH
2) 4.87 (s, 2H, Ar-CH
2-Cl), 4.74 (s, 2H, Ar-CH
2-Cl), 3.89 (s, 2H, Ar-OCH
3), 3.46 (s, 2H, Ar-CH
2-C=C).
13C NMR (400 MHz, CDCl
3) δ(ppm): 150.3 (C-Ar), 141.6 (C-Ar), 136.0 (-CH=), 129.8 (C-Ar), 127.6 (C-Ar), 125.9 (C-Ar), 116.3 (=CH
2), 113.4 (C-Ar), 56.1 (OCH
3), 39.9 (-CH
2-), 38.5 (-CH
2-), 37.7 (-CH
2-). EA found: C 55.35%, H 5.22%; calcd for (C
12H
14O
2Cl
2): C 55.19%, H 5.40%.
2,3,4-Tris (chloromethyl)-6-methoxyphenol (2e). Following the general procedure, guaiacol acetate (1e, 10 mmol) was stirred with PF (42 mmol), conc. HCl (37%, 25 mL), and DCM (2.5 mL) at 40 °C for 6 h. After extraction with DCM, the product was obtained as a dark brown solid in 60% yield. 1H NMR (400 MHz, CDCl3) δ(ppm): 6.87 (s, H, Ar-H), 6.03 (s, H, Ar-OH), 4.85 (s, 2H, Ar-CH2-Cl), 4.84 (s, 2H, Ar-CH2-Cl), 4.68 (s, 2H, Ar-CH2-Cl), 3.93 (s, 3H, Ar-OCH3). 13C-NMR (600 MHz, CDCl3) δ(ppm): 146.8 (C-Ar) 145.1 (C-Ar), 128.9 (C-Ar), 128.7 (C-Ar), 123.4 (C-Ar), 112.9 (C-Ar), 56.3 (OCH3), 43.9 (CH2); 38.7 (CH2), 36.5 (CH2). EA found: C 44.65%, H 3.95%; calcd for (C10H11O2Cl3): C 44.56%, H 4.11%. Mp n.d. (decomposed).
2,6-Bis (chloromethyl)-4-methylphenol (2f). Following the general procedure, p-cresol (1f, 10 mmol) was stirred with PF (22 mmol), conc. HCl (37%, 5.4 mL), and DCM (0.5 mL) at 40 °C for 6 h. After extraction with DCM, the product was obtained as a white waxy solid, in 76% yield. 1H NMR (600 MHz, CDCl3) δ(ppm): 7.09 (s, 2H, mAr-H), 5.55 (br s, 1H, Ar-OH), 4.66 (s, 4H, Ar-CH2-Cl), 2.28 (s, 3H, Ar-CH3). 13C-NMR (400 MHz, CDCl3) δ(ppm): 36.5 (CH2), 56.0 (OCH3), 113.3 (C-Ar). EA found: C 52.88%, H 5.07%; calcd for (C9H10OCl2): C 52.71%, H 4.91%. Mp n.d. (decomposed).
2.3. Amidation
General procedure. The chloromethyl substrate (10 mmol), dissolved in tetrahydrofuran (THF, 10 mL), was added dropwise to formamide (FA) (20, 40, or 60 mmol, for one, two, or three chloromethyl groups, respectively), which was kept at 60 °C inside a round-bottom flask. The reaction mixture was then stirred at reflux for 3 h. Once cooled to RT, the mixture was washed with brine (3 × 10 mL) and dried over sodium sulfate. The product was obtained by vacuum-drying.
N-(5-formyl-2-hydroxy-3-methoxybenzyl)formamide (3a): Following the general procedure, 2a (10 mmol) and FA (20 mmol) afforded the monoamide product as a brown thick liquid, in 99% yield. 1H NMR (400 MHz, CDCl3) δ(ppm): 9.84 (s, 1H, Ar-C(=O)H), 8.17 (s, 1H, N-C(=O)H), 7.51 (d, 1H, Ar–H), 7.42 (d, 1H, Ar–H), 6.47 (br s, 1H, Ar–OH), 5.33 (s, 2H, Ar–CH2-N), 3.99 (s, 3H, O-CH3). EA found: C 57.54%, H 5.13%, N 6.83%; calcd for (C10H11NO4): C 57.41%, H 5.30%, N 6.70%.
