KNO₃-Loaded Coffee Husk Ash as a Heterogeneous Alkali Catalyst for Waste Frying Oil Valorization into Biodiesel

Abstract

In this study, a heterogeneous basic catalyst was synthesized from a catalyst composite material (CCM) of coffee husk ash and char mixture (A/C) impregnated with KNO₃ and employed to transesterify crude waste frying oil (WFO). The effect of CCM calcination temperature (CCMCT) (500–700 °C) on the catalyst physicochemical properties was investigated. A differential scanning calorimeter was used to examine potential phase changes during the calcination of A/C and CCM. The catalysts from each CCMCT were characterized by X-ray diffraction (XRD), Brunauer–Emmet–Teller surface area analyzer, scanning electron microscopy (SEM), SEM with energy-dispersive X-ray diffractometer, colorimeter, and attenuated total reflectance Fourier transform infrared (ATR-FTIR) spectrometer. The methoxy functional group FTIR peak integral value and the dynamic viscosity of the biodiesel synthesized by each catalyst were used to determine the qualitative WFO conversion. Furthermore, the quantitative WFO conversion was determined using nuclear magnetic resonance (¹H NMR) analysis. Crystallinity, elemental composition, basicity, and morphology of catalysts were highly dependent on the CCMCT. Without transesterification condition optimization (reaction temperature of 45 ± 2.5 °C, catalyst loading of 3 wt %, methanol to oil molar ratio of 12:1, and reaction time of 1 h), a higher catalytic performance (72.04% WFO conversion) was reached using a catalyst from the CCMCT of 600 °C. When using a coffee husk ash catalyst without KNO₃ impregnation (C-00-600), the WFO conversion was only 52.92%. When comparing the C-25-600 and C-00-600 catalysts, it was observed that KNO₃ impregnation had a substantial impact on the catalyst crystallinity, basicity, and morphology.

Introduction

From an economic and environmental pollution perspective, relying on fossil fuels as a primary source of energy is unreliable in the future generation.¹ That is why the world’s attention is shifting toward biofuels.² Among biofuels, biodiesel is the best alternative to replace fossil diesel fuel. The benefits of biodiesel include lower overall exhaust emission gases and toxicity, biodegradability, sourcing from renewable and domestic feedstocks, low sulfur content, superior flashpoint, improved combustion efficiency, and a 78% reduction in net CO₂ emissions based on life cycle assessment.³⁻⁵

Unless non-edible and low-grade fat and oils are employed in the transesterification reaction, large-scale biodiesel production cannot be feasible.⁶ Therefore, engaging waste frying oils (WFOs) to produce biodiesel is economical, minimizes the competition between food and fuel, and reduces pollution from WFO residue.^2,7 However, the transesterification of low-grade feedstocks such as WFO is ineffective without a suitable catalyst. Homogeneous, heterogeneous, and enzyme catalysts were investigated. Alkali homogeneous catalysts are being used in most existing biodiesel plants. However, soap formation, which makes product separation challenging, rendered their utilization unsuitable for feedstocks containing significant FFA (>1%).⁸ Even though acidic catalysts hinder soap formation via simultaneous esterification and transesterification, they are corrosive to reactors and do not provide sufficient catalysis.⁹ The enzyme catalysts are nonpolluting and with high selectivity. They have downsides including a longer reaction time, high cost, and deactivation in the presence of methanol.^10,11 Nowadays most investigations are converging toward heterogeneous solid catalysts because of their reusability and renewability.^12,13 Similarly, the use of heterogeneous catalysts can also increase the sustainability of biodiesel production for future generations and reduce the production cost of biodiesel.^2,12 Furthermore, while using heterogeneous catalysts, greater purity glycerin can be obtained with fewer dissolved ions.¹⁴

The main barrier to the use of heterogeneous catalysts is their high cost. Exploration of a cheap precursor that gives a high-performing catalyst is necessary to overcome the economic constraints in the next generation for the mass production of biodiesel.¹⁵ Because of catalyst typical traits, including thermal and mechanical stability and low cost, metal oxides are frequently employed in the synthesis of heterogeneous catalysts.¹⁶ The catalytic activity depends on the base or acid strength and active site concentration.¹⁷ The catalyst texture and degree of crystallinity also matter.¹⁸ These catalyst properties depend on the catalyst synthesis condition and precursor composition. Although it was not always the case, most metallic oxides showed better stability and catalytic activity when mixed with other oxides or integrated with porous supports.¹⁹⁻²⁴

Mixed oxide catalysts can be synthesized from chemical and biological sources. In this field, heterogeneous catalysts made from agricultural wastes with little economic value have emerged, and they are projected to lower the cost of biodiesel synthesis.²⁵ Lignocellulosic biomass (LB), one of the several precursors for mixed oxides, is eco-friendly and renewable. The synthesis of heterogeneous catalysts can use LB-derived ash directly as a catalyst or active site support.^26,27 Ash from LB can be used as an alkali heterogeneous catalyst or precursor if it is rich in alkali metal (particularly K, Ca, and Mg) compounds.^28,29 Based on catalyst synthesis conditions (mainly calcination), these compounds can be transformed into a variety of forms, including metal oxides. During basic catalyst synthesis, the basicity (basic strength) of a catalyst is very important parameter.²⁶ K₂O is more strongly basic than CaO and MgO. KNO₃ is a possible source of K₂O which increases the basicity of the catalyst.^30,31 The K and Ca compounds have a synergetic impact on transesterification and higher K content in ashes ensures stronger basicity.²⁵ For instance, Babajide et al. 2010³² prepared a basic catalyst by KNO₃ (which later decomposed into K₂O during calcination) impregnation to silica-rich ash. Similar phenomena were also observed from KNO₃-loaded zeolite.³³ Also, the decomposition of K₂CO₃ and CaCO₃ in the LB ash may be facilitated due to a chemical modification.³⁴

In addition to the calcination condition, the composition of LB-derived ash depends strongly on the LB type.²⁶ Various LBs were investigated for alkali catalyst synthesis. Waste Musa paradisiaca,²⁹Mangifera indica peel,²⁸ pineapple leaves,³⁵ and waste Sesamum indicum plant³⁶ are a few examples. In this study, coffee husk (CH) ash was proposed as a basic catalyst precursor based on its reported chemical composition [K₂O (46.46%) and CaO (17.7%)].³⁷ KNO₃ was also incorporated into the CH ash to make the catalyst composite material (CCM). The influence of CCM calcination temperature on the physicochemical properties and performance of CH-derived catalyst was studied in particular. To the best of our knowledge, the catalytic activity of CH ash and its derivative has been rarely investigated.

