[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Next Article in Journal
Enhancing Circular Practices in Olive Oil Production: The Role of Green Finance
Previous Article in Journal
Sustainable Style: Unraveling the Trends and Future of Green Marketing in the Textile and Apparel Industry
You seem to have javascript disabled. Please note that many of the page functionalities won't work as expected without javascript enabled.
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sustainability
Volume 17
Issue 1
10.3390/su17010293
sustainability-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

One-Pot Combined Hydrodistillation of Industrial Orange Peel Waste for Essential Oils and Pectin Recovery: A Multi-Objective Optimization Study

by
Jacopo Paini
1,
Giusi Midolo
2,
Francesca Valenti
3 and
Gianluca Ottolina
1,*
1
Institute of Chemical Sciences and Technologies “Giulio Natta”, National Research Council, Via Mario Bianco 9, 20131 Milan, Italy
2
Department of Agriculture, Food and Environment, University of Catania, Via S. Sofia 100, 95123 Catania, Italy
3
Department of Agricultural and Food Sciences, Alma Mater Studiorum-University of Bologna, Viale Giuseppe Fanin 50, 40127 Bologna, Italy
*
Author to whom correspondence should be addressed.
Sustainability 2025, 17(1), 293; https://doi.org/10.3390/su17010293
Submission received: 13 November 2024 / Revised: 16 December 2024 / Accepted: 27 December 2024 / Published: 3 January 2025
(This article belongs to the Special Issue Sustainable Use of Bioactive Compounds: Green Approaches Towards Healthier Outcomes)
Browse Figures
Figure 1
<p>Simplified scheme of the sequential and combined processes for the integrated recovery of limonene and pectin from OPW.</p> "> Figure 2
<p>Response surface plots of selected responses from the models described in <a href="#sustainability-17-00293-t002" class="html-table">Table 2</a> at the optimal RT of 130 min: (<b>A</b>) Essential oil mass (EO) against the effect of LSR and pH, 3D surface, and contour plot; (<b>B</b>) limonene purity (LP) against the effect of LSR and pH, 3D surface, and contour plot; (<b>C</b>) pectin mass (PM) against the effect of LSR and pH, 3D surface, and contour plot; (<b>D</b>) electric consumption (EC) against the effect of RT and the combined effect of LSR and pH; (<b>E</b>) desirability plot with the resulting optimal point at fixed residence time.</p> "> Figure 3
<p>Mass balance flowchart of the combined hydrodistillation process performed at the optimal conditions.</p> ">
Versions Notes

Abstract

:
Sustainable waste management for orange peel waste is a global concern that presents a concomitant opportunity. In this study, a combined process was developed to optimize the simultaneous recovery of pectin, essential oils, and sugars from industrial orange peel waste. The sequential recovery process was used as a benchmark, while a one-pot combined process was optimized through the design of the experiments. A multi-objective desirability function was computed to maximize process performance while balancing opposing optimal conditions. The aim was to find a model able to confidently predict yields while reducing the process environmental footprint, potentially giving the necessary multi-product flexibility in modern biorefining. As a result, the combined process under optimal conditions, liquid-to-solid ratio of 2.5, pH value of 3.7, and residence time of 130 min, yielded 0.52% of essential oils and 11% of pectin on a dry basis. The environmental factor 18 is relevant to the fine chemicals industry, which is the target sector of this study. Finally, the process mass balance was calculated, demonstrating the opportunity to further enhance process environmental sustainability and efficiency by upgrading the resulting solid fraction.

