Influence of Heterogeneous Catalysts and Reaction Parameters on the Acetylation of Glycerol to Acetin: A Review

Glycerol, a polyhydric alcohol, is currently receiving greater attention worldwide in view of its glut in the market occasioned by the recent upsurge in biodiesel production. The acetylation of glycerol to acetin (acetyl glycerol) is one of the many pathways of upgrading glycerol to fine chemicals. Acetin, which could be mono, di, and or triacetin, has versatile applications in the cosmetics, medicines, food, polymer, and fuel industries as a humectant, emulsifier, plasticizer, and fuel additive and so it is of high economic value. Given the critical role of catalysts in green chemistry, this paper reports the influence of the different heterogeneous catalysts used in glycerol acetylation. It also reviewed the influence of catalyst load, temperature, molar ratio, and the time on the reaction.


Introduction
Glycerol, also known as propane-1,2,3-triol, is a polyhydric alcohol with three hydrophilic hydroxyl groups, each attached to the three carbon atoms. The hydroxyl groups are mainly responsible for the properties and applications of glycerol [1,2]. Glycerol is soluble in water and other polar solvents. It is a clear, colourless, odourless, viscous, non-toxic, and hygroscopic liquid with a high boiling point (290 • C) and low melting point (18 • C) [3][4][5].
Glycerol has versatile applications in medical, pharmaceutical, personal care, and food industries as sweetener, flavour, preservative, emulsifier, humectant, antioxidant, and additive. It is also used in the automotive and chemical industries as a plasticizer, lubricant, and cryoprotectant [6][7][8].
The distribution of some of these applications varies in different literature reports. The percentage distribution, as found in various literature articles, is shown in Figure 1 [9]. It has been reported that glycerol is among the 12 most important bio-based chemicals worldwide [10], with the global market expected to reach USD 2.52 billion by 2020 [11]. Glycerol can be obtained via synthetic or natural means. The synthetic route is usually through propylene oxidation or chlorination, while the natural route involves the hydrolysis of vegetable oil and or the transesterification of fatty acids or vegetable oil as well as the fermentation of yeast [4,12,13]. The synthesis of glycerol via propylene is based on petroleum feedstock (Scheme 1) and is no longer popular due to the global shift to less carbon emission. The hydrolysis of oil, also known as saponification, involves the reaction of triglyceride with alkaline hydroxide (Scheme 2). It is an old method that dates back to 2800 BC and accounted for most of the commercial needs over the years [14,15]. However, production through this method is no longer meeting the global need. The fermentation method is limited in industrial applications. The transesterification method involves the reaction of triglyceride (oil) with alcohol in the presence of a catalyst, which results in the production of biodiesel (fatty acid esters) and glycerol (Scheme 3). The transesterification method is now the most popular because of the large quantity of biodiesel produced. Biodiesel is considered to be an environmentally benign renewable energy that is biodegradable, non-toxic, and renewable with less obnoxious emissions [11,16]. The transesterification method also produced glycerol as a by-product, which is known as crude glycerol due to its low-grade purity. The recent increase in biodiesel production has led to a proportional increase in the crude glycerol. It has been reported that for every 100 kg of biodiesel produced, 10 kg (10%) of glycerol is obtained as the by-product leading to surplus in the market [12,17,18]. Recent literature indicates that by the year 2020, the global production of glycerol will move up to 41.9 billion litres [19]. This number is expected to increase with the recent prediction that biodiesel will account for over 70% of the global transportation fuel by 2040 [20]. This forecast is supported by many deliberate policies put in place by different countries and organizations to encourage biodiesel production and its utilization, such as the implementation of mandatory biodiesel blending targets, tax exemptions, government support, investment subsidies, and research and development programs [20]. Figure 2 shows the progressive increase in biodiesel production for the last ten years [21]. The above policies have led to the diversification of biodiesel feedstocks depending on the country, as indicated in Table 1. The impact of excess glycerol produced from the transesterification process is already felt in the market [22].  Glycerol can be obtained via synthetic or natural means. The synthetic route is usually through propylene oxidation or chlorination, while the natural route involves the hydrolysis of vegetable oil and or the transesterification of fatty acids or vegetable oil as well as the fermentation of yeast [4,12,13]. The synthesis of glycerol via propylene is based on petroleum feedstock (Scheme 1) and is no longer popular due to the global shift to less carbon emission. The hydrolysis of oil, also known as saponification, involves the reaction of triglyceride with alkaline hydroxide (Scheme 2). It is an old method that dates back to 2800 BC and accounted for most of the commercial needs over the years [14,15]. However, production through this method is no longer meeting the global need. The fermentation method is limited in industrial applications. The transesterification method involves the reaction of triglyceride (oil) with alcohol in the presence of a catalyst, which results in the production of biodiesel (fatty acid esters) and glycerol (Scheme 3). The transesterification method is now the most popular because of the large quantity of biodiesel produced. Biodiesel is considered to be an environmentally benign renewable energy that is biodegradable, non-toxic, and renewable with less obnoxious emissions [11,16]. The transesterification method also produced glycerol as a by-product, which is known as crude glycerol due to its low-grade purity. The recent increase in biodiesel production has led to a proportional increase in the crude glycerol. It has been reported that for every 100 kg of biodiesel produced, 10 kg (10%) of glycerol is obtained as the by-product leading to surplus in the market [12,17,18]. Recent literature indicates that by the year 2020, the global production of glycerol will move up to 41.9 billion litres [19]. This number is expected to increase with the recent prediction that biodiesel will account for over 70% of the global transportation fuel by 2040 [20]. This forecast is supported by many deliberate policies put in place by different countries and organizations to encourage biodiesel production and its utilization, such as the implementation of mandatory biodiesel blending targets, tax exemptions, government support, investment subsidies, and research and development programs [20]. Figure 2 shows the progressive increase in biodiesel production for the last ten years [21]. The above policies have led to the diversification of biodiesel feedstocks depending on the country, as indicated in Table 1. The impact of excess glycerol produced from the transesterification process is already felt in the market [22]. Glycerol can be obtained via synthetic or natural means. The synthetic route is usually through propylene oxidation or chlorination, while the natural route involves the hydrolysis of vegetable oil and or the transesterification of fatty acids or vegetable oil as well as the fermentation of yeast [4,12,13]. The synthesis of glycerol via propylene is based on petroleum feedstock (Scheme 1) and is no longer popular due to the global shift to less carbon emission. The hydrolysis of oil, also known as saponification, involves the reaction of triglyceride with alkaline hydroxide (Scheme 2). It is an old method that dates back to 2800 BC and accounted for most of the commercial needs over the years [14,15]. However, production through this method is no longer meeting the global need. The fermentation method is limited in industrial applications. The transesterification method involves the reaction of triglyceride (oil) with alcohol in the presence of a catalyst, which results in the production of biodiesel (fatty acid esters) and glycerol (Scheme 3). The transesterification method is now the most popular because of the large quantity of biodiesel produced. Biodiesel is considered to be an environmentally benign renewable energy that is biodegradable, non-toxic, and renewable with less obnoxious emissions [11,16]. The transesterification method also produced glycerol as a by-product, which is known as crude glycerol due to its low-grade purity. The recent increase in biodiesel production has led to a proportional increase in the crude glycerol. It has been reported that for every 100 kg of biodiesel produced, 10 kg (10%) of glycerol is obtained as the by-product leading to surplus in the market [12,17,18]. Recent literature indicates that by the year 2020, the global production of glycerol will move up to 41.9 billion litres [19]. This number is expected to increase with the recent prediction that biodiesel will account for over 70% of the global transportation fuel by 2040 [20]. This forecast is supported by many deliberate policies put in place by different countries and organizations to encourage biodiesel production and its utilization, such as the implementation of mandatory biodiesel blending targets, tax exemptions, government support, investment subsidies, and research and development programs [20]. Figure 2 shows the progressive increase in biodiesel production for the last ten years [21]. The above policies have led to the diversification of biodiesel feedstocks depending on the country, as indicated in Table 1. The impact of excess glycerol produced from the transesterification process is already felt in the market [22].  Annual total biodiesel production for the last ten (10) years. Data obtained from REN21 [21]. Table 1. Common feedstocks of biodiesel production by different countries [23][24][25][26].  Year Scheme 2. Alkaline hydrolysis of oil to produce glycerol (saponification reaction). Y = Alkali metal.

Scheme 3.
Transesterification of triglyceride to biodiesel fatty acid esters and glycerol.

Scheme 3.
Transesterification of triglyceride to biodiesel fatty acid esters and glycerol.
One of the high valued products receiving attention by researchers and industrialists in recent times is oxygenated fuel additives. These additives are usually blended with gasoline, diesel, and or biodiesel to improve their cold flow properties, viscosity, octane number, thermal stability, reduce gum formation and particulate emission [53][54][55]. The oxygenated fuel additives from glycerol include glycerol esters or acetyl glycerol (acetin), glycerol ethers, and glycerol formal (solketals and acetals) [3,[55][56][57][58].
Scientific literature articles have revealed few reviews on the synthesis of acetin in recent years [59][60][61], unlike other oxygenated fuel additives, namely ethers, solketals, and acetals [22,[62][63][64][65][66][67][68][69]. Thus, this paper reviews the recent research findings on acetin production from glycerol with a particular focus on the heterogeneous catalysts used and how they influenced the reaction. Unlike earlier reviews, this review further emphasized the conditions in which the catalysts were used with detailed findings documented to serve as an appropriate guide for subsequent studies in this area. This review also presents some of the findings in pictorial form for easy understanding, unlike earlier reviews.

Acetin (Glycerol Esters or Acetyl Glycerol)
Acetin is obtained via the reaction of glycerol with acetic acid (or any carboxylic acids), acetic anhydride, or through the transesterification of either triglycerides or glycerol with methyl acetate [3,54,70]. There are three types of acetin depending on the number of the hydroxyl group substituted Given the low commercial value of the produced glycerol, its surplus, and possible environmental issues, efforts are ongoing to convert it to high-value products, which is expected to improve its commercial viability, biodiesel economics, and eliminate the perceived environmental concern of the surplus. The conversion of glycerol to the high valued products is usually done through various catalytic reactions such as hydrogenolysis [29,30], reforming [31,32], dehydration [33,34], carboxylation [35,36], oxidation [37,38], oligomerization [39,40], esterification or acetylation [41,42], etherification [43,44], acetalization [45,46], pyrolysis and gasification [47,48]. A good number of literature reviews are available to show the research activities in most of the reactions above [7,9,34,[49][50][51][52].
One of the high valued products receiving attention by researchers and industrialists in recent times is oxygenated fuel additives. These additives are usually blended with gasoline, diesel, and or biodiesel to improve their cold flow properties, viscosity, octane number, thermal stability, reduce gum formation and particulate emission [53][54][55]. The oxygenated fuel additives from glycerol include glycerol esters or acetyl glycerol (acetin), glycerol ethers, and glycerol formal (solketals and acetals) [3,[55][56][57][58].
Scientific literature articles have revealed few reviews on the synthesis of acetin in recent years [59][60][61], unlike other oxygenated fuel additives, namely ethers, solketals, and acetals [22,[62][63][64][65][66][67][68][69]. Thus, this paper reviews the recent research findings on acetin production from glycerol with a particular focus on the heterogeneous catalysts used and how they influenced the reaction. Unlike earlier reviews, this review further emphasized the conditions in which the catalysts were used with detailed findings documented to serve as an appropriate guide for subsequent studies in this area. This review also presents some of the findings in pictorial form for easy understanding, unlike earlier reviews.

Acetin (Glycerol Esters or Acetyl Glycerol)
Acetin is obtained via the reaction of glycerol with acetic acid (or any carboxylic acids), acetic anhydride, or through the transesterification of either triglycerides or glycerol with methyl acetate [3,54,70]. There are three types of acetin depending on the number of the hydroxyl group substituted from the glycerol atom with the acetyl or acyl group. Hence, the reaction is referred to as acetylation. The acetin includes monoacetin (monoacetyl glycerol), diacetin (diacetyl glycerol) and triacetin (triacetyl glycerol) [71][72][73]. Their structures are shown in Scheme 4. All types of acetin have a wide range of applications.
Appl. Sci. 2020, 10, x FOR PEER REVIEW 5 of 33 from the glycerol atom with the acetyl or acyl group. Hence, the reaction is referred to as acetylation. The acetin includes monoacetin (monoacetyl glycerol), diacetin (diacetyl glycerol) and triacetin (triacetyl glycerol) [71][72][73]. Their structures are shown in Scheme 4. All types of acetin have a wide range of applications. Monoacetin and diacetin are used in the cosmetics, medicines, and food industries as humectants and emulsifiers [12,74]. Monoacetin is also used as a tanning agent and in the production of explosives [75]. Both monoacetin and diacetin are monomers for the production of biodegradable polyesters [76]. Triacetin is used as a solvent in various reactions, as a plasticizer of cellulosic polymers and copolymers [59,77]. Triacetin and diacetin are, most importantly, also used as biofuel additives to improve viscosity and cold flow properties as well as to reduce greenhouse gas emissions [3,[78][79][80][81][82]. In view of the excellent applications of triacetin, its price is high and remains stable with a growing demand of 5-10% yearly [59]. Hence, effort in the diversification of triacetin production is apt and should be welcomed.

Mechanism of Acetylation of Glycerol
Various articles report the plausible reaction mechanism in the acetylation of glycerol with acetic acid [58,[83][84][85][86][87]. The mechanism proceeds via the activation (protonation) of the carbonyl group of the acetic acid by a strong acid catalyst (usually Bronsted-type acid). The resultant acylium ion makes it more susceptible to nucleophilic attack by the oxygen atom of the hydroxyl molecule attached to glycerol, thereby forming a tetrahedral intermediate and leading to the transfer of hydrogen to the next hydroxyl molecule (attached to the carbonyl carbon) to form a water molecule. On the loss of the water molecule, a monoacetin (monoacetyl glycerol) is formed, and on further reaction with acetic acid following a similar mechanism, diacetin (diacetyl glycerol) is formed, and finally, triacetin (triacetyl glycerol) is formed as shown in Scheme 5. Monoacetin and diacetin are used in the cosmetics, medicines, and food industries as humectants and emulsifiers [12,74]. Monoacetin is also used as a tanning agent and in the production of explosives [75]. Both monoacetin and diacetin are monomers for the production of biodegradable polyesters [76]. Triacetin is used as a solvent in various reactions, as a plasticizer of cellulosic polymers and copolymers [59,77]. Triacetin and diacetin are, most importantly, also used as biofuel additives to improve viscosity and cold flow properties as well as to reduce greenhouse gas emissions [3,[78][79][80][81][82]. In view of the excellent applications of triacetin, its price is high and remains stable with a growing demand of 5-10% yearly [59]. Hence, effort in the diversification of triacetin production is apt and should be welcomed.

Mechanism of Acetylation of Glycerol
Various articles report the plausible reaction mechanism in the acetylation of glycerol with acetic acid [58,[83][84][85][86][87]. The mechanism proceeds via the activation (protonation) of the carbonyl group of the acetic acid by a strong acid catalyst (usually Bronsted-type acid). The resultant acylium ion makes it more susceptible to nucleophilic attack by the oxygen atom of the hydroxyl molecule attached to glycerol, thereby forming a tetrahedral intermediate and leading to the transfer of hydrogen to the next hydroxyl molecule (attached to the carbonyl carbon) to form a water molecule. On the loss of the water molecule, a monoacetin (monoacetyl glycerol) is formed, and on further reaction with acetic acid following a similar mechanism, diacetin (diacetyl glycerol) is formed, and finally, triacetin (triacetyl glycerol) is formed as shown in Scheme 5. The acetylation of glycerol with acetic anhydride in a strong acidic medium proceeds via two possible mechanisms [88][89][90]. The first plausible mechanism, which is less energetic, involves the protonation of the carbonyl oxygen atom, which provides room for nucleophilic attack leading to the formation of a tetrahedral intermediate (quaternary carbon atom) which requires space. This is exemplified in the reaction Scheme 6. While in the second plausible mechanism, the protonation occurs on the oxygen atom attached to the carbonyl group leading to the formation of acylium ion (acylation) due to the presence of a bulky group as shown in Scheme 7. The acetylation of glycerol with acetic anhydride in a strong acidic medium proceeds via two possible mechanisms [88][89][90]. The first plausible mechanism, which is less energetic, involves the protonation of the carbonyl oxygen atom, which provides room for nucleophilic attack leading to the formation of a tetrahedral intermediate (quaternary carbon atom) which requires space. This is exemplified in the reaction Scheme 6. While in the second plausible mechanism, the protonation occurs on the oxygen atom attached to the carbonyl group leading to the formation of acylium ion (acylation) due to the presence of a bulky group as shown in Scheme 7.
The conversion of glycerol to acetin can be influenced by several experimental parameters, just like most reactions. The main parameters include the nature of the catalyst and the catalyst amount (load), temperature, reactants and their mole ratios, as well as the time of the reaction. These parameters are reviewed below.  The conversion of glycerol to acetin can be influenced by several experimental parameters, just like most reactions. The main parameters include the nature of the catalyst and the catalyst amount (load), temperature, reactants and their mole ratios, as well as the time of the reaction. These parameters are reviewed below.

Catalyst and Catalyst Load
The presence of a catalyst in the reaction medium is to lower the activation energy and to form the desired product. In acetylation, the nature of the catalyst plays an important role, just like other reactions. Glycerol acetylation is mostly catalysed with acid catalysts [80,91]. These catalysts could be homogeneous or heterogeneous [92]. Although some studies have reported a high yield of acetylated products using homogeneous acid catalysts such as sulphuric, nitric, hydrofluoric, hydrochloric, p-toluenesulfonic acids, as well as phosphoric acids [19,76,93], they are bedevilled with many environmental and technical drawbacks. Some of these include difficulty in separating the homogeneous catalyst from the product, the corrosion of equipment, difficulty in waste disposal, and toxicity of the reagents [22,58,73]. With the above disadvantages, efforts have generally shifted to the use of heterogeneous catalysts that are less toxic, highly selective, more stable, easy to separate, greener, and more sustainable [78,94,95]. The major advantage of heterogeneous catalysts is that it affords scientists the ability to manipulate the surface area and the acid density to suit a particular reaction. The reusability of heterogeneous catalysts aids their industrial applications irrespective of the type of reactor used [96]. A lot of heterogeneous catalysts have been used in glycerol acetylation, some of which include ion-exchange and functionalized resins [82,[97][98][99], zeolites and functionalized zeolites [73,100,101], heteropoly acids and supported heteropoly acids [93,[102][103][104]], metal oxides including mixed oxides and supported mixed oxides [105][106][107][108], montmorillonite (K-10 clay) and modified montmorillonite [87,91,109], mesoporous silica and functionalized silica [94,110,111]

Catalyst and Catalyst Load
The presence of a catalyst in the reaction medium is to lower the activation energy and to form the desired product. In acetylation, the nature of the catalyst plays an important role, just like other reactions. Glycerol acetylation is mostly catalysed with acid catalysts [80,91]. These catalysts could be homogeneous or heterogeneous [92]. Although some studies have reported a high yield of acetylated products using homogeneous acid catalysts such as sulphuric, nitric, hydrofluoric, hydrochloric, p-toluenesulfonic acids, as well as phosphoric acids [19,76,93], they are bedevilled with many environmental and technical drawbacks. Some of these include difficulty in separating the homogeneous catalyst from the product, the corrosion of equipment, difficulty in waste disposal, and toxicity of the reagents [22,58,73]. With the above disadvantages, efforts have generally shifted to the use of heterogeneous catalysts that are less toxic, highly selective, more stable, easy to separate, greener, and more sustainable [78,94,95]. The major advantage of heterogeneous catalysts is that it affords scientists the ability to manipulate the surface area and the acid density to suit a particular reaction. The reusability of heterogeneous catalysts aids their industrial applications irrespective of the type of reactor used [96]. A lot of heterogeneous catalysts have been used in glycerol acetylation, some of which include ion-exchange and functionalized resins [82,[97][98][99], zeolites and functionalized zeolites [73,100,101], heteropoly acids and supported heteropoly acids [93,[102][103][104]], metal oxides including mixed oxides and supported mixed oxides [105][106][107][108], montmorillonite (K-10 clay) and modified montmorillonite [87,91,109], mesoporous silica and functionalized silica [94,110,111], activated carbon and functionalized biomass-derived carbon [112][113][114]. However, it is important to indicate that the focus of most of the studies has been the effect of the type of catalyst, loading, and reusability. However, other salient features that also aid the efficiency of catalysts in the acetylation of glycerol include the pore size and volume, surface acidity, and preparation techniques [41,87,105,115,116]. The findings of various researchers in glycerol acetylation using different catalysts are hereby reviewed in a cluster of catalyst type.

Montmorillonite (K-10 Clay) and Modified Montmorillonite
Montmorillonite is an aluminohydroxysilicate (an octahedral AlO 6 layer sandwiched between two tetrahedral SiO 4 layers), which is the main constituent of most clay minerals. Most clay minerals are swellable (and also reversible) and have high ion-exchange capacity, and hence are widely used in catalysis generally [117]. The representative structure of montmorillonite clay is shown in Figure 4. Recent advances have shown various modifications of montmorillonite, thereby improving its Bronsted and Lewis acids and pore size distribution [118]. These modifications might have informed their deployment as a catalyst and catalyst support in glycerol acetylation. Kakasaheb et al. [91] reported the synthesis of a series of sulphuric acid-modified Montmorillonite (K-10) catalysts (10, 20, and 30% (w/w) H 2 SO 4 /K10). They were used for the acetylation of bio-glycerol with acetic acid at reaction conditions of 120 • C, 0.4 g catalyst loading, the glycerol-to-acetic acid molar ratio of 1:12, and for 5 h. Furthermore, 20% (w/w) H 2 SO 4 /K10 performs better with 99% glycerol conversion and 23%, 59%, 15%, and 2% yield towards monoacetin, diacetin, triacetin, and diglycerol triacetate. The obtained results were found to correlate with the acidity and other textural properties of the catalysts. The reusability of 20% (w/w) H 2 SO 4 /K10 catalyst was fairly good because, after three reaction cycles, only the partial deactivation of the catalyst was noticed, which was partly attributed to coke deposition and a loss of active acid sites. In a related study, the organic acids-treated montmorillonite clay performed better as a catalyst in acetylation than the untreated [87]. The organic acids used include methanesulfonic acid (MSA), p-toluenesulfonic acid (pTSA), and phenoldisulfonic acid (PDSA), and over 90% glycerol conversion was obtained in each case when compared with the untreated montmorillonite clay with only about 60% glycerol conversion. The selectivity towards triacetin was favoured with the organic acids-treated clay in the following order: PDSA > MSA > p-TSA > untreated clay. The performance was attributed to the enhancement of the catalyst's acidity and pore characteristics (increase in pore volume, especially around the acid sites) arisen from the organic acid's treatment. In another study, the modification of montmorillonite with metals enhanced the selectivity towards diacetin. The lanthanum cation-exchanged montmorillonite (La 3+ -mont), was found to be efficient with 98% glycerol conversion and selectivity of > 99% towards diacetin only. However, when the La 3+ was substituted with Ce 4+ , Sc 3+ , Ti 4+ and Fe 3+ , selectivity towards monoacetin and triacetin improved at 120 • C, glycerol-to-acetic acid molar ratio of 1:2, and reaction time of 24 h [109] which is not cost effective.

Metals and Metal Oxides
Metals and metal oxides mostly used in catalysis in recent times are multi-metals or multiphase oxides and are of transition metal origin. Because of the synergy of various components, the resultant metals and metal oxides catalysts usually exhibit Bronsted and Lewis acid sites, stability, regeneration capacity, and were found to be active over a wide range of temperatures [120,121]. Hence, researchers have also deployed these catalysts in glycerol acetylation with different findings. The use of Cu and Ni metals and Cu-Ni bimetals supported on γ-Al2O3 were compared with sulphuric acid-treated alumina (H2SO4/γ-Al2O3) catalysts in the esterification of glycerol with acetic acid at 110 °C, atmospheric pressure, the molar ratio of 1:9, and the catalyst loading of 0.25 g [122]. The results revealed that the 2 M H2SO4/γ-Al2O3 catalyst showed the best activity with 97% glycerol conversion and a selectivity of 27%, 49.9%, and 23.1% mono-, di-, and triacetin after a 5 h reaction time. The catalyst was also reported not to lose activity even after three consecutive runs. The performance of the 2 M H2SO4/γ-Al2O3 catalyst was attributed to the high acid strength of 2.51 mmol g −1 when compared to the other catalysts tested. In a similar use of bimetallic-supported catalyst, Ramalingam et al. [90] reported the use of Ag-Cu-doped rice husk silica-alumina (RHS-Al)-based biomass catalyst under similar reaction conditions to the glycerol-to-acetic acid molar ratio of 1:10, 110 °C and a catalyst load of 0.80 g. The results showed high catalytic activity with 97.5% glycerol conversion and selectivity towards mono-, di-, and triacetin (3.7, 58.4, and 37.9%) over 1Ag-10Cu/RHS-Al catalyst indicating that the bimetallic catalyst performed better than the single metal catalyst and was attributed to the synergistic effect between the metals. The use of several metal oxides (Bi2O3, Sb2O3, SnO2, TiO2, Nb2O5 and Sb2O5) was also investigated at the glycerol-to-acetic acid molar ratio of 1:6 and a temperature of 120 °C. Only Sb2O5 (antimony pentoxide) resulted in good activity and selectivity towards diacetin after 1 h [123]. Glycerol conversion was 96.8% while selectivity towards mono-, di-and triacetin were 33.2, 54.2 and 12.6%, respectively. Incidentally, it

Metals and Metal Oxides
Metals and metal oxides mostly used in catalysis in recent times are multi-metals or multiphase oxides and are of transition metal origin. Because of the synergy of various components, the resultant metals and metal oxides catalysts usually exhibit Bronsted and Lewis acid sites, stability, regeneration capacity, and were found to be active over a wide range of temperatures [120,121]. Hence, researchers have also deployed these catalysts in glycerol acetylation with different findings. The use of Cu and Ni metals and Cu-Ni bimetals supported on γ-Al 2 O 3 were compared with sulphuric acid-treated alumina (H 2 SO 4 /γ-Al 2 O 3 ) catalysts in the esterification of glycerol with acetic acid at 110 • C, atmospheric pressure, the molar ratio of 1:9, and the catalyst loading of 0.25 g [122]. The results revealed that the 2 M H 2 SO 4 /γ-Al 2 O 3 catalyst showed the best activity with 97% glycerol conversion and a selectivity of 27%, 49.9%, and 23.1% mono-, di-, and triacetin after a 5 h reaction time. The catalyst was also reported not to lose activity even after three consecutive runs. The performance of the 2 M H 2 SO 4 /γ-Al 2 O 3 catalyst was attributed to the high acid strength of 2.51 mmol g −1 when compared to the other catalysts tested. In a similar use of bimetallic-supported catalyst, Ramalingam et al. [90] reported the use of Ag-Cu-doped rice husk silica-alumina (RHS-Al)-based biomass catalyst under similar reaction conditions to the glycerol-to-acetic acid molar ratio of 1:10, 110 • C and a catalyst load of 0.80 g. The results showed high catalytic activity with 97.5% glycerol conversion and selectivity towards mono-, di-, and triacetin (3.7, 58.4, and 37.9%) over 1Ag-10Cu/RHS-Al catalyst indicating that the bimetallic catalyst performed better than the single metal catalyst and was attributed to the synergistic effect between the metals. The use of several metal oxides (Bi 2 O 3 , Sb 2 O 3 , SnO 2 , TiO 2 , Nb 2 O 5 and Sb 2 O 5 ) was also investigated at the glycerol-to-acetic acid molar ratio of 1:6 and a temperature of 120 • C. Only Sb 2 O 5 (antimony pentoxide) resulted in good activity and selectivity towards diacetin after 1 h [123]. Glycerol conversion was 96.8% while selectivity towards mono-, di-and triacetin were 33.2, 54.2 and 12.6%, respectively. Incidentally, it was Sb 2 O 5 that showed strong Bronsted acid sites was able to protonate adsorbed pyridine which might have influenced selectivity towards diacetin. The Sb 2 O 5 catalyst was found to be active even after six reaction cycles. More recently, Zhang et al. [107] synthesized a diatomite-loaded H 2 SO 4 /TiO 2 catalyst and was used in the esterification reaction of glycerol with oleic acid. In a 6 h reaction time, 59.6% diacetin was obtained at a much higher temperature of 210 • C, low catalyst load (0.1%), and low molar ratio of glycerol-to-oleic acid (1:2). The reusability of the catalyst was tested five times. The diacetin obtained was about 50% in the first three cycles but decreased in the fourth cycle by 12.5%, and 10% in the fifth cycle. The author attributed the results to the blockage of active sites by the bulky triacetin and the leaching of active sites from the catalyst. Recently, Kulkarni et al. [108] deployed unsulphated and sulphated CeO 2 -ZrO 2 mixed oxide catalysts in glycerol acetylation with acetic acid and reported that the sulphated CeO 2 -ZrO 2 mixed oxide catalyst exhibited better performance when compared with the unsulphated catalyst. Within 3 h reaction time, 99.1% of glycerol conversion was achieved with 57.28% and 21.26% selectivity towards diacetin and triacetin at 100 • C. The catalyst was reused in three consecutive cycles without a significant loss of activity.

Mesoporous Silica and Functionalized Silica
Mesoporous silica is an ordered porous network material with SiO 4 tetrahedral as the building block, illustrated in Figure 5. It is made of weak Bronsted acid oxides and hence has a low catalytic value. However, because the surface is amenable to functionalization and other modifications, resulting in a highly stable material with a high surface area and large pore size, it is currently receiving significant attention as catalyst support in various reactions to enhance the reactivity and selectivity of products, especially of large molecules like fatty acids and their esters [59,117,124].
Ghoreishi and Yarmo [110] reported the influence of a sulphated silica catalyst, prepared with the sol-gel technique with different sulphate loading percentages (5, 10, 15, and 20%) in the esterification of glycerol with acetic acid at a molar ratio of 1:6, at a temperature of 50 • C and with 0.2 g catalyst load in 6 h. The results revealed that the catalytic activity increased with an increase in the acidic properties, which in turn was found to increase with the increased sulphuric acid loading. The 20% sulphated silica exhibited the highest catalytic activity with 96.88% glycerol conversion and a selectivity of 51.9, 45.27, and 2.11% towards mono-, di-, and triacetin. The authors further reported a strong correlation between the textural properties (surface area and pore volume) and the sulphur content of the catalysts. The decrease in both the surface area and pore volume was found to correlate with an increase in sulphur content, attributable to the blockage of pores by the active non-porous species. The authors further studied the effect of catalyst loading on the reaction, and the results revealed a slight increase in glycerol conversion with an increase in sulphated silica catalyst loading from 0.2 to 0.8 g. The selectivity to monoacetin decreased with an increase in catalyst loading, while the selectivity to diacetin and triacetin improved marginally. Khayoon et al. [82] reported a much higher selectivity towards diacetin and triacetin with 34 and 55%, when a 3% yttrium supported on mesoporous silicate material (SBA-3) prepared via hydrothermal method was deployed as a catalyst at a glycerol-to-acetic acid molar ratio of 1:4 and a temperature of 110 • C. A 100% glycerol conversion was achieved in 2.5 h. The catalyst was found to be stable after being reused in three reaction cycles. The high triacetin produced was attributed to the strong acidity, and the high surface area with the large pore size, which makes the diffusion of both the substrates and products possible. However, in the fourth cycle, the glycerol conversion and triacetin were reduced to 80% and 50, respectively, due to mass transfer limitations. The authors further reported that only 44% glycerol conversion was achieved with a catalyst load of 0.05 g. However, when the catalyst load was increased to 0.20 g, the glycerol conversion increased to 100%. The subsequent increase in the catalyst load did not improve the conversion, an indication of reaching the optimum limit. In a similar vein, Costa et al. [73] also reported the excellent performance of SBA-15-functionalized with propylsulfonic group (Pr-SO 3 H-SBA-15) as a catalyst but at a higher temperature of 120 • C and a glycerol-to-acetic acid mole ratio of 1:6. In comparison with zeolites (H-ZSM-5 and H-Beta), the Pr-SO 3 H-SBA-15 catalyst exhibited a 96% glycerol conversion and a selectivity of 13, 55, and 32% mono-, di-and triacetin in 2.5 h reaction time. The good performance of the Pr-SO 3 H-SBA-15 catalyst was attributed to adequate balance between the amount of acid site and its distribution on the surface of the mesoporous structure. Unlike earlier studies, the Pr-SO 3 H-SBA-15 catalyst could not maintain the activity in consecutive cycles, even with treatment in between. In a comparable study, Testa et al. [92] deployed the synthesized solid acid catalysts, namely propyl-SO 3 H functionalized amorphous silica (SAS), propyl-SO 3 H-functionalized mesoporous silica SBA-15 (SSBA), titania-doped silica (hexagonal mesoporous silica, HMS) (Ti10HMS), sulphated zirconia (SZ and calcined SZ-470) and niobic acid nanoparticles (Nb 2 O 5 ·nH 2 O) in the esterification reaction of glycerol with acetic acid in a molar ratio of 1:3 at 105 • C. The SAS and SSBA catalysts showed the best results. Though the complete glycerol conversion (100%) was achieved within 1 h, the triacetin of 49% and 33% were achieved after 3 h with SAS and SSBA, respectively. Both catalysts were found to be very stable, with no leaching of the sulfonic groups. The excellent activities of the two catalysts were attributed to their acid capacities, including the strength and the surface density of the acid site.
Appl. Sci. 2020, 10, x FOR PEER REVIEW 12 of 33 between the amount of acid site and its distribution on the surface of the mesoporous structure. Unlike earlier studies, the Pr-SO3H-SBA-15 catalyst could not maintain the activity in consecutive cycles, even with treatment in between. In a comparable study, Testa et al. [92] deployed the synthesized solid acid catalysts, namely propyl-SO3H functionalized amorphous silica (SAS), propyl-SO3H-functionalized mesoporous silica SBA-15 (SSBA), titania-doped silica (hexagonal mesoporous silica, HMS) (Ti10HMS), sulphated zirconia (SZ and calcined SZ-470) and niobic acid nanoparticles (Nb2O5·nH2O) in the esterification reaction of glycerol with acetic acid in a molar ratio of 1:3 at 105 °C. The SAS and SSBA catalysts showed the best results. Though the complete glycerol conversion (100%) was achieved within 1 h, the triacetin of 49% and 33% were achieved after 3 h with SAS and SSBA, respectively. Both catalysts were found to be very stable, with no leaching of the sulfonic groups. The excellent activities of the two catalysts were attributed to their acid capacities, including the strength and the surface density of the acid site.

Carbon and Functionalized Carbon-Based Materials
The use of carbon-based catalysts in acetylation has been reported as well. Several literature articles have indicated the synthesis of a functionalized catalyst from biomass materials via carbonization and sulfonation, but only few reported their application in glycerol acetylation. Okoye et al. [114] reported the use of crude glycerol as the precursor material in carbon-based catalyst synthesis and was found to be active in glycerol acetylation. They also reported that an increase in catalyst load from 1 to 2 wt% increased glycerol conversion (>90%) and remained relatively constant even when the catalyst load was increased up to 4 wt%. This observation was attributed to active site saturation. An increased catalyst load favours the selectivity to di-and triacetin due to increased active sites, while a lower catalyst load favours monoacetin with decreased glycerol conversion. The authors also observed that the synthesized carbon-based catalyst was reused seven times, and the glycerol conversion, as well as the product selectivity, remained constant, indicating high stability. Similarly, sulfonated carbon material derived from Pongamia pinnata de-oiled seed cake was synthesized and used for the same reaction but with different acetylating agents (lauric and oleic acids) [101]. The catalyst was found to perform better when compared to zeolites H-Y, H-ZSM-5, and liquid H2SO4. Glycerol conversion ranged from 80 to 95% with monoacetin as the major product (70-80%) within 24 h. The authors also reported that the sulfonated catalyst increased the conversion of glycerol, which was subsequently found to depend on the catalyst load. Up to 96% glycerol conversion was obtained at 7.5 wt% catalyst load. However, 5 wt% was considered to be the optimum load because it was not significantly different from the former. The performance of the sulfonated catalyst was found to correlate with its acid density, pore size, and the dimensions of the reacting molecules. The authors also confirmed the reusability of the sulfonated carbon catalyst in three successive esterification cycles with oleic acid. The results indicated that the catalyst was deactivated by only 14% but retained the active site (-SO3H) by almost 90% after the third reaction cycle with a slight improvement in the selectivity of monoacetin, which was the desired product. The slight loss was attributed to the possible restricted access to the active sites due to adsorbed carbonaceous impurities on the surface of the catalyst, and the high stability was due to the stability of the C(sp 2 )-SO3H bond. Similarly, Konwar et al. [115] deployed the same catalyst but with acetic anhydride.

Carbon and Functionalized Carbon-Based Materials
The use of carbon-based catalysts in acetylation has been reported as well. Several literature articles have indicated the synthesis of a functionalized catalyst from biomass materials via carbonization and sulfonation, but only few reported their application in glycerol acetylation. Okoye et al. [114] reported the use of crude glycerol as the precursor material in carbon-based catalyst synthesis and was found to be active in glycerol acetylation. They also reported that an increase in catalyst load from 1 to 2 wt% increased glycerol conversion (>90%) and remained relatively constant even when the catalyst load was increased up to 4 wt%. This observation was attributed to active site saturation. An increased catalyst load favours the selectivity to di-and triacetin due to increased active sites, while a lower catalyst load favours monoacetin with decreased glycerol conversion. The authors also observed that the synthesized carbon-based catalyst was reused seven times, and the glycerol conversion, as well as the product selectivity, remained constant, indicating high stability. Similarly, sulfonated carbon material derived from Pongamia pinnata de-oiled seed cake was synthesized and used for the same reaction but with different acetylating agents (lauric and oleic acids) [101]. The catalyst was found to perform better when compared to zeolites H-Y, H-ZSM-5, and liquid H 2 SO 4 . Glycerol conversion ranged from 80 to 95% with monoacetin as the major product (70-80%) within 24 h. The authors also reported that the sulfonated catalyst increased the conversion of glycerol, which was subsequently found to depend on the catalyst load. Up to 96% glycerol conversion was obtained at 7.5 wt% catalyst load. However, 5 wt% was considered to be the optimum load because it was not significantly different from the former. The performance of the sulfonated catalyst was found to correlate with its acid density, pore size, and the dimensions of the reacting molecules. The authors also confirmed the reusability of the sulfonated carbon catalyst in three successive esterification cycles with oleic acid. The results indicated that the catalyst was deactivated by only 14% but retained the active site (-SO 3 H) by almost 90% after the third reaction cycle with a slight improvement in the selectivity of monoacetin, which was the desired product. The slight loss was attributed to the possible restricted access to the active sites due to adsorbed carbonaceous impurities on the surface of the catalyst, and the high stability was due to the stability of the C(sp 2 )-SO 3 H bond. Similarly, Konwar et al. [115] deployed the same catalyst but with acetic anhydride. Additionally, 100% triacetin selectivity was achieved in 20-50 min with both mesoporous sulfonated carbon and zeolite H-Y catalysts. The authors further revealed that the selectivity to triacetin was influenced mainly by the pore structure and surface acid site density of the catalysts. Sulfonated carbon catalyst made from the mixture of carbonized rice husk and sulphuric acid was reported by Carvalho et al. [75]. After 5 h of reaction, 90% glycerol conversion and selectivity of 11, 52, and 37% towards mono, di, and triacetin were obtained, respectively. The improved selectivity towards di-and triacetin was attributed to the high total acidity (5.8 mmol·g −1 ). Surprisingly, non-functionalized biochar (Karanja seed shells) was also able to catalyse the esterification of glycerol with acetic acid at 120 • C, at a molar ratio of 1:5, a catalyst load of 0.2 g, and >80% glycerol conversion was reported after 1 h. However, the major product's selectivity was monoacetin. The reaction was also attributed to moderate to strong acidic sites exhibited by the biochar despite the non-functionalization [125]. The use of a two-step reaction system with a sulfonic acid-functionalized glycerol-based carbon catalyst produced 100% triacetin [126]. In the first step, acetic acid was used, and in the second step, acetic anhydride was added to complete the reaction. The developed catalyst was found to be highly stable and recyclable. The authors further studied the effect of the catalyst loading (5-20 wt%) on the reaction and revealed that the catalyst load of 5 wt% led to 100% glycerol conversion with 22, 67, and 11% mono-, di-and triacetin selectivity after 1 h, respectively, and when compared with the control, the selectivity was 42, 54 and 4%, for mono-, di-and triacetin, respectively, which indicates the slow pace of the esterification reaction without the catalyst. However, when the catalyst load was increased to 20 wt%, the glycerol conversion increased to 100%, and the selectivity to monoacetin decreased while the di-and triacetin increased simultaneously. The trend was attributed to catalyst acidity, which enhanced the surface interactions of the glycerol with acetic acid. The reusability and stability of the catalyst were found to be excellent even after five reaction cycles at a 1:4 glycerol-to-acetic acid molar ratio, at 115 • C and with a 1 h reaction time. No deactivation and leaching under the reaction conditions were observed. Synthesized ordered mesoporous carbon via hard template method and its modification with concentrated sulphuric or 4-aminobenzenesulfonic acid were tested in glycerol acetylation [116]. Both showed good activity with the material modified via diazonium cation exhibiting high glycerol conversion and selectivity to triacetin under similar conditions. Figure 6 is an illustration of a sulfonated carbon-based catalyst.
Appl. Sci. 2020, 10, x FOR PEER REVIEW 13 of 33 Additionally, 100% triacetin selectivity was achieved in 20-50 min with both mesoporous sulfonated carbon and zeolite H-Y catalysts. The authors further revealed that the selectivity to triacetin was influenced mainly by the pore structure and surface acid site density of the catalysts. Sulfonated carbon catalyst made from the mixture of carbonized rice husk and sulphuric acid was reported by Carvalho et al. [75]. After 5 h of reaction, 90% glycerol conversion and selectivity of 11, 52, and 37% towards mono, di, and triacetin were obtained, respectively. The improved selectivity towards diand triacetin was attributed to the high total acidity (5.8 mmol·g −1 ). Surprisingly, non-functionalized biochar (Karanja seed shells) was also able to catalyse the esterification of glycerol with acetic acid at 120 °C, at a molar ratio of 1:5, a catalyst load of 0.2 g, and >80% glycerol conversion was reported after 1 h. However, the major product's selectivity was monoacetin. The reaction was also attributed to moderate to strong acidic sites exhibited by the biochar despite the non-functionalization [125]. The use of a two-step reaction system with a sulfonic acid-functionalized glycerol-based carbon catalyst produced 100% triacetin [126]. In the first step, acetic acid was used, and in the second step, acetic anhydride was added to complete the reaction. The developed catalyst was found to be highly stable and recyclable. The authors further studied the effect of the catalyst loading (5-20 wt%) on the reaction and revealed that the catalyst load of 5 wt% led to 100% glycerol conversion with 22, 67, and 11% mono-, di-and triacetin selectivity after 1 h, respectively, and when compared with the control, the selectivity was 42, 54 and 4%, for mono-, di-and triacetin, respectively, which indicates the slow pace of the esterification reaction without the catalyst. However, when the catalyst load was increased to 20 wt%, the glycerol conversion increased to 100%, and the selectivity to monoacetin decreased while the di-and triacetin increased simultaneously. The trend was attributed to catalyst acidity, which enhanced the surface interactions of the glycerol with acetic acid. The reusability and stability of the catalyst were found to be excellent even after five reaction cycles at a 1:4 glycerol-to-acetic acid molar ratio, at 115 °C and with a 1 h reaction time. No deactivation and leaching under the reaction conditions were observed. Synthesized ordered mesoporous carbon via hard template method and its modification with concentrated sulphuric or 4-aminobenzenesulfonic acid were tested in glycerol acetylation [116]. Both showed good activity with the material modified via diazonium cation exhibiting high glycerol conversion and selectivity to triacetin under similar conditions. Figure 6 is an illustration of a sulfonated carbon-based catalyst.

Ion-Exchange and Functionalized Resins
Ion-exchange resins are usually co-polymers of polystyrene and divinylbenzene with exchangeable ions or functional groups. They exist commonly as Amberylsts, Nafion, Dowex-50W, and sulfonic acid resin. Figure 7 shows the structure of some resin catalysts. Several reports indicate the deployment of ion exchange resins in its original and functionalized forms as well as catalyst support in glycerol acetylation. Lacerda et al. [99] compared the use of acidic anhydride and acetic acid in glycerol acetylation over amberlyst-15. A 100% glycerol conversion was achieved with both acetylating agents but with varying selectivity to mono-, di-, and triacetin. The conditions of the reaction with the acetic anhydride were much milder when compared with that of acetic acid. The reaction of glycerol with acetic anhydride was at a molar ratio of 1:3, a temperature of 60 °C, and a

Ion-Exchange and Functionalized Resins
Ion-exchange resins are usually co-polymers of polystyrene and divinylbenzene with exchangeable ions or functional groups. They exist commonly as Amberylsts, Nafion, Dowex-50W, and sulfonic acid resin. Figure 7 shows the structure of some resin catalysts. Several reports indicate the deployment of ion exchange resins in its original and functionalized forms as well as catalyst support in glycerol acetylation. Lacerda et al. [99] compared the use of acidic anhydride and acetic acid in glycerol acetylation over amberlyst-15. A 100% glycerol conversion was achieved with both acetylating agents but with varying selectivity to mono-, di-, and triacetin. The conditions of the reaction with the acetic anhydride were much milder when compared with that of acetic acid. The reaction of glycerol with acetic anhydride was at a molar ratio of 1:3, a temperature of 60 • C, and a catalyst load of 5 wt% glycerol, which led to 100% glycerol conversion with a selectivity of 0.7, 1.3, and 98.1% mono, di, and triacetin in 2 h. While the reaction of glycerol with acetic acid produced 100% glycerol conversion with a selectivity of 3.5, 8.7, and 87.8% for mono, di, and triacetin, respectively, this was at a higher molar ratio of 1:6 and a higher temperature of 120 • C with the same catalyst load and time. Similar results were reported by Silver et al. [88]  . However, its catalytic activity in acetylation reaction was inferior. The authors suggested that in addition to PrSO 3 H-SBA-15 and amberlyst-15 catalysts exhibiting moderate to high Bronsted acid strength (1.2 and 4.9 mmol·g −1 ), they also have the configuration and spatial arrangement of their surface acid moieties that contributed to their excellent activities, unlike the other catalysts. The two-step acetylation of glycerol with acetic acid and acidic anhydride over amberlyst-35 resin catalyst was compared with zeolite (HY and HZSM-5) catalysts [80]. At the optimum conditions of 105 • C, the glycerol-to-acetic acid molar ratio of 1:9, 0.5 g catalyst loading, and 4 h reaction time in the first step, the glycerol conversion was almost 100% with a selectivity of 25.9% towards triacetin with the amberlyst-35 catalyst. The catalyst performance was attributed to its high acid capacity when compared to the rest of the catalysts used. With the addition of acetic anhydride in the second step, the selectivity of triacetin improved to almost 100% within 15 min. Amberlyst-35 was found to be the most excellent catalyst for the conversion as no significant deactivation occurred after being reused five times. Even when the reaction was extended to 12 h, the catalyst was found to be very stable. The same authors also investigated the influence of the catalyst load at a fixed molar ratio of glycerol-to-acetic acid of 1:6, 105 • C, and 4 h reaction time. The glycerol conversion was almost 100% with the presence of 0.25 g of the catalyst. While without the catalyst, the conversion was 73.6%. The selectivity to monoacetin decreased with an increase in catalyst amount up to 0.5 g, while the di-and triacetin increased. However, no appreciable change was noticed in the selectivity of mono-, di-, and triacetin when the catalyst load was further increased to 1.0 g. The authors concluded that much of the catalyst added did not substantially improve its activity. However, it has been established by several researchers that ion exchange resins are not thermally stable. They become easily deactivated above 150 • C [59,98].
Appl. Sci. 2020, 10, x FOR PEER REVIEW 14 of 33 catalyst load of 5 wt% glycerol, which led to 100% glycerol conversion with a selectivity of 0.7, 1.3, and 98.1% mono, di, and triacetin in 2 h. While the reaction of glycerol with acetic acid produced 100% glycerol conversion with a selectivity of 3.5, 8.7, and 87.8% for mono, di, and triacetin, respectively, this was at a higher molar ratio of 1:6 and a higher temperature of 120 °C with the same catalyst load and time. Similar results were reported by Silver et al. [88] using acetic anhydride but at a lower reaction time of just 20 min. Kim et al. [41] also used amberlyst-15 catalyst in comparison with several other solid catalysts including silica-alumina, zeolite (HUSY), dodecamolybdophosphoric acids supported on Nb2O5 (HPMo/Nb2O5) and mesoporous SBA-15 (HPMo/SBA-15), sulphated ceria-zirconia (SCZ), propylsulfonic acid functionalized SBA-15 (PrSO3H-SBA-15), sulfonic acid functionalized SBA-15 (SO3H-SBA-15) and microcrystalline cellulose (SO3H-Cell) catalysts in glycerol acetylation with acetic acid at a molar ratio of 1:6, a temperature of 80 °C and the catalyst load of 5 wt% with respect to glycerol. After a long reaction time of 8 h, PrSO3H-SBA-15, amberlyst-15, and SO3H-SBA-15 catalysts exhibited superior performance in glycerol conversion and selectivity to higher acetins. However, the SO3H-SBA-15 catalyst showed significant deactivation due to the loss of acidity. The excellent performance by the PrSO3H-SBA-15 and amberlyst-15 catalysts could not be entirely due to the strength of Bronsted acid because the sulphated ceria-zirconia (SCZ) exhibited the higher Bronsted acid strength (2.9 mmol·g −1 ) compared to PrSO3H-SBA-15 (1.2 mmol·g −1 ). However, its catalytic activity in acetylation reaction was inferior. The authors suggested that in addition to PrSO3H-SBA-15 and amberlyst-15 catalysts exhibiting moderate to high Bronsted acid strength (1.2 and 4.9 mmol·g −1 ), they also have the configuration and spatial arrangement of their surface acid moieties that contributed to their excellent activities, unlike the other catalysts. The two-step acetylation of glycerol with acetic acid and acidic anhydride over amberlyst-35 resin catalyst was compared with zeolite (HY and HZSM-5) catalysts [80]. At the optimum conditions of 105 °C, the glycerol-to-acetic acid molar ratio of 1:9, 0.5 g catalyst loading, and 4 h reaction time in the first step, the glycerol conversion was almost 100% with a selectivity of 25.9% towards triacetin with the amberlyst-35 catalyst. The catalyst performance was attributed to its high acid capacity when compared to the rest of the catalysts used. With the addition of acetic anhydride in the second step, the selectivity of triacetin improved to almost 100% within 15 min. Amberlyst-35 was found to be the most excellent catalyst for the conversion as no significant deactivation occurred after being reused five times. Even when the reaction was extended to 12 h, the catalyst was found to be very stable. The same authors also investigated the influence of the catalyst load at a fixed molar ratio of glycerol-to-acetic acid of 1:6, 105 °C, and 4 h reaction time. The glycerol conversion was almost 100% with the presence of 0.25 g of the catalyst. While without the catalyst, the conversion was 73.6%. The selectivity to monoacetin decreased with an increase in catalyst amount up to 0.5 g, while the diand triacetin increased. However, no appreciable change was noticed in the selectivity of mono-, di-, and triacetin when the catalyst load was further increased to 1.0 g. The authors concluded that much of the catalyst added did not substantially improve its activity. However, it has been established by several researchers that ion exchange resins are not thermally stable. They become easily deactivated above 150 °C [59,98].

Heteropoly Acids and Supported Heteropoly Acids
Heteropoly acids (HPA) are complex compounds that exhibit very strong Bronsted acidity but with low specific surface area and low thermal stability [103,130]. They also exhibit homogeneous behaviour (highly soluble in polar media) in most reactions. However, with intense research, the protons of HPA can now be substituted by incorporating metal ions, especially transition metals. The Keggin structure of heteropoly acids (HPA) is illustrated in Figure 8. Other materials such as silica, carbon, zeolite, resin, metal oxides, etc., have also been used to immobilized HPA as support, thereby improving their surface area, thermal stability, as well as reduce their solubility and leaching in reaction medium [59,71,102,118]. The modifications have greatly improved the catalytic activity of HPA and have been used in glycerol acetylation. Zhu et al. [131] reported the complete crude glycerol conversion (100%) and selectivity of 6.4%, 61.3%, and 32.3% towards mono-, di-, and triacetin, respectively, at 120 • C in 4 h over a zirconia-supported H 4 SiW 12 O 40 (HSiW/ZrO 2 ) catalyst. There was no catalyst deactivation noticed even after four reaction cycles and was attributed to good surface Bronsted acid site and hydrothermal stability. Similarly, 100% glycerol conversion was achieved within situ-generated supported silicomolybdic heteropolyanions from the sol-gel synthesized MoO 3 /SiO 2 catalyst. As the mol% of MoO 3 supported on SiO 2 increased, the conversion remained the same, but the selectivity varies. 1 mol% MoO 3 gave selectivity of 56, 28, and 17% mono-, di-, and triacetin. A 10 mol% MoO 3 produced a selectivity of 36, 31, and 33% mono-, di-, and triacetin, respectively. While 20 mol% MoO 3 produced 17, 33, and 50% mono-, di-, and triacetin, respectively, in 8 h reaction time, at 100 • C, with a glycerol-to-acetic acid molar ratio of 1:10 and a catalyst load of 10 wt% with respect to glycerol. However, 76% selectivity towards triacetin was obtained when the temperature and reaction time were increased to 118 • C and 20 h, respectively [132]. Sandesh et al. [89] studied various solid catalysts, namely cesium phosphotungstate (CsPWA), amberlyst-15, H-beta, sulphated zirconia (SZ), and montmorillonite K-10, in glycerol acetylation or esterification with acidic anhydride and acetic acid. The reaction of glycerol with acidic anhydride was carried out under a milder condition of 1:3 mole ratio, 30 • C and a catalyst load of 1 wt% with respect to glycerol when compared with that of acetic acid, a 1:8 mole ratio, 85 • C and catalyst load of 7 wt%. The activity of the catalysts in both reactions is in the order SZ < H-beta < K10 < A-15 < CsPWA. The CsPWA catalyst exhibited the highest glycerol conversion of >98% in both reactions with combined di-and triacetin selectivity of 99.1% with acetic anhydride and 75% with acetic acid. The high glycerol conversion was attributed to the presence of high Bronsted acidity while that of selectivity was found to depend on the nature of the active sites on the catalyst. The authors went further to study the effect of the CsPWA catalyst load on the glycerol esterification with acetic acid. The result showed increased glycerol conversion from 56 to 98% when the catalyst load was increased from 3 to 7 wt% in a reaction time of 2 h. The selectivity to di-and triacetin also increased from 31 and 0 to 59 and 16%, respectively. This observation was attributed to the presence of more active sites with a high catalyst load. However, when the catalyst amount was increased further to 9 wt%, no appreciable change was observed, and the authors concluded that the maximum glycerol conversion and product selectivity were attained with the catalyst load of 7 wt%. The CsPWA catalyst exhibited good reusability with minimal decreased activity after three reaction cycles. The activity of CsPWA in glycerol esterification was also confirmed by Veluturla et al. [121], but the maximum glycerol conversion and esters yield were obtained at a 5 wt% catalyst load. Most of the earlier works prefer the use of lower catalyst loads irrespective of the catalyst used, and the pattern of the results was similar [98,102,133]. Polymeric material and polyvinylpyrrolidone (PVP) were separately impregnated with phosphotungstic, silicotungstic, and phosphomolybdic acids, and each used as a catalyst in the conversion of glycerol to acetin using acetic acid as the acetylating reagent [104]. It was observed that the PVP impregnated with phosphotungstic acid (PVP-DTP) exhibited the highest selectivity to triacetin (34%), and 100% glycerol conversion was also achieved at the optimum conditions. The performance was attributed to its high acidity. However, the catalyst was found to be highly hydrophobic. Even when the authors compared the PVP-DTP catalyst with the commercial montmorillonite KSFO catalyst under the same conditions, the former performed better.

Others
The use of a magnetic solid acid catalyst (Fe-Sn-Ti(SO4 2-)-400) in glycerol acetylation with acetic anhydride was reported by Sun et al. [134]. A 100% glycerol conversion and 99.0% selectivity to triacetin were achieved at a molar ratio of 1:6, at 80 °C, and within 30 min. The catalyst was reused three times, and the activity remained unchanged. The authors attributed the activity of the catalyst to the high surface area and pore volume. However, when the acetic acid was used as the acetylating agent under similar reaction conditions, the glycerol conversion reduced drastically to 32.3%, while the only product was monoacetin with a selectivity of 100%. However, when the reaction temperature increased to 140 °C, the glycerol conversion improved up to 75.7% with monoacetin selectivity of 90.3%. Meanwhile, Sandesh et al. [135] showed that acetin could also be produced simultaneously with biodiesel by reacting glycerol with methyl acetate over Ca and Sn-mixed metal hydroxides catalysts. The CaSn(OH)6 catalyst performed much better when compared to other hydroxy stannates and metal oxides like MgSn(OH)6, ZnSn(OH)6, SrSn(OH)6, Ca(OH)2, CaO and MgO. The high glycerol conversion (78.2%) was attributed to the high basicity exhibited by the CaSn(OH)6 catalyst contrary to the expectation. The authors also reported that this transesterification method is cost-effective. The use of acid-functionalized ionic liquids as a catalyst in glycerol acetylation has been reported to be of high catalytic activity [136][137][138]. A >95% yield of triacetin was obtained in glycerol acetylation with acetic acid over 1-methyl-3-(3-sulfopropyl)-imidazolium hydrogen sulphate ([HSO3-pmim][HSO4]) as the catalyst after 6 h of reaction time at 120 °C [136]. The catalyst was reused ten times, and yet the activity remained at >91%. A similar performance was reported by Liu et al. [137], but the selectivity to diacetin (51.4%) was the main product after 0.5 h at 100 °C and 0.5 wt% catalyst load. Similarly, Keogh et al. [138] reported a range of nitrogen-based Brønsted-acidic ionic liquids (alkyl-pyrrolidone and alkylamine cations) with reasonable activity. N-methyl-2-pyrrolidinium hydrogen sulphate [H-NMP][HSO4] was found to be the most active and cost-effective catalyst with >99% glycerol conversion and 42.3% selectivity towards triacetin, and >95% of combined di-and triacetin selectivity within 1 h. Non-porous materials such as hydroxylated magnesium fluoride were deployed in glycerol acetylation with acetic acid (1:3) at 100 °C, and the result indicates >90% glycerol conversion with diacetin as the major product (≈60%) followed by triacetin (≈30%) after 22 h reaction time, unlike the uncatalyzed reaction that produced only 30% glycerol conversion with monoacetin as the major product (>90%). However, the reaction time was reduced from 4 h to 0.5 h when non-conventional activation methods, namely microwave or ultrasound irradiation, were used [139]. The authors attributed the excellent performance of the catalyst to the density of acid sites on the external surface of the catalyst, while the selectivity was influenced by the nature of acid sites (Lewis and Brønsted).
Findings from the above catalytic studies showed that an increase in catalyst loading to a certain level (amount) improved glycerol conversion, especially to diacetin and triacetin, due to the increased number of available and accessible active acid site in the reaction mixture. Additional findings also revealed that the presence of the Bronsted acid site on the catalyst promotes optimal glycerol acetylation. The effect on monoacetin production is minimal, and this can be seen in the pictorial

Others
The use of a magnetic solid acid catalyst (Fe-Sn-Ti(SO 4 2− )-400) in glycerol acetylation with acetic anhydride was reported by Sun et al. [134]. A 100% glycerol conversion and 99.0% selectivity to triacetin were achieved at a molar ratio of 1:6, at 80 • C, and within 30 min. The catalyst was reused three times, and the activity remained unchanged. The authors attributed the activity of the catalyst to the high surface area and pore volume. However, when the acetic acid was used as the acetylating agent under similar reaction conditions, the glycerol conversion reduced drastically to 32.3%, while the only product was monoacetin with a selectivity of 100%. However, when the reaction temperature increased to 140 • C, the glycerol conversion improved up to 75.7% with monoacetin selectivity of 90.3%. Meanwhile, Sandesh et al. [135] showed that acetin could also be produced simultaneously with biodiesel by reacting glycerol with methyl acetate over Ca and Sn-mixed metal hydroxides catalysts. The CaSn(OH) 6 catalyst performed much better when compared to other hydroxy stannates and metal oxides like MgSn(OH) 6 , ZnSn(OH) 6 , SrSn(OH) 6 , Ca(OH) 2 , CaO and MgO. The high glycerol conversion (78.2%) was attributed to the high basicity exhibited by the CaSn(OH) 6 catalyst contrary to the expectation. The authors also reported that this transesterification method is cost-effective. The use of acid-functionalized ionic liquids as a catalyst in glycerol acetylation has been reported to be of high catalytic activity [136][137][138]. A >95% yield of triacetin was obtained in glycerol acetylation with acetic acid over 1-methyl-3-(3-sulfopropyl)-imidazolium hydrogen sulphate ([HSO 3 -pmim][HSO 4 ]) as the catalyst after 6 h of reaction time at 120 • C [136]. The catalyst was reused ten times, and yet the activity remained at >91%. A similar performance was reported by Liu et al. [137], but the selectivity to diacetin (51.4%) was the main product after 0.5 h at 100 • C and 0.5 wt% catalyst load. Similarly, Keogh et al. [138] reported a range of nitrogen-based Brønsted-acidic ionic liquids (alkyl-pyrrolidone and alkylamine cations) with reasonable activity. N-methyl-2-pyrrolidinium hydrogen sulphate [H-NMP][HSO 4 ] was found to be the most active and cost-effective catalyst with >99% glycerol conversion and 42.3% selectivity towards triacetin, and >95% of combined di-and triacetin selectivity within 1 h. Non-porous materials such as hydroxylated magnesium fluoride were deployed in glycerol acetylation with acetic acid (1:3) at 100 • C, and the result indicates >90% glycerol conversion with diacetin as the major product (≈60%) followed by triacetin (≈30%) after 22 h reaction time, unlike the uncatalyzed reaction that produced only 30% glycerol conversion with monoacetin as the major product (>90%). However, the reaction time was reduced from 4 h to 0.5 h when non-conventional activation methods, namely microwave or ultrasound irradiation, were used [139]. The authors attributed the excellent performance of the catalyst to the density of acid sites on the external surface of the catalyst, while the selectivity was influenced by the nature of acid sites (Lewis and Brønsted). Findings from the above catalytic studies showed that an increase in catalyst loading to a certain level (amount) improved glycerol conversion, especially to diacetin and triacetin, due to the increased number of available and accessible active acid site in the reaction mixture. Additional findings also revealed that the presence of the Bronsted acid site on the catalyst promotes optimal glycerol acetylation. The effect on monoacetin production is minimal, and this can be seen in the pictorial representation in Figure 9. The performance of some selected catalysts in recent studies is summarized in Table 2.
representation in Figure 9. The performance of some selected catalysts in recent studies is summarized in Table 2.

Temperature
Temperature is an important factor that influences the equilibrium of the reaction, which also has a significant impact on the rate of reaction, glycerol conversion, and product selectivity. It is generally reported that the acetylation of glycerol with acetic acid is mainly an endothermic reaction process, and hence requires a large amount of heat [61]. However, since the reaction is catalyst driven, care is taken to avoid the deactivation of thermally unstable catalysts by using moderate temperatures [80]. A good number of researchers have reported the influence of temperature in glycerol acetylation, and their findings are reviewed below.
Ghoreishi and Yarmo [110] varied the temperature from 50 °C to 110 °C in the esterification of glycerol with acetic acid in a stirred batch reactor equipped with a reflux system, and the results showed a slight increase in glycerol conversion with an increase in temperature. The authors further reported that diacetin and triacetin were the major products at high temperature while monoacetin was the predominant product at a lower temperature. In a similar study, using a carbon-synthesized catalyst from crude glycerol, Okoye et al. [114] reported that at 100 °C, the glycerol conversion was almost 100% with the corresponding selectivity of 50.97%, 45.98%, and 3% towards mono-, di-, and triacetin, respectively. However, when the temperature increased to 105,110,115, and 120 °C, the monoacetin selectivity decreased, while there was a corresponding increase in diacetin at 105 and 110 °C, respectively. However, at 115 and 120 °C, the triacetin selectivity increased while diacetin and monoacetin continued to decline. The highest combined selectivity of the di and triacetin of 88% was obtained at 110 °C, with a 2 wt% catalyst load, the glycerol-to-acetic acid molar ratio of 1:3, and in 3 h. The authors explained that the increased temperature enhanced the endothermic process, which further promoted the deprotonation of the second and third hydroxyl groups of glycerol over the active sites of the catalyst, hence promoting the formation of di-and triacetin. Similar results were reported by Carvalho et al. [75]. The di-and triacetin selectivity of 54 and 16% were obtained when the temperature was increased up to 120 °C over a sulfonated carbonized rice husk catalyst. However, the maximum glycerol conversion was achieved at 150 °C and subsequently decreased when the

Temperature
Temperature is an important factor that influences the equilibrium of the reaction, which also has a significant impact on the rate of reaction, glycerol conversion, and product selectivity. It is generally reported that the acetylation of glycerol with acetic acid is mainly an endothermic reaction process, and hence requires a large amount of heat [61]. However, since the reaction is catalyst driven, care is taken to avoid the deactivation of thermally unstable catalysts by using moderate temperatures [80]. A good number of researchers have reported the influence of temperature in glycerol acetylation, and their findings are reviewed below.
Ghoreishi and Yarmo [110] varied the temperature from 50 • C to 110 • C in the esterification of glycerol with acetic acid in a stirred batch reactor equipped with a reflux system, and the results showed a slight increase in glycerol conversion with an increase in temperature. The authors further reported that diacetin and triacetin were the major products at high temperature while monoacetin was the predominant product at a lower temperature. In a similar study, using a carbon-synthesized catalyst from crude glycerol, Okoye et al. [114] reported that at 100 • C, the glycerol conversion was almost 100% with the corresponding selectivity of 50.97%, 45.98%, and 3% towards mono-, di-, and triacetin, respectively. However, when the temperature increased to 105, 110, 115, and 120 • C, the monoacetin selectivity decreased, while there was a corresponding increase in diacetin at 105 and 110 • C, respectively. However, at 115 and 120 • C, the triacetin selectivity increased while diacetin and monoacetin continued to decline. The highest combined selectivity of the di and triacetin of 88% was obtained at 110 • C, with a 2 wt% catalyst load, the glycerol-to-acetic acid molar ratio of 1:3, and in 3 h. The authors explained that the increased temperature enhanced the endothermic process, which further promoted the deprotonation of the second and third hydroxyl groups of glycerol over the active sites of the catalyst, hence promoting the formation of di-and triacetin. Similar results were reported by Carvalho et al. [75]. The di-and triacetin selectivity of 54 and 16% were obtained when the temperature was increased up to 120 • C over a sulfonated carbonized rice husk catalyst. However, the maximum glycerol conversion was achieved at 150 • C and subsequently decreased when the temperature was further increased to 180 • C. This observation was attributed to the non-availability of the acetic acid > 150 • C due to vaporization. It was concluded that the suitable temperature to improve the esterification of glycerol with acetic acid to di-and triacetin was 120 • C. Testa et al. [92] reported that at 60 • C, the glycerol conversion was 44% in 3 h with monoacetin as the main product. However, when the temperature was increased to 105 • C, the glycerol conversion was complete with 33% triacetin after 3 h of the reaction. Further increase in the temperature up to 120 • C did not change both the glycerol conversion and selectivity towards triacetin, contrary to the findings of Carvalho et al. [75] as indicated above, which may be due to the catalyst used (sulfonic acid-functionalized amorphous silica (SAS) and mesoporous silica (SSBA)). Konwar et al. [101] reported the effect of temperature on the esterification of glycerol with lauric acid and oleic acid. Results revealed that as the temperature increased from 100 to 150 • C, the rate of esterification increased for both agents with decreased selectivity towards monoacetin and increased selectivity towards di-and triacetin due to a secondary reaction. By optimizing the esterification reaction of glycerol with acetic acid, the influence of temperature was investigated by Liao et al. [80] over the amberlyst-35 catalyst. The temperature range investigated was from 95 to 115 • C at a fixed acetic acid to glycerol molar ratio of 6, with a catalyst weight of 0.5 g. Almost 100% glycerol conversion was achieved after 4 h. The selectivity to monoand diacetin decreased (55.5 to 47.3% and 28.0 to 24.3%) while triacetin increased (16.5 to 28.3%) with an increase in temperature. This performance was attributed to a consecutive esterification reaction. Similarly, Mufrodi et al. [140] reported an increased glycerol conversion with increased temperature (100-120 • C) over a sulphuric acid catalyst, however, above 118 • C, the conversion reduced due to the evaporation of the acetylating agent (acetic acid). The increased temperature also increased the selectivity to mono-, di-, and triacetin. The authors further reported that for every 5 • C rise in temperature, the triacetin selectivity increased. The increase from 1.96% to a maximum of 13.69% at 115 • C was recorded. However, a further increase in the temperature reduced triacetin to 12.93%, which was attributed to the evaporation of acetic acid. The use of iron oxide supported on mesoporous aluminosilicate (Fe/Al-SBA-15) catalyst at 100 • C did not show any glycerol conversion even after 8 h of reaction until at 120 • C when the quantitative conversion of glycerol was achieved (>99%) with a selectivity of 71% diacetin and 28% triacetin. Further increase in temperature to 140 • C led to a slight improvement of diacetin (80%), and a reduction of triacetin (20%) which was attributed to the low thermal stability of the triacetin. However, the acetylating agent used was levulinic acid (1:4) in a continuous stirring (1200 rpm) reactor. The formation of diacetin as the major product could also be explained based on the lower activity of the secondary hydroxyl group, thereby preventing further reaction to form triacetin. Meanwhile, monoacetin was not observed in the product due to the equal activity of the two terminal hydroxyl groups. In another study, Khayoon et al. [82] also reported a substantial increase in glycerol conversion from 65 to 100% with increased temperature from 90 to 110 • C. Further increase in temperature did not improve the conversion. The selectivity to triacetin was found to depend on temperature. At lower temperatures (<90 • C), only about 7% of triacetin was produced, but when it was increased to 110 • C the triacetin increased to 55%. Further increase in temperature to 120 • C does not affect the triacetin selectivity. Similar findings have been reported by Goncalves et al. [141] and Dosuna-Rodriguez et al. [142]. However, contrary to the above, Rafi et al. [125] revealed that increasing the temperature from 80 to 120 • C increased the glycerol conversion from 25 to 85%. However, further increase to 140 • C did not show any appreciable enhancement. The selectivity to mono-, di-, and triacetin followed the same pattern as reported previously. Di-and triacetin increased continuously with an increase in temperature, while monoacetin decreased, and this observation was related to the increased activity of the biochar. However, the use of lauric acid in glycerol esterification over layered double hydroxide (LDH, Mg-Al-CO 3 ) catalyst accommodated the use of high temperatures. At 100 • C, only 25% of glycerol conversion was achieved in 2 h, but when the temperature was increased to 180 • C, glycerol conversion increased to 99%. Nevertheless, the selectivity towards monoacetin (>50%) was largely favoured [143]. Sandesh et al. [89] also investigated glycerol esterification at lower temperatures (65-95 • C) over the CsPWA catalyst. Results indicate that in 2 h, molar ratio of 1:8, and temperature of 65 • C, the glycerol conversion was low (65%). However, as temperature increased to 75 • C, glycerol conversion increased to 92%. While at 85 • C, the conversion reached 98%, which appears to be the maximum because no further improvement was observed even when the temperature was further increased to 95 • C. The selectivity of di-and triacetin also improved with increased temperature while the monoacetin decreased substantially. The authors attributed the low performance of the reaction at low temperature to a low-level formation of acylium ion by the acetic acid, and as the temperature increased, the acylium ion formation also improved leading to high performance. A similar finding was reported using the same type of catalyst but with a temperature of 110 • C as the most desirable [121]. The influence of temperature in glycerol acetylation has been earlier reported in several literature articles [102,133,144,145], and the results obtained were similar to the findings reported above.
The findings from various researchers are in line with the Arrhenius equation (Equation (1)), which establishes a relationship between the temperature, activation energy, and rate of reaction. Increased temperature directly increased the number of collisions of the reactant, which is directly proportional to the rate of reaction [60,146]. The increased temperature promotes rapid glycerol conversion, with a corresponding increase in diacetin and triacetin selectivity at the expense of monoacetin. This observation is exemplified in the pictorial representation in Figure 10: where K is the reaction rate constant, A is the Arrhenius constant related to the frequency factor, Ea is the activation energy, R is the gas constant, and T is the temperature.
Appl. Sci. 2020, 10, x FOR PEER REVIEW 19 of 33 conversion increased to 92%. While at 85 °C, the conversion reached 98%, which appears to be the maximum because no further improvement was observed even when the temperature was further increased to 95 °C. The selectivity of di-and triacetin also improved with increased temperature while the monoacetin decreased substantially. The authors attributed the low performance of the reaction at low temperature to a low-level formation of acylium ion by the acetic acid, and as the temperature increased, the acylium ion formation also improved leading to high performance. A similar finding was reported using the same type of catalyst but with a temperature of 110 °C as the most desirable [121]. The influence of temperature in glycerol acetylation has been earlier reported in several literature articles [102,133,144,145], and the results obtained were similar to the findings reported above.
The findings from various researchers are in line with the Arrhenius equation (Equation (1)), which establishes a relationship between the temperature, activation energy, and rate of reaction. Increased temperature directly increased the number of collisions of the reactant, which is directly proportional to the rate of reaction [60,146]. The increased temperature promotes rapid glycerol conversion, with a corresponding increase in diacetin and triacetin selectivity at the expense of monoacetin. This observation is exemplified in the pictorial representation in Figure 10: where K is the reaction rate constant, A is the Arrhenius constant related to the frequency factor, Ea is the activation energy, R is the gas constant, and T is the temperature.

Reaction Time
For any reaction to reach equilibrium, the period of contact of the reactants with the catalyst plays an important role. Reports in the literature also showed the influence of time in glycerol acetylation. Ghoreishi and Yarmo [110] reported that in glycerol acetylation with acetic acid at a molar ratio of 1:3, 50 °C, and 200 mg catalyst load (10% sulphated silica), monoacetin was favoured at a lower reaction time (2 h). However, higher acetins were favoured as the reaction proceeds with time, attributable to the continuous conversion of monoacetin to diacetin and triacetin. The glycerol conversion increased from about 60% in 2 h to almost 100% in 6 h. The effect of time in glycerol

Reaction Time
For any reaction to reach equilibrium, the period of contact of the reactants with the catalyst plays an important role. Reports in the literature also showed the influence of time in glycerol acetylation. Ghoreishi and Yarmo [110] reported that in glycerol acetylation with acetic acid at a molar ratio of 1:3, 50 • C, and 200 mg catalyst load (10% sulphated silica), monoacetin was favoured at a lower reaction time (2 h). However, higher acetins were favoured as the reaction proceeds with time, attributable to the continuous conversion of monoacetin to diacetin and triacetin. The glycerol conversion increased from about 60% in 2 h to almost 100% in 6 h. The effect of time in glycerol conversion was also reported by Okoye et al. [114] over glycerol-based carbon catalyst and concluded that the glycerol conversion increased with an increase in reaction time from 1 to 3 h. In 1 h, about 70% glycerol conversion was achieved with high monoacetin selectivity, but in 3 h, the glycerol conversion increased to almost 100% with higher di-and triacetin selectivity but low monoacetin selectivity. Chandrakala et al. [126] also reported the effect of the reaction time on the esterification of glycerol with acetic acid (1:4) in the presence of 5 wt % glycerol-based carbon catalyst at 115 • C for various time of 1 to 4 h. Results indicate a complete glycerol conversion within 1 h with a selectivity of 22% monoacetin, 67% diacetin, and 11% triacetin, respectively. When the reaction time was increased up to 4 h, the triacetin increased to 25%, with a corresponding slight decrease in mono-and diacetin. Similar findings were reported by Khayoon et al. [82] over the 3% yttrium on mesoporous silicate material support (3%/SBA-3), where the glycerol conversion was 62% after 0.5 h, and the conversion was completed after 1 h. The selectivity to monoacetin decreased, while the diacetin increased with time. Additional findings showed that triacetin commenced after 0.5 h into the reaction, which was attributed to the direct acetylation of monoacetin, and after 1.5 h, some of the produced diacetin also transformed into triacetin, hence the observed decrease in diacetin. The authors concluded that the complete glycerol conversion with combined selectivity of di-and triacetin of 89% was achieved. The biochar catalyst also showed excellent activity with the increased reaction time to a maximum of 4 h, which in turn improved the selectivity towards di-and triacetin with a concurrent decrease in the monoacetin selectivity [125]. The use of sulfonated carbonized willow catkins by Tao et al. [113] also showed a similar trend. The glycerol conversion increased rapidly to 100% with the increased reaction time from 0.25 to 2 h. The extension of time up to 7 h led to a slight decrease in glycerol conversion, which the authors attributed to the drop in glycerol concentration and the accumulation of water as the by-product. However, the selectivity to higher acetin improved significantly with time. In 7 h, the combined selectivity of di-and triacetin reached 95.3%. The catalyst performance was attributed to the acid property and the nature of the surface arisen from the sulfonation, which might have provided good access to the active sites for the hydrophilic glycerol. Similar findings have been reported by several researchers despite the use of different catalysts [89,102,133,144]. Figure 11 shows the pictorial representation of some selected findings of the influence of reaction time on glycerol conversation and selectivity towards mono-, di-and triacetin as reported in some recent research articles. The reviewed literature showed a linear relationship between reaction time and increased acetin formation. Increased time favours higher selectivity towards diacetin and triacetin, respectively.
Appl. Sci. 2020, 10, x FOR PEER REVIEW 20 of 33 conversion was also reported by Okoye et al. [114] over glycerol-based carbon catalyst and concluded that the glycerol conversion increased with an increase in reaction time from 1 to 3 h. In 1 h, about 70% glycerol conversion was achieved with high monoacetin selectivity, but in 3 h, the glycerol conversion increased to almost 100% with higher di-and triacetin selectivity but low monoacetin selectivity. Chandrakala et al. [126] also reported the effect of the reaction time on the esterification of glycerol with acetic acid (1:4) in the presence of 5 wt % glycerol-based carbon catalyst at 115 °C for various time of 1 to 4 h. Results indicate a complete glycerol conversion within 1 h with a selectivity of 22% monoacetin, 67% diacetin, and 11% triacetin, respectively. When the reaction time was increased up to 4 h, the triacetin increased to 25%, with a corresponding slight decrease in mono-and diacetin. Similar findings were reported by Khayoon et al. [82] over the 3% yttrium on mesoporous silicate material support (3%/SBA-3), where the glycerol conversion was 62% after 0.5 h, and the conversion was completed after 1 h. The selectivity to monoacetin decreased, while the diacetin increased with time. Additional findings showed that triacetin commenced after 0.5 h into the reaction, which was attributed to the direct acetylation of monoacetin, and after 1.5 h, some of the produced diacetin also transformed into triacetin, hence the observed decrease in diacetin. The authors concluded that the complete glycerol conversion with combined selectivity of di-and triacetin of 89% was achieved. The biochar catalyst also showed excellent activity with the increased reaction time to a maximum of 4 h, which in turn improved the selectivity towards di-and triacetin with a concurrent decrease in the monoacetin selectivity [125]. The use of sulfonated carbonized willow catkins by Tao et al. [113] also showed a similar trend. The glycerol conversion increased rapidly to 100% with the increased reaction time from 0.25 to 2 h. The extension of time up to 7 h led to a slight decrease in glycerol conversion, which the authors attributed to the drop in glycerol concentration and the accumulation of water as the by-product. However, the selectivity to higher acetin improved significantly with time. In 7 h, the combined selectivity of di-and triacetin reached 95.3%. The catalyst performance was attributed to the acid property and the nature of the surface arisen from the sulfonation, which might have provided good access to the active sites for the hydrophilic glycerol. Similar findings have been reported by several researchers despite the use of different catalysts [89,102,133,144]. Figure 11 shows the pictorial representation of some selected findings of the influence of reaction time on glycerol conversation and selectivity towards mono-, diand triacetin as reported in some recent research articles. The reviewed literature showed a linear relationship between reaction time and increased acetin formation. Increased time favours higher selectivity towards diacetin and triacetin, respectively.

The Reactants and Their Molar Ratios
The most common reactants used in the conversion of glycerol to acetin include acetic acid and acetic anhydride [126,134,135]. However, it has been reported that acetic acid is, thermodynamically, resistant to acetylation due to the formation of a tetrahedral intermediate. The formation of water (in situ), may also deactivate the catalyst and reverse the reaction [41,110]. Nevertheless, acetic acid is still preferred because it is cheaply available and less toxic despite the positive Gibbs free energy exhibited. Acetylation with acetic anhydride proceeds quickly to the product, attributed to the ease of formation of acylium ion [80]. However, the use of acetic anhydride is of grave concern because of its application in the production of narcotics. Currently, it is banned in some countries [114]. It was also reported that the rate of reaction of glycerol with acetic anhydride is much higher than with acetic acid, and because of its speed, it is much more difficult to handle [67,80]. Several journal articles also reported the use of other carboxylic acids, such as oleic, levulinic, lauric, propionic, and butanoic acids in acetylation [58,84,143,148]. Most of these carboxylic acids have shown good glycerol conversion and selectivity to monoacetin but limit the selectivity to higher acetins (diacetin and triacetin). The limitation was attributed to their bulky structure. The performance of these acetylating agents in the reaction can also be influenced by their molar ratios. In terms of the stoichiometric equation, 3 mol of acetic acid is needed to react completely with 1 mol of glycerol to produce monoacetin and subsequently to diacetin and triacetin as the final product [60]. However, excess acetic acid shifts the equilibrium towards higher diacetin and triacetin, respectively.
An increase in the molar ratio of acetic acid to glycerol led to an increase in glycerol conversion [110]. In addition, 96.88% glycerol conversion and selectivity of 51.90, 45.27, and 2.11% monoacetin, diacetin, and triacetin were obtained at an acetic acid to glycerol molar ratio of 6 over a 20% sulphated silica catalyst. When the molar ratio of 3 was used over the same catalyst, the glycerol conversion reduced to 90.12% with a selectivity of 58.83, 37.10, and 3.24% monoacetin, diacetin, and triacetin, respectively. However, when 10% sulphated silica catalyst was used, monoacetin selectivity was favoured with 81.12%, while diacetin was 18.32%, and only traces of triacetin were observed at a molar ratio of 3. The difference in the catalyst performance was attributed to the acid density and the availability of the acetylating agent. That is, a catalyst with a higher acid density favours higher acetin (di-and triacetin) formation. A similar observation was reported with a crude glycerol-derived carbon catalyst [114]. At an acetic acid to glycerol molar ratio of 1, over 70% of monoacetin was obtained with 55% glycerol conversion. However, when the molar ratio was varied from 3 to 9, glycerol conversion increased to 100% with a high production of di-and triacetin (46%, 43%) and low production of monoacetin (11%). When the molar ratio was > 4, the triacetin production decreased, and it was attributed to the likely effect of excess acetic acid leading to mass transfer resistance. This resistance limits the sufficient interaction or contact between the glycerol and the catalyst. Isahak et al. [13] reported the esterification of glycerol with oleic acid over a mixture of silicotungstic acid-ionic liquid catalyst at 100 • C for 8 h. High glycerol conversion (96.4%) was achieved with 96% selectivity towards monoacetin. Only 4% of diacetin was reported. A good yield of glycerol ester was also reported by Sari et al. [148] using the same reactants (glycerol and oleic acid) but with a different catalyst made of methyl ester sulfonic acid. Liao et al. [80] also reported the influence of acetic acid to glycerol molar ratio (2:1, 3:1, 6:1, 9:1) at fixed conditions of 105 • C, 0.5 g catalyst load and in 4 h over amberlyst-35 catalyst. The results indicate a higher conversion of glycerol and a higher selectivity towards triacetin with an increased amount of acetic acid. This observation was attributed to the improvement in the acid strength with the availability of more acetic acid, which subsequently improved the catalytic activity. In the work of Mufrodi et al. [140], it was reported that based on the stoichiometric calculation, three moles of acetic acid requires one mole glycerol to produce one mole of triacetin. However, in the experimental work, the increase in the molar ratio of acetic acid to glycerol (3, 4, 5, and 6) increased the glycerol conversion by 0.29% for each addition of 1 mol acetic acid. The maximum glycerol conversion of 98.50% was achieved. The increase in molar ratio also affected the selectivity as indicated by the decreased monoacetin and increased di-and triacetin, respectively. The influence of the acetic acid to glycerol molar ratio in esterification was also evaluated by Carvalho et al. [75] over a sulfonated carbonized rice husk catalyst. The results indicate an increase in the formation of di-and triacetin with a simultaneous decrease in monoacetin as the molar ratio increases. The authors opined that the observation was due to the continuous conversion of monoacetin to the higher acetins; however, the glycerol conversion was not significantly affected until at a molar ratio of 5. The glycerol conversion was 90%, and a further increase in the molar ratio to 9 did not significantly improve the conversion. However, the selectivity to di-and triacetin improved greatly, but for economic and environmental reasons, the authors felt that there was no need to use a high volume of acetic acid given the amount of products produced. Similarly, a report by Chandrakala et al. [126] showed a marginal improvement of diacetin from 67 to 73% and triacetin from 11 to 20% when the molar ratio of acetic acid to glycerol was varied from 4:1 to 10:1 over 5 wt% of a glycerol-based carbon catalyst at 115 • C for 1 h. However, when the reaction time was increased to 4 h, the triacetin was increased to 40% at a molar ratio of 10:1. This finding is different from the findings of Rafi et al. [125], who used unfunctionalized biochar as a catalyst. About 75% glycerol conversion was obtained at a mole ratio of 3:1. The conversion increased up to 89% when the molar ratio was varied to 5:1 but at a higher temperature of 120 • C. The improvement was also observed in the combined di-and triacetin selectivity (60%). Further variation in the molar ratio did not produce any appreciable change in the glycerol conversion. The difference in the above performance may be due to the nature of the catalyst and the different operating temperatures used. Similarly, the higher conversion of glycerol (99%) was obtained when glycerol was esterified with lauric acid (3:1) but at a higher temperature of 180 • C in 2 h over layered double hydroxide (LDH, Mg-Al-CO 3 ) catalyst with a selectivity of approximately 58, 30 and 12% for mono-, di-and tri-acetins, respectively [143].
The use of different carboxylic acids, namely acetic acid, propionic acid, and 1-butanoic acid as esterification agents in glycerol acetylation over silver-exchanged phosphotungstic acid (Ag 1 PW) catalyst (1 wt% catalyst load) at 120 • C, with a 1:10 molar ratio (glycerol to carboxylic acids), was reported by Zhu et al. [84]. In less than 20 min, acetic acid showed 96.8% glycerol conversion with 48.4, 46.4, and 5.2% selectivity towards mono-, di-, and triacetin, propionic acid achieved 70.9% glycerol conversion with 55, 43.1, and 1.9% selectivity towards mono-, di-and triacetin, respectively, while 1-butanoic acid produced 64.3% glycerol conversion with 53.9, 46 and 0.1% selectivity towards mono-, di-, and triacetin, respectively. The very low selectivity towards triacetin could be attributed to steric hindrance arising from the structures of the carboxylic acids (acetic acid, propionic acid, and 1-butanoic acid) because the trend indicates decreased glycerol esterification with increasing chain length. The influence of the acetic acid to glycerol molar ratio (4:1, 6:1, and 8:1) was investigated at a fixed reaction temperature of 110 • C over 3% Y/SBA-3 by Khayoon et al. [82]. The best result was obtained with a molar ratio of 4:1, which resulted in the complete conversion of glycerol with the corresponding selectivity of 11% monoacetin, 34% diacetin, and 55% triacetin, respectively. The good selectivity towards triacetin at a 4:1 molar ratio was also attributed to the excellent textural characteristics, the crystal phase stability of the catalyst, and the availability of the acetylating agent. It was a general expectation that excess acetylating agents could shorten the time required to reach the reaction equilibrium and enhance the formation of higher acetins due to consecutive acetylation reactions, but that was not the case. The authors attributed it to the hydrolysis effect of the co-produced water. Contrary to the above finding, Kotbagi et al. [132] showed higher glycerol conversion at a higher molar ratio of acetic acid to glycerol. The molar ratio of 3, 5, and 10 resulted in a glycerol conversion of 67, 91, and 100%, respectively, after 8 h of reaction using a supported silicomolybdic heteropolyanions catalyst at 10 wt% with respect to glycerol. The selectivity to mono-, di-, and triacetin also varies with the molar ratio. At a molar ratio of 3, monoacetin was favoured with 81% selectivity, while the selectivity to di and triacetin was 16 and 3%, respectively. However, at a molar ratio of 10, triacetin was favoured with 50% selectivity, followed by diacetin with 33% while the monoacetin decreased to 17%. This result was similar to the earlier work reported by Chandrakala et al. [126], confirming the conversion of monoacetin to the higher acetins (di-and triacetin) in line with the earlier stated mechanism of the reaction. Sandesh et al. [89] also reported that the molar ratio of acetic acid-to-glycerol of 8 produced the highest glycerol conversion of 98% with the CsPWA catalyst. When a molar ratio of 10 was used, no noticeable change in the conversion was reported. The selectivity of the desired product also followed the same pattern. The molar ratio of 4 and 6 did not produce much of di-(35% each) and triacetin (0 and 6%) until at a molar ratio of 8 (59% diacetin and 16% triacetin), which showed that high quantity of acetic acid increased its accessibility to glycerol. This finding was corroborated by Veluturla et al. [121] with a maximum glycerol conversion of 98.2% at an acetic acid to glycerol molar ratio of 9 over a similar catalyst. Several literature articles previously reported a similar pattern of increased glycerol conversion with a variant molar ratio [34,83,102,133,145]. Figure 12 shows the pictorial representation of some selected findings of the influence of molar ratios of the reactants on the glycerol conversion and selectivity towards mono-, di-, and triacetin as reported by some recent studies.
Appl. Sci. 2020, 10, x FOR PEER REVIEW 23 of 33 that the molar ratio of acetic acid-to-glycerol of 8 produced the highest glycerol conversion of 98% with the CsPWA catalyst. When a molar ratio of 10 was used, no noticeable change in the conversion was reported. The selectivity of the desired product also followed the same pattern. The molar ratio of 4 and 6 did not produce much of di-(35% each) and triacetin (0 and 6%) until at a molar ratio of 8 (59% diacetin and 16% triacetin), which showed that high quantity of acetic acid increased its accessibility to glycerol. This finding was corroborated by Veluturla et al. [121] with a maximum glycerol conversion of 98.2% at an acetic acid to glycerol molar ratio of 9 over a similar catalyst. Several literature articles previously reported a similar pattern of increased glycerol conversion with a variant molar ratio [34,83,102,133,145]. Figure 12 shows the pictorial representation of some selected findings of the influence of molar ratios of the reactants on the glycerol conversion and selectivity towards mono-, di-, and triacetin as reported by some recent studies.

Conclusions
From the above review, it is evident that there is an increased interest in glycerol acetylation to acetin, which may not be unconnected with the surplus production of glycerol and the drive to improve the economics of biodiesel production. It may also be attributed to the high-value application of acetin in the cosmetic, pharmaceutical, medical, food, and polymer industries as a humectant, emulsifier, plasticizer, and most importantly, as a biofuel additive to improve the viscosity and cold flow properties of gasoline, diesel, and bio-diesel. This review paper summarizes the different heterogeneous catalysts deployed in acetylation reaction. This review indicates that apart from the popularly used acetylating agents (acetic acid and acetic anhydride), other agents include oleic, levulinic, lauric, propionic, and butanoic acids. This paper also clearly indicates that the nature of the catalyst, catalyst loading, temperature, reactants molar ratio, and time of the reaction influences the glycerol conversion to mono-, di-, and triacetin with the temperature and reactants molar ratio been the most influential.
Finally, there is the need for continual research in this area to identify new inexpensive, reusable, and environmentally friendly catalytic materials and milder conditions of reaction to favour high selectivity to triacetin. The need for further studies on catalyst deactivation and the kinetic of acetylation is recommended to facilitate the understanding of the process for industrial applications.