Materials and Methods

Materials

WFO was obtained from local pan-fried fish sellers and coffee (Yirga Cheffe) husk was collected from local agro-processing units around Dilla, Ethiopia. Chemicals such as potassium nitrate (98%), phenolphthalein indicator (≥99%), methanol (≥98%), toluene (99.8%), benzoic acid (≥99.5%), potassium hydroxide (85%), sodium hydroxide (97%), and ethanol (≥98%) were purchased from local retailers in Addis Ababa Ethiopia.

Methods

WFO Pretreatment and Characterization

The crude WFO was heated and filtered to remove solid impurities and contaminants. The filtered WFO was further heated to 90 °C for 1 h to remove moisture.³⁸ Then, it was poured into a clean container and maintained sealed for further use. The treated WFO acid and saponification values were determined using ISO 660:2020 and AOCS 920.160 standard procedures, respectively.

CH Preparation and Characterization

A dried fine CH powder was prepared by washing the raw CH with tap water, sun drying for 15 days, oven drying at 105 °C for 3 h, and grinding with a coffee grinder (NIMA NM-124, China). A fine CH powder (≤250 μm) was employed during CH characterization. The CH moisture content was determined by the test approach given in the ASTM standard E1756-08. The ASTM standards E1755-01 and E872-82 were used to determine the ash and volatile matter content of the CH, respectively.

Catalyst Synthesis

The CCM was prepared by a capillary impregnation of KNO₃ onto ash and char mixture (A/C). The A/C was prepared by calcination of CH powder (60 g in a dished crucible) in a muffle furnace (Nabertherm P330, Germany) at 600 °C for 2 h. Adding excess distilled water during KNO₃ impregnation may cause the leaching of K₂CO₃.³⁹ Therefore, titration of the A/C with distilled water was first performed to determine the A/C pore volume. Then, the impregnation was done by mixing the KNO₃ with 12 mL of distilled hot water (35 °C) for 5 min, followed by 12 g of A/C addition into the solution. In comparison to the ash content of A/C, the KNO₃ weight percentage was kept at 25%. After impregnation, the CCM was dried in a static air oven at 105 °C for 4 h. Finally, the CCM was thermally activated with a heating rate of 10 °C/min in the muffle furnace. During the calcination process, the furnace vent system was opened for calcination products and air exchange.

To determine the effect of calcination temperature on the catalyst physicochemical properties, CCM was calcined for 3 h (at 500, 550, 600, and 650 °C) and 2 h (at 700 °C). The catalysts were designated as C-25-500, C-25-550, C-25-600, C-25-650, and C-25-700. For comparison purposes, catalysts were also prepared by calcination of the ash and char mixture (A/C) at 600 and 700 °C for 3 and 2 h, respectively, without KNO₃ impregnation. These catalysts were designated by C-00-600 and C-00-700, respectively. The first and second numbers after the letter C in the catalyst designation stand for KNO₃ weight percentage and CCM calcination temperatures, respectively.

Solid Materials and Product Characterization

For the A/C and CCM, DSC (SKZ1052B, China) analysis under a nitrogen (120 mL/min) environment was done. The temperature range was 25–700 °C at a heating rate of 10 °C/min. A spectrophotometer (CM-600d, Japan) was employed to measure the color of the catalysts. Before performing a color test of each catalyst sample, the spectrophotometer was subjected to zero calibration followed by white calibration. Following calibrations, the average values of three measurements of each sample’s CIE-L*a*b* color space values were recorded. The crystalline phases of the A/C, CCM, and catalysts were distinguished by an X-ray diffractometer (Shimadzu XRD-7000, Japan), with Cu Kα radiation (λ = 0.15406 nm) and operating at 40 kV and 30 mA. The 2θ values range from 10 to 80° with a step size of 0.02° and a scanning speed of 3°/min. To find out which crystalline phases were present, peak 2θ values were contrasted with 2θ values found in the literature. The surface area of catalysts was determined by the surface area analyzer (Horiba Instruments, Inc. SA-9600 series, USA) using the Brunauer–Emmet–Teller (BET) method. Before the surface area analysis, each catalyst was heated overnight at 110 °C in an oven after proper size reduction with mortar and pestle. Furthermore, the catalyst samples were degassed by the surface area analyzer degassing process step at 150 °C for 3 h. Nitrogen was adsorbed by the catalyst in a liquid form by keeping the bath at 77 K. The linear BET curve was generated by measuring the number of N₂ molecules adsorbed at three relative pressures (0.1, 0.2, and 0.3).

The catalysts surface morphology was characterized by a scanning electron microscope (JCM-6000Plus, German). The accelerating voltage was 15 kV. A scanning electron microscope with energy-dispersive X-ray (SEM–EDX) analysis was employed to determine the elemental composition of catalysts. When analyzing catalyst and intermediate products for the identification of functional groups, a Fourier transform infrared spectrometer (Thermo Scientific Nicolet, IiS50 ABX smart iTX) with an attenuated total reflectance ATR (diamond crystal) and DTGS KBr detector was utilized. The scanning range was from 400 to 4000 cm^–1 with a resolution of 2 cm^–1 and 32 numbers of scanning.

Catalysts Basicity Measurement

To assess the total basicity of the catalysts, the Hammett indicator (phenolphthalein) and 0.1 N benzoic acid in toluene as a titer were utilized.⁴⁰ For the first 30 min, a suspension of 0.15 g of catalyst in phenolphthalein solution in toluene (2 mL, 0.1 mg/mL) was agitated. Finally, the benzoic acid solution in toluene (0.01 M) was used to titrate the suspension. The titration endpoint was identified by the mixture’s transformation from pink to colorless.