1. Introduction

Citrus fruits are widely consumed worldwide, with about 100 million tons processed annually. Among citrus fruits, oranges account for 60% of the production, making them the most widely consumed citrus fruit. Juice processing generates a highly perishable byproduct, primarily composed of peel, pulp, and seeds, which accounts for more than half of the fruit’s total weight known as orange peel waste (OPW) [1,2,3]. The cell wall of orange fruits mainly consists of three types of carbohydrates, the pectin, cellulose, and hemicellulose, together accounting for up to 70% of the peel dry weight, serving as the primary reservoir for sugars and sugar-derived fermentation products [4,5]. While there may be variations based on the cultivar, the carbohydrate content of OPW is typically characterized by pectin (13.0–42.5%), cellulose (9.2–37.1%), and hemicellulose (4.3–31.1%). Current waste management practices include using OPW as cattle feed. However, its nutritional value is negligible due to its low protein content (i.e., less than 10%). Moreover, landfill disposal is not recommended as it can contribute to groundwater contamination and greenhouse gas (GHG) emissions due to the high chemical and biological oxygen demand [6]. Indeed, research investigations have highlighted the potential environmental challenges associated with OPW. One concern is the risk of water pollution due to biomaterials found in the peels, including sugar, pectin, and essential oils. Conversely, the strategic utilization of these byproducts enhances potential profits while simultaneously safeguarding the environment.
Pectin is a structural polysaccharide present in plants with α-(1 → 4) glycosidic link. The term pectin encompasses several polymers characterized by different molecular weight, chemical configuration, and neutral sugar content. This versatile biomolecule is utilized in the food industry due to its notable gelling properties. Furthermore, due to its biocompatibility, abundance, and cost-effectiveness, the potential as a chemical platform for advanced materials has been recently explored [7]. Advanced applications comprehend biomedical and tissue engineering, electronics, and environmental remediation. Commercial pectin is generally obtained from citrus fruits such as lime, grapefruit, and orange. Therefore, the processing OPW to recover pectin presents a compelling advantage due to its associated reliability and economic value. Conventionally, pectin recovery from fruit peels is achieved through acid hydrolysis. Dilute strong acids, including hydrochloric, nitric, phosphoric, and sulfuric acids, are employed at a mild temperature (ca. 80 °C) to achieve the hydrolysis of pectin side chains [5]. The hydrolysis process primarily targets arabinose residues, followed by galactose and rhamnose residues. Due to the covalent linkage of side chains to the reducing ends of hemicellulose, acid hydrolysis facilitates the recovery of pectin from the hemicellulose/cellulose matrix. Under relatively low pH conditions (i.e., <3), carboxyl groups of pectin are in the ionized form, enhancing the gelling tendency. However, under harsh treatment conditions, including too low pH, high temperature, and extended residence time, the biomolecule may undergo fragmentation, leading to the production of low-quality pectin. Therefore, carefully controlled conditions are crucial for isolating this biomolecule with sufficient quality.
The outermost pigmented layer, or flavedo, of orange peels is abundant in essential oil, in which limonene is usually the predominant compound (90–98% of essential oil). It is a cyclic monoterpene known for its widespread applications in the food, cosmetic, and agricultural industries, serving as flavoring, perfuming, and bio-herbicide agents, as well as a solvent for the extraction of lipids from foods [8]. Hydrodistillation using steam is the most common and environmentally friendly method for extracting essential oils. This method is commonly employed in the chemical industry; it separates and purifies different components of a mixture while allowing for the recycling of byproducts. Therefore, hydrodistillation minimizes waste by enhancing resource utilization efficiency. During the process, the unavoidable presence of an azeotrope leads to a point where the vapor and the liquid phase have the same analyte–solvent composition [9]. Hydrodistillation takes advantage of using water as the solvent, the most environmentally friendly option. This methodology exploits the azeotropic boiling point of a water–limonene mixture (i.e., 97.4 °C) [10]. However, the oxygenation of some monoterpenes, like limonene transforming into carveol, carvone or linalool, is likely to occur during this process [5]. Limonene is an indispensable green molecule, but the production is still limited [11]. Despite its remarkable production cost and high commercial price, limonene is evidently a relevant biomolecule, having also been included in the European Commission’s compilation of the 50 most economically valuable bio-based compounds [12]. Moreover, due to its antimicrobial properties [5,8], this biomolecule may hinder the biotechnological upgrade of OPW (e.g., anaerobic digestion or fermentation). Therefore, the recovery of limonene from OPW would not only generate additional profits but also remove a detrimental compound for its further biotechnological upgrade.
Process development is generally carried out through modelling and algorithmic optimizations, which are recognized for their superior efficiency in terms of cost, time, and material compared to the classic “best-guess approach” [13,14]. The design of experiments (DoE) is a valuable methodology to rationally describe a chemical system through a model. DoE employs a multivariate approach to understand the relationship between selected variables (i.e., process controlling factors) and responses (i.e., product yield or quality) using a mathematical model. This model can predict, with known precision, the response within the experimental domain [15]. To achieve this, it is crucial to initially identify the most influential control factors (screening experimental design); then, determine the most appropriate design for the investigation’s objective (implementation experimental design). After having identified the most influencing factors, a second run of experiments will be conducted to optimize the process. Process optimization is usually a multivariate problem which can be solved through a desirability function [16]. This function essentially transforms each estimated response into a desirability value. Then, the geometric mean of these values provides the overall desirability of the combined response levels to attain the best process conditions.
While the existing literature on OPW valorization primarily focuses on recovering either limonene or pectin individually, in this study, a novel approach is introduced within the context of biomass deconstruction. In this research, the aim is to explore and optimize a process for the simultaneous recovery of simple sugars, pectin, and limonene within an integrated closed loop biorefinery framework.

2. Materials and Methods

2.1. Materials and Reagents

Blood orange peel waste belonging to Tarocco variety (Citrus sinensis) from organic farming was provided by Ortogel Spa, Sicily (Italy). During the industrial juice production process, limonene was mechanically recovered. Therefore, in this study, we wanted to quantify the additional amount of limonene that could be retrieved. The material was initially pretreated with a manual pulper adding distilled water. After separation, an aliquot of the solid fraction was dried in an oven at 50 °C for 48 h to determine the moisture content, while the rest was stored in the freezer until further use. Analytically pure-grade reagents and solvents were utilized.

2.2. Sugar Composition

After the pretreatment, the liquid fraction, Figure 1, was characterized in terms of sugars (sucrose, fructose, and glucose) and quantified by HPLC coupled with ELSD (evaporative light scattering detector). A Rezex monosaccharides RHM Ca2+ column (Phenomenex) was utilized as stationary phase at 80 °C, with ultrapure water as mobile phase at 0.6 mL/min flow rate. Quantification of the characterized sugars was performed by means of a calibration curve with the corresponding standards.

2.3. Sequential Extraction Process

Limonene was recovered by hydrodistillation with a Clevenger apparatus, and pectin was recovered from the remaining slurry after hydrodistillation, as shown in Figure 1. Approximately 100 g of pulped biomass in distilled water, liquid-to-solid ratio (LSR) of 3.5, was heated up to the boiling point for 7 h, the distillate was recovered with chloroform, dried over anhydrous sodium sulfate, and roto-evaporated. The yield of the recovered extracted oils was 9.2 ± 1.4 mgDW/gDW with an average purity of 95.8 ± 1.1%. Quantification of the limonene was performed by means of calibration curve with the corresponding standard. Limonene content in essential oil was evaluated by GC-FID (Agilent Technologies, Santa Clara, CA, USA), while the effective presence of limonene was verified by both GC-MS (Thermo Fisher Scientific, Waltham, MA, USA) and authentic sample analysis.
The temperature program for both analyses was as follows: 80 °C for 1 min, then increased by 8 °C/min to 160 °C, and then increased by 25 °C/min to 240 °C. Operating conditions were detector and inlet temperature at 250 °C, constant flow of 1.2 mL/min, and split of 60:1. Chromatographic (GC-FID) separation was achieved using a DB 5MS column (30 m × 0.25 mm i.d., 0.25 μm film thickness; Mega, Legnano, Italy) with hydrogen as carrier gas. Chromatographic (GC-MS) separation was achieved using Mega-1 MS column (30 m × 0.25 mm i.d., 0.25 μm film; Mega, Italy) with helium as carrier gas.
For pectin recovery, the remaining slurry was acidified to a pH of 1.5 with HCl and heated at 90 °C for 1.5 h in water bath, then filtered with a cheesecloth. The solutions were added as NaOH to a pH of 3.5 and 1.5 volumes of ethanol. After 12 h at 4 °C, the flocculated pectin was recovered by centrifugation and dried at 50 °C for 48 h achieving a 0.17 ± 0.02 gDW/gDW yield of recovered pectin.

