Abstract
Fructose, glucose, and sucrose were converted to organic acids in the presence of aqueous aluminum nitrate (Al(NO3)3) to develop a technically viable route for upgrading sugarcane molasses. Reactions were carried out in a microwave reactor and a muffle oven with conventional heating (convective heat transfer) using a sealed glass tubes and a hydrothermal stainless-steel autoclave as reaction vessels, respectively. Conversion was evaluated for different reaction times and temperatures. Lactic acid predominated as the product from the retro-aldol chain splitting of fructose, reaching a 67.5 % molar yield using 2.67 mmol·L-1 Al(NO3)3 (4 wt% based on the carbohydrate dry mass) in a stainless-steel reactor with conventional heating. Sucrose required hydrolysis, glucose isomerization, retro-aldol chain splitting, dehydration, tautomerization, and 1,2-H migration to produce lactic acid in molar yields approaching those obtained from fructose (65.5 %). Besides lactic acid, formic and levulinic acids were produced in variable amounts through a fructose dehydration pathway, having 5-(hydroxymethyl)-furfural (HMF) as reaction intermediate. The use of a stepwise heating regime was a critical parameter to achieve high product yields and good lactic acid selectivity in these reaction systems.
Download PDF
Full Article
Conversion of Carbohydrates to Organic Acids in the presence of Aluminum Nitrate in Aqueous Medium
Priscila D. Fernandes 
,a Rafael D. Lima ,a Gustavo R. Gomes ,b Daniel S. Rampon ,c and Luiz P. Ramos ,a,d,*
Fructose, glucose, and sucrose were converted to organic acids in the presence of aqueous aluminum nitrate (Al(NO3)3) to develop a technically viable route for upgrading sugarcane molasses. Reactions were carried out in a microwave reactor and a muffle oven with conventional heating (convective heat transfer) using a sealed glass tubes and a hydrothermal stainless-steel autoclave as reaction vessels, respectively. Conversion was evaluated for different reaction times and temperatures. Lactic acid predominated as the product from the retro-aldol chain splitting of fructose, reaching a 67.5 % molar yield using 2.67 mmol·L-1 Al(NO3)3 (4 wt% based on the carbohydrate dry mass) in a stainless-steel reactor with conventional heating. Sucrose required hydrolysis, glucose isomerization, retro-aldol chain splitting, dehydration, tautomerization, and 1,2-H migration to produce lactic acid in molar yields approaching those obtained from fructose (65.5 %). Besides lactic acid, formic and levulinic acids were produced in variable amounts through a fructose dehydration pathway, having 5-(hydroxymethyl)-furfural (HMF) as reaction intermediate. The use of a stepwise heating regime was a critical parameter to achieve high product yields and good lactic acid selectivity in these reaction systems.
DOI: 10.15376/biores.20.1.1037-1058
Keywords: Water-soluble sugars; Acid catalysis; Dehydration; Retro-aldol; Lactic acid
Contact information: a: Research Center in Applied Chemistry (CEPESQ), Department of Chemistry, Federal University of Paraná – UFPR, P. O. Box 19032, Curitiba, PR, 81531-980, Brazil; b: Institute of Chemistry, State University of Campinas – UNICAMP, P. O. Box 6154, Campinas, SP, 13083-970, Brazil; c: Laboratory of Polymers and Catalysis (LAPOCA), Department of Chemistry, Federal University of Paraná – UFPR, P. O. Box 19032, Curitiba, PR, 81531-980, Brazil; d: Graduate Program in Chemistry, Federal University of Paraná – UFPR, P. O. Box 19032, Curitiba, PR, 81531-980, Brazil
*Corresponding author: luiz.ramos@ufpr.br
GRAPHICAL ABSTRACT
INTRODUCTION
The biorefinery converts renewable feedstocks to biofuels, bioenergy, sustainable platform chemicals, and value-added biobased materials, and this can play a major role in sustainable process development (Takkellapati et al. 2018). Biorefineries have two strategic goals: the energy goal of replacing petroleum derivatives in favor of widely available sustainable (preferably renewable) raw materials, and the economic goal of establishing a strong biobased industry to support the upcoming transition to a low-carbon bioeconomy (Bozell and Petersen 2010; Igbokwe et al. 2022). Among a range of valuable biorefinery products, organic acids represent an important market share of technically viable carbohydrate derivatives. Nearly two-thirds of biobased building blocks are made of organic acids, showing the importance of these chemicals to improve the sustainability footprint of our chemical industry (Serrano-Ruiz et al. 2010).