The leachable basicity determination was started by shaking 0.5 g of catalyst in 50 mL of distilled water for 1 h at room temperature. Then, the solution was filtered to get a clean filtrate free of catalyst particles. The clear filtrate was mixed with a methanol solution of phenolphthalein (5 mL, 0.1 mg/L) and titrated using a toluene solution of benzoic acid (0.01 M).⁴⁰

Transesterification Reaction and Post-Treatment Process

For each catalyst, an activity test was conducted by the transesterification of the crude WFO using a reaction condition stated by Vyas et al. 2009,⁴¹ Ayoola et al. 2020,³⁰ and a preliminary study. A reflux setup comprising a 500 mL three-necked round-bottomed flask reactor and a tap water-cooled condenser was used to carry out the transesterification procedures. A 3 wt % catalyst was first stirred in the reactor with methanol for 30 min at 400 rpm by keeping the reaction temperature at 45 ± 2.5 °C. Once the 30 min methoxide formation time is elapsed, WFO (16.67 g) which was heated to the reaction temperature was added to the reaction mixture and the rpm increased to 600. The transesterification reaction duration was 1 h with an alcohol-to-oil molar ratio of 12:1.³¹ After completion of the stipulated reaction time followed by cooling, the reaction mixture was centrifuged at 1200 rpm for 15 min. The separated phase was aged overnight for further separation of glycerol. The upper layer was extracted by a pipette and washed with hot water (at 55 °C, 2:1 distilled water to biodiesel volume ratio, 20 min, and 4 cycles) to remove the remaining impurities. Then, the biodiesel was dried at 110 °C for 24 h in a static air oven to remove the remaining water.⁴² Purified and dried biodiesel was labeled BID-x depending on the catalyst (C-25-x) type employed, where x belongs to the CCMCT in the range of 500 to 700 °C. The biodiesel synthesized using the C-00-600 catalyst was designated by BID-00-600. The purified biodiesel was then stored in an airtight container for subsequent analysis.

Qualitative and Quantitative Analysis of WFO Conversion

The FTIR spectra regions 1188–1200 cm^–1 (because of −O–CH₃ stretching) and 1425–1447 cm^–1 (because of CH₃ asymmetric bending) were nominated for qualitative analysis of biodiesel yield.^43,44 In these regions, biodiesel had a peak but not the WFO. The dynamic viscosity of biodiesel in mPa s at 40 °C measured using a Rheometer (MCR 102 SN82129736, Austria) was also employed to evaluate the qualitative WFO conversion. The lower viscosity and higher peak integral values imply higher WFO conversion.^44,45 Therefore, the CCM calcination temperature for a high WFO conversion yielding catalyst was selected based on the above qualitative WFO conversion information.

The ¹H NMR analysis was performed for quantitative WFO conversion determination using a Bruker AVANCE TM 400 MHz NMR spectrometer. Deuterated chloroform (CDCl₃) and tetramethyl silicate were used as a solvent and internal standard, respectively.⁴⁶ The quantitative WFO conversion into methyl ester was determined by comparing methoxy protons of the methyl ester and α-methylene protons (eq 1).⁴⁷

1

where X (%) is the conversion of triglycerides of WFO to its fatty acid methyl ester, A is the integral value of methoxy protons of methyl ester, and B is the integral value of the α-methylene protons of methyl ester.

Results and Discussion

Oil Characterization

The acid value of the WFO was 1.56 mg KOH/g oil (i.e., % FFA as oleic acid = 0.78%). It is in the range of WFO acid value as reported by Sarno and Iuliano.⁴⁸ A solid gel was produced while transesterifying the WFO in this study with KOH may be due to its acid value (i.e., acid value > 1 mg_KOH/g_oil).⁴⁹ While utilizing the newly synthesized heterogeneous alkali catalysts, however, the problem was not noticed. The saponification value of the oil was 196.8 mg_KOH/g_oil, which is similar to that of waste palm frying oil reported somewhere.⁵⁰ Based on the empirical correlation,⁵¹ the WFO molecular weight was determined to be 861.75 g/mol.

Proximate Analysis

The moisture, ash, volatile matter, and fixed carbon content of dried CH were 2.42 ± 0.02, 5.26 ± 0.21, 70.02 ± 0.25, and 22.32 ± 0.16% respectively. When it comes to obtaining catalysts from agricultural waste by calcination, the ash amount is crucial. The ash content of the CH in this study was near to the pineapple leaves (6%) and higher than cupuacu seeds (2.6%) ash contents.³⁵ The relation between the other proximate analysis values with LB ash-based catalysts physicochemical properties is not investigated. The volatile matter content of the CH ash (70.02 ± 0.25%) is an indicator for the potential of CH to be used as a source of bio-oil and biogas through pyrolysis and anaerobic digestion process, respectively.^52,53 High peak temperature can also be attained via the combustion of CH.⁵⁴ The fixed carbon content of CH (22.32 ± 0.16%) makes CH a candidate precursor to biochar synthesis for the application of heterogeneous catalyst active site support⁵⁵ and soil fertility amendment.⁵⁶ After calcination of the CH (60 g) at 600 °C for 2 h, 17.83 g (29.72%) of ash and char mixture (A/C) was obtained. The mixture (A/C) contains 17.7% ash and 82.3% char. In the process of (A/C) preparation, energy can be extracted from CH combustion for power generation.⁵⁷ As a result, maximizing the energy extraction while also enhancing the A/C physicochemical characteristics for KNO₃ impregnation is crucial.

