Catalytic Conversion of Glycerol to Lactic Acid Over Cu-Based Catalysts

Abstract

Copper (I, II) oxide powders were tested for glycerol conversion to lactic acid under alkaline conditions. Fresh and spent catalysts were characterized using powder X-ray diffraction, Scanning Electron Microscopy/Energy Dispersive X-ray Spectroscopy, the BET-nitrogen adsorption method and FTIR spectroscopy. In all cases, an almost complete in situ reduction of Cu (I, II) oxides into metallic Cu was observed, even after one catalytic run. Moreover, all the samples of spent catalysts showed similar catalytic activity regardless of their initial form and particle size. Commercial copper powders, prepared copper powders and in situ reduced copper catalysts were tested under the same conditions to compare their catalytic activity. It was shown that the in situ reduced copper catalyst had similar activity to the specially prepared copper powders and much higher activity compared to the commercial copper powders. The in situ reduced copper catalyst exhibited rather high stability. The glycerol conversion and lactic acid selectivity were about 98% and 70%, respectively, after ten catalytic cycles.

1. Introduction

Lactic acid (LA) and its derivatives are widely used in food, cosmetic, pharmaceutical and chemical industries [1,2,3]. The most promising line of LA processing is the production of biodegradable polymers. Polylactic acid (PLA) is widely used in medicine, including advanced delivery systems for the controlled release of drugs, implants (scaffolds) and skin closure biomaterials. Polymers based on LA can also be used for the production of biodegradable packaging and mulch films for agriculture [4,5]. The recent market size value of PLA is estimated at USD 3.51 billion while revenue forecast for 2030 is set to be over USD 7 billion [6].

Currently, fermentation of various natural feedstock is used for industrial LA production [7]. However, fermentation processes are connected with rather low productivity and efficiency, high cost of enzymes, waste generation and difficulty of product purification [8,9].

Meanwhile, there is a great worldwide demand to utilize the glycerol produced as a biodiesel by-product (about 10% of the total volume of biodiesel) [10,11]. According to forecasts [12], the amount of crude glycerol produced will soon reach 1.2 million tons. Thus, catalytic conversion of glycerol in to value-added chemical products (in particular, in LA) looks highly beneficial [13].

LA can be produced from glycerol using various methods, including hydrothermal conversion [14,15], hydrogenolysis [16] and selective oxidation [17,18,19]. The hydrothermal processes require high reaction temperatures and lead to high energy consumption and by-product formation. The oxidative glycerol conversion to LA in the presence of Pt and Au-based catalysts proceeds under milder conditions [18,19]. However, the high cost of these catalysts restricts their use in industrial processes.

Recently, great attention was focused on copper (I, II) oxide catalysts for glycerol conversion to LA in alkaline solutions. In contrast to the hydrothermal processes, using bulk or supported copper (I, II) oxide allowed for achieving high LA selectivity and high glycerol conversion even at low temperatures and requires a low NaOH/glycerol molar ratio [4,20,21,22,23,24,25]. According to previously published articles, the usage of Cu₂O allows for achieving 100% glycerol conversion and achieves an LA selectivity of about 70–80% [22,26,27]. Practically the same results were achieved while using CuO [28], CuO/ZrO₂ [25], CuO-ZnO-Al₂O₃ [29] and Cu nanoparticles [23,30,31] as catalysts. Unfortunately, it is rather difficult to compare results from the mentioned articles due to the different process conditions (noted in the review [4]). Moreover, there is very little information available in the literature about the stability of copper (I, II) oxide catalytic systems, especially for glycerol conversion into LA [22,27].

In this study, copper (I, II) oxides and metallic copper were prepared using various methods and tested for their activity and stability for glycerol conversion into LA in an alkaline media.

2. Results and Discussions

GC–MS analysis of the obtained reaction liquid-phase products showed that LA was the main reaction product in all cases (see Supporting Information, Figures S1 and S2). The liquid-phase by-products were diglycerol (DG), 1,2-propylene glycol (PG) and acetic acid (AA). In addition, formic, methacrylic and propanoic acids, and 2,3-butanediol were also found in small amounts (with a total selectivity of less than 2%). Carboxylic acids existed in the solution in the form of sodium salts. The quantity of by-produced DG using Cu-based catalysts was significantly reduced in comparison to using only sodium hydroxide (25% versus 2~6%). In our opinion, this was caused by the reversibility of the glycerol oligomerization reaction in the presence of basic catalysts and a significant equilibrium shift towards the decomposition of the dimer into glycerol by increasing the rate of LA formation in the presence of the Cu-based catalysts [28]. Condensation products were also detected using HPLC–MS chromatography for all catalysts. The detected ion masses were in the range of 300–900 Da.