Lactic and levulinic acids are some of the biomass-derived chemicals with the greatest economic potential (Dutta and Bhat 2021; Li et al. 2022). Lactic acid has received great attention due to its use in the synthesis of poly(lactic acid) (PLA), an industrially compostable and biocompatible high molecular mass biopolymer that plays an important role in replacing plastics derived from the petrochemical industry (Djukić-Vuković et al. 2019; Megías-Sayago et al. 2021). Also, lactic acid can be used in other areas for environmental, ecological, medical, and pharmaceutical applications. For instance, in the latter case, the (S) isomer can be used to produce cosmetics, ointments, anti-acne solutions, humectants, and controlled-release medications (Hofvendahl and Hahn-Hägerdal 2000; Ramot et al. 2016; Ojo and de Smidt 2023). Levulinic acid can be used as a precursor to produce herbicides such as aminolevulinic acid, pesticides, pharmaceuticals, green solvents, plasticizers, cosmetics, and intermediates (Bozell and Petersen 2010; Bazoti et al. 2023).
Sustainable conversion processes preferably based on low-price, widely available catalytic systems have been sought to produce organic acids such as lactic and levulinic acids from water-soluble carbohydrates. For instance, lactic, levulinic, and formic acids have been used as sources to produce glucose, fructose, and sucrose using different homogeneous and heterogeneous Brønsted-Lowry and Lewis acid catalytic systems (Table 1).
Table 1. Examples of Brønsted-Lowry and Lewis Acid Catalysts for the Conversion of Carbohydrates to Levulinic and Lactic Acids
Sucrose is extracted from either sugarcane or sugar beet and is available all year-round in the form of molasses, particularly in tropical countries, while glucose and fructose can be found in the form of starch (corn and other starchy syrups) and inulin hydrolysates (agave syrups), respectively. Cellulose may also be a valuable and widely available carbohydrate source for organic acid production, but its high recalcitrancy requires harsher reaction conditions to release glucose in high yields for catalytic conversion (Tang et al. 2014; Deng et al. 2018; Tallarico et al. 2019; Xu et al. 2020; Yan et al. 2023).
When sucrose is used as the starting reagent, the reaction begins by its hydrolysis to glucose and fructose (Fig. 1). Since hydrolysis depends on the protonation of the glycosidic oxygen, Brønsted-Lowry acids are normally used as catalysts. However, sucrose hydrolyses can take place under mild reaction conditions because fructose is a good leaving group due to its conformational instability. Next, glucose is isomerized to fructose, a step that is accelerated by the presence of a Lewis acid catalyst (Binder and Raines 2009). Then, fructose is converted to organic acids in two different reaction pathways: dehydration to 5-(hydroxymethyl)-furfural (HMF) followed by its rehydration to formic and levulinic acids (Bozell and Petersen 2010; Serrano-Ruiz et al. 2010; Osmundsen 2013; Gomes et al. 2017), and retro-aldol chain splitting forming glyceraldehyde and dihydroxyacetone (C3 building blocks) in tautomeric equilibrium, followed by glyceraldehyde dehydration to pyruvaldehyde and an 1,2-H migration (intramolecular Canizzaro reaction) to yield lactic acid (Orazov and Davis 2015; Albuquerque et al. 2017; Tang et al. 2018, 2019). This retro-aldol reaction pathway works best in the presence of a Lewis acid catalyst (Marianou et al. 2018; Hossain et al. 2021).
Fig. 1. Proposed steps for sucrose conversion to organic acids
In dehydration, water molecules are eliminated from fructose to form HMF in a mechanism that involves cyclic intermediates or ring opening reactions (Ståhlberg et al. 2010). HMF rehydration begins with the breakdown of the furan ring aromaticity and the formation of a carbocation (Osmundsen 2013; Zhang et al. 2015). Subsequently, successive steps involving the addition and elimination of water molecules occur until the structure is completely opened. In the final stage of this mechanism, formic acid is eliminated, and levulinic acid is formed in equimolar moieties. In total, two water molecules are incorporated to form levulinic and formic acids (Zhang et al. 2015). Such dehydration and rehydration reactions occur at high temperatures in aqueous media even in the absence of an acid catalyst. However, better yields and faster reaction kinetics are obtained in the presence of a strong Brønsted-Lowry acid catalyst (Osmundsen 2013; Zhang et al. 2015).