2.4. Combined Extraction Process

The pulped biomass was soaked overnight with 300 mL of distilled water according to the predefined LSR (Table 1). Thus, the volume in the reactor headspace was maintained at a constant while controlling the liquid-to-solid ratio by adding different amounts of material. The combined process was performed in a mechanically stirred glass reactor (100 rpm) with sealed headspace connected to a Clevenger apparatus and a condenser for limonene hydrodistillation with cyclopentyl methyl ether (CPME) recovery. Before the combined process began, the pH was determined and adjusted accordingly. Then, the reactor was heated with an oil bath set to 120 °C to ensure water boiling. Electric consumption of both processes was monitored using a wattmeter and the value was recorded at the end of the residence time. The heating phase, lasting about 1 h, was distinguished from the energy consumption needed for the residence time.

2.5. Recovery and Characterization of Pectin

At the end of the process, the reactor was cooled, and the resulting slurry was filtered with a cheesecloth. Pectin was precipitated in the liquid fraction by adding 1.5 volume of 95% ethanol and stored for 24 h at 4 °C. The pectin was recovered by centrifugation and dried at 50 °C for 12 h, and the yield was calculated as follows:
P e c t i n   y i e l d g D W / g O P W _ D W = W e i g h t   o f   d r y   p e c t i n   [ g ] W e i g h t   o f   d r y   O P W   [ g ]
Dry pectin was purified by means of ultrafiltration, 0.5 g was dissolved in 200 mL of distilled water and ultra-filtrated with a 10 kDa cutoff membrane, 3 × 200 mL yielding 72 ± 8% of purified pectin on average. The purified pectin recovered under optimal conditions was analyzed to determine the degree of esterification.

Pectin Degree of Esterification

Fourier transform infrared spectroscopy (FTIR) analysis was carried out to determine the degree of esterification and the identity of the pectin extracted from OPW. The spectra were obtained using a FTIR spectrometer with multiple reflection attenuated total reflectance (ATR) (Jasco, Hachiōji, Japan). Thirty-two scans were taken for each sample in the range 4000 to 500 cm−1 with a resolution of 2 cm−1. The degree of esterification was determined following the methods already described [17,18]. The determination was performed considering the area of the ester carbonyl peak at 1740 cm−1 calculated with a valley-to-valley baseline with the following formula:
D E % = A 1740 A 1740 + A 1630 ∗ 100
To confirm the results obtained by FTIR, the degree of esterification was also enzymatically determined using pectate lyase, E.C. 4.2.2.2, from Aspergillus sp. (Megazyme). Commercial standards with known esterification degrees (from citrus fruit: esterification ≥ 85%, 55–70%, 20–40%.and polygalacturonic acid; Merck) were assayed to build the calibration curve. Standards and samples were solubilized (0.5 g/mL) in 50 mM Tris-HCl buffer (pH 8). The solutions were mixed with 500 µL of 50 mM Tris-HCl buffer, 1 mM CaCl2 (pH 8), 500 µL deionized water, and 500 µL of diluted enzyme (0.01 U in 50 mM Tris HCl buffer). Enzymatic reaction conditions were 40 °C for 30 min, at the end of which, the enzymatic product was evaluated through absorbance at 235 nm [19].

2.6. Multi-Objective Optimization of the Combined Process

Central composite factorial design (CCD) and statistical analysis were performed using the rsm package compiled in R-environment [20]. Three factors and their operational ranges investigated were as follows: (i) LSR with levels from 2.50 to 4.10; (ii) pH with levels ranging from 2.00 to 4.00; and (iii) process residence time (RT) in minutes from 90 to 150, as depicted in Table 1.
The responses selected for the multi objective optimization were process energy consumption EC [kWh], dry weight mass of the essential oil EO [mg], limonene purity LP [%], and dry weight mass of pectin PM [g]. A second order polynomial model was proposed to explain each experimental response, as follows:
Y = b 0 + ∑ i = 1 3 b i x i + ∑ i = 1 3 b i i x i 2 + ∑ i = 1 3 ∑ j = i + 1 3 b i j x i x j
where Y represents the responses; b0, is the intercept; bi, bii, and bij are the regression coefficients; and xi and xj are the independent variables in coded values. Models of the selected responses were evaluated by graphical inspection of the residual graphs and analysis of variance (ANOVA) to select the factor statistically significant with a 90% confidence level (Tables S1–S4, Figures S1–S4). Prediction performance of the models was further statistically analyzed by calculating lack of fit test, coefficient of determination (R2), mean absolute error (MAE), root mean square error (RMSE), and Mallows’ Cp [21].
The multi-objective optimization of the models was achieved by solving the desirability function using the dedicated package in R environment [22]. The identified optimal solution was experimentally validated in triplicate.