N,N′-((5-formyl-4-hydroxy-1,3-phenylene)bis(methylene))diformamide (3b): Following the general procedure, 2b (10 mmol) and FA (40 mmol) afforded the diamide product as a pale yellow thick liquid, in 98% yield. 1H NMR (600 MHz, CDCl3) δ(ppm): 11.43 (s, 1H, Ar-OH), 9.92 (s, 1H, Ar-C(=O)H), 8.17 (s, H, o-N-C(=O)H), 8.13 (s, H, pC-N-C(=O)H), 7.65 (d, 1H, Ar–H), 7.61 (d, 1H, Ar–H), 5.31 (s, 2H, Ar–oCH2-N), 5.19 (s, 2H, Ar–pCH2-N). EA found: C 56.05%, H 5.21%, N 11.99%; calcd for (C11H12N2O4): C 55.93%, H 5.12%, N 11.86%.
N-(5-acetyl-2-hydroxybenzyl)formamide (3c): Following the general procedure, 2c (10 mmol) and FA (20 mmol) afforded the monoamide product as a pale pink thick liquid, in 97% yield. 1H NMR (400 MHz, CDCl3) δ(ppm): 8.16 (s, H, N-C(=O)H), 7.96 (d, 1H, Ar-H), 7.89 (dd, 1H, Ar-H), 6.97 (d, 1H, Ar-H), 5.87 (br s, 1H, Ar-OH), 5.27 (s, 2H, Ar-CH2-N), 2.56 (s, 3H, CH3-(C=O)-Ar). EA found: C 62.36%, H 5.65%, N 7.40%; calcd for (C10H11NO3): C 62.17%, H 5.74%, N 7.25%.
N,N′-((5-acetyl-2-hydroxy-1,3-phenylene)bis(methylene))diformamide (3c2): Following the general procedure, 2c2 (10 mmol) and FA (40 mmol) afforded the monoamide product as a pale pink thick liquid, in 92% yield. 1H NMR (400 MHz, CDCl3) δ(ppm): 8.16 (s, 2H, N-C(=O)H), 7.99 (s, 2H, Ar-H), 5.75 (br s, 1H, Ar-OH), 5.28 (d, 2H, Ar-CH2-N), 2.57 (s, 3H, CH3-(C=O)-Ar). EA found: C 57.43%, H 5.78%, N 11.02%; calcd for (C12H14N2O4): C 57.59%, H 5.64%, N 11.19%.
N,N′-((6-allyl-3-hydroxy-4-methoxy-1,2-phenylene)bis(methylene))diformamide (3d): Following the general procedure, 2d (10 mmol) and FA (20 mmol) afforded the diamide product as a dark thick liquid, in 94% yield. 1H NMR (400 MHz, CDCl3) δ(ppm): 8.10 (s, 1H, N-C(=O)H), 8.06 (s, 1H, N-C(=O)H), 6.75 (s, 1H, Ar-H), 5.91 (m, 1H, C-CH=C), 5.43 (s, 2H, Ar-CH2-N), 5.28 (s, 2H, Ar-CH2-N), 5.07 (dd, 1H, -C=CH2), 4.97 (dd, 1H, -C=CH2), 3.91 (s, 3H, Ar-OCH3), 3.44 (dt, 2H, Ar-CH2-C=C). EA found: C 61.03%, H 5.69%, N 9.96%; calcd for (C14H16N2O4): C 60.86%, H 5.84%, N 10.14%.
N,N′,N″-((4-hydroxy-5-methoxybenzene-1,2,3-triyl)tris(methylene))triformamide (3e). Following the general procedure, 2e (10 mmol) and FA (60 mmol) afforded the triamide product as a dark thick liquid, in 96% yield. 1H NMR (400 MHz, CDCl3) δ(ppm): 8.10 (s, H, N-C(=O)H), 8.09 (s, H, N-C(=O)H), 8.06 (s, H, N-C(=O)H), 6.98 (s, 1H, Ar-H), 6.14 (s, 1H, Ar-OH), 5.45 (s, 2H, Ar-CH2-N), 5.38 (s, 2H, Ar-CH2-N), 5.30 (s, 2H, Ar-CH2-N), 3.94 (s, 3H, Ar-OCH3). EA found: C 52.70%, H 5.73%, N 14.11%; calcd for (C13H17N3O5): C 52.88%, H 5.80%, N 14.23%.
N,N′-((2-hydroxy-5-methyl-1,3-phenylene)bis(methylene))diformamide (3f): Following the general procedure, 2f (10 mmol) and FA (40 mmol) afforded the diamide product as a whitish thick liquid, in 93% yield. 1H NMR (400 MHz, CDCl3) δ(ppm): 8.13 (s, 2H, N-C(=O)H), 7.70 (s, 1H, Ar-OH), 7.14 (s, 2H, Ar-H), 5.22 (s, 4H, Ar-CH2-Cl), 2.28 (s, 3H, Ar-CH3). EA found: C 59.60%, H 6.22%, N 12.58%; calcd for (C11H14N2O3): C 59.45%, H 6.35%, N 12.61%.