Metal salts of the MqAx type, after encountering water, dissociate and immediately hydrate, forming 
species plus hydronium ions ( 
). With this, Brønsted-Lowry acids are formed in situ, and their acidity depends on the strength of the anion. Meanwhile, aluminum halides such as AlCl3 dissociate in water forming hydrochloric acid (HCl, pKa = –6.3), Al2(SO4)3 and Al(NO3)3 release sulfuric (H2SO4, pKa1 = –3.0) and nitric acids (HNO3, pKa = –1.4), respectively. Hence, hexose dehydration depends on the Brønsted-Lowry acid strength released in situ. On the other hand, hydrolysis of the aluminum ion can generate numerous mononuclear (e.g., 
and polynuclear (e.g., 
species, whose formation depends on the metal concentration and reaction pH (Akitt 1989). At pH close to 3, hydrolysis of the aluminum ion mainly generates the mononuclear hexaaquo aluminum(III) species [Al(H2O)6]3+. As the pH rises, species such as [Al(OH)(H2O)5]2+, [Al(OH)2(H2O)4]+ and Al(OH)3 are formed, as well as polynuclear species (Baes and Mesmer 1976). However, Al(OH)3 flocculates because it is moderately soluble in water and, depending on the pH and Al3+ concentration, side reactions may occur, generating less stable polymeric and colloidal species (Martell and Motekaitis 1992). Among the abovementioned species, the ion has the greatest Lewis acid character, presenting a hydrolysis constant K equal to 1.1 (K1.1). The pK1.1 value of Al3+ varies between 4.9 and 5.5 (Richens 1997; Akitt 1989).
According to Fang et al. (2023), the Lewis acid strength is not the only factor influencing carbohydrate conversion to lactic acid, but also the formation of intermediate complexes with metal cations derived from homogeneous catalysts. It has been hypothesized that small cationic radii favor the binding capacity of metal ions and fructose hydroxyl groups, facilitating the cleavage of carbon bonds and splitting the molecule to form lactic acid.
Hossain et al. (2021) hypothesized that cationic hydroxyl-aluminum complexes formed in situ, with the general formula [Al(OH)h](3-h)+, are the actual catalytic species in reactions involving aluminum salts due to their strong Lewis acid character. For instance, [Al(OH)h](3-h)+ were critical for the conversion of dihydroxyacetone to lactic acid via pyruvaldehyde in aqueous medium.
In this work, fructose, glucose, and sucrose were converted to organic acids in the presence of Al(NO3)3 in aqueous media under both conventional (convective) and microwave heating systems. Both retro-aldol chain splitting and dehydration pathways were evaluated in the presence and absence of a catalyst or a catalyst precursor under different temperatures, reaction times, and heating regimes. The working hypothesis was that, compared to other aluminum-based Lewis acids such as AlCl3 and Al(NO3)3 would be more selective for lactic acid production through the retro-aldol chain splitting pathway. In addition, the use of microwave heating would accelerate the reaction kinetics without interfering with the reaction mechanism (Szabolcs et al. 2013).
EXPERIMENTAL
Materials
Fructose, glucose, sucrose, furfural (99 %), acetic acid (>99.7 %), levulinic acid (99 %), lactic acid (99 %), HMF (99 %), and Al(NO3)3∙9H2O were obtained from Sigma-Aldrich Brazil (Jurubatuba, SP, Brazil). Other reagents and organic solvents were purchased from local suppliers in analytical and chromatographic or spectrometric grade, respectively. All chemicals were used as received without any further treatment.
Methods
Thermal conversion
The thermal conversion of fructose (66.7 mmol∙L-1), glucose (66.7 mmol∙L-1), and sucrose (33.2 mmol∙L-1) was assessed in a CEM microwave reactor (North Carolina, US) using a 10 mL sealed glass tube and a series of 50 mL hydrothermal stainless-steel autoclaves with a removable inner polytetrafluoroethylene (PTFE) chamber (SS-PTFE reactor) that were placed inside a muffle oven. Reactions were performed at 200 °C for 120 min and product analyses were carried out by HPLC at 65 °C using H2SO4 8 mmol∙L-1 as the mobile phase as described below.
