A Three-Step, Gram-Scale Synthesis of Hydroxytyrosol, Hydroxytyrosol Acetate, and 3,4-Dihydroxyphenylglycol

Hydroxytyrosol and two other polyphenols of olive tree, hydroxytyrosol acetate and 3,4-dihydroxyphenylglycol, are known for a wide range of beneficial activities in human health and prevention from diseases. The inability to isolate high, pure amounts of these natural compounds and the difficult and laborious procedures for the synthesis of them led us to describe herein an efficient, easy, cheap, and scaling up synthetic procedure, from catechol, via microwave irradiation.


Introduction
Polyphenols are a wide family of compounds found in plant-based foods, such as cranberries, grapes, olives, and walnuts, and have many diverse biological activities. One of the most important components of the polyphenol family is hydroxytyrosol (HT), a simple o-diphenol found in leaves and fruits of olive tree, virgin olive oil, as well as its wastewaters. It is derived from the hydrolysis of oleuropein, during the maturation of olives, storage of oil, and preparation of table olives. Two other important polyphenols of olive tree are hydroxytyrosol acetate and 3,4-dihydroxyphenylglycol (DHPG) (Figure 1).

Results and Discussion
The great interest of this high-added value natural compound triggered the development of various protocols in order to recover HT from olive industry wastewaters, but usually, it is isolated as a mixture with other phenolic compounds. Additionally, a few procedures describe the isolation of HT through the hydrolysis of oleuropein, a natural compound found mainly in olive tree leaves and other sources such as Ligustrum Vulgare [47][48][49][50][51]. Nevertheless, these approaches, although eco-friendly, are costly, often require the use of harmful solvents in extensive amount and usually result in low yield.
Due to the described problems, there has been a growing interest in developing an efficient synthetic route in order to obtain pure HT, but the synthesis is problematic since this compound is relatively soluble in water and polar organic solvents, therefore, the complete isolation through extraction from the reaction media is difficult [52]. Furthermore, HT is easily oxidized and relatively unstable during silica gel purifications and long-term storage [52]; thus adequate precautions must be taken (acid pH and absence of oxidants are crucial), concerning the synthesis and purification. Numerous synthetic approaches (more than 85 references in Reaxys database and 18 patents) [46][47][48][50][51][52][53][54][55][56][57][58][59][60][61][62][63][64][65][66][67][68][69] for the synthesis of HT have been described; however all these methods appear to have many disadvantages, i.e., low yields, laborious purifications, high cost, and many byproducts ( Figure 2). The starting material for most of these synthetic procedures is tyrosol, through an oxidative procedure [51,[57][58][59]62,64,69], though, the major disadvantage of the abovementioned approaches consists mainly of the use of non-eco-friendly oxidative reactants and the relatively high price of tyrosol. Alternatively, the synthesis of HT has been presented by the reduction of 3,4-dihydroxyphenylacetic acid but the commercial price of the acid (or the price of the corresponding nitrile) is too high [48,52,60,66]. Additionally, the isolation of HT from the reaction media is difficult (because of the water solubility), thus the overall yield is often low. Interestingly, only three synthetic approaches use catechol as starting material, even though this molecule is 6400-fold cheaper than tyrosol (Sigma-Aldrich 2019) ( Table 1). and other sources such as Ligustrum Vulgare [47][48][49][50][51]. Nevertheless, these approaches, although ecofriendly, are costly, often require the use of harmful solvents in extensive amount and usually result in low yield. Due to the described problems, there has been a growing interest in developing an efficient synthetic route in order to obtain pure HT, but the synthesis is problematic since this compound is relatively soluble in water and polar organic solvents, therefore, the complete isolation through extraction from the reaction media is difficult [52]. Furthermore, HT is easily oxidized and relatively unstable during silica gel purifications and long-term storage [52]; thus adequate precautions must be taken (acid pH and absence