2.4. Other Synthetic Operations
Formamide (FA) was synthesized by means of a procedure from the literature [
39], as follows. Ammonium formate (10 g, 158 mmol), placed in a round-bottom flask equipped with Dean–Stark apparatus on its top, was heated to 150 °C and then slowly to 180 °C. The volume of evolved water was used as a measure of process conversion. The heating was stopped when 2.8 mL (156 mmol) of water was collected in the Dean–Stark trap. The obtained product was used without further purification.
Guaiacol acetate (
1e) was synthesized by means of a procedure from the literature [
40]. Guaiacol (2.20 mL, 20 mmol), Ac
2O (2.85 mL, 25 mmol), and TEA (3.48 mL, 25 mmol) were mixed at 0 °C, then the reaction mixture was brought to 60 °C and left stirring for three hours. The reaction mixture was diluted with distilled water (10 mmol) and left stirring vigorously for 10 min. The reaction mixture was diluted with ethyl acetate (20 mL), and the organic phase was washed with 2M aqueous H
2SO
4 (3 × 5 mL), 10%
m/
v aq. NaHCO
3 solution (3 × 10 mL), and saturated NaCl aqueous solution (3 × 10 mL), before being anhydrified over MgSO
4 and vacuum-dried. The product was then used without further purification as a pale yellow liquid.
1H NMR (400 MHz, CDCl
3)
δ(ppm): 7.20 (dq, 1.80 Hz, 7.50 Hz, 1H, Ar-H), 7.04 (dd, 1.68 Hz, 7.82 Hz, 1H, Ar-H), 6.96 (ddd, 1.20 Hz, 8.17 Hz, 15.76 Hz, 2H, Ar-H), 3.83 (s, 3H, Ar-OCH
3), 2.32 (s, 3H, -OC(=O)CH
3). EA found: C 64.93% H 5.92%; calcd for (C
9H
10O
3): C 65.05% H 6.07%.
3. Results and Discussion
The chemical upgrading of biomass-sourced phenolics is a promising approach to obtaining industrially relevant building blocks with enhanced properties. To minimize exposure to harmful chemicals during the chloromethylation process, paraformaldehyde (PF) was selected as the alkylating agent instead of 37% aqueous formaldehyde. This substitution also reduced the amount of water in the reaction medium, resulting in a faster reaction.
Vanillin (
1a) was chosen as the initial starting material, as it is one of the main phenolic monomers obtainable through the oxidation of lignin [
41]. Initial experiments were conducted at 25 °C using one molar equivalent of PF, along with a small amount of DCM to improve the uniformity of the reaction medium. A five- to eight-fold molar excess of hydrochloric acid relative to PF was found to be necessary to obtain a chloromethylated product free of alcohol byproducts. Initially, substrate conversion was limited, reaching only 51%, with the selective formation of the monoalkylated product
2a (
Table 1, entry 1) [
35].
Many established Blanc–Quelet procedures employ a metal salt catalyst, in addition to 37% aqueous HCl, to promote the reaction. A typical example is ZnCl
2, which is particularly effective for deactivated substrates [
42]. However, in light of sustainability considerations, we opted for a metal-free approach. Concentrated H
2SO
4, a cost-effective and efficient alternative, was introduced from entry 2 onward. Its addition significantly enhanced the formation of
2a, with further improvements observed by extending the reaction time (entry 3) or increasing the temperature (entry 4). Subsequent investigations (entries 5–7) revealed that two molar equivalents of H
2SO
4 yielded the best results. Notably, omitting DCM (entries 5–7) led to a faster reaction and slightly improved yields in a shorter time frame. Under vigorous stirring, the chloromethylation of vanillin can be efficiently performed in water at 60 °C, achieving completion in just 6 h. No dichloromethylated products were detected in any of the experiments, even when increased amounts of PF were used, suggesting a specific deactivation of the positions
ortho to the aldehyde group.
The amidation of
2a was subsequently investigated through the
N-alkylation of formamide. When carried out in a small amount of THF under reflux conditions and using an excess of formamide, a clean and complete conversion to compound
3a was achieved (
Figure 2).