Effect of gradual heating on carbohydrate conversion using Al(NO3)3·9H2O
Dehydration of carbohydrates in aqueous media was carried out in the presence of 2.67 mmol·L-1 Al(NO3)3·9H2O as the catalyst precursor, which initially corresponded to 4 wt % based on the carbohydrate dry mass. Experiments were performed with fructose, glucose (both at 66.7 mmol∙L-1), and sucrose (33.2 mmol∙L-1) using the CEM microwave reactor (MWR) and the SS-PTFE reactor described above. In the latter situation, heating was based on convective heat transfer, and the temperature profile was determined outside and inside one of the SS-PTFE reactors using a thermocouple. The reaction time was 120 min in both systems. However, the MWR took only 2 min to reach the setpoint temperature of 200 °C, whereas the heating time for the SS-PTFE lasted approximately 90 min. For this reason, one additional experiment was carried out in which the temperature was increased manually in the MWR to simulate the heating ramp of the muffle oven. Once the proposed reaction time was reached, the reaction vessels were cooled down to ambient temperature and aliquots were withdrawn, diluted with ultrapure water to approach 3 mg∙mL-1 in relation to theoretical yields, filtered to pass a 0.45 m PTFE syringe filter, and analyzed by high performance liquid chromatography (HPLC). Reaction yields were calculated based on the stoichiometric amount of each product (see below for details). When the reaction was carried out in duplicates, the results were reported as the mean value with the corresponding experimental error.
Influence of time on reaction conversion in the presence of Al(NO3)3
The influence of reaction time on carbohydrate conversion was investigated in the SS-PTFE reactor and in sealed glass tubes heated by microwave irradiation using 2.67 mmol·L-1 of the catalyst precursor. Fructose, glucose, and sucrose concentrations were the same as used previously and all reactions were carried out at 180 °C for 5 to 90 min in the microwave oven. The SS-PTFE reaction system was only used for fructose and the reaction times ranged from 30 to 240 min.
Kinetic study
Kinetic studies were performed using the two first-order equations (Eqs. 1 and 2) to adjust the reaction profiles,
where Ct is the product concentration at a given reaction time, is the initial reagent concentration, Cinf is the infinite concentration, kobs is the observed rate constant for pseudo-first order reactions, and t is the reaction time. Equation 1 was used for reagent consumption, while Eq. 2 was used for products that behave as intermediates (e.g., HMF is produced by dehydration and later converted to formic and levulinic acids). For this reason, both CO and Cinf values are duplicated in Eq. 2, indicating two reaction stages: subscript 1 for product formation and subscript 2 for product consumption. Data were processed with the OriginPro 2018 software (version 95E).
Chromatographic analysis
Carbohydrate and organic acid analyses were carried out by HPLC using a Shimadzu (Kyoto, Japan) LC-20AD HPLC workstation equipped with a SIL-10AF autosampler and two detection systems: differential refractometry (Shimadzu RID-10A) and diode array detector for UV spectrophotometry (Shimadzu SPD‑M10AVP). The chromatographic column was an Hi-Plex-H (Agilent, 300 × 7.7 mm; 8 mm) that was operated at 65 °C using H2SO4 8 mmol L-1 as mobile phase in a flow rate of 0.6 mL min‑1. Quantification was carried out by external calibration using calibration curves (R2 > 0.99) ranging from 0.1 to 3.0 mg∙L-1 for the following analytes: fructose, glucose, sucrose, HMF, furfural, lactic and, levulinic acid, acetic acid, and formic acid. Carbohydrates and organic acids were quantified by differential refractometry (RID-10A, Appendix Fig. S1), while furan compounds were monitored by UV spectrophotometry at the 280 nm wavelength (SPD‑M10AVP). In some cases, the HPLC oven was turned off and analyses were carried out at ambient temperature (around 25 °C) to prevent sucrose hydrolysis. For this, reactions were carried out with sucrose (33.2 mmol∙L-1) under microwave heating without adding an exogenous acid catalyst for 90 min at 180 °C (thermal treatment). Then, reaction aliquots were taken and analyzed by HPLC. In addition, standard solutions containing sucrose, and a mixture of sucrose, glucose, and fructose were also analyzed for comparison. Organic acids and HMF yields were calculated with respect to their corresponding theoretical yield, which was based on the stoichiometric amount that could have been produced from the total carbohydrate content of the starting material. Equations 3 and 4 show how HMF and organic acid yields were calculated from carbohydrates (fructose, glucose, and sucrose),
where HMFexp and OrgAcexp are the amounts of HMF or organic acid that were determined experimentally, mCarb is the initial mass of fructose, glucose or sucrose, and SF is the stoichiometric factor of carbohydrates conversion into HMF, levulinic, formic, and lactic acids (0.70, 0.64, 0.25, and 0.50, respectively). These factors were determined by dividing the molar mass of the analyte by the molar mass of the carbohydrate used for conversion.
RESULTS AND DISCUSSION
Sucrose Analytical Stability
Sucrose undergoes hydrolysis under relatively mild acidic conditions (Steinbach et al. 2018). Since reaction aliquots were analyzed by HPLC at 65 °C using a cation-exchange resin as the stationary phase and H2SO4 8 mmol∙L-1 as the mobile phase, it was important to demonstrate that glucose and fructose release in the reaction medium was not due to acid hydrolysis during HPLC analysis, a behavior that was observed already in the authors’ earlier studies (data not shown). This was resolved by performing HPLC of a sucrose standard solution at ambient temperature (25 °C), using otherwise identical analytical conditions. The blue line in Fig. 2 demonstrates that sucrose remained stable, since glucose and fructose were not detectable in the HPLC profile. Then, sucrose (33.2 mmol∙L-1) was thermally treated in the MWR at 180 °C for 90 min, and the product was subjected to the same analytical procedure. The red line in Fig. 2 shows that sucrose was almost completely hydrolyzed even in the absence of any reaction catalyst (that is, under thermal conversion). Hence, as observed later in this research, the absence of sucrose in the reaction products could not be attributed to hydrolysis during HPLC, using a mild acid mobile phase. This condition had to be used because it delivered the best resolution among sugars, furans, and organic acids, allowing for a more accurate qualitative and quantitative analyses of reaction products (see Fig. S1 in the Appendix).
Fig. 2. HPLC analysis of sucrose at ambient temperature before and after thermal treatment at 180 °C for 90 min under microwave irradiation
Thermal Conversion
The thermal conversion of fructose (66.7 mmol∙L-1), glucose (66.7 mmol∙L-1), and sucrose (33.2 mmol∙L-1) was assessed in the MWR at 200 °C for 120 min, and the results are shown in Table 2. As expected, sucrose was not detected in any of the reaction products because it was hydrolyzed to fructose and glucose. Fructose, either pure or released by hydrolysis, was almost completely consumed during the heat treatment, while glucose remained partially unconverted in the reaction medium. Sucrose generated 26.61 mol % HMF by hydrolysis and dehydration, while HMF acid catalyzed rehydration released 7.43 mol % formic acid and 2.22 mol % levulinic acid. Also, the retro-aldol/1,2-H migration pathway yielded 2.83 mol % lactic acid from sucrose.
Table 2. Conversion, Product Yields, and Selectivities for the Thermal Treatment of Fructose, Glucose, and Sucrose *
n.d., not detected; * Reactions were carried out for 120 min under microwave (MW) irradiation and in a SS-PTFE reactor system using a setpoint temperature of 200 °C.
The same reaction control was also carried out in the SS-PTFE reaction system, in which the setpoint temperature was 200 °C. Compared to MWR, both glucose and fructose were more extensively consumed, generating slightly higher yields of the three organic acids mentioned above. However, carbohydrate conversion to organic acids remained relatively low, justifying the need for addition of an exogenous acid catalyst. Also, the thermal treatment turned the reaction medium darker as time went by, showing that carbohydrates were gradually lost to side reaction forming humins. Such dark water-insoluble polymeric materials are known to arise from condensation reactions involving dehydration co-products such as furan compounds and organic acids (Deng et al. 2018).
Influence of Temperature on Carbohydrate Conversion
The influence of temperature on carbohydrate conversion using Al(NO3)3 as the catalyst precursor was investigated using MWR at 170, 180, 190, 200, and 215 °C for a fixed reaction time of 90 min. Table 3 shows the yield of products derived from fructose at different reaction temperatures, while Fig. 3 provides the HPLC profile of MWR reaction products at 180 °C using the Agilent Hi-Plex-H column and detection by differential refractometry.