of oxidants are crucial), concerning the synthesis and purification. Numerous synthetic approaches (more than 85 references in Reaxys database and 18 patents) [46][47][48][50][51][52][53][54][55][56][57][58][59][60][61][62][63][64][65][66][67][68][69] for the synthesis of HT have been described; however all these methods appear to have many disadvantages, i.e., low yields, laborious purifications, high cost, and many byproducts ( Figure  2). The starting material for most of these synthetic procedures is tyrosol, through an oxidative procedure [51,[57][58][59]62,64,69], though, the major disadvantage of the abovementioned approaches consists mainly of the use of non-eco-friendly oxidative reactants and the relatively high price of tyrosol. Alternatively, the synthesis of HT has been presented by the reduction of 3,4dihydroxyphenylacetic acid but the commercial price of the acid (or the price of the corresponding nitrile) is too high [48,52,60,66]. Additionally, the isolation of HT from the reaction media is difficult (because of the water solubility), thus the overall yield is often low. Interestingly, only three synthetic approaches use catechol as starting material, even though this molecule is 6400-fold cheaper than tyrosol (Sigma-Aldrich 2019) ( Table 1). Concerning the synthesis of HT acetate, several synthetic procedures have been described, such as selective transesterification of the primary hydroxyl using suitable Lewis acids or protection of the phenolic hydroxyls before the acetylation [38,46,52,[57][58][59]62,64,[67][68][69][70][71][72][73]. Nevertheless, all of the above procedures are problematic since they use tyrosol or HT as staring material, two expensive compounds. Also, referring to the protection/deprotection approach, many steps are involved, thus the overall yield is relatively low. Regarding the synthesis of DHPG, only one synthetic procedure has been described with many steps, low yield, and vigorous conditions [74]. Concerning the synthesis of HT acetate, several synthetic procedures have been described, such as selective transesterification of the primary hydroxyl using suitable Lewis acids or protection of the phenolic hydroxyls before the acetylation [38,46,52,[57][58][59]62,64,[67][68][69][70][71][72][73]. Nevertheless, all of the above procedures are problematic since they use tyrosol or HT as staring material, two expensive compounds. Also, referring to the protection/deprotection approach, many steps are involved, thus the overall yield is relatively low. Regarding the synthesis of DHPG, only one synthetic procedure has been described with many steps, low yield, and vigorous conditions [74].
Prompted by the above, herein we described a simple and efficient synthesis of HT, with inexpensive catechol as staring material, without using extreme experimental conditions or laborious purifications. Additionally, the synthesis of HT acetate and DHPG was illustrated by engaging an intermediate from the procedure of HT synthesis. scale-up conditions. There are only three other cases in which catechol is used as starting material, though, the developed approach possessed the highest yield, with only three steps involved. Besides, only limited amount of solvents, for the reaction and the purification procedures, were needed. scale-up conditions. There are only three other cases in which catechol is used as starting material, though, the developed approach possessed the highest yield, with only three steps involved. Besides, only limited amount of solvents, for the reaction and the purification procedures, were needed.   (5), via a onestep reaction, upon treatment with potassium acetate in dry DMF at 70 °C, which yielded up to 85% (Scheme 1). It is important to notice that the presented procedure was straightforward, with no protection of the phenolic groups needed. The overall yield of the procedure was up to 73%, with minimum purifications, since the only silica gel column chromatography was performed on the final step. Into the best of our knowledge, this was the only approach that used catechol for the synthesis of HT-acetate, since all the other procedures use tyrosol or HT as starting material (Table 1). In addition, we can assert that the presented procedure for the synthesis of HT acetate could be used for the synthesis of various HT lipophilic esters, in order to study their biological activity.