3. Results and Discussion

Orange peels, a byproduct of juice production, represent a valuable resource for recovering two useful products, pectin and limonene. Traditionally, these processes are separate. Limonene is extracted during the mechanical juicing process, while pectin is chemically extracted from the remaining peels. In this study, the aim was to investigate the simultaneous recovery of residual limonene still present in orange peel waste after mechanical squeezing and pectin extraction. Furthermore, limonene is known to have antimicrobial properties and OPW biotechnological upgrade is expected to be inhibited [5,8]. Even if essential oils have already been extracted during the juice production process, the combined process was optimized to evaluate its potential considering a closed-loop resource biorefinery [23]. To achieve this goal, in this study, the following was also examined: (i) the recovery of simple sugars obtained from maceration in water after a pulping pretreatment; (ii) the sequential process of limonene and pectin recovery; and (iii) the combined process (Figure 1).
The aim of this study was to test and optimize a process for the recovery of value-added products that can be utilized as raw materials for several supply chains. The liquid fraction obtained after the pulping pretreatment described in Figure 1 is rich in sugars (i.e., sucrose, fructose, sucrose) that can be fermented to obtain value-added products. Indeed, the fraction contained almost 40% by weight of these sugars with a sucrose/glucose/fructose weight ratio of 1.0/3.3/5.8. Hence, the liquid was considered as an important product for its utilization as a fermentative medium. A preliminary study was performed through two separate processes of OPW to recover limonene and pectin. In the first case, called the sequential process (Figure 1), residual limonene, and pectin were recovered in two different stages, instead of in the combined process in Figure 1, the hydrodistillation was carried out in the acidic conditions which are typical for the recovery of the pectin. The resulting process models and statistical analysis workflow is presented and analyzed by response surface methodology. Then, overall optimal conditions of the combined process were tested in triplicate and the recovered products characterized. Finally, the process mass balance was given and compared with the sequential process. It is worth noting that the final residue still contains large amounts of organic material that could be further converted into recoverable bioproducts, thus possibly improving the process. Due to this, the environmental factor (E factor) was calculated to assess the chance for further improvement [24].

3.1. Sequential Extraction Process

While OPW management is critical for achieving sustainable growth due to its valuable bioproducts with diverse applications, the industrialization of OPW upgrading through bioproduct recovery faces challenges. Concerns regarding process efficiency and economic profitability currently limit its widespread adoption [25]. Pectin is a unique biomaterial for several industries like food processing and packaging, nutraceuticals, and cosmetics. Due to this cross-sectorial industrial demand, the pectin market is quickly expanding and pectin recovery from OPW is undeniably a sustainable idea [6].
In the sequential process, 17.3 ± 0.1% of pectin yield was obtained (Figure 1); the yield is in line with rates from orange peels in the literature (i.e., ca. 20.0%) [26,27].
In this study, OPW was already treated through the mechanical extraction of limonene during the industrial juice production process. Therefore, the yields obtained through hydrodistillation in the sequential process were assumed to be the maximum essential oil yields which can be potentially recovered. Despite this, the sequential hydrodistillation yielded 0.93 ± 0.1% of essential oils using chloroform as solvent in extractive work-up. The yield obtained is slightly lower in comparison with the literature, for instance, Chen et al. obtained about 1.0% of essential oils from orange peels [28]. Furthermore, Hilali et al. proposed a solar distillation apparatus to recover essential oils, achieving a maximum yield rate of 1.05% in 120 min compared to a conventional hydrodistillation maximum yield rate of 1.03% after 190 min [29]. The authors demonstrated that essential oil recovery using conventional hydrodistillation was complete after 190 min at room temperature (RT). Therefore, the experimental window for the RT factor in the combined process optimization of this study was chosen to align with these findings.
Nevertheless, higher yields can potentially be obtained by changing the methodology for essential oil recovery. Another type of extraction used is solid–liquid extraction, which is a widely used industrial process that dissolves the components of a solid material using a liquid solvent (maceration). Battista et al. obtained a maximum yield rate of about 1.7% of essential oils through Soxhlet extraction with hexane at 60 °C. Yields were inversely proportional to the operating temperature, in which the yield dropped from 1.3% to 0.28% by changing the temperature from 85 °C to 100 °C, respectively [11]. Despite the potential higher yields in essential oils, the use of hexane as solvent may raise some doubt about the process sustainability not only in economic terms, but also from an environmental perspective. Hexane is a petrochemical solvent which is toxic not only for the environment but also human health. In the EU, hexane use in industries is restricted by several international regulations for authorizing (EC 1907/2006) and safely disposing of this solvent (96/61/EC) [30]. Hence, the environmentally friendlier CPME instead of chloroform was chosen as solvent to separate the diluent (i.e., water), considering the potential future scaling-up and industrial developments, while simultaneously reducing the environmental footprint of the process.
Process environmental assessment must also consider electric consumption, where in the sequential process, electric consumption reaches on average 11.2 ± 1.0 kWh. The consumption is related to the energy required for the hydrodistillation and for maintaining the water bath at the necessary temperature for the pectin recovery step. Therefore, the novelty of this study is the coupling of the heating phases related to the two steps and the evaluation of the balance between process electric consumption while considering the bioproduct thermal degradation through yields.