At 170 °C, fructose was not completely consumed, with 25.9% remaining unreacted in the reaction medium (Table 3). At 180 °C, the selectivity for retro-aldol increased, resulting in the identification of at least one reaction intermediate (glyceraldehyde) and increased lactic acid yields (Fig. 3). However, at temperatures above 190 °C, the dehydration pathway was favored, with HMF acting as a reaction intermediate to produce levulinic and formic acids in almost equivalent molar yields. According to Choudhary et al. (2013), the selectivity for fructose dehydration increases with increasing reaction temperatures, indicating that the apparent activation energy for HMF formation is higher than that of the retro-aldol pathway. In addition, once produced, HMF was rapidly consumed due to its high reactivity and thermal instability (Lei et al. 2014). The presence of glucose was also observed in all reaction temperatures, and this was a result of the glucose/fructose isomerization equilibrium under the applied experimental conditions.
Table 3. Fructose Conversion at Different Temperatures for 90 min Using Microwave Heating and 2.67 mmol·L-1 Al(NO3)3∙9H2O
n.d., not detected.
Fig. 3. HPLC analysis of MWR reaction products using the Agilent Hi-Plex-H chromatographic column at 65 °C, 8 mmol∙L-1 H2SO4 as the mobile phase, and detection by differential refractometry. Reactions were carried out for 90 min at 180 °C using Al(NO3)3 as the catalysts precursor.
Bicker et al. (2005) studied the conversion of fructose in stainless steel tube reactors at 260 °C using Zn(SO4) as catalyst. Fructose was almost completely consumed in 140 s, yielding lactic acid as the main reaction product, followed by HMF in much lower yields. Glucose formation by fructose isomerization was also observed, along with the release of reaction intermediates such as glyceraldehyde, dihydroxyketone, and pyruvaldehyde. Hence, the retro-aldol mechanism prevailed over dehydration, probably due to the use of high temperatures and short reaction times. Also, it seems that HMF rehydration did not occur because both levulinic and formic acids formation was not reported.
Choudhary et al. (2013) studied the kinetics of fructose conversion in thick-walled glass vials using conventional heating in an oil bath and CrCl3 as catalyst at 140 °C. Fructose conversion was complete after 160 min, but no lactic acid was formed, meaning that the reaction pathway was dominated by dehydration to HMF followed by rehydration to levulinic and formic acids. Fructose isomerization was also observed in this reaction system forming glucose and mannose in equimolar amounts.
The same trend was observed for glucose conversion at different temperatures (Table 4). However, glucose was not completely consumed in any of the applied reaction temperatures. While the highest temperatures (e.g., 215 °C) led to the best levulinic and formic acids yields from glucose, fructose required lower temperatures (190 °C) to produce these organic acids predominantly. Incomplete conversion and low reaction yield from glucose demonstrate the greater stability of its pyranosidic ring compared to the furanosidic ring of fructose (Zhang et al. 2015). For this same reason, glucose must undergo isomerization to fructose prior to its dehydration to HMF (Rasmussen et al. 2014) or retro-aldol conversion to lactic acid (Marianou et al. 2018).
Table 4. Glucose Conversion at Different Temperatures for 90 min using Microwave Heating and 2.67 mmol·L-1 Al(NO3)3∙9H2O
n.d., not detected.
Table 5 shows the effect of temperature on sucrose conversion. The observed trends in product formation were very similar to that of fructose (Table 3), except for the presence of glucose from sucrose hydrolysis. The two main reaction pathways competed at the lowest temperature range (170 to 180 °C), with lactic acid being the main reaction product at yields lower than those obtained directly from fructose. HMF and its rehydration derivatives (formic and levulinic acids) became predominant at 190 °C, while the selectivity for the latter increased considerably at temperatures above 200 °C. Once again, this was expected because HMF is thermally unstable, and its gradual disappearance is related to its rehydration to formic and levulinic acids (Fig. 1) (Lei et al. 2014). Hence, HMF is rapidly formed and gradually consumed during the reaction course to produce organic acids plus water-insoluble humins. Glucose remained present in the reaction mixture even at the highest reaction temperature, while fructose was almost completely consumed as already observed in Table 3.
Table 5. Sucrose Conversion at Different Temperatures for 90 min Using Microwave Heating and 2.67 mmol·L-1 Al(NO3)3∙9H2O