Sonocogreen Decoration of Clinoptilolite by CaO Nanorods as Ecofriendly Catalysts in the Transesterification of Castor Oil into Biodiesel; Response Surface Studies.

A CaO/clinoptilolite green nanocomposite (CaO/Clino) was synthesized by a green modification technique using calcium nitrate and green tea extract. The CaO/Clino nanocomposite promises a total basicity of 4.82 mmol OH/g, surface area of 252.4 m2/g, and ion exchange capacity of 134.3 mequiv/100 g, which qualifies the product as an effective catalyst in the transesterification of castor oil. The transesterification performance of the CaO/Clino catalyst was addressed statistically based on the response surface methodology and central composite rotatable design, considering the essential experimental parameters. Based on the interaction effect between the studied variables, the CaO/Clino catalyst can achieve an experimental biodiesel yield of 93.8% after 2.5 h at 120 °C with 3.5 wt % catalyst loading and 15:1 ethanol/castor oil molar ratio. The optimization function of the design suggested enhancement in the performance of the CaO/Clino catalyst to achieve a yield of 95.4% if the test time interval increased to 2.65 h and the ethanol content increased to 16:1 as a molar ratio to castor oil. The produced biodiesel over CaO/ClinO has acceptable technical qualifications according to the international requirements (EN 14214 and ASTM D-6751). The synthetic green CaO/Clino nanocomposite has better recyclability as a heterogeneous catalyst and higher activity than some investigated catalysts in literature.

Introduction

The essential challenge that faces humanity in the contemporary world is the safety of our environment from the toxic emission in the atmosphere and the environment.[1] Clean and renewable energy resources have become the main interest for scientific, economic, and health organizations as the best solution to control the catastrophic emissions and overcome drawbacks of coal- and petroleum-based fuels.[2,3] Recently, different forms of biofuels and biodiesel were extracted from green precursors including edible and nonedible plant oils, vegetable oils, algae extracts, and animal fats.[1,4−7]

The biodiesel products such as biofuels have no toxic emissions, excellent biodegradability, low CO2 production, and low shoots as byproducts.[8,9] Moreover, they exhibit promising lubricity, viscosity, octane number, and flash point, which qualify them as a direct fuel for the engines or as a mixing blend with petrodiesel.[10,11] Nonedible oils and the waste products of cooking oil were recommended as economical precursors for biodiesel production.[12,13] Pongamia oil, mahua oil, castor oil, linseed oil, and jatropha oil are commonly assessed nonedible oils in the production of biodiesel.[3,7,13,14] From the previously reported nonedible oils, castor oil was investigated extensively for its low cost, high oil content, and the wide distribution of its plants in the different agricultural environments and countries in the world.[13,14]

Biodiesel is fatty acid alkyl esters derived mainly from the oil precursors by transesterification reactions in the existence of alcohol and catalyst (homogenous or heterogeneous).[15,16] In the later periods, the heterogeneous phases of the used catalysts were preferred for several economic, environmental, and technical considerations.[3,17,18] They can be produced at low cost, require simple isolation and separation techniques, possess valuable regeneration capacity, are a pure biodiesel product, and exhibit promising ecofriendly properties and no corrosive effect.[16,19,20] Several heterogeneous catalysts were introduced for the transesterification of castor oil as Ni/doped ZnO,[21] Fe doped ZnO,[22] Li–MgO,[23] Sr–MgO,[24] MgO/MCM-41,[25] and La2O3-modified Y-zeolite.[26]

The availability, the production cost, the activity, the dispersion properties, and the safety are the essential points that direct the orientation of the researcher for the commercial production of the heterogeneous catalysts.[3] The modified forms of minerals and rocks, especially the species which were functionalized by alkali ions (K+, Ca2+, Na+, and Mg2+), were presented as potential basic catalysts that can be produced at low cost and have high international reserves.[3,19] Natural zeolites such as clinoptilolite, mordenite, and sodalite were studied as catalysts in several transesterification systems either as raw materials or in their modified forms.[3,26−28] The modification of such natural zeolitic minerals by alkalis and acids and by the integration of them in composites with nanometal oxides caused significant enhancement of their physiochemical properties and their catalytic performances.[12,13,27]

Natural zeolite minerals such as clinoptilolite are microporous materials of alkali hydrate aluminum silicate composition, high natural reserves, high thermal stability, surface area, significant safety, mechanical strength, and ion replacement capacities, which are the controlled factors of the catalysts.[29,30] Unfortunately, there are little efforts that have been presented to study the catalytic activity of clinoptilolite in any transesterification reactions either in its pure form or the modified form. Moreover, the application of the green treatment methods in the modification of clinoptilolite, especially by alkali bearing salts, was not investigated sufficiently as promising techniques to produce effective and ecofriendly basic catalysts.[3] The green modification techniques which depend on the produced green extracts from plants, vegetables, leaves, and algae were recommended as safe and low-cost methods.[31] The resulted extracts act as strong reducing and stabilizing agents, which gives the formed materials biodegradability, significant reactivity, low agglomeration probabilities, and high surface area.[32]

CaO and the Ca2+-bearing materials were demonstrated also as one of the most promising groups of heterogeneous catalysts.[19,33] They are available materials of low fabrication cost, high basicity, high safety, and low alcohol solubility.[19,34] Therefore, the present study involved a novel synthesis of the green nanocomposite form clinoptilolite decorated with CaO nanorods (CaO/Clino). The synthesis process was conducted by the green alkali-modified process using the green tea extract as a reducing agent at room temperature. The composite was evaluated as an ecofriendly basic catalyst in the transesterification of castor oil. The transesterification study was conducted based on the response surface methodology (RSM) and the central composite rotatable design (CCRD), considering the time interval, CaO/Clino loading, alcohol-to-castor oil molar ratio, and temperature as the essential variables.