GC–TCD analysis of the gas phase after the process showed that the main gas product was hydrogen (more than 95% from volume), which was formed as a result of the glycerol→LA main reaction, as well as a result of the dehydrogenation of PG [32].

Thus, the process of glycerol transformation into LA can be shown as follows (Figure 1):

The reaction samples were analyzed to determine the total carbon content before and after synthesis because the carbon dioxide produced could be absorbed in the sodium hydroxide solution and converted to sodium carbonate. The value of the carbon balance corresponded to the ratio between the amount of total carbon in the product solution and its amount in the initial mixture. The average value of the two parallel measurements was taken as a result. The total amount of carbon-containing components (CO, CO₂, CH₄) was less than 0.05% by volume. Such a small value of carbon dioxide found in the gas phase can be explained by its consumption to form carbonates in an alkaline medium [33].

Table 1. Comparison of different fresh catalysts in glycerol conversion into LA.

Catalyst	Conversion of Glycerol, XGly, %	Selectivity, Seli, %					Carbon Balance, %
		DG	LA	PG	AA	Others
Control	>1.0	0	0	0	0	0	99.8
CuO, no base Cu2O-1, no base Cu2O-2, no base Cu2O-3, no base CP, no base	>1.0	0	0	0	0	0	99.6
Control, only base *	24.9	25.3	43.4	1.6	4.0	25.7	95.3
CuO *	86.3	2.3	67.9	3.9	2.2	23.7	89.4
Cu2O-1 *	91.8	4.9	69.5	3.3	2.5	19.8	98.1
Cu2O-2 *	88.6	6.6	63.3	4.3	3.5	22.3	95.5
Cu2O-3 *	86.5	3.8	62.8	6.6	7.3	19.5	97.5
CP *	90.1	5.0	66.7	4.6	1.8	21.9	96.2
Cu powder commercial	80.2	5.0	61.6	6.7	3.2	23.5	97.8

Reaction conditions: C_GLY = 1.0 mol·L⁻¹, solvent water, T = 240 °C, 6 h, stirring speed = 1000 rpm, catalyst/glycerol = 0.027 g/g. * NaOH/glycerol = 1.1 mol/mol.

shows the conversion of glycerol and the corresponding product selectivity after 6 h.

As can be seen in Table 1, the conversion of an aqueous glycerol solution was less than 1% in the synthesis carried out without adding a base. In the presence of only sodium hydroxide in a molar-to-glycerol ratio of 1.1, the glycerol conversion was about 23% and the main products were LA and DG.

All Cu-based catalysts showed high activity for the glycerol conversion into LA in the base media (Table 1). There was a steep increase in the glycerol conversion (from 23 to ~90%) and the selectivity of LA formation (from 43 to ~70%) in comparison to the homogeneous catalytic process during the process carried out in the presence of fresh copper-containing catalysts. The most selective catalyst for LA was Cu₂O-1 (69.5%).

Table 2. Results of aqueous phase glycerol conversion into LA in the presence of spent catalysts.

Catalyst	Conversion of Glycerol, XGly,%	Selectivity, Seli, %					Carbon Balance, %
		DG	LA	PG	AA	Others
CuO	88.5	2.8	66.2	4.1	2.6	24.3	95.4
Cu2O-1	88.6	4.7	69.5	4.9	1.9	19.0	96.8
Cu2O-2	88.6	5.6	64.6	4.7	2.8	22.3	96.5
Cu2O-3	86.5	4.1	67.6	6.1	5.3	16.9	98.3
CP	89.4	4.0	69.7	4.9	2.8	18.6	97.4

Reaction conditions: C_GLY = 1.0 mol·L⁻¹, solvent water, T = 240 °C, 6 h, stirring speed = 1000 rpm, NaOH/Glycerol = 1.1 mol/mol, catalyst/glycerol = 0.027 g/g.

shows the retesting results of the spent copper-containing catalysts.

As can be seen from Table 2, all samples of spent catalysts showed similar catalytic activity regardless of their initial form.

Freshly prepared and spent catalysts were characterized using instrumental analysis methods.