Continuing the investigation with salicylaldehyde (
1b), which was selected as the second bio-based phenolic substrate, the application of the developed chloromethylation protocol resulted in incomplete conversions at 40 °C. This is likely due to the lower activation of the aromatic ring, attributed to the absence of the methoxy group present in vanillin. However, unlike the previous case, a mixture of mono- and dichloromethylated derivatives was obtained. Increasing the reaction temperature to 70 °C proved effective in achieving complete conversion and enhanced selectivity toward the dialkylated product
2b, a trend consistent with previous reports [
36]. Some attempts to attain the selective preparation of a monoalkylated product by limiting the amount of PF were unsuccessful.
The converging directing effects of the substrate′s functional groups—
ortho/para for the hydroxyl and
meta for the aldehyde—facilitated electrophilic substitution at both positions with comparable speed (
Figure 3).
The amidation of 2b cleanly afforded the diamide 3b under conditions analogous to those previously established.
Piceol, or 4-hydroxyacetophenone (
1c), a bio-based phenolic compound found in the needles and mycorrhizal roots of Norway spruces [
43], was selected as the third case study. Its substitution pattern suggests a reactivity profile similar to that of
1b, with a synergistic directing effect exerted by the carbonyl and hydroxyl functional groups. Indeed, its chloromethylation proceeded toward both mono- and dialkylation; however, unlike the previous case, both products could be obtained with excellent selectivity. This is likely due to the less deactivating nature of the ketone group in
1c, compared to the aldehyde in
1b, which allowed the formation of the monochloromethylated product
2c (
Figure 4) at temperatures as low as 40 °C. In contrast, higher amounts of PF and higher temperatures were necessary to obtain the doubly alkylated product
2c2.
Amidation of the monochloro-(2c) and dichloro-(2c2) compounds occurred with high selectivity, using a molar excess of formamide (2.0 and 4.0 eq., respectively) and operating in refluxing THF, in accordance with the previously established conditions.
When considering bio-based phenolics, it is important to note that most of them lack electron-withdrawing groups, making them more prone to Ar-S
E reactions. On the one hand, this favors chloromethylation, which could proceed even without sulphuric acid activation. On the other hand, it increases the likelihood of undesired coupling between two arene units and PF. For instance, eugenol (
1d) is known to undergo condensation with formaldehyde and phosphoric acid, resulting in the formation of a bis-eugenol adduct [
44]. As expected, applying the developed chloromethylation conditions to
1d (6 h, 25 °C) led to a complex mixture of chloroalkylated products, alongside a minor amount of coupling products. Therefore, the removal of H
2SO
4 was considered, resulting in improved selectivity towards chloromethylation products. However, the limited water solubility of eugenol led to the formation of a heterogeneous system, characterized by a sticky organic phase, complicating process understanding. This resulted in yields that were highly sensitive to minor experimental variations, along with the production of undesired by-products. This issue prompted the addition of a small amount of DCM, which significantly improved the repeatability of the experiments and allowed better control over the reaction outcome. Similar to the reaction on
1b, selective monoalkylation of eugenol was not achieved, even with substoichiometric amounts of PF. Instead, selective dialkylation toward
2d was observed when more than 2.5 equivalents of aldehyde were used (
Figure 5). Amidation of
2d, conducted under the same conditions as those for previous cases, proceeded without issues.
Guaiacol is a major product of the hydrogenolysis of lignin [
45] and wood pyrolysis [
46]. Several studies have documented guaiacol′s tendency to undergo condensation with formaldehyde [
47], a property often exploited for the production of bio-based thermosets [
48]. Despite successfully applying a H
2SO
4-free protocol with eugenol, chloromethylation of guaiacol proved to be more challenging. The significant activation of the aromatic ring towards Ar-S
E reactions, partially enhanced by its small steric hindrance and compounded with the peculiar water solubility, led to very unselective conversion. Mixtures of polyalkylated products, along with minor amounts of condensation byproducts, were consistently obtained, regardless of PF equivalents or the use of low temperatures (0 °C).
To partially deactivate the aromatic ring and adjust the substrate solubility away from the aqueous phase (similar to the behavior of eugenol), guaiacol was first converted to its
O-acyl derivative using Ac
2O (as detailed in the Materials and Methods section). The chloromethylation of this derivative was then investigated in the presence of DCM. Guaiacol acetate (
1e) exhibited a strong tendency to form polyalkylated products, even when substoichiometric amounts of PF were used. However, to achieve better conversions of
1e, more than 2.5 equivalents of PF were required, resulting in the exclusive formation of the trialkylated product
2e. We soon realized that simultaneous deprotection of the acetate group occurred, which can be attributed to the aqueous acidic environment of the chloromethylation process. It is likely that the first alkylation step proceeded faster than the deprotection, generating a more apolar intermediate capable of undergoing further alkylations in the organic phase. As a result, complete conversion of
1e into
2e was achieved using 4.0 equivalents of PF (
Figure 6).
In addition to the expected activation of the
para- positions relative to both the acetoxy and methoxy groups, the observed selectivity is explained by the significantly higher propensity for alkylation
ortho- to the hydroxy group, rather than
ortho- to the methoxy group. Indeed, no isomers of
2e were detected in any of the experiments. The first introduced
para-CH
2Cl substituent, exhibiting ′chameleon-like’ behavior [
49], likely contributed to slowing the substitution at the position adjacent to the methoxy group. Transformation of
2e into the triamide
3e proceeded smoothly, following the same method applied to other benzyl chlorides.
p-Cresol (
1f) is commonly extracted from coal tar but can also be obtained through biomass hydrothermal decomposition [
50] and is present in tobacco smoke [
51]. It is used in the production of antioxidants, such as di-
tert-butylhydroxy toluene [
52], and in the production of novolac resins where
ortho-methylol groups are first introduced by reaction with formaldehyde under alkaline conditions [
53]. As a ′moderately activated′ bio-based phenolic, and given that chloromethylation of this substrate has been scarcely investigated [
38], we decided to include it within our studies. Despite its significantly lower activation compared to guaiacol,
p-cresol showed a strong tendency for dichloromethylation, with high selectivity for product
2f (
Figure 7).
Several attempts to obtain monoalkylated products by reducing reaction time, temperature, or PF equivalents resulted in unsatisfactory selectivity and partial conversions of 1f. This behavior is similar to that of eugenol, not only in terms of its propensity for dialkylation but also in the beneficial effect of adding a small amount of DCM. This addition improved repeatability and reduced contamination from byproducts. The amidation of 2f successfully yielded the diamide 3f, following the same protocol developed for other benzyl chlorides.
4. Conclusions
In this paper, the chloromethylation of a series of bio-based phenolics was developed, demonstrating a practical and straightforward approach. Most chloromethylations on solid substrates can occur “on water”, i.e., without the need for organic solvents. Moreover, as most chloromethyl products are solid, their isolation can often be achieved through a simple (solventless) filtration. In other cases, such as liquid or substrates with low water solubility (e.g., 1d, 1e, 1f), the system benefits from the addition of small amount of DCM, resulting in satisfactory selectivity and repeatability.
The success of the chloromethylation and its selectivity depend on the modulation of the chloromethylating reagent (PF:37% HCl ~ 1:6) in relation to the substitution pattern of the substrate, which governs its activation towards Ar-SE processes. Phenolics devoid of deactivating groups (1d, 1e, 1f) were promptly chloromethylated by the standard PF:37% HCl (~1:6) reagent at 25 °C, although highly activated substrates often attain limited selectivity (as observed for 2d and 2e). Instead, deactivated phenolics, such as those containing one aldehyde (1a, 1b) or ketone (1c) function, require reagent activation, obtained through the addition of H2SO4 (~0.9 eq with respect to PF) and, usually, a higher operating temperature.
The number of chloromethyl groups incorporated into the product is primarily determined by the nature of the substrate. For instance, deactivated substrates such as 1a and 1c favor the formation of monoalkylated derivatives, whereas in other cases, such as 1b, high selectivity for the dialkylated product is observed. Piceol (1c), characterized by a balanced activation for Ar-SE, was the only case where a control of mono- vs. dialkylation was obtained by setting the process temperature. Most of the activated substrates investigated—such as 1d and 1f—exhibited a strong tendency toward dialkylation. However, substrates with excessive activation, such as guaiacol, led to complex mixtures containing significant amounts of condensation byproducts. A controlled process for the formation of the trialkylated product 2e was only achieved by reducing the reactivity of guaiacol through conversion into its O-acyl derivative (1e).
The developed method, working with low solvent amounts, is suitable for upscaling. Considering the case of piceol, the production of
2c was characterized by a volume-time output (VTO) of 83 L·h/kg, a PMI of around 8, and an E factor around 7. The obtained benzyl chlorides (
2a–
f), some of those previously not described, are characterized by a broad synthetic versatility and therefore represent valuable electrophilic building blocks for the chemical industry. Various applications in synthesis as well as in materials chemistry can be suggested based on the literature describing the reactivity of benzyl chlorides. As an application, their selective conversion to the corresponding
N-formylamides (
3a–
f) was proven, through which the synthetic usefulness of the chloromethylated intermediates could be extended (see
Figure 1 and the related description). Thanks to the high electrophilicity typical of these benzyl chlorides, the amidation process seamlessly proceeded to a clean and selective conversion.