Regarding the synthesis of 3,4-dihydroxyphenylglycol (7), a slightly different methodology was developed, which involved the initial hydrolysis of compound 2 to provide alcohol 6. Although this hydrolysis is already described in the literature via HCOONa/HCOOH system [40], in our hands proceeded almost quantitively, by microwave heating, in distilled water. The reaction was scalable up to grams, without any by-products formed, while the purification was very simple, by trituration with ethyl acetate. Finally, target DHPG (7) was prepared through catalytic hydrogenation in methanol, in an 82% overall yield. The overall yield of the proposed procedure was up to 70%, far better than the previous procedure, with only three steps involved (Table 1) [74]. Interestingly, reduction of keto compound 6 over palladium on activated carbon, in the presence of a few drops of perchloric or hydrochloric acid, in abs. ethanol, afforded ether 8 [81], presumably through protonation of the intermediate alcohol and subsequent nucleophilic substitution by ethanol. It should be noted that this compound, even as a racemic mixture, is very interesting, since it is a stable Synthesis through dibenzyloxy-HT, acetylation and subsequent deprotection. 3 (5), via a onestep reaction, upon treatment with potassium acetate in dry DMF at 70 °C, which yielded up to 85% (Scheme 1). It is important to notice that the presented procedure was straightforward, with no protection of the phenolic groups needed. The overall yield of the procedure was up to 73%, with minimum purifications, since the only silica gel column chromatography was performed on the final step. Into the best of our knowledge, this was the only approach that used catechol for the synthesis of HT-acetate, since all the other procedures use tyrosol or HT as starting material (Table 1). In addition, we can assert that the presented procedure for the synthesis of HT acetate could be used for the synthesis of various HT lipophilic esters, in order to study their biological activity.
Regarding the synthesis of 3,4-dihydroxyphenylglycol (7), a slightly different methodology was developed, which involved the initial hydrolysis of compound 2 to provide alcohol 6. Although this hydrolysis is already described in the literature via HCOONa/HCOOH system [40], in our hands proceeded almost quantitively, by microwave heating, in distilled water. The reaction was scalable up to grams, without any by-products formed, while the purification was very simple, by trituration with ethyl acetate. Finally, target DHPG (7) was prepared through catalytic hydrogenation in methanol, in an 82% overall yield. The overall yield of the proposed procedure was up to 70%, far better than the previous procedure, with only three steps involved (Table 1) [74]. Interestingly, reduction of keto compound 6 over palladium on activated carbon, in the presence of a few drops of perchloric or hydrochloric acid, in abs. ethanol, afforded ether 8 [81], presumably through protonation of the intermediate alcohol and subsequent nucleophilic substitution by ethanol. It should be noted that this compound, even as a racemic mixture, is very interesting, since it is a stable 3 (5), via a onestep reaction, upon treatment with potassium acetate in dry DMF at 70 °C, which yielded up to 85% (Scheme 1). It is important to notice that the presented procedure was straightforward, with no protection of the phenolic groups needed. The overall yield of the procedure was up to 73%, with minimum purifications, since the only silica gel column chromatography was performed on the final step. Into the best of our knowledge, this was the only approach that used catechol for the synthesis of HT-acetate, since all the other procedures use tyrosol or HT as starting material (Table 1). In addition, we can assert that the presented procedure for the synthesis of HT acetate could be used for the synthesis of various HT lipophilic esters, in order to study their biological activity.
Regarding the synthesis of 3,4-dihydroxyphenylglycol (7), a slightly different methodology was developed, which involved the initial hydrolysis of compound 2 to provide alcohol 6. Although this hydrolysis is already described in the literature via HCOONa/HCOOH system [40], in our hands proceeded almost quantitively, by microwave heating, in distilled water. The reaction was scalable up to grams, without any by-products formed, while the purification was very simple, by trituration with ethyl acetate. Finally, target DHPG (7) was prepared through catalytic hydrogenation in methanol, in an 82% overall yield. The overall yield of the proposed procedure was up to 70%, far better than the previous procedure, with only three steps involved (Table 1) [74]. Interestingly, reduction of keto compound 6 over palladium on activated carbon, in the presence of a few drops of perchloric or hydrochloric acid, in abs. ethanol, afforded ether 8 [81], presumably through protonation of the intermediate alcohol and subsequent nucleophilic substitution by ethanol. It should be noted that this compound, even as a racemic mixture, is very interesting, since it is a stable To overcome the solubility and stability limitations of HT, we envisioned a three-step synthetic procedure depicted in Scheme 1. The first step of the synthesis involved Friedel-Craft acylation of catechol (1) with chloroacetyl chloride in nearly quantitative yield. The preparation of chloride 2 has been previously reported through many different procedures [75,76]; however, the method reported herein, via microwave irradiation, was straightforward, simple, high yielding, and cost-effective. Thus, after vacuum evaporation of the excess of POCl 3 and trituration with water, the crude product was used without any further purification. It should be noted that the evaporation of the excess of POCl 3 before the trituration with water was crucial, otherwise, and if the reaction mixture is poured into ice-water (an experimental usually used for reactions with POCl 3 ), the overall yield is remarkably reduced (about 40%) even after extraction of the reaction media. Prompted by the above, herein we described a simple and efficient synthesis of HT, with inexpensive catechol as staring material, without using extreme experimental conditions or laborious purifications. Additionally, the synthesis of HT acetate and DHPG was illustrated by engaging an intermediate from the procedure of HT synthesis.