2.1. Catalyst Characterization

2.1.1. SEM Analysis

The morphology and microstructure of the powders were investigated using SEM. Figure 2 displays the SEM images of the fresh and spent powder samples.

Based on the SEM images, the average particle sizes were determined (

Table 3. Characteristics of catalyst particles before and after catalytic test.

Catalyst Sample	Surface Area, BET, m2/g		Average Particle Size, nm
	Fresh	Spent	Fresh	Spent
CuO	1.9 ± 0.1	1.2 ± 0.1	119 ± 25	1124 ± 342
Cu2O-1	5.1 ± 0.3	2.2 ± 0.0	190 ± 67	437 ± 112
Cu2O-2	0.9 ± 0.2	0.8 ± 0.2	289 ± 106	396 ± 145
Cu2O-3	3.5 ± 0.1	1.0 ± 0.2	543 ± 142	586 ± 150
CP	4.8 ± 0.1	3.2 ± 0.3	693 ± 205	1043 ± 241

).

The SEM images (Figure 2) display the grains containing several agglomerated domains with different orientations. Table 3 shows the results of the average particle sizes of the grains before and after synthesis. After synthesis, the used catalysts formed additional agglomerates, which were larger than in the freshly prepared samples. The starting powders were agglomerates ranging from submicrons to a few microns, and the spent powders had agglomerates of up to a few microns.

Table 3 also shows some decrease in the specific surface area of the catalysts. This fact can be caused by additional agglomeration into bulk clusters with a change in the microstructure of the powders. However, despite the observed tendency of increasing particle sizes and a decrease in the specific surface area of the catalysts, the catalytic activity and selectivity to LA of fresh and used catalysts were similar (Table 1 and Table 2).

2.1.2. XRD Analysis

The XRD method implies phase identification of the catalyst powders before and after the process (Figure 3), and calculates the size of the crystalline domains (see Supporting Information Equation (S1)).

All the diffraction peaks can be indexed with a lattice constant and compared to the Joint Committee on Powder Diffraction Standards (JCPDS). The phase composition of the synthesized catalysts corresponds to CuO (a), Cu₂O (b, c, d) and Cu⁰ (e).

The strong and sharp peaks indicate that the obtained crystals are highly crystalline. The sizes of the crystalline domains, calculated using the XRD method, were in the range of 16–52 nm for all the obtained samples (the values of the crystallite sizes of the powders are shown in Supporting Information Table S1).

In general, after the syntheses, the average crystal size slightly decreased for the oxide powders and remained the same for the copper (CP) catalysts. This circumstance can be associated with changes in the phase composition after catalytic testing: copper oxides were reduced Cu²⁺ → Cu⁺ → Cu⁰; CuO turned into a three-phase system, containing in addition to copper (II) oxide, Cu₂O and metallic copper. Cu₂O-1 and Cu₂O-3 were reduced to metallic copper, and Cu₂O-2 partly recovered to metallic copper. The CP phase composition before and after the synthesis did not change.

Thus, the results of XRD analysis indicate an in situ reduction of Cu(II) and Cu(I) into metallic copper under the process conditions.

2.1.3. FTIR Analysis

The results of the FTIR analysis confirmed that the produced Cu₂O-1 powder after synthesis transformed into copper powder, which is in agreement with the results of the XRD. The FTIR spectrum of the fresh and spent Cu₂O-1 catalysts are shown in Figure 4.

A transmittance active peak in the range of 600 to 650 cm⁻¹ (at around 630 cm⁻¹) attributed to the characteristic Cu(I)-O vibrations in the Cu₂O [34] was detected. After the reaction, this band disappeared, which confirms the formation of Cu⁰ particles only.

2.1.4. EDS Analysis

The Cu₂O-1 catalyst sample (which showed maximum activity and selectivity) was analyzed using the EDS method after several test cycles. The results of the analysis are shown in Figure 5. The EDS spectrum of the Cu₂O-1 catalyst before the synthesis corresponds to Cu₂O. At the same time, only Cu elements were detected in the samples after conducting catalytic tests for one and 10 cycles. These results also agreed with the XRD and the FTIR analyses.

2.2. Effect of Reaction Conditions on the Conversion of Glycerol over In Situ Cu Catalyst

The influence of temperature, the NaOH/glycerol ratio, catalyst loading, glycerol concentration and reaction time on glycerol conversion and selectivity of LA formation was studied in order to determine the optimal conditions for carrying out the process in the presence of a Cu catalyst prepared using the in situ reduction of Cu₂O-1.