Thermal Analysis of Support and CCM

For A/C and CCM, a DSC thermogram in the range of 25–700 °C was generated (Figure 1). The early two endothermic peaks for both A/C and CCM correspond to the vaporization of water and impurities melting transition, respectively.^58,59 The decomposition of organic components induced endothermic peaks at 275.5 °C for CCM and 254.5 °C for A/C.⁵⁹ The melting of the impregnated KNO₃ in the CCM is responsible for the difference between the two thermograms in this region. The next region was due to the decomposition of inorganic constituents, mainly carbonates.⁵⁹ The exothermic peaks at 639.2 and 645.4 °C for A/C and CCM, respectively, are indicators of carbonated hydroxyapatite (CHAp) crystallization.⁶⁰ According to Kim et al. 2000,⁶¹ the recrystallization of CHAp begins at about 600 °C, and then, the crystal size grows as temperature increases. The impregnation of KNO₃ onto A/C was shifted and broadened the CHAp crystallization peak (Figure 1). The CHAp crystallization peak temperature was shifted from 639.2 (A/C) to 645.4 °C (CCM) due to the impregnation of KNO₃ onto A/C. Peak broadening reveals the presence of impurity within the formed CHAp crystals.

Figure 1.

DSC thermogram of materials that contain ash and char mixture (A/C) and CCM.

Catalyst Synthesis Study

At first, KNO₃ impregnation was onto CH ash [synthesized by calcination of CH at 550 °C for 5 h to prepare a catalyst composite material (KNO₃/CHA)]. However, the calcination of KNO₃/CHA for 1 h at 650 °C resulted in a hard solid that was attached to the crucible and was not flowable due to agglomeration (Table 1). This phenomenon obliged changing the CCM synthesis method to KNO₃ impregnation onto ash and char mixture [KNO₃/(A/C)]. The hard solid aggregate formation was not detected while calcining the CCM. Fine powder solid catalysts were obtained from CCM calcination at 500, 550, 600, and 650 °C for 3 h. However, melting was observed approximately after 2 h calcination of CCM at 700 °C (Table 1). For A/C (without KNO₃ impregnation) calcination at 700 °C, the melting problem did not arise (Table 1, 700^b). The reason for melting may be a reduced CCM melting point due to KNO₃ impregnation in addition to the possible CH thermal decomposition products (K₂O, Na₂O, and P₂O₅).⁶² The product obtained after CCM calcination (catalyst) was observed to be heterogeneous in nature. A thin hard layer on the catalyst top surface was observed just after calcination (before homogenization with mortar and pestle) at 550 °C and higher calcination temperatures (Table 1). This layer may be formed by elements (K, Ca, and Na) volatilization and condensation and may resist the easy evacuation of gas products, protecting the CCM from easy combustion. Furthermore, CO₂ residence time could be increased which encourages its reaction with Ca₁₀ (PO₄)₆(OH)₂ to form CHAp (Table 2i). Based on the physicochemical property evaluations and literature reports, possible reactions during the catalyst synthesis process are hypothesized (Table 2).

Table 1. Color Change of Catalysts Heated under Different CCM Calcination Temperatures.

^aKNO₃/CHA calcination.

^bWithout KNO₃ impregnation.

^cCalcination time.

^dKNO₃ wt %.

^eHard solid and cannot be homogenized.

Table 2. Summary of Proposed Reactions Takes Place during KNO₃ Impregnation onto Ash and Char Mixture (A/C) Sample and Calcination of Impregnated Ash and Char Mixture Sample (CCM) at Various Calcination Temperatures.

			during calcination of CCM at various temperatures (°C)
	proposed reactions (rows a up to k)	during KNO3 impregnation onto A/C (#)	500	550	600	650	700	ref.
a	MgCO3 → MgO + CO2	×	√	×	×	×	×	(63)
b	FeS + O2→ Fe3O4 + Fe2O3 + SO2	×	√	×	×	×	√	(63)
c	C + O2 → CO2	×	aLA	√	√	√	√	(63)
d	CaCO3 → CaO + CO2	×	aLA	aLA	√	√	√	(63), (64)
e	K2CO3 → K2O + CO2	×	×	aLA	√	√	√	(64)
f	10Ca(OH)2+3P2O5→ Ca10(PO4)6(OH)2 + 9H2O	√	CR	CR	CR	CR	CR	(61)
g	10CaCO3 + 6H3PO4→ Ca10(PO4)6 (OH)2 + 10CO2+8H2O	√	CR	CR	CR	CR	CR	(65)
h	10Ca(OH)2+6H3PO4 →Ca10(PO4)6 (OH)2 + 8H2O	√	CR	CR	CR	CR	CR	(65)
i	Ca10(PO4)6(OH)2 + CO2 →Ca10 (PO4)6(OH) (CO3) + H2O	×	×	×	×	√	×	(60)
j	FeO + O2 → Fe2O3	×	×	×	×	×	√	(63)
k	Fe3O4 + O2 → Fe2O3	×	×	×	×	×	√	(63)

×: the reaction didn’t happen, aLA: the reaction is proceeding with a little amount, √: the reaction is taking place in a significant amount, CR: the reaction is completed below the stated temperature, ^#: during CCM preparation.

Effect of CCM Calcination Temperature on the Elemental Composition of Catalysts

The EDX results indicated that the catalysts contained metals such as K, Ca, Mg, Na, and Al (Table 3). The peaks for Au and Pd in the EDX spectra (Figure 2) were induced from the sputter layer. The elemental makeup of catalysts was highly altered with the CCM calcination temperature intensity change. Table 3 results show that the potassium (K) amount in the catalyst increased from 8.25 to 24.24 wt % as the CCM calcination temperature increased from 500 to 600 °C. However, an increase in the calcination temperature to 650 °C decreased the amount of K in the catalyst. This result is attributed to the formation of CHAp (Table 2i). Furthermore, calcination of CCM at 700 °C enhanced the decomposition of CHAp which results in an increase of K amount in the catalyst as compared to CCM calcination at 600 °C. Based on the elemental analysis of the catalyst K is the dominant metal in all calcination conditions (Table 3). In the previous studies also the dominant metal present was K in walnut shell,²⁶ Waste Musa paradisiacal plant,²⁹ and waste M. indica peels²⁸ ashes. For a step change of CCM calcination temperature from 500 to 550 °C, the carbon content of the catalyst decreased significantly (from 38.21 to 9.51%). This could be due to a combined effect of temperature changes and the carbon structure oxidative degradation due to KNO₃ decomposition.⁶⁶ The KNO₃ decomposition products may accelerate the combustion process. The increased C content for C-25-650 after continual decrement with temperature increment (Table 3) may be associated with carbonated hydroxyapatite formation and crystallization at 650 °C as stated during the DSC analysis.