3.2. One-Pot Combined Hydrodistillation Process Optimization

Generally, essential oils from vegetal matrices are extracted through hydrodistillation or steam distillation while pectin is traditionally extracted at 80 °C in acidic environment for about 1 h [25,30]. To guarantee process energy savings, the development of a one-step process is recommended. However, the advantages of a one-step process may be offset by low recovery yields [5]. Indeed, these molecules are thermally sensitive and highly prone to be chemically modified leading to the degradation of target compounds. Essential oils are recovered at temperatures above 106 °C, but extremely high temperatures may degrade them [25]. To this end, a process optimization for the simultaneous recovery of essential oils and pectin is needed to achieve the highest compound recovery, at the same time minimizing their degradation.
In this study, a full factorial experimental design was used with a liquid-to-solid ratio (LSR), pH, and residence time (RT) as the process controlling conditions to achieve a multi-objective optimization. The experimental domain was chosen to explore the broadest possible region while ensuring the feasibility of the process from an industrial perspective. Process control conditions were selected considering the most influential parameters in both extractions. For example, pectin extraction is governed primarily by pH, while LSR affects the hydrodistillation of essential oils [31] Finally, RT was considered as a process condition to have energy control of the process, since the setpoint temperature remained fixed in all experiments.
Several responses were selected and independently modeled by considering the principle of parsimony [32]. Performance of each model was assessed through the calculation of different statistical metrics and the response surface given in Table 2 and Figure 2. The four models display a good reliable predicting ability as shown by the values of R2 and adjusted R2 (R2Adj). Another statistical metric to consider is that based on residuals (i.e., MAE and RMSE) which need to be as low as possible. A high coefficient of determination with low MAE and RMSE is an acceptable model performance with a trustworthy predicting ability [33]. The last metric in Table 2 is Mallows’ Cp value which is an indicator for the overall model quality by identifying the most accurate model without overfitting, choosing the relative lowest value among the different possible models [34]. Moreover, ANOVA was conducted to evaluate the significance of the models and the results showed that all the models were significant with a level of p < 0.05 (Tables S1–S4, Figures S1–S4).

3.2.1. Model Explaining Essential Oil Mass as Response (EO)

The essential oil (EO) mass model incorporated quadratic terms for both LSR and pH, which control this process parameter. The ANOVA model demonstrates that it is statistically significant with a lack-of-fit test of 0.330. The Mallows’ Cp value indicates that the model is not overfitted and the coefficient of determination (R2) explains 75% of response variability. The suboptimal values of R2Adj RMSE, and MAE reported in Table 2 show that model variability is relatively high. The reason was attributed to the combined effect of the experimental error associated with the roto-evaporation step and the relatively low mass measurement of essential oils. Despite this, the model adequately explains the response. As a result, the effects of the two controlling parameters and their interactions are visually inspected with two-dimensional (2D) contour plots and three-dimensional (3D) response surfaces. The second order model for essential oil mass, EO, as a function of pH and LSR obtained at the optimal RT (i.e., 130 min) is shown in Figure 2A. The resulting 2D and 3D plots are specular to the response surface presented for essential oil yield obtained by ultrasound-assisted acid hydrolysis [31]. This is because the parameters selected in this study (i.e., LSR and pH) are the opposite of those used by Karanicola (solid loading [%] and acid concentration [%], respectively) [35]. The circular shape of the contour plot indicates the non-significant influence of the interaction of the two independent variables (i.e., LSR and pH). EO mass ranged from 10.5 mg (LSR 3.3, pH 3) to 148.5 mg (LSR 2.5, pH 4).
Figure 2A shows that the increase in LSR is unfavorable for EO mass; this may be explained by the two mechanisms of the increasing hydrolytic reactions due to the excessive water content and the heating of OPW [36]. Therefore, the EO exchange rate is influenced by solvent volume and high LSR increases their diffusion due to the different concentrations between the OPW and the solvent [37].

3.2.2. Model Explaining Limonene Purity as Response (LP)

The model for the limonene purity is governed by both LSR, pH linear terms, and pH quadratic term; the pH has the highest influence on the response with an increase in the response at increasing pH (positive influence). The model is statistically significant, low p value, and 83% of response variability is explained with the R2Adj value of 0.795. The lack-of-fit test is 0.306, meaning that the model fits the data well. The model performance benchmark for limonene purity (RMSE, MAE, and Mallows’ Cp) in Table 2 confirmed a model with a trustworthy predicting ability with encouraging performance indicators for the multi-objective optimization. It was observed that a pH below 2 leads to unwanted reactions of limonene, decreasing its purity in essential oils from 93.4% (LSR 2.5, pH 4) to 7.5% (LSR 4.1, pH 2). This was demonstrated by the contour plot for the limonene purity model in Figure 2B. The hypothesis was also confirmed by the fact that terpenes, such as limonene, are unstable under acidic conditions due to the hydration of double bonds [38]. Therefore, in this study, it is revealed that essential oil mass and limonene purity are optimized by maintaining a high pH with a low LSR.

3.2.3. Model Explaining Pectin Mass as Response (PM)

The pectin mass model (PM) is statistically significant, low p value, and the statistical values in Table 2 indicate an acceptable MAE and RMSE. The addition of the LSR quadratic term resulted in the lowest Mallows’ Cp value for the model explaining the pectin mass. Furthermore, the model explains 81% of pectin mass variability (R2) with an R2Adj of 0.770. Pectin mass is equally influenced by LSR and pH linear terms. The negative values of the coefficients indicates that the response increases as both LSR and pH have low values (negative influence). The 2D and 3D plots for the PM model are shown in Figure 2C and demonstrate no interaction among LSR and pH as independent variables. The overall trend confirms that pectin mass varies from 1.1 g to 3.1 g and is maximized at low LSR and pH, i.e., LSR 2.5 and pH of 2. Similar process conditions were reported in the optimization of pectin yield and esterification degree from orange peel [39]. The contrasting optimal pH trends for LP and PM models highlight the necessity for process optimization to maximize the simultaneous recovery of both products.