2.2.1. Effect of the Reaction Temperature

The effect of the reaction temperature on the glycerol conversion to LA under alkaline conditions was examined by conducting a series of relevant experiments in the temperature range of 210–250 °C. Under a lower temperature condition, the glycerol expenditure rate is too low and under a higher temperature, the side reactions take a more proactive role accompanied by a by-product yield increase.

Figure 6 and Figure 7 show the glycerol conversions and product selectivity over the in situ Cu catalyst at different reaction temperatures.

Glycerol conversion increased from 52% to 91% with an increase in the reaction temperature from 210 to 250 °C. In general, a temperature increase caused a minor by-product yield increase. LA selectivity within the set range was almost free of temperature influence.

2.2.2. Effect of NaOH/glycerol Molar Ratio

The NaOH/glycerol molar ratio immensely affects the glycerol conversion rate [22]. An increase in the alkali concentration led to an increase in glycerol conversion to about 90% when using an in situ Cu catalyst (Figure 8). The LA selectivity was first increased from 57% to 65%, and then slightly decreased to 58%. The maximum LA selectivity was achieved with a NaOH/glycerol ratio of ~1.1 mol/mol. A further increase in the base excess related to the glycerol substrate led to the LA selectivity decreasing. This fact can be explained due to the increase in the rate of side reactions with hydroxyl ions, as in the case of the process temperature increase [15].

2.2.3. Effect of Initial Glycerol Concentration

The dependence of the process parameters on the initial concentration of glycerol in an aqueous solution is shown in Figure 9 and Figure 10.

In the case of a homogeneous process (only a base without an in situ Cu catalyst), glycerol conversion slightly increased and a noticeable increase in the selectivity of DG from 6% to 38% was observed with an increase in the initial concentrations of glycerol and alkali. In the presence of an in situ Cu catalyst, the dependence of the glycerol conversion had a pronounced extremum at 1.0 mol·L⁻¹, while the selectivity of LA remained more or less constant.

2.2.4. Effect of Catalyst Loading

The dependence of the glycerol conversion and the LA selectivity on the catalyst loading is shown on Figure 11.

Increasing the catalyst loading by up to 8.2% (0.0815 g/g based on glycerol) allowed for increasing the LA selectivity by about 72% with glycerol conversion close to 100%. A further increase in the catalyst concentration did not lead to a significant change in the glycerol conversion and the selectivity to LA.

2.3. Reuse (Stability Study) of In Situ Cu Catalyst

The reusability of a Cu catalyst prepared using the in situ reduction of Cu₂O-1 during the conversion of glycerol to LA was investigated. To test the catalyst’s stability, after reactor cooling, the reaction mixture was sampled with a pipette equipped with a filter and the entire catalyst remained at the bottom of the reactor. Then, a freshly prepared glycerol solution and alkali were loaded into the reaction vessel and the next cycle of catalytic transformation started.

Figure 12 shows the results of the Cu in situ catalyst recycling test.

In accordance with the process conditions used in this study, no significant differences in activity and selectivity to LA were found with ten consecutive tests of the same amount of spent catalyst. In situ, the Cu catalyst produced a 98 ± 1% conversion of glycerol and a 73 ± 2% selectivity for LA.

It was expected that with the repeated use of the reduced catalyst, the amount of hydrogen in the gas–liquid system should increase since it would not be spent on catalyst reduction. As a result, H₂ will more actively interact with the glycerol in the reaction of hydrogenolysis, which should ultimately increase the selectivity to PG. Unlike the expectations, the average selectivity of PG formation within 10 cycles was 4.6%, which corresponds to the values of the selectivity of PG formation in the initial comparative testing of various catalysts (Table 2), ranging from 3.3 to 6.6%. The results can be explained by the limited solubility of hydrogen in water at the given reaction temperatures and pressures (240 °C, 3.4 MPa), and the high rate of the target reaction glycerol→LA, which helps to reduce the concentration of the potential substrate (glycerol) for the implementation of the hydrogenolysis route.