To overcome the solubility and stability limitations of HT, we envisioned a three-step synthetic procedure depicted in Scheme 1. The first step of the synthesis involved Friedel-Craft acylation of catechol (1) with chloroacetyl chloride in nearly quantitative yield. The preparation of chloride 2 has been previously reported through many different procedures [75,76]; however, the method reported herein, via microwave irradiation, was straightforward, simple, high yielding, and cost-effective. Thus, after vacuum evaporation of the excess of POCl3 and trituration with water, the crude product was used without any further purification. It should be noted that the evaporation of the excess of POCl3 before the trituration with water was crucial, otherwise, and if the reaction mixture is poured into ice-water (an experimental usually used for reactions with POCl3), the overall yield is remarkably reduced (about 40%) even after extraction of the reaction media. The next step concerns the reduction of the carbonyl group; still all the initial attempts, under several conditions, provided complex mixtures. Among the variety of conditions used for the reduction of compound 2, only the reduction with triethylsilane in the presence of a Lewis acid gave the desired chloro compound 3 [77][78][79][80]. More precisely, of the various Lewis acids and reaction conditions that were tested, the use of a fourfold excess of trifluoroacetic acid or a fivefold excess of boron trifluoride etherate and a twofold excess of triethylsilane derived the desirable catechol 3 in 95% or 92%, respectively. The reaction proceeded smoothly with no formation of byproducts. It is remarkable to note that the crude of the reaction was readily purified, after vacuum evaporation of the volatiles, by just trituration with boiling cyclohexane and vacuum evaporation of the solvent.
Finally, chloride 3 was readily converted to the corresponding HT (4) upon hydrolysis with distilled water, via microwave heating in a Milestone Start E apparatus. Reaction conditions were optimized, varying applied power and temperature, measured by fiber optic contact thermometer. The best results were obtained using programmed irradiation at 700 watts, at 101 • C, at open vessel conditions, and HT was obtained in >92% yield.
It is important to note that by the above-mentioned conditions, because of the absence of basic media (actually the reaction media was acidic due to the hydrochloric acid produced by the reaction) or oxidants, the reaction proceeded readily. Furthermore, the isolation was very simple, and no chromatography purification was needed. After completion of the reaction, the mixture was extracted with dichloromethane, and the aqueous phase was vacuum evaporated at low temperature or lyophilized, in order to obtain pure HT. Scaling up to 10 g of the above-mentioned synthetic procedure afforded reproducible overall yields. Higher scale-ups, up to 120 g showed the same excellent results only by modifying the time reaction conditions. The final step of the hydrolysis of chloride 3 to HT (4) was also accomplished via conventional heating, but much more time was needed. In particular, the hydrolysis of 23 g of chloride 3 to HT (4) required 4 h, instead of 32 h of conventional heating. Additionally, the yield of the reaction was lower (65% vs. 94.8% by MW heating) probably due to the instability of HT. Most importantly, the purification of HT, synthesized by MW-assisted hydrolysis, was very simple allowing the overall workup simplification, whilst the same purification of the product, of the conventional heating procedure, was laborious.