Results and discussion

Characterization of the CaO/Clino Green Catalyst

XRD Analysis

The clinoptilolite mineral shows several characteristic peaks at about 9.88, 11.18, 13, 14.9, 17.36, 17.5, 19, 22.36, 22.7, 24, 25, 26, 26.87, 28.1, 30, 31, 32, 32.9, and 34° (JCPDS-card no., 98-000-2606) (Figure A). This demonstrated the purity of the studied natural zeolite sample as a single phase of clinoptilolite with a crystallite size of 56.18 nm. After the green modification of clinoptilolite by calcium nitrate, the pattern validated strong reeducation and disappearance for the zeolite identification peaks (Figure B). Its peaks were observed as much reduced and shifted peaks at 22.71, 28.15, and 30°. Additionally, the formation of CaO was confirmed by detecting intense diffraction peaks at about 32.2° (JCPDS-card no., 00-004-0777) (Figure B). However, the loaded CaO has an essential impact on reducing the detected peaks of clinoptilolite; the expected leaching for Si and Al ions from the zeolite lattice during the reaction has a significant effect in destroying its crystalline structure.

Figure 1

XRD patterns of raw clinoptilolite (A) and the synthetic CaO/Clino green nanocomposite (B).

Scanning Electron Microscopy and Energy-dispersive X-ray Analyses

The surficial features and the morphologies are essential parameters that have a strong effect on the textural properties as well as the reactivity of the solid particles. The raw particles of clinoptilolite were detected in the scanning electron microscopy (SEM) images as compacted particles in cluster forms composed of several flakey grains of observable smooth surfaces (Figure A,B). The modification reactions have a strong effect on the feature of clinoptilolite and appeared in new morphology (Figure C,D). The clinoptilolite surface appeared to be completely covered with nanorods of CaO that have highly oriented properties and form parallel rows to each other (Figure C,D). This has a very strong impact in inducing the surface area and the dispersion properties of the modified product during its application as a catalyst.

Figure 2

SEM images and EDX analysis of raw clinoptilolite (A,B) and SEM images and EDX analysis of the CaO/Clino green nanocomposite (C,D).

The EDX analysis of raw clinoptilolite validated its composition of Si (36.28%), O (46.64%), K (3.43%), Al (7.21%), Fe (3.01%), and Ca (1.98%) (Figure A). The EDX analysis of the CaO/Clino green nanocomposite verified enrichment in the Ca and O content to 7.31 and 50.77%, confirming the formation of CaO as a new phase, in agreement with the concluded results from the X-ray diffraction (XRD) pattern (Figure C). The decline in the Si (28.49%) as well as Al (5.85%) content is an indication of their partial leaching or exchanging by Ca2+ ions during the modification reaction under the influence of either the alkalinity conditions or the reducing effect of the green extract.[35−37]

FTIR Analysis

The Fourier transform infrared spectroscopy (FTIR) spectrum of clinoptilolite showed the presence of T–O and O–T–O (T = Al and Si) (467 1/cm), T–O bonds of the TO4 tetrahedron (1041 1/cm), zeolitic water (1640 1/cm), and OH (3442 1/cm) as the essential functional groups[38,39] (Figure A). The spectrum of the sample after the modification effect (CaO/Clino) displayed noticeable deviation in the previously stated positions of clinoptilolite-related bands in addition to significant intensification in the OH-related band (Figure B). This validated enhancement in the total basicity with the alkali green modification reactions of clinoptilolite. This was reported widely for such types of reactions, as the alkaline solutions have an etching effect on the aluminosilicate minerals which induced the exposure of the basic siloxane groups such as Si–OH and Al–OH[40] (Figure B). Moreover, a new identification band for Ca–O (742 and 517 1/cm) and C–O stretching (1438 1/cm) was detected in the spectrum of CaO/Clino, supporting the previous results about the formation of calcium oxide[19] (Figure B). The C–O group was detected as a carbonation product during the synthesis process.[41]

Figure 3

FTIR spectra of raw clinoptilolite (A) and the CaO/Clino green nanocomposite (B).

Textural and Physicochemical Properties

The essential textural and physicochemical properties of clinoptilolite as well as the CaO/Clino green nanocomposite are presented in Table . The presented measurements validated slight decline in the surface area (258 m2/g), the pore volume (0.041 cm3/g), and the pore diameter (18.3 nm) of clinoptilolite; they were 252.4 m2/g, 0.028 cm3/g, and 15.4 nm for the CaO/Clino green nanocomposite, respectively. This reflected the role of the loaded CaO nanoparticles in filling the structural pores of the zeolite particles, as can be detected from the N2 adsorption desorption isotherm curve (Figure S1).

Table 1

Textural and Physicochemical Parameters of Clino and CaO/Clino Green Nanocomposites

parameters	clinoptilolite	CaO/Clino	spent CaO/Clino
specific surface area (m2/g)	258	252.4	247.4
total volume (mL/g)	0.041	0.028	0.016
average pore size (nm)	18.3	15.4	10.4
cation exchange capacity (mequiv/100 g)	132	134.3	128.5
basicity (mmol OH/g)	3.86	4.82	3.45

On the other hand, the clinoptilolite ion exchange capacity as well as total basicity enhanced after the modification reaction (Table ). The ion exchange capacity slightly improved from 132 mequiv/100 g for clinoptilolite to about 134.3 mequiv/100 g for CaO/Clino, and the total basicity increased significantly from 3.86 mmol OH/g for clinoptilolite to 4.82 mmol OH/g for the CaO/Clino green nanocomposite. Such improvement in the basicity as well as the ion exchange properties is related to the partial destruction and etching effect of the alkaline solutions on the basic units and functional groups of clinoptilolite.

Response Surface Results of Biodiesel Production

Analyses of the Variances and Validation of the Approaches

Conducting the regression analysis for the variance (ANOVA) of the built design is a valuable indication of the validation of the representative model. The built design included random experiments suggested based on the statistical functions of the CCRD (Table ). The effects of the selected variables [(A) time interval, (B) CaO/Clino loading, (C) temperature, and (D) ethanol-to-oil ratio] as well as the interaction effect between them were described according to the second-order quadratic polynomial model.

Table 2

Experimental Runs of the Design and the Determined Value of the Response (Biodiesel Yield)

std	run	time (h) (A)	CaO/Clino loading (wt %) (B)	temperature (°C) (C)	ethanol/oil ratio (%) (D)	biodiesel yield (%) (Y)
11	1	1.00	5.00	60.00	18.00	58.2
5	2	1.00	2.00	120.00	12.00	56.8
21	3	2.50	3.50	60.00	15.00	73.6
4	4	4.00	5.00	60.00	12.00	53.8
19	5	2.50	2.00	60.00	15.00	63.4
7	6	1.00	5.00	120.00	12.00	59.7
22	7	2.50	3.50	120.00	15.00	93.8
16	8	4.00	5.00	120.00	18.00	66.4
23	9	2.50	3.50	90.00	12.00	74.5
1	10	1.00	2.00	60.00	12.00	52.2
2	11	4.00	2.00	60.00	12.00	55.4
15	12	1.00	5.00	120.00	18.00	63.3
25	13	2.50	3.50	90.00	15.00	84.3
20	14	2.50	5.00	90.00	15.00	80.5
18	15	4.00	3.50	90.00	15.00	80.6
24	16	2.50	3.50	90.00	18.00	81.7
14	17	4.00	2.00	120.00	18.00	64.2
12	18	4.00	5.00	60.00	18.00	60.6
13	19	1.00	2.00	120.00	18.00	60.8
9	20	1.00	2.00	60.00	18.00	54.3
3	21	1.00	5.00	60.00	12.00	50.7
6	22	4.00	2.00	120.00	12.00	59.3
26	23	2.50	3.50	90.00	15.00	84.3
8	24	4.00	5.00	120.00	12.00	62.8
17	25	1.00	3.50	90.00	15.00	63.7
10	26	4.00	2.00	60.00	18.00	47.5

Fitting of the actually determined biodiesel yields and the predicted values from the addressed statistical model demonstrated high fitting relation with the determination coefficient higher than 0.93 (Figure S1A). The excellent fitting results validated the significance of the quadratic polynomial model to describe the behaviors of the reactions. Additionally, the normal probability regression plotting for the studentized residuals of the presented biodiesel yields as responses verified normal properties for the model in representing the results (Figure S1B). Moreover, the recognized prediction deviation curve for the suggested test declared the significant prediction accuracy of the selected model (Figure ). The deviation in values of the responses (biodiesel yield) distributed regularly on the positive as well as the negative side of the reference line within the range of −4.14 to 6.64 (Figure ).

Figure 4

Standard deviation of the predicted biodiesel yields for the selected 26 tests.

The values of model-F, sum of squares, lack of fit, and model-Prob > F were delivered from the analysis of the model ANOVA and used to evaluate the adequacy of the used regression model (Table S1). Based on the presented value of model-F (10.24), the model is significant and the probability of the noise effect did not exceed 0.02% (Table ). The values of model-Prob > F are less than 0.05 for all the studied model terms, which reflect the significance of them and nonlinear relations between the response (biodiesel yield) and the affected variables (Table ). The noticeable agreement between the values of Pred R-squared (0.91) and Adj R-squared (0.83) and Adeq precision (10.10) is an adequate signal and reveals the qualification of the selected model to navigate the design space (Table ). Considering the significance of the model, the polynomial regression equation was estimated from the studied model to represent the relation between the four variables and the essential response (biodiesel yield) (eq ).where Y is the response (biodiesel yield), A refers to the interval of the reaction, B refers to the CaO/Clino loading, C refers to the temperature, and D refers to the ethanol-to-oil molar ratio.