Table 3. Elemental Composition of Catalysts from SEM–EDX Analysis.

	composition (wt %)
catalyst	O	C	K	Ca	P	Mg	S	Cl	Na	Al	Pd	Au
C-25-500	51.26	38.21	8.25	1.44	0.33	0.28	0.13	0.09	0.00	0.00	0.00	0.00
C-25-550	64.5	9.51	18.33	4.39	1.92	0.59	0.45	0.22	0.09	0.00	0.00	0.00
C-25-600	61.92	8.68	24.24	3.10	0.77	0.98	0.30	0.00	0.00	0.00	0.00	0.00
C-25-650	64.7	10.22	21.21	1.99	0.25	1.66	0.00	0.00	0.00	0.00	0.00	0.00
C-25-700	61.62	6.72	25.91	3.55	1.16	0.02	0.47	0.29	0.13	0.12	0.00	0.00

Figure 2.

SEM–EDX analysis of catalysts obtained at various calcination temperatures: (a) C-25-500, (b) C-25-550, (c) C-25-600, (d) C-25-650, and (e) C-25-700 catalysts.

Catalyst Color Change with CCM Calcination Temperature

The catalyst color change as a function of CCM calcination temperature was visible (Table 1). Additionally, the spectrophotometer’s CIE L*a*b* values of color measurement were used to quantify the catalysts color intensity (Figure 3). According to the L* values, the catalyst’s brightness increased as the calcination temperature rose from 500 to 600 °C and maximum at 600 °C. The lower brightness at 500 °C was mostly owing to the presence of more carbon material (Table 3). The carbon combustion (Table 2c) increased the brightness, and further temperature increments to 650 and 700 °C lowered the catalyst brightness, possibly due to the increased yellow and red color intensity (b* and a* values, respectively). The b* intensity rise with CCM calcination temperature was possibly due to the increased K₂O (pale yellow color) amount because of potassium compounds decomposition. When compared to the C-25-700 catalyst, the yellow color intensity decrement for C-00-700 indicates that KNO₃ caused more K₂O formation. The red color intensity also increased with temperature and reached the maximum at 700 °C (for CCM calcination). The formation of carbonated hydroxyapatite crystals could also be a factor in the color intensity changes. The rise in red color intensity may also be due to the formation of iron oxide (Fe₂O₃) from the reaction of a trace amount of iron compounds with oxygen (Table 2j,k).⁶³ The C-00-700 was less bright than the C-25-700, with a reduction in red and yellow color intensity, implying that the lack of KNO₃ resulted in less carbon material combustion.

Figure 3.

Color analysis of catalyst obtained at different calcination temperatures using spectrophotometry.

XRD Analysis

The XRD patterns of A/C, CCM, C-00-600, C-25-600, and C-25-650 (Figure 4) were compared to ash catalyst XRD patterns in the reported literature. One of the significant differences between A/C and CCM with the catalysts was the absence of the XRD hump of the amorphous framework from the catalysts XRD pattern. The incorporation of potassium nitrate (KNO₃) into A/C was recognized by XRD peak 2θ values at 23.74, 33.88, and 41.1 for CCM. However, the KNO₃ peak was not visible in the XRD patterns of the catalysts may be due to KNO₃ conversion into K₂O.³² The K₂O peaks are visible in the XRD patterns of the catalysts. XRD peaks appeared around the 2θ value of 52.3 which do not occur in the XRD pattern of the catalysts, revealing the presence of carbon molecules in A/C and CCM.²⁸ In addition to K₂CO₃, the strongest peak at 32.28 for the C-25-650 catalyst also belongs to the carbonated hydroxyapatite crystals.⁶⁷ The absence of fairchildite [K₂Ca (CO₃)₂] crystal peaks in the CCM XRD pattern could be due to fairchildite transitioning to butschliite and subsequently leaching to calcite during KNO₃ impregnation. No CaCO₃ peaks were observed for C-00-600 and C-25-600 catalysts. This happened may be due to CaCO₃ conversion into CaO (i.e., the decomposition temperature of CaCO₃ decreases in the presence of K₂CO₃⁶⁴) or to hydroxyapatite (Table 2g). The kaolinite [Al₂Si₂O₅(OH)₄] crystal peak (around 2θ values of 24) disappearance for C-25-650 was due to its conversion to an amorphous metakaolin.³⁷ The CaCO₃ peak was seen in the C-25-650 XRD pattern due to the adsorption of CO₂ from the atmosphere by CaO. The intensity of major peaks increased except for fairchildite when comparing the XRD pattern of KNO₃ modified and unmodified catalysts (C-00-600 and C-25-600). As stated in the literature, the increased intensity, and narrowed diffraction peaks are indicators of increased crystallinity and crystal size.⁶⁸ One of the positive impacts of KNO₃ was the enhanced crystallinity (high crystallinity can sometimes be a sign of increased catalyst activity¹⁸). However, somewhere stated that the crystal size has an impact on catalysts BET surface area.⁶⁹ Because of carbonate substitutions (during CHAp formation) in the hydroxyapatite lattice (lattice strain), the X-ray diffraction peaks of C-25-650 particles are broader and shorter than those of C-25-600 particles, indicating its weaker crystallinity and small crystal size.⁶⁷

Figure 4.

Comparison of XRD patterns of the mixture of ash and char (A/C), KNO₃-impregnated A/C (CCM), and catalysts obtained from A/C (C-00-600) and CCM (C-25-600 and C-25-650).

Peaks of XRD pattern for C-25-600 (the catalyst that yielded higher WFO conversion) were assigned to their corresponding compounds as follows. At a 2θ value of 25.84, peaks for SiO₂ appeared. The K₂O peaks were observed at 48.32, 50.34, 51.88, 56.76, and 57.54. The peaks at 13.23, 19.48, 20.54, 26.86, 28.06, 33.36, 34.00, and 43.06 belong to K₂Ca(CO₃)₂.²⁵ The presence of the K₂CO₃ was revealed by the XRD peaks at 2θ values of 12.80, 26.86, 29.78, 31.04, 32.28, 34.00, 40.80, and 41.80. Diffraction peaks at 54.22 and 67.44 were due to the CaO presence. Alkali oxides SrO and MgO revealed their existence by inducing XRD peaks at 2θ values of 29.78 and 62.5. Hydrated potassium carbonate (K₂CO₃·1.5H₂O) was also realized based on a similar 2θ value (45.30) reported by Olatundun et al.²⁵ Basumatary, et al. (2021) reported comparable 2θ values for K₂CO₃, K₂O, and KCl in the banana trunk, peel, and rhizome ash catalysts.²⁹ The study by Miladinović et al.²⁶ on a walnut shell catalyst was also considered in assigning 2θ values for SiO₂, CaO, MgO, and K₂O. Nath et al.³⁶ also reported equivalent 2θ values of CaO, KCl, K₂O, and SrO. During the allocation of 2θ values of SiO₂, CaO, and KCl, a report by Basumatary et al.⁷⁰ for heterogeneous catalyst synthesized from sugarcane bagasse was also utilized. A catalyst from waste M. indica peel ash had also similar 2θ values of SiO₂ and CaO.²⁸ The presence of hydroxyapatite crystals was recognized based on the report by Pavankumar et al.⁶⁷ Finally, from the XRD analysis, it was observed that potassium (K) and calcium (Ca) were the abundant components in the form of their oxide, chloride, and carbonate form and played a role in the transesterification of the WFO. The presence of K₂CO₃·1.5H₂O and K₂Ca (CO₃)₂ suggests the ash’s high catalytic potential.²⁵

FTIR Analysis

Comparative examination of CCM and A/C FTIR spectra was used to look into functional group identification and alterations caused by KNO₃ impregnation (Figure 5). The peaks at 1623 (C=C) indicate the presence of aromatic groups.⁷¹ The incorporation of KNO₃ to the A/C was recognized by the new peaks at 1556 and 1367 (N–O vibrations) for CCM.³² The bands at 862 and 867 belong to CO₃ group bending³⁹ and 823, 698, 695, 825, 613, and 565 for A/C and CCM are due to metal–O stretching vibrations.²⁹ The vibration at 3741 for CCM was due to OH stretching vibration in the amorphous hydroxyapatite structure.⁶⁰ The presence of the phosphate group was recognized by P=O and P–O bond stretching vibrations at 1180 and 1130, respectively, for A/C.⁷² These two bands migrated to 1001 and 973, which correspond to the triply degenerate asymmetric and nondegenerate symmetric stretching modes of PO₄^3– (P–O bond), respectively, due to the possibility of the amorphous hydroxyapatite formation (Table 2f–i).⁶⁰

Figure 5.

FTIR spectra of the mixture of ash and char (A/C) and KNO₃ impregnated A/C (CCM).

The catalysts prepared at different calcination temperatures have similar FTIR spectrum patterns with a few differences (Figure 6). The broad peak between 3485 and 2645 for catalysts was due to the stretching mode of OH ions of adsorbed water. The band at 2973 belongs to C–H stretching because of organic constituents.⁷³ Due to KNO₃ breakdown, the N–O stretching band is not visible in the catalysts FTIR spectra.³² Peaks due to the bending mode of O–H–O of lattice water appeared in 1741⁷⁴ and the intensity increased with temperatures between 500 and 650 °C. However, it disappeared at 700 °C due to dehydroxylation.⁶⁰ Dehydroxylation of carbonated hydroxyapatite (CHAp) at 700 °C was also stated elsewhere.⁷⁵ The presence of carbonates was identified by their C–O stretching and bending vibrations peaks at 1444 and 1371, respectively.²⁹ These C–O stretching bands decrease in intensity with temperature due to the decomposition reaction of carbonates. The bands at 1114 and 1039 correspond to the stretching vibration mode of the phosphate group in hydroxyapatite.⁷³ Metal–oxygen vibratory stretching peaks appeared at 563, 619, and 704 due to oxides (K₂O, CaO, and MgO) and carbonates which are responsible for facilitating the transesterification reaction.^25,29 Peaks at 563 and 619 are may also characteristic vibrational modes of the phosphate group.⁷³ The CO₃ group was identified by its characteristic bending peak at 871.³⁹

Figure 6.

FTIR spectra of catalysts obtained from the calcination of impregnated A/C at different calcination temperatures.

BET Surface Area Analysis

It was realized that the BET surface area is inversely proportional to the CCM calcination temperature (Table 4). The highest surface area was for the C-25-500 catalyst. The reduced surface area, when compared to catalysts from previous studies (Table 4), is probably due to pore blockage by particle agglomeration and sintering. The crystal size growth of the catalyst (C-25-600 from XRD analysis) may also initiate pore blockage and reduced the BET surface area.⁶⁹ There was difficulty in determining the BET surface area of the C-25-650 catalyst may be due to its morphology (Figure 7). Because of the stated reason, the surface area may become below the threshold of the sensitivity of the surface area analyzer.

Table 4. Comparison of BET Surface Area and Basicity of the Catalysts with Previous Works.

catalyst name	BET surface area (m2/g)	total basicity (mmol/g)	leachable basicity (mmol/g)	ref.
C-25-500	2.251	0.75	3.11	this work
C-25-550	2.132	0.93	3.50
C-00-600	n.da	1.05	3.32
C-25-600	0.773	1.6	3.93
C-25-650	n.da	1.93	3.94
C-25-700	0.585	2.9	4.21
walnut shell ash	8.8			(26)
waste Sesamum indicum plant ash	3.66			(36)
M. paradisiaca peel ash	4.1	1.43		(29)
trunk ash	6.4	1.59
rhizome ash	7.0	1.39
Mg9Al1(10)		0.51	0.21	(40)
Mg8Zn1Al1		0.13	4.28

^aNot determined.

Figure 7.

SEM micrograph of (a) ash and char mixture (A/C) calcined at 600 °C (C-00-600) and KNO₃ impregnated A/C calcined at different temperatures and (b) C-25-550, (c) C-25-600, (d) C-25-650, and (e) C-25-700 observed at different magnifications.

Basicity Analysis

The total basicity (TB) and leachable basicity (LB) values increased with calcination temperature (Table 4). The TB values are close to the values reported for ash and synthetic mixed oxide base catalysts (Table 4). In all cases, the TB is lower than leachable basicity because of catalysts very low surface area (<3 m²/g). A catalyst with a low surface area masks the alkali sites inside the structure and prevents toluene titration, whereas water washing allows those sites to be leached.⁴⁰ Therefore, the increase in TB with calcination temperature was due to an alkaline site concentration increment on the catalyst surface. However, for higher FFA-containing feedstock, high basic site presence on a catalyst surface will reduce the biodiesel yield.¹⁵ The lower biodiesel yield for the C-25-700 catalyst may be associated with these phenomena in addition to its BET surface area. The increase in leachable basicity with the calcination temperature was due to carbonate decomposition to their corresponding highly water-soluble alkali oxides such as CaO and K₂O.⁶⁴ The carbonate decomposition at lower calcination temperature was may be facilitated due to the presence of carbon for K₂CO₃ and K₂CO₃ for CaCO₃ in the CCM.^64,76 The KNO₃ modification may boost the catalyst basicity in two ways: that is, by facilitating the K₂CO₃ and CaCO₃ decomposition and by acting as a source of K₂O during its decomposition.

Catalyst Morphology Analysis

For each CCM calcination temperature, different morphological structures were observed in the SEM image (Figure 7). The highly porous C-25-550 catalyst, with a particle size of 200 μm, had a unique flake morphology due to pore opening (Figure 7b). The pore opening is caused by oxidative carbon structural (i.e., aliphatic groups) degradation, as indicated by the higher carbon content difference between C-25-500 and C-25-550 in the EDX analysis (Table 3). Pore opening was also observed in biochar activated with an oxidizing agent H₂O₂.⁶⁶ The morphology of C-00-600 (Figure 7a) and C-25-600 (Figure 7c) at a particle size of 200 μm is also porous. The effect of KNO₃ impregnation was exposed by the reduced porosity of C-25-600 when compared to C-00-600. The C-25-650 (Figure 7d) particles were the most tightly packed and large agglomerates among the catalysts may be due to sintering.⁶⁴ Agglomeration may also be associated with CHAp crystallization.⁷⁷ Most of the particles in C-25-650 are large and have nonporous surfaces. The formation of large particles reduces the catalyst surface area significantly. Increasing the calcination temperature from 650 to 700 °C partially recovered the spongy morphology of the catalyst, which may be related to structural and compositional changes.²⁶ There is a close similarity between the morphology of the C-25-600 and ash catalyst from Cocoa pod husk-plantain peel blend, especially at a particle size of 10 μm.²⁹

Catalyst Performance Test

Figure 8 shows the proposed reaction mechanism for the newly synthesized heterogeneous basic catalyst during triglyceride transesterification. In the first step, methoxides are formed by the interaction of methanol mainly with K₂CO₃, K₂O, and CaO which acts directly as a strong nucleophile. The methoxide attacks the carbonyl group of the triglyceride in step 2 to generate a tetrahedral reactive intermediate (step 3). In step 4, the intermediate product is converted into 1 mol of fatty acid methyl ester and diglyceride anion.⁷⁸ Then, the transesterification reactions continue with the reaction of diglyceride and monoglyceride with other methoxide molecules to form 2 mol of fatty acid methyl ester (FAME). The diglyceride and monoglyceride anions are neutralized by the proton on the catalyst surface to form a glycerol molecule.

Figure 8.

KNO₃-loaded CH catalytic transesterification proposed reaction mechanism of WFO.

The regions 1425–1447 cm^–1 (top right corner, CH₃ asymmetric bending) and 1188–1200 cm^–1 (top left corner, O–CH₃ stretching) were absent in oil but present in biodiesel (Figure 9). During the qualitative biodiesel yield analysis, the peak integral values within the 1188–1200 cm^–1 region were used. A maximum peak integral value of 0.38 was obtained for BID-600. The peak integral value for biodiesel produced using a catalyst without KNO₃ modification (BID-00-600) was 0.27. This implies that the KNO₃ modification had a positive effect on the catalyst activity. The qualitative WFO conversion was also analyzed using biodiesel viscosity measurement (Figure 10). Higher oil conversion implies lower biodiesel viscosity. The lowest viscosity was for BID-600. At 40 °C the viscosity of WFO and BID-600 were 38.56 and 7.46 mPa s, respectively. This was also evidence for WFO conversion into a mixture of fatty acid methyl ester (biodiesel). The lower viscosity value for BID-600 implies the higher activity of the C-25-600 catalyst.

Figure 9.

FTIR spectra of WFO and biodiesel synthesized using calcined ash and char mixture (A/C) and impregnated A/C and their FTIR spectra expansion within the region of 1180–1220 cm^–1 (top left corner) and 1420–1460 cm^–1 (top right corner).

Figure 10.

Effect of temperature and catalyst type used for transesterification reaction of WFO on the viscosity of produced biodiesel.

Both viscosity and FTIR peak integral analysis confirmed the higher activity of the C-25-600 catalyst. The catalysts activity was in the order of C-25-600 > C-25-550 > C-25-650 > C-25-500 > C-25-700. Therefore, a CCM calcination temperature of 600 °C was selected for the catalyst synthesis.

Analysis of WFO Methyl Ester Using ¹H NMR

The ¹H NMR spectrum for biodiesel produced using a catalyst without KNO₃ modification of the CH ash (BID-00-600) and with KNO₃ modified CH ash (BID-600) are presented in (Figure 11a) and (Figure 11b), respectively. The peaks were identified by their corresponding integral values (at the bottom of the peaks) and chemical shifts. At a chemical shift of 2.24 to 2.28 and 2.19 to 2.23 biodiesel characteristic peaks [α-methylene protons adjacent to the carbonyl carbon to esters (−CH₂–CO₂R)] were observed with an integral value of 1.31 and 1.24. The singlet sharp peaks with integration values of 1.04 and 1.34 during a chemical shift at 3.62 and 3.57 ppm were due to methoxy (−CO–O–CH₃) protons that indicated the biodiesel formation. The quartet characteristics peaks designated by integral values of 1.00 correspond to olefinic protons (−CH=CH−). At a chemical shift of 2.73 and 2.69 ppm, a triplet with integral values of 0.27 and 0.28 was due to bisallylic (−C=C–CH₂–C=C−) groups of unsaturated fatty acids. The carbonyl carbon β-methylene protons (−CH₂–C–CO₂R) appearance was recognized by their triplet with integral values of 1.47 and 1.45. Signals between 4.08 and 4.28 ppm (glycerol moiety protons of the WFO) are due to incomplete conversion of triglycerides to fatty acid methyl ester.⁷⁹

Figure 11.

¹H NMR spectra synthesized biodiesel obtained by transesterification reaction using calcined: ash and char mixture (A/C) [BID-00-600 (a)] and impregnated A/C [BID-600 (b)].

According to eq 1, the WFO conversion from ¹H NMR analysis for BID-00-600 and BID-600 were 52.92 and 72.04%, respectively. It was a 19.12% conversion difference. Both the ¹H NMR and FTIR analysis confirmed that the modified CH catalyst (C-25-600) has better performance than the unmodified one (C-00-600). The difference in WFO conversion was from the enhanced basicity and crystallinity of the CH ash because of its modification with KNO₃.

The ¹³C NMR spectra of BID-00-600 and BID-600 are presented in Figure 12a,b, respectively. The ¹³C NMR spectrum exhibits signals at 62.20, 69.10 ppm (Figure 12a), and 61.65, 68.67 ppm (Figure 12b), which are respectively attributed to the unconverted WFO and the methylene and methine carbons of the glycerine moiety. The methoxy carbon (−COO–CH₃) of the methyl esters signal appeared at 51.10 ppm for both spectra, confirming the conversion of the WFO into biodiesel. On the other hand, the olefinic carbons contained in the biodiesel were indicated by the signals at 127.62–129.71 ppm in the ¹³C NMR spectra of BID-00-600 and 127.67–129.91 ppm in the ¹³C NMR spectra of BID-600. In the ¹³C NMR study, signals at 172.31, 173.01, and 174.10 ppm (Figure 12a), and 172.45, 172.73, and 173.71 ppm (Figure 12b) also appeared. In both spectra, the first two signals are due to the carbonyl carbon of the triglyceride and the third ones are because of the carbonyl carbon of biodiesel. The signals at 22.31–33.72 ppm (Figure 12a) and 22.5–33.83 ppm (Figure 12b) and 13.62 (Figure 12a) and 13.79 ppm (Figure 12b) ppm are indicating the methylene and methyl carbons of the fatty acid moiety of biodiesel, respectively. The ¹³C NMR spectrum signal intensity that belongs to biodiesel carbon was higher for BID-600 than BID-00-600 and the reverse is true for triglyceride carbon signals. This implies that the WFO conversion is higher when using the C-25-600 catalyst than C-00-600 during the WFO transesterification.

Figure 12.

¹³C NMR spectra of synthesized biodiesel obtained by the transesterification reaction using calcined ash and char mixture (A/C) [BID-00-600 (a)] and impregnated A/C [BID-600 (b)].

Comparison of KNO₃-Loaded CH Ash Catalyst with Other Catalysts

Table 5 presents a comparison of the synthesized catalyst with some reported heterogeneous catalysts. Although the biodiesel yield achieved with other catalysts in Table 5 was higher than the yield achieved in the current study, the reaction conditions were higher. When using KNO₃-loaded CH catalyst, 72.04 wt % yield of biodiesel was obtained at a lower temperature and in a shorter reaction time as compared to other reported catalysts (Table 5). Therefore, the synthesized catalyst has promising potential in the transesterification reaction to produce biodiesel.

Table 5. Comparison of Catalytic Activity of KNO₃-Loaded CH Ash Catalyst with Reported Solid Alkali Catalysts for Transesterification Reaction.

		reaction conditions
catalyst	feedstock	temp. (°C)	catalyst amount (wt %)	methanol to oil molar ratio	time (h)	FAME yield (wt %)	ref.
C-25-600	WFO	45	3	12:1	1	72.04	this work
CaO derived from Mereterix mereterix	used oil	60	3	6.03:1	3	89	(30)
CaO/SCBAa	Jatropha seed oil	65	10	9:1	4.75	92.84	(70)
KNO3/(CaO–MgO)	Neem seed oil	50–55	0.9	12:1	24	78	(31)
KNO3/Al2O3	Jatropha seed oil	70	6	12:1	6	84	(41)

^aSCBA: sugar cane bagasse ash.

Conclusions

CH’s potential as a precursor in the synthesis of a heterogeneous alkali catalyst for the transesterification of WFO was revealed in this study. The activity of the catalyst was highly dependent on the CCM calcination temperature. Maximum WFO conversion (72.04%) was attained at a calcination temperature of 600 °C. Lower WFO conversions below and above 600 °C CCM calcination temperatures were due to lower basicity and reduced surface area of catalysts, respectively. The combined effect of morphology, crystallinity, and basicity, all of which were highly influenced by the calcination temperature, determined the activity of the catalysts. Increasing the calcination temperature increased the catalyst basicity because of the thermal decomposition of compounds into their corresponding oxides (mainly K₂O and CaO). However, the rise in the CCM calcination temperature reduced the catalyst surface area due to the melting and sintering. The KNO₃ impregnation (modification) affected the basicity and morphology of the catalysts. It also had a significant role in altering the catalyst crystallinity. When compared to the unmodified catalyst (C-00-600), the modified catalyst (C-25-600) increased the WFO conversion by 19.12%. Furthermore, the transesterification reaction parameters optimization and reusability of the produced catalyst investigation are in progress.

The authors declare no competing financial interest.