3.2.4. Model Explaining Process Electric Consumption as Response (EC)

Process residence time had the highest effect in the process electric consumption model. Furthermore, the interaction between LSR and pH slightly influences the response. The ANOVA of the model was statistically significant with a p value. The lack-of-fit statistic showed a p value of 0.310 demonstrating that the model was not significantly different from the observed data. The model showed the highest performance, explaining almost 89% of variability in the process electric consumption. Indeed, the RMSE and MAE low values demonstrated a particular model predicting ability and the Mallows’ Cp value is lower than the number of coefficients indicating a relatively precise model without overfitting. The results from the analysis demonstrate a good model predicting ability. Therefore, the response surface of the model is an acceptable representation of the response trend at varying statistically relevant factors. The contour plot of the electric consumption model, represented in Figure 2D, shows that an increase in the interaction between LSR and pH positively affects electric consumption.
In this study, the fact that process RT is not statistically significant for both products (i.e., pectin and limonene, Table 2) goes against other works found in the literature [35,40]. However, RT is still an important variable for optimizing the combined process since EC is a contributive aspect of the overall process of energetic consumption. To conclude, the analyses of the models demonstrate the need for a trade-off among process conditions to optimize it in terms of the selected responses.

3.3. Process Optimization and Validation of the Optimal Point

To achieve optimal process conditions while reducing energy consumption, the models were transformed into a desirability function. The aim was to maximize the production of limonene and pectin while minimizing energy consumption. Desirability constraints were carefully chosen based on limonene purity (LP) as the most restrictive. Indeed, for each of the selected models the range was from 1 to 200 mg for EO, from 96 to 99.9% for LP, and from 1 to 10 g for PM. The desirability plot shown in Figure 2E as a function of LSR and pH as independent variables at the optimal RT fixed at 130 min shows that the conditions trade-off was identified at an LSR of 2.5 and a pH value of 3.7. The plot also shows that at low pH values, the desirability dramatically decreases, this is because stricter desirability constraints were imposed on the LP model, which is heavily influenced by the pH value, as shown in Table 2. Optimal conditions have been experimentally validated in triplicate and compared against the predicted values obtained from the desirability function. Table 3 summarizes the optimal outputs along with the results obtained from the preliminary sequential extraction process described in Section 3.1.
Overall, the combined process underperforms compared to the sequential process. This is attributed to the different reaction times (RTs): 7 h for the sequential process versus 130 min for the combined process at optimal conditions. Furthermore, each model, except for the LP model, predicted a response within the experimental standard deviation, confirming the good predictive ability of the models (Table 2).
The chromatographic analysis of the essential oils (LP) at the optimal process conditions is shown in Figure S5, demonstrating the presence of other compounds that might result from limonene pH-dependent reactions and co-distillation.
Finally, the recovered pectin obtained at the optimal conditions was qualitatively analyzed by FTIR and the resulting spectrum in a range of 4000–400 cm−1 is represented in Figure S6. Bands at approximately 3500 and 2800 cm−1 were associated with the hydroxyl (-OH) and the -CH bond vibrations, respectively. The presence of several peaks in the latter represents the -CH bond variants (i.e., -CH, -CH2, -CH3) that might be related with to the methyl ester groups of pectin. Then, the peak at around 1740 cm−1 was related to the esterified carbonyl groups of pectin; meanwhile, the peak at 1620 cm−1 was attributed to the non-esterified carboxylate ion stretching and the band at around 1500 cm−1 was assigned to benzene ring vibration [41,42]. Even if OPW contains less than 5% of protein, the shoulders at 1650 cm−1 and 1540 cm−1 were associated to the presence of proteins since they are representative of the amide group [43]. Despite this, it was possible to practically determine the degree of esterification which was slightly greater than 50% [17,18]. The degree of esterification was also confirmed by the pectate lyase enzymatic digestion (51.6%); hence, it was concluded that the pectin recovered can be considered as highly methoxylated.

Mass Balance

The proposed combined process was analyzed through mass balance to assess the conversion of OPW into the value-added products. The combined process, as depicted in Figure 1 and detailed in Figure 3, yielded 5.2 mg/gDW, (i.e., 0.5%) essential oils with an average limonene purity of 92% at the optimal process conditions (Table 3). In comparison, the sequential process yielded almost twice the amount of EOs with a slightly higher purity (i.e., 96%). After solid liquid separation, pectin yield was about 11%, representing approximately two thirds of the yield obtained from the sequential process (i.e., 16–17% approximately, see Table 3). In addition, the aim of this work was to evaluate the chance of finding the optimal conditions for limonene and pectin recovery during the same process. It is important to highlight that the optimal RT of the combined process was 130 min more than three times lower than the RT in the sequential hydrodistillation, without considering the RT needed for the separated pectin extraction step. Therefore, an effective optimization was achieved in terms of processing time and energy. The model presented here shows the chance of predicting the yields with acceptable confidence while considering process electric consumption. This feature can potentially give the necessary multi-product flexibility in modern biorefining to accommodate market needs. Sustainability can be viewed as another multivariate optimization study by selecting a few metrics to guide resource efficiency. Several metrics have been proposed aimed at increasing sustainability by focusing on the total mass of material used or on waste generated for the process (e.g., E factor). If the increase in resource efficiency has been advised from a business perspective because it would maximize value, then minimizing waste is the final goal in assessing process environmental impact [44]. The environmental factor (E factor) is calculated through the difference in the input dry mass and the mass of target products, divided by the target product mass [24]. Process environmental impact is preferred when the E factor approaches zero. The combined process mass balance flowchart, shown in Figure 3, shows that half of the dry weight input material goes on average in the solid fraction.
This fraction is supposed to be mainly composed of cellulose, which represents an additional valuable resource, potentially increasing process sustainability. Therefore, the E factor was calculated considering the solid fraction as target product. The resulting value was therefore 18, in line with the E factor related to the fine chemicals industrial segment, which would be the target sector for the products obtained in this study. Furthermore, a life cycle assessment (LCA) of the combined process is presented in a dedicated study [45].