The overall yield of proposed approach was up to 85%, with only three steps involved, whilst the purification was easy with no silica gel chromatography needed (Table 1). Nevertheless, the main advantage of the procedure concerned the use of catechol as staring material, an inexpensive compound, instead of tyrosol, which is usually used, and the reproducibility of the method, even in scale-up conditions. There are only three other cases in which catechol is used as starting material, though, the developed approach possessed the highest yield, with only three steps involved. Besides, only limited amount of solvents, for the reaction and the purification procedures, were needed.
Taking advantage of the intermediate 3, we were also able to prepare HT acetate (5), via a one-step reaction, upon treatment with potassium acetate in dry DMF at 70 • C, which yielded up to 85% (Scheme 1). It is important to notice that the presented procedure was straightforward, with no protection of the phenolic groups needed. The overall yield of the procedure was up to 73%, with minimum purifications, since the only silica gel column chromatography was performed on the final step. Into the best of our knowledge, this was the only approach that used catechol for the synthesis of HT-acetate, since all the other procedures use tyrosol or HT as starting material (Table 1). In addition, we can assert that the presented procedure for the synthesis of HT acetate could be used for the synthesis of various HT lipophilic esters, in order to study their biological activity.
Regarding the synthesis of 3,4-dihydroxyphenylglycol (7), a slightly different methodology was developed, which involved the initial hydrolysis of compound 2 to provide alcohol 6. Although this hydrolysis is already described in the literature via HCOONa/HCOOH system [40], in our hands proceeded almost quantitively, by microwave heating, in distilled water. The reaction was scalable up to grams, without any by-products formed, while the purification was very simple, by trituration with ethyl acetate. Finally, target DHPG (7) was prepared through catalytic hydrogenation in methanol, in an 82% overall yield. The overall yield of the proposed procedure was up to 70%, far better than the previous procedure, with only three steps involved (Table 1) [74]. Interestingly, reduction of keto compound 6 over palladium on activated carbon, in the presence of a few drops of perchloric or hydrochloric acid, in abs. ethanol, afforded ether 8 [81], presumably through protonation of the intermediate alcohol and subsequent nucleophilic substitution by ethanol. It should be noted that this compound, even as a racemic mixture, is very interesting, since it is a stable form of DHPG, with better lipophilicity, and could be useful to study the therapeutic role and potentiality of DHPG.

General Experimental Procedures
All chemicals were purchased from Alfa Aesar (Ward Hill, MA, USA). Melting points were determined on a Büchi apparatus and were uncorrected. 1 H NMR, 13 C NMR and 2D spectra were recorded on a Bruker Avance III 600 spectrometer (Bruker BioSpin GmbH, Rheinstetten, Germany), in deuterated solvents and were referenced to TMS (δ scale). The signals of 1 H and 13 C spectra were unambiguously assigned by using 2D NMR techniques: 1  2-chloro-1-(3,4-dihydroxyphenyl)ethanone (2): A solution of catechol 1 (100 g, 0.94 mol) and chloroacetyl chloride (84.9 mL, 1.04 mol) in phosphoryl chloride (264.2 mL, 2.82 mol) was stirred under microwave irradiation (450 W) at 105 • C for 3 h. After completion of the reaction, the excess of phosphoryl chloride was vacuum evaporated, poured into ice-water, and the resulting solid product was filtered off and dried to afford 162.6 g (93%) of the title compound, which was used for the next step without any further purification. 1

Conclusions
In summary, a new, large-scale, straightforward synthetic approach for the preparation of HT as well as HT acetate and DHPG has been described in the present work, using simple, low-cost, and scalable procedures, starting from inexpensive, commercially available catechol. The established procedure for the synthesis of HT acetate could be used for the synthesis of other lipophilic esters of HT. Additionally, a synthetic approach for the synthesis of compound 8 has been proposed, which could be used for the development of new lipophilic and stable analogs of DHPG. All the reactions proceeded smoothly, with no byproducts formed, while the purification of the compounds was very simple, mostly by solvent trituration. The synthetic procedures were based in microwave irradiation, which enabled more precise control of the reaction conditions and contributed to higher yields, thus, the new synthetic approaches could be an alternative for the industrial preparation of HT, HT acetate and DHPG.