3. Materials and Methods

3.1. Materials

Glycerol (99.9%), LA (85.0%), PG (99.5%), AA (99.7%), NaOH (98.0%), copper sulfate pentahydrate (CuSO₄∙5H₂O, 99.0%), sodium borohydride (NaBH₄, 96.0%), copper acetate monohydrate (99.0%), polyvinylpyrrolydone (powder, average Mw~55 k) were purchased from Sigma (Perth, WA, USA). Copper powder 150 mesh (99.5%; particle size above 89 μm), copper formate tetrahydrate (98.0%) were purchased from Alfa Aesar (Ward Hill, MA, USA). DG (80.0%) was purchased from TCI Europe NV (Zwijndrecht, Belgium). Glucose 1-hydrate (99.9%) and ascorbic acid (99.9%) were purchased from PanReac AppliChem (Darmstadt, Germany). All chemicals were used without further purification. Deionized water was used in all experiments.

3.1.1. Preparation of Catalysts

CuO

CuO was prepared by a deposition–precipitation method. In total, 2.0 g of copper acetate monohydrate was dissolved in 300 mL of water, after that the solution pH was adjusted to 11.9 by slowly adding of NaOH solution (0.5 M) with continuous stirring (800 rpm, magnetic stirrer). The resulting precipitate was stirred for 1 h at 60 °C (800 rpm), and then it was washed several times with deionized water to eliminate Na⁺ and CH₃COO⁻. The obtained product was dried to constant weight at 40 °C, and exposed to calcination at 400 °C for 4 h.

Cu₂O

A series of Cu₂O powders were obtained by aqueous reduction method in the presence of several reduction agents: glucose (named Cu₂O-1), ascorbic acid (named Cu₂O-2), and NaBH₄ (Cu₂O-3, respectively) without adding any other organic modificators.

NaOH aqueous solution (0.2 mol·L⁻¹, 200 mL) was dripped into aqueous CuSO₄∙5H₂O solution (Cu²⁺ concentration is 0.1 mol·L⁻¹, 250 mL) with continuous stirring (800 rpm). Three such solutions were obtained. Then, a glucose solution in water (0.4 mol·L⁻¹, 100 mL for preparation Cu₂O-1), ascorbic acid solution in water (0.4 mol·L⁻¹, 100 mL for preparation Cu₂O-2), and NaBH₄ solution in water (0.2 mol·L⁻¹, 100 mL for preparation Cu₂O-3) were added to the resulting solution. The solutions were stirred at 40 °C for 30 min (800 rpm) to obtain a red-brown color. Precipitation oxides (Cu₂O) were separated from the solution by centrifugation (5000 rpm, 10 min), washed several times with an aqueous–ethanol mixture, and dried in vacuum at room temperature to constant weight.

Copper Powder (CP)

Metallic Cu powder was prepared with NaBH₄ by aqueous reduction method without adding any organic modifiers [28]. An aqueous solution of NaOH (0.2 mol∙L⁻¹, 40 mL) was added with stirring to an aqueous solution of CuSO₄∙5H₂O (Cu²⁺ concentration was 0.2 mol∙L⁻¹, 60 mL). Then, NaBH₄ in water (0.4 mol·L⁻¹, 100 mL) was added dropwise (50 mL∙min⁻¹) to the resulting solution. The solution was mixed at 40 °C for at least 1 h to obtain a brown color. The Cu precipitate was separated from the solution by centrifugation, washed several times with an aqueous–ethanol solution and dried in vacuum at room temperature to constant weight.

3.1.2. Characterization of Catalysts

The KBr pellet technique was applied to determine IR specters of the catalyst samples. IR spectra were recorded in air at room temperature using Shimadzu (Kyoto, Japan) IRAffinity-1 spectrometer in the region of wavenumbers from 400 to 4000 cm⁻¹ with a resolution of 0.5 cm⁻¹.

The catalyst’s surface morphology and the average particle size were determined using scanning electron microscope Hitachi-S2500 (Brisbane, QLD, Australia) equipped with energy-dispersive spectrometer (JNCA).

X-ray diffraction (XRD) patterns were recorded with a Shimadzu XRD-6100 diffractometer using Cu Kα radiation (λ = 0.1541 nm) in the 2θ range from 10° to 80° with a step size of 0.02°.

The surface area of catalyst was determined by using the adsorption desorption method by the standard Brunauer–Emmett–Teller (BET) method using Sorbi (Newton Stewart, UK) MS instrument (META). All the samples were preliminarily kept in vacuum at 110 °C for 2 h before analysis.

3.2. Catalytic Activity Test

The glycerol conversion reaction was carried out in a stainless-steel autoclave with the capacity 0.35 L equipped with a sampler, thermostatic jacket, and thermocouple. In a typical experiment, 0.25 L of an aqueous glycerol solution and calculated amounts of NaOH and catalyst were placed into the reactor. The reaction mixture was heated to the required temperature. The stirring speed was set at 1000 rpm.

After reaching the reaction temperature, the initial liquid sample of reaction mixture was taken and the reaction initial time was set. During the process, the reaction liquid samples were taken at predetermined time intervals.

After reacting for a prescribed period, the reactor was cooled down and the spent catalyst was separated from the reaction mass by centrifugation. The isolated powder was washed several times with deionized water, dried in vacuum at room temperature to constant weight and analyzed.

The catalytic tests for all catalysts were repeated at least twice. Relative deviations did not exceed ±5%.

3.3. Analysis

The liquid samples were rapidly cooled to room temperature, centrifuged and acidified with sulfuric acid aqueous solution (40 wt%) to pH values of about 2 before analysis.

The identification of liquid products was performed by GC–MS, using the Shimadzu GC-2010 instrument equipped with GCMS-QP2010 mass spectrometer and VB-1701 capillary column (VICI). Condensation products were identified by HPLC–MS, using Agilent 1200 HPLC system equipped with Agilent (Santa Clara, CA, USA) 6310 mass detector and C18 column (250 × 4.6 mm).

The total carbon balance was determined using vario EL cube analyzer (Elementar Analysensysteme, Windows^® based vario EL cube operating software with LIMS integration and auto sleep and wake-up function for automated and unattended overnight operation).

The quantitative analysis of liquid products was performed by HPLC equipped with a RI detector (Waters, Milford, MA, USA) using Rezex ROA-Organic acid column (00H-0138-K0, 300 × 7.8 mm, particle size 8 μm, Phenomenex, Torrance, CA, USA,) and 0.025 M aqueous sulfuric acid solution (0.5 mL/min) as eluent at 65 °C. The substrate and products (glycerol, DG, LA, PG, AA) concentrations were identified and quantified on the basis of standard solutions used as external standards.

The gas products were analyzed by GC, using Chromos GC–1000 instrument (Chromos Engineering, Dzerzhinsk, Russia) equipped with thermal conductivity detector (TCD) using HayeSep N and NaX columns. The calibration gas mixtures were used as external standards.

The glycerol conversion (${\mathrm{X}}_{\mathrm{Gly}}$) was evaluated according to the following equation:

$${\mathrm{X}}_{\mathrm{Gly}},\%=\frac{\mathrm{mole} \mathrm{of} \mathrm{consumed} \mathrm{glycerol}}{\mathrm{mole} \mathrm{of} \mathrm{initial} \mathrm{glycerol}}\times 100$$

(1)

The product i selectivity (Sel_i) was evaluated according to the following equation:

$${\mathrm{Sel}}_{\mathrm{i}},\%=\frac{({\mathrm{C}}_{\mathrm{i},\mathrm{j}} - {\mathrm{C}}_{\mathrm{i},0})\times (\mathrm{carbon} \mathrm{number} \mathrm{of} \mathrm{product} \mathrm{molecule})}{(\mathrm{carbon} \mathrm{number} \mathrm{of} \mathrm{glycerol})\times (\mathrm{mole} \mathrm{of} \mathrm{consumed} \mathrm{glycerol})}\times 100$$

(2)

where C_i,0 and C_i,j—product (i) initial and current concentration (mol·L⁻¹), respectively.

4. Conclusions

In this paper, different copper-containing catalysts were tested as catalysts for glycerol conversion into LA under alkaline conditions. The obtained values of glycerol conversion and LA selectivity in the presence of all copper-containing catalysts were relatively constant and amounted to 86–92% and 66–69%, respectively.

Using various catalyst analysis methods (XRD, FTIR, EDS) before and after the process, an in situ reduction of copper oxides into metallic Cu was shown. The Cu powders in situ obtained under hydrothermal LA synthesis are not inferior to the specially prepared copper powders in terms of their catalytic properties and retain their activity and selectivity in the conversion of glycerol into LA at least for 10 cycles.

The effect of the reaction conditions over in situ Cu catalysts was described (temperature, catalyst loading, NaOH/glycerol molar ratio and glycerol concentration) for glycerol conversion and product yields. The use of an in situ Cu catalyst allows for achieving a 98 ± 1% conversion of glycerol and a 73 ± 2% LA selectivity under the optimal conditions.