4. Conclusions

In this study, DoE was applied to optimize LSR, pH, and RT for the simultaneous recovery of limonene and pectin through a combined one-pot acid hydrodistillation process. The aim was to find a compromise between the potentially differing optimal conditions for maximizing the recovery of both limonene and pectin. This was achieved through multi-objective optimization and modeling for improved process management. Even if the yield obtained for limonene and pectin are lower in the combined process than the sequential one, the whole process is faster and more efficient in terms of energy consumption due to the strict requirements set in the desirability function. The combined hydrodistillation was further optimized to reduce process electric consumption while assessing the quantitative and qualitative characteristics of the recovered products in terms of purity and degree of esterification. The resulting model of the combined process has a satisfactory prediction ability of the outcomes, paving the way for multi-purpose biorefineries. Further improvements in this model could require the integration of machine learning models with DoE to better address industrial needs.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/su17010293/s1, Figure S1: essential oil (EO) mass model diagnostic plots; Figure S2: limonene purity (LP) model diagnostic plots; Figure S3: pectin mass (PM) model diagnostic plots; Figure S4: process electric consumption (EC) model diagnostic plots; Figure S5: essential oil extract GC/FID chromatogram obtained at the optimal conditions. LSR 2.5; pH 3.7; RT 130 min; Figure S6: FTIR profile of pectin obtained at the optimal conditions. LSR 2.5; pH 3.7; RT 130 min. Relevant bands are highlighted and named; Table S1: ANOVA and model assessment of essential oils (EOs) mass model; Table S2: ANOVA and model assessment of limonene purity (LP) model; Table S3: ANOVA and model assessment of pectin mass (PM) model; Table S4: ANOVA and model assessment of process electric consumption (EC) model.

Author Contributions

Conceptualization, J.P. and G.O.; methodology, J.P., G.M. and F.V.; validation, J.P. and G.O.; formal analysis, J.P., G.M. and F.V.; investigation, J.P.; resources, F.V. and G.O.; data curation, G.M. and F.V.; writing—original draft preparation, J.P.; writing—review and editing, F.V. and G.O.; visualization, J.P. and G.O.; supervision, G.O.; project administration, G.O.; funding acquisition, G.O. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by Fondazione Cariplo, Milan, Italy [grant number 2020-1070] CIRCLE—CItrus waste ReCycLing for added value products.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Data are contained within the article or Supplementary Materials.

Conflicts of Interest

The authors declare no conflicts of interest. The funders had no role in the design of this study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