Table 3

ANOVA for the Studied CCD

source	sum of squares	DF	mean square	F value	prob > F	significance
model	3459.34	14	247.10	10.24	0.0002	significant
A	53.04	1	53.04	2.20	<0.0001
B	67.42	1	67.42	2.79	<0.0001
C	410.10	1	410.10	16.99	<0.0001
D	56.18	1	56.18	2.33	0.0003
A2	346.56	1	346.56	14.36	0.0301
B2	125.63	1	125.63	5.20	0.3448
C2	6.63	1	6.63	0.27	0.0925
D2	77.43	1	77.43	3.21	<0.0001
AB	5.52	1	5.52	0.23	0.0449
AC	6.50	1	6.50	0.27	0.7796
AD	6.00	1	6.00	0.25	0.2798
BC	2.91	1	2.91	0.12	0.2941
BD	21.16	1	21.16	0.88	0.1853
CD	3.61	1	3.61	0.15	0.5847
residual	265.54	11	24.14
lack of fit	265.54	10	26.55
pure error	0.000	1	0.000
cor. total	3724.88	25
std. dev.	4.91		R2	0.9287
mean	65.63		adj. R2	0.8380
C.V.	7.49		pred. R2	0.9140
PRESS	1437.63		adeq. precision	10.103

Influence of the Affecting Variables and the Interaction Effect

Time Interval and Temperature

The interaction influence of the reaction interval and the adjusted temperature on the transesterification of castor oil into biodiesel over the CaO/Clino nanocomposite is presented by 3D regression curves in Figure A. The interaction impact on the activity of CaO/Clino as catalysts and the resulted biodiesel yields was evaluated, considering the essential variables at their intermediate values (3.5 wt % CaO/Clino loading value and 15:1 as an ethanol-to-oil molar ratio).

Figure 5

Interaction effect between the essential transesterification variables on the biodiesel yield; (A) test time interval and the adjusted temperature, (B) test time interval and CaO/Clino loading, (C) test time interval and used ethanol-to-oil ratio, (D) adjusted temperature and the CaO/Clino loading, (E) adjusted temperature and ethanol-to-oil ratio, and (F) used ethanol-to-oil ratio and the CaO/Clino loading.

The presented results validated enhancement in the experimentally measured biodiesel yields with expanding the test interval from 1 to 2.5 h at any adjusted temperature value (60, 90, and 120 °C) (Figure A). Conducting the test at the lower level of the temperature (60 °C), the yield increased from 55.4 to 73.6% by extending the test interval from 1 to 2.5 h. At 90 °C, the yield percentage increased from 63.7 to 84.3% by expanding the test period from 1 to 2.5 h. Adjusting the temperature at its upper level (120 °C) resulted in biodiesel yields of 73.5 and 93.8% at the test interval of 1 and 2.5 h, respectively. Such results validated the effective role of temperature in the intensification of the influence of the reaction time intervals, and the best impact can be detected at the upper level of the temperature (120 °C). Conducting the reactions for longer intervals induced strongly the miscibility, homogeneity, and interaction chances between the different reactants which enhance the catalytic performance and the conversion rate of castor oil over the CaO/Clino nanocomposite.[34]

Although the longtime intervals have a positive impact on the catalytic activity of CaO/Clino, expanding the interval to 4 h resulted in a reversible effect. The achieved yields declined significantly to 67.5, 80.6, and 83.4% at the experimental temperatures of 60, 90, and 120 °C, respectively. The reversible properties of transesterification as a chemical reaction depend strongly on the adjusted interval of the test. Conducting the tests for time intervals higher than the required conditions might cause hydrolysis for the produced fatty acid ethyl ester (FAEE) esters again to their starting free fatty acids.[1,42]

The previously determined yields at the different temperature values validated its positive influence in inducing the activity of the CaO/Clino catalyst and the conversion rate of castor oil for all the inspected time intervals achieving the best effect by conducting the test for 2.5 h. The endothermic nature of such type of reactions leads the high temperature to have a strong role in accelerating the castor oil transesterification rate.[1] Moreover, the reaction system that is of high temperature displayed a reduction in the oil viscosity and the immiscibility between the liquid reactants (ethanol and castor oil), which induce the homogeneity of the reactions and the interaction chances.[19]

Reaction Time and CaO/Clino Loading

The interaction between CaO/Clino loading and the time interval of the conducted tests was followed for its direct impact on inducing the catalytic reaction either at the short intervals or by using low quantities of the catalyst. The interaction impact on the activity of CaO/Clino and the resulted biodiesel yields were evaluated, considering intermediate values of temperature (90 °C) and ethanol-to-oil molar ratio (15:1). As can be estimated from the 3D curve, the applied CaO/Clino loadings from 2 to 3.5 wt % play a valuable role in intensifying the transesterification rate as a function of the test time interval (Figure B). Also, the influence of the CaO/Clino loading depends strongly on the adjusted time interval of the test, considering the previously estimated best interval (Figure B).

Conducting the test for 1 h resulted in biodiesel yields of 56.7, 63.7, and 60.4% in the presence of CaO/Clino loadings of 2, 3.5, and 5 wt %, respectively. After 2.5 h, the applied loadings of 2, 3.5, and 5 wt % are associated with 70.3, 84.3, 80.56% biodiesel yields, respectively. After 4 h, the obtained yields from the same CaO/Clino loading in order are 63.9, 80.6, and 69.07%. These results indicated a positive effect on the increase in the CaO/Clino loadings from 2 to 3.5 wt %. This related to the predicted increase in the surface area as well as the catalytic sites of CaO/Clino using higher quantities of it which accelerate the conversion rates and efficiencies.[4,7] On the other hand, using the CaO/Clino at loading of 5 wt % caused desirable influence in the efficiency of the reactions and the obtained biodiesel yields. This validated that the best experimental load in the studied system is 3.5 wt % and the over content in the CaO/Clino particles can increase the viscosity and the mass transfer resistance which reduce the mixing homogeneity and the interaction chances.[1,3]

Reaction Time and Ethanol-to-Oil Ratio

The 3D response surface diagram was used to represent the interaction effects between the time intervals of the tests and the selected ethanol-to-castor oil molar ratio (Figure C). The interaction impact on the activity of CaO/Clino and the resulted biodiesel yields was evaluated, considering the intermediate values of temperature (90 °C) and CaO/Clino loading (3.5 wt %) (Figure C). Performing the test for 1 h results in biodiesel yields of 54.4, 63.7, and 58.4 % by using the ethanol content with 12:1, 15:1, and 18:1 as the ethanol-to-castor oil molar ratios. After 2.5 h, the yields at the same ethanol content in order are 74.5, 84.3, and 81.7%, while the time interval of 4 h resulted in yields of 68.08, 80.6, and 70.3%. These yields demonstrated that 15:1 is the best ethanol molar ratio during the transesterification of the castor oil over the CaO/Clino nanocomposite and the higher ratio (18:1) is not preferred in the addressed system. Generally, the alcohol molecules are essential components to confirm the free fatty acids in FAEE.[7] Therefore, including the ethanol at a suitable ratio is vital to preserving the balance of the reaction in addition to its effect in accelerating the mass transfer rate by lowering the immiscibility properties of the system.[3] Although the ethanol content is an essential component to complete the reaction, the over content of it has a significant adverse influence in the determined yields as well as the conversion rate. This behavior was discussed in the literature and was assigned to four reasons involving (1) the role of excess ethanol molecules in deactivating the CaO/Clino active sites, (2) the excess ethanol content will balk the chemical balance of the reaction, (3) such free ethanol molecules caused dissolving for the present glycerol byproducts which are of reversible effect on the reaction, and (4) such free ethanol molecules might be converted into emulsifier centers under the expected inversion of the ethanol polar groups.[1,26,43]

Temperature and the CaO/Clino Loading

The influence of CaO/Clino loading in the efficiency of the castor oil transesterification reaction was investigated at different temperature values to evaluate the interaction effect on the process (Figure D). The interaction impact on the activity of CaO/Clino and the resulted biodiesel yields were addressed at the intermediate values of time (2.5 h) and ethanol-to-castor oil molar ratio (15:1) (Figure D). At 60 °C, the obtained biodiesel from castor oil over CaO/Clino increased by 63.4, 73.6, and 69.5% by upgrading the CaO/Clino loading by 2, 3.5, and 5 wt %, respectively. Conducting the tests at 90 °C resulted in yields of 74.3, 84.3, and 80.56% for the same investigated loadings in order. At 120 °C, the use of CaO/Clino at catalyst loading of 2, 3.5, and 5 wt % upgraded the resulted yields to 77.54, 93.8, and 80.6%, respectively. Such results declare the effective role of the temperature in inducing the catalytic effect of CaO/Clino as a function of the used quantities. Also, the results validated that the best effect of temperature can be obtained by using 3.5 wt % of CaO/Clino as catalyst loading.

Temperature and the Ethanol-to-Oil Ratio

The interaction influence of temperature at different molar ratios of ethanol as the alcohol used was evaluated, considering the test time intervals and the CaO/Clino loading at their intermediate values of 2.5 h and 3.5 wt %, respectively (Figure E). The conducted experiments at 60 °C with 12:1, 15:1, and 18:1 ethanol/castor oil molar ratios achieved yields of 65.13, 73.6, and 68.3%, respectively. At 90 °C, applying the previously reported ethanol ratios in order resulted in yields of 63.7, 84.3, and 80.6%. The experiments which were investigated at 120 °C showed biodiesel yields of 79.5, 93.8, and 84.8% with ethanol-to-castor oil molar ratios of 12:1, 15:1, and 18:1, respectively. This displayed the growing significance of the role of the ethanol content with the rising temperature. Also, the results validated that the best effect of temperature can be obtained at the 15:1 ethanol-to-castor oil ratio.

Ethanol-to-Oil Ratio and the CaO/Clino Loading

The impact of the ethanol ratio in prompting or deactivating the activity of CaO/Clino, considering the used quantities of it in the system, was studied at the intermediate values of the test time interval (2.5 h) and the temperature (90 °C) (Figure F). Adjusting the molar ratio of ethanol at 12:1 resulted in measured yields of 65.15, 74.5, and 69.7% with 2, 3.5, and 5 wt % of CaO/Clino as catalyst loadings, respectively. At the ethanol molar ratio of 15:1, the presence of CaO/Clino at loadings of 2, 3.5, and 5 wt % increase the measured yields to 70.3, 84.3, and 80.56%, respectively. Adjusting the ethanol ratio at 18:1 in the presence of the same CaO/Clino loading in order resulted in measured biodiesel yields of 69.4, 81.7, and 75.6%. The previously presented yields reflected a promising effect in enhancing the basic catalytic activity of CaO/Clino in the range of 12:1 to 15:1 molar ratio and significant deactivation effect at the studied molar ratio of 18:1.

Statistical Optimization

Considering the advantageous options of the quadratic programming and the optimization function of the used software (Design Expert, version 7), the predicted optimum conditions for the best generation of biodiesel from castor oil over CaO/Clino are presented in Table . The optimization conditions were predicted to be limited by the selected levels for the studied variables in addition to the adjusted scheme constraints of the optimization test (Table ). The predicted solutions from the design agree well with the experimentally investigated best conditions (Table ). The suggested solutions have the ability to enhance the achieved yield to 95.4% when the experiment is conducted for 2.65 h at 120 °C using the CaO/Clino at the loading of 3.5 wt % and using the ethanol-to-castor oil molar ratio of 16:1 (Table ). Also, the yield can be raised to 94.1% by using the conditions of 2.6 h, 3.7 wt %, 120 °C, and 15:1 as the test time interval, the used CaO/Clino loading, temperature, and ethanol-to-oil ratio, respectively (Table ).

Table 4

Scheme Constraints of the Optimization Test and Suggested Optimization Solutions for the Best Biodiesel Yields from Castor Oil over the CaO/Clino Catalyst

optimization test scheme constraints
name		lower limit	upper limit	lower coded	upper coded	importance
time	is in range	1 h	4 h	1	1	3
temperature	is in range	60 °C	120 °C	1	1	3
methanol/oil ratio	is in range	12:1	18:1	1	1	3
catalyst loading	is in range	2 wt %	5 wt %	1	1	3

Recyclability Properties

Synthesis of the catalyst with high recyclability is critical for further evaluation of it on the commercial scale. The separated grains for the CaO/Clino nanocomposite after each test were washed three times using methanol as excellent organic solvents for the castor oil or the glycerol molecules that accumulated over its surface. The washing step continued for 15 min, and then they were dried for 12 h at 80 °C to be ready for the next reusing cycle. The experimental conditions in the completed five reusing cycles are 2.5 h, 3.5 wt %, 15:1, and 120 °C as the test time interval, the CaO/Clino loading, the ethanol-to-castor oil molar ratio, and the adjusted temperature, respectively (Figure ).

Figure 6

Recyclability of the CaO/Clino catalyst in the transesterification of castor oil.

The results validated the excellent recyclability properties of the CaO/Clino nanocomposite as a catalyst in the transesterification of castor oil (Figure ). The results showed biodiesel yields higher than 93% for the first two reusing cycles, higher than 90% for four reusing cycles, and higher than 88% for the five reusing cycles (Figure ). The noticeable decline in the determined yields with the consistent repeating of the reusing cycles reflected the increase in the quantities of the adsorbed castor oil or glycerol on the surface of CaO/Clino which hinders the catalytic effect of its active sites.[2] This was also supported by the observed decline in the surface area and pore volume of the spent CaO/Clino catalyst after the five recyclability runs in addition to the significant decrease in the basicity of the catalyst (Table ). Moreover, the FTIR spectrum of the spent catalyst showed a new identification band at 1750 cm–1 and the C=O functional groups, which led to the suggestion of the adsorption of the glycerol and/or the formed esters on the surface of the catalyst (Figure S2). Also, the possible leaching of the Ca2+ ions during the recyclability tests was followed by detecting their content in the biodiesel samples using ICP. It was determined that the content of the leached Ca2+ ions increased from 21 to 52 ppm with increasing the recyclability cycles from cycle 1 to cycle 5. Such leaching for the Ca2+ ions has a direct impact in reducing the efficiency of the CaO/Clino catalyst with increasing the numbers of the reusing cycles.

Technical Qualifications of the Biodiesel Samples

The technical qualifications of the estimated biodiesel from castor oil in the existence of CaO/Clino as the catalyst were evaluated, considering the recommended specifications of ASTM D-6751 and EN 14214 as the commonly used international standards (Table ). The reported properties strongly match the reported specifications for the two standards except for the viscosity as well as the density that appear to be slightly higher values (Table ). This was reported widely for the transesterification of castor oil and assigned to the saturation of the biodiesel sample with the hydroxyl groups of the ricinoleic acid.[13,26] Such a drawback can be solved by integrating the product in blends with the other forms of petrodiesel which will be have a strong influence in reducing the corrosion and wear effect on the engines.[13,26] The determination of the cetane index at a value of more than 45 qualifies the sample to be used as safe fuel in the traditional engines with little toxic emissions.[17] Additionally, the determined flashpoint of the sample reflected the safety of the product for the handling and transportation processes (Table ).

Table 5

Technical Qualification of the Resulted Biodiesel Sample from the Castor Oil over the CaO/Clino Nanocomposite

contents	unit	ASTM D-6751	EN 14214	biodiesel
viscosity	mm2/s	1.9–6	3.5–5	6.2
moisture content	wt (%)	<0.05	<0.05	0.031
flash point	°C	>93	>120	158
calorific value	MJ/kg		>32.9	44.7
cloud point	°C	–3 to 15		5.12
pour point	pp	–5 to 10		3.2
cetane number		≥47	≥51	68.5
density	g/cm3	0.82–0.9	0.86–0.9	0.93
Na + K	mg/kg	≤5	≤5	2.42
acid value	Mg/KOH/g	≤0.5	≤0.5	0.37
iodine value			≤120	82.2

Comparison with Other Catalysts

The activity of the CaO/Clino green nanocomposite as a catalyst in the transesterification of castor oil was compared with the activities of other studied catalysts in literature. Based on the presented biodiesel yields, CaO/Clino shows higher activity than some MgO-based catalysts, ZnO-based catalysts, and lithium-modified zeolite, shell-based calcium oxide, and KOH (Table ). This showed the promising basic catalytic activity of the prepared CaO/Clino to be applied as an effective basic catalyst in the transesterification of nonedible oils as well as edible oils.

Table 6

Comparison between the Catalytic Activity of the CaO/Clino Nanocomposite and Other Catalysts in Literature as a Function of the Obtained Yields

catalyst	time	T (°C)	ethanol/oil ratio	loading (wt %)	yield (%)	references
Mussel shell base	180 min	60	6:1	2	91.17	(45)
iron(II)-doped zinc oxide	50 min	55	12:1	14	91	(21)
alkaline-modified anthill	120 min	64.8	8:1	3	86.34	(46)
MgO/MCM-41	24 h	60		4	85	(25)
Ni-doped ZnO	60 min	55	8:1	11	95.2	(22)
Sr/MgO	30 min	65	12:1	5	93	(24)
MgO	8.3 h	60	6:1	2	18	(47)
NaY zeolite-supported La2O3	50 min	70	15:1	10	84.6	(26)
Li–MgO	2 h	60	12:1	9	93.9	(48)
KOH	60 min	60	12:1	1.25	94.4	(49)
CaO/Clino(experimental)	2.5 h	120	15:1	3.5	93.8	this study
CaO/Clino(theoretical)	2.65 h	120	16:1	3.5	95.4	this study

Mechanism

The catalytic mechanism of the composite can be explained based on the structure and the essential active sites of it (Figure ). The used catalyst represents integration between two essential components including clinoptilolite substrate and green CaO nanorods (Figure ). The clinoptilolite substrate as natural zeolite has microporous properties and potassium ion-rich chemical structure that are effective basic sites during the transesterification reactions. The green treatment of the clinoptilolite by calcium nitrate as alkali solution induced the reactivity as well as the basicity of it either by leaching the structural Si and Al ions or by etching the siloxane groups.[44] The essential alkali ions (K+ and Ca2+) in the structural units of clinoptilolite especially the K+ ions play a major role in the efficiency of the reaction. During the initial steps, the ethanol molecules adsorbed on the surface of the CaO/Clino nanocomposite by its external active sites and this stage also involved strong abstraction of the hydrogen protons of the ethanol molecules (Figure ). The occurred exchange between the ethanol molecules and the alkali ions of clinoptilolite resulted in the formation of CH3O– (alkoxide anion). Additionally, a portion of the existing free fatty acids was adsorbed by the exposed siloxane groups of clinoptilolite (OH earing groups), which resulted in the formation of physical bonds between the carbonyl groups and the siloxane groups. Then, the carbonyl groups were attacked by the generated alkoxide anions (CH3O−) after their protonation, which resulted in the formation of the esters (FAEE) in the system[1] (Figure ). On the other hand, the acidity of the adsorbed ethanol molecules on the surface of CaO/Clino caused the common production of methoxide moieties on its surface (Step 1) (Figure ).

Figure 7

Suggested transesterification mechanism of castor oil over the CaO/Clino green nanocomposite as the basic catalyst.

The other component of the composite is the green synthesized CaO nanorods that are the essential component in the catalyst (Figure ). The generated CaO-related moieties on the surface of the composite prompt the formation of the methoxide moieties. This occurred normally after the stabilization of the alcohol protons and the active basic sites of CaO by the acidic sites of zeolite that are represented by basal oxygen of the structural silica tetrahedron or the alumina octahedron (Figure ).

Conclusions

The green composite from clinoptilolite supported with CaO nanorods (CaO/Clino) was synthesized as a basic catalyst in the transesterification of castor oil. It has enhanced basicity (4.82 mmol OH/g) and promising surface area (252.4 m2/g) and is applied effectively in the conversion of castor oil based on statistical central composite design. Based on the suggested experiments and the interaction effect of the variables, the catalyst realized a biodiesel yield of 93.8 % after 2.5 h at 120 °C with 3.5 wt % CaO/Clino loading and 15:1 ethanol/castor oil molar ratio. Theoretically, the yield can be increased to 95.4% by raising the time to 2.65 h and the ethanol content to 16:1. Additionally, the generated diesel has suitable technical qualifications according to the international requirements. The composite shows significant recyclability as a heterogeneous catalyst and higher activity than some investigated catalysts in the literature.

Experimental Procedures

Materials

Natural zeolite (clinoptilolite) of chemical composition [Al2O3 (11.52%), SiO2 (68.39%), Na2O (0.38%), CaO (1.65%), K2O (4.06%), Fe2O3 (2.676%), MgO (0.483%)] with loss of ignition of 10.99% was used in the synthesis processes. Calcium nitrate tetrahydrate (Ca(NO3)2·4H2O) (Sigma-Aldrich, >97% purity) and extract from commercial green tea were applied in the green treatment reactions. Commercial castor oil and ethanol (Sigma-Aldrich, 99.8%) were applied in all the transesterification tests.

Synthesis of the CaO/Clinoptilolite Green Nanocomposite (CaO/Clino)

As a starting step, the clinoptilolite (Clino) fractions were activated mechanically by continuous grinding for 8 h based on the report by Shaban et al.[50] After that, about 3 g of the Clino sample after the grinding was mixed homogenously with an aqueous solution of calcium nitrate tetrahydrate (Ca(NO3)2·4H2O) (2.5 g Ca salt + 50 mL water) using a magnetic stirrer at 500 rpm and under the sonication waves (240 W). The mixing process continued for about 90 min at room temperature (25 °C) and pH 8. Then, the 50 mL of the green tea extract was added to the Clino/Ca salt mixture (50 mL) and stirred again for 120 min, followed by sonication treatment for an additional 120 min at the same pH and temperature. The extract was prepared by boiling the green tea leaves (5 g) within 100 mL of distilled water for 5 min at 100 °C. After the previous step, the mixture was left for 24 h to confirm the green precipitation of Ca or CaO as well as provide a suitable chance for the interaction between the Ca2+ ions and the surface of clinoptilolite in the presence of the green extract (Figure ). Finally, the solid precipitate was separated from the solution, washed extensively with distilled water, and inserted into the oven at 70 °C for 12 h to be ready for the characterization and the transesterification tests (Figure ).

Figure 8

Schematic diagram for the green synthesis of the CaO/Clino catalyst.

Characterization

A PANalytical X-ray diffractometer (Empyrean) with Cu Kα radiation was used to follow the crystal properties of clinoptilolite before and after the modification in the range of 5 to 70° and at a scanning speed of 5°/min. SEM (Gemini, Zeiss-Ultra 55) was applied to inspect the possible changes in the surface features of clinoptilolite at an accelerating voltage of 30 kV. EDX and Fourier transform infrared spectrometer (FTIR-8400S) were applied to detect the elemental composition and the essential functional groups, respectively. The major oxides of the raw clinoptilolite samples were detected using the Panalytical Axios Advanced XRF technique, at Nuclear Material Authority, Egypt. Texturally, the changes in the surface area and the porosity were observed using a surface area analyzer (SA3100; Beckman Coulter). The total basicity was measured according to the report by Abukhadra and Sayed,[51] by mixing the catalyst with HCl solution for 24 h (50 mL acid + 20 mL water). Then, about 25 mL of the solution was isolated and titrated with 0.1 M NaOH solution in the presence of phenolphthalein. The ion exchange capacities of Clino, as well as CaO/Clino, were determined by the BaCl2 method.[52]

Transesterification System

All the transesterification tests were performed using reactors of a Teflon-lined stainless steel autoclave with a total volume of 150 mL. A magnetic stirrer with a digital hot plate was used for mixing and as the source of the required temperature. The essential variables which were selected to evaluate the transesterification of castor oil over CaO/Clino are the time, the ethanol content, the CaO/Clino loading, and temperature. The experiments were completed based on the suggested variables from the built statistical design using Design Expert Software (version 6.0.5). The design was built based on the RSM and central composite statistical design (CCD). The minimum and maximum values of the variables were selected considering the obtained results from previously accomplished tests (Table S1).

The conducted procedures involved filtration and heating of 40 g castor oil (70 °C for 10 min) to confirm the purification of the sample from any solid suspensions or water molecules. This followed by mixing the oil with the CaO/Clino catalyst at certain loading for an additional 10 min with systematic increase in the temperature until it reached the selected value for the experiment. This was followed by the addition of ethanol at the selected molar ratio for a certain time interval according to the suggested values from the statistical design.

After the test was completed, the catalyst was separated by centrifugation and the liquid components were poured directly into different types of glass separating funnels. The samples were left for 24 h to ensure the complete separation of the glycerol byproducts from the resulted esters (biodiesel) at the bottom of the separating funnel. Then, the separated biodiesel was treated by the temperature at 70 °C for about 5 h to ensure the complete evaporation of the free ethanol molecules. The content of the formed FAEE was measured using a gas chromatography device (Agilent 7890A), and its value was applied directly to calculate the achieved biodiesel yield, according to eq

The biodiesel yield being the target response of the design, its value was fitted with a second-order polynomial equation, eq , as the descriptive equation for the influence of the variables on its valuesY is the achieved yield as the target response, X and X refer to the input variables, β0 is a constant, β is the coefficient linear term, β is the coefficient quadratic term, β is the coefficient cross-product term, and K is the number of the variables.

Analysis of the Biodiesel Samples

The used gas chromatography technique is of Agilent-7890A-type, and the essential analytical procedures using it involved first controlled dilution of extracted biodiesel products using n-hexane. The FAME content in the samples was determined after that utilizing the Agilent-7890A Series gas-chromatograph system connected with split/splitless injection system, flame ionization detector, and DB WAX capillary column (30 m × 0.25 m × 0.25 μm) that contains H2 gas carrier and a split ratio of 100:1. The temperature of the detector as well as the injector was fixed at 280 °C during the analysis steps of the produced biodiesel samples. The temperature of the oven was adjusted first at 120 °C, and then the value was raised regularly at fixed rates of 10 °C for about 50 min until it was 260 °C, which is the maximum temperature. The determination of the FAME involved fixed injection of the biodiesel products at a fixed rate of using of 1 μL in the presence of methyl heptadecanoate as an internal standard.