References

  1. Citrus Fruit Statistics—European Commission. Available online: https://agriculture.ec.europa.eu/data-and-analysis/markets/overviews/market-observatories/fruit-and-vegetables/citrus-fruit-statistics_en (accessed on 21 November 2024).
  2. Citrus: World Markets and Trade|USDA Foreign Agricultural Service. Available online: https://fas.usda.gov/data/citrus-world-markets-and-trade (accessed on 25 July 2024).
  3. FAO. Citrus Fruit Fresh and Processed Statistical Bulletin 2020; FAO: Roma, Italy, 2021. [Google Scholar]
  4. Jeong, D.; Park, H.; Jang, B.-K.; Ju, Y.; Shin, M.H.; Oh, E.J.; Lee, E.J.; Kim, S.R. Recent Advances in the Biological Valorization of Citrus Peel Waste into Fuels and Chemicals. Bioresour. Technol. 2021, 323, 124603. [Google Scholar] [CrossRef] [PubMed]
  5. Kim, I.J.; Jeong, D.; Kim, S.R. Upstream Processes of Citrus Fruit Waste Biorefinery for Complete Valorization. Bioresour. Technol. 2022, 362, 127776. [Google Scholar] [CrossRef] [PubMed]
  6. Sharma, P.; Vishvakarma, R.; Gautam, K.; Vimal, A.; Kumar Gaur, V.; Farooqui, A.; Varjani, S.; Younis, K. Valorization of Citrus Peel Waste for the Sustainable Production of Value-Added Products. Bioresour. Technol. 2022, 351, 127064. [Google Scholar] [CrossRef] [PubMed]
  7. Benassi, L.; Alessandri, I.; Vassalini, I. Assessing Green Methods for Pectin Extraction from Waste Orange Peels. Molecules 2021, 26, 1766. [Google Scholar] [CrossRef] [PubMed]
  8. Siddiqui, S.A.; Pahmeyer, M.J.; Assadpour, E.; Jafari, S.M. Extraction and Purification of D-Limonene from Orange Peel Wastes: Recent Advances. Ind. Crops Prod. 2022, 177, 114484. [Google Scholar] [CrossRef]
  9. Zhong, J.; Cheng, H.; Dai, Y.; Jiao, Y.; Wang, K.; Xin, L.; Zhang, Y.; Zhu, Z.; Cui, P.; Lu, Y.; et al. Design and Multiple Performance Evaluation of Green Sustainable Process for Azeotropes Separation via Extractive Distillation. ACS Sustain. Chem. Eng. 2023, 11, 16849–16881. [Google Scholar] [CrossRef]
  10. Valentini, F.; Vaccaro, L. Azeotropes as Powerful Tool for Waste Minimization in Industry and Chemical Processes. Molecules 2020, 25, 5264. [Google Scholar] [CrossRef]
  11. Battista, F.; Remelli, G.; Zanzoni, S.; Bolzonella, D. Valorisation of Residual Orange Peels: Limonene Recovery, Volatile Fatty Acids and Biogas Productions. ACS Sustain. Chem. Eng. 2020, 8, 6834–6843. [Google Scholar] [CrossRef]
  12. European Commission; Joint Research Centre; Ronzon, T.; Lammens, T.; Spekreijse, J.; Vis, M.; Parisi, C. Insights into the European Market for Bio-Based Chemicals: Analysis Based on 10 Key Product Categories; Publications Office of the European Union: Luxembourg, 2019; ISBN 978-92-79-98420-4. [Google Scholar]
  13. Montgomery, D.C. Design and Analysis of Experiments; John Wiley & Sons, Inc.: Hoboken, NJ, USA, 2012; Volume 3, ISBN 978-1-119-11347-8. [Google Scholar]
  14. Taylor, C.J.; Pomberger, A.; Felton, K.C.; Grainger, R.; Barecka, M.; Chamberlain, T.W.; Bourne, R.A.; Johnson, C.N.; Lapkin, A.A. A Brief Introduction to Chemical Reaction Optimization. Chem. Rev. 2023, 123, 3089–3126. [Google Scholar] [CrossRef]
  15. Benedetti, B.; Caponigro, V.; Ardini, F. Experimental Design Step by Step: A Practical Guide for Beginners. Crit. Rev. Anal. Chem. 2022, 52, 1015–1028. [Google Scholar] [CrossRef]
  16. Derringer, G.; Suich, R. Simultaneous Optimization of Several Response Variables. J. Qual. Technol. 1980, 12, 214–219. [Google Scholar] [CrossRef]
  17. Gnanasambandam, R.; Proctor, A. Determination of Pectin Degree of Esterification by Diffuse Reflectance Fourier Transform Infrared Spectroscopy. Food Chem. 2000, 68, 327–332. [Google Scholar] [CrossRef]
  18. Kyomugasho, C.; Christiaens, S.; Shpigelman, A.; Van Loey, A.M.; Hendrickx, M.E. FT-IR Spectroscopy, a Reliable Method for Routine Analysis of the Degree of Methylesterification of Pectin in Different Fruit- and Vegetable-Based Matrices. Food Chem. 2015, 176, 82–90. [Google Scholar] [CrossRef]
  19. Hansen, K.M.; Thuesen, A.B.; Soderberg, J.R. Enzyme Assay for Identification of Pectin and Pectin Derivatives, Based on Recombinant Pectate Lyase. J. AOAC Int. 2001, 84, 1851–1854. [Google Scholar] [CrossRef]
  20. Lenth, R.V. Response-Surface Methods in R, Using Rsm. J. Stat. Soft. 2009, 32, 1–17. [Google Scholar] [CrossRef]
  21. Mallows, C.L. Some Comments on CP. Technometrics 1973, 15, 661–675. [Google Scholar] [CrossRef]
  22. Kuhn, M. Desirability: Function Optimization and Ranking via Desirability Functions; R Package Vignette: Madison, WI, USA, 2016. [Google Scholar]
  23. Błaszczyk, A.; Sady, S.; Pachołek, B.; Jakubowska, D.; Grzybowska-Brzezińska, M.; Krzywonos, M.; Popek, S. Sustainable Management Strategies for Fruit Processing Byproducts for Biorefineries: A Review. Sustainability 2024, 16, 1717. [Google Scholar] [CrossRef]
  24. Sheldon, R.A. The E Factor: Fifteen Years On. Green Chem. 2007, 9, 1273. [Google Scholar] [CrossRef]
  25. Leong, Y.K.; Chang, J.-S. Valorization of Fruit Wastes for Circular Bioeconomy: Current Advances, Challenges, and Opportunities. Bioresour. Technol. 2022, 359, 127459. [Google Scholar] [CrossRef]
  26. Chen, Q.; Hu, Z.; Yao, F.Y.-D.; Liang, H. Study of Two-Stage Microwave Extraction of Essential Oil and Pectin from Pomelo Peels. LWT-Food Sci. Technol. 2016, 66, 538–545. [Google Scholar] [CrossRef]
  27. Fidalgo, A.; Ciriminna, R.; Carnaroglio, D.; Tamburino, A.; Cravotto, G.; Grillo, G.; Ilharco, L.M.; Pagliaro, M. Eco-Friendly Extraction of Pectin and Essential Oils from Orange and Lemon Peels. ACS Sustain. Chem. Eng. 2016, 4, 2243–2251. [Google Scholar] [CrossRef]
  28. Chen, Y.; Wu, J.; Xu, Y.; Fu, M.; Xiao, G. Effect of Second Cooling on the Chemical Components of Essential Oils from Orange Peel (Citrus sinensis). J. Agric. Food Chem. 2014, 62, 8786–8790. [Google Scholar] [CrossRef]
  29. Hilali, S.; Fabiano-Tixier, A.S.; Ruiz, K.; Hejjaj, A.; Ait Nouh, F.; Idlimam, A.; Bily, A.; Mandi, L.; Chemat, F. Green Extraction of Essential Oils, Polyphenols, and Pectins from Orange Peel Employing Solar Energy: Toward a Zero-Waste Biorefinery. ACS Sustain. Chem. Eng. 2019, 7, 11815–11822. [Google Scholar] [CrossRef]
  30. Ozturk, B.; Winterburn, J.; Gonzalez-Miquel, M. Orange Peel Waste Valorisation through Limonene Extraction Using Bio-Based Solvents. Biochem. Eng. J. 2019, 151, 107298. [Google Scholar] [CrossRef]
  31. Lucchesi, M.E.; Chemat, F.; Smadja, J. Solvent-Free Microwave Extraction of Essential Oil from Aromatic Herbs: Comparison with Conventional Hydro-Distillation. J. Chromatogr. A 2004, 1043, 323–327. [Google Scholar] [CrossRef]
  32. Laird, J. The Law of Parsimony. Monist 1919, 29, 321–344. [Google Scholar] [CrossRef]
  33. Jha, A.K.; Sit, N. Comparison of Response Surface Methodology (RSM) and Artificial Neural Network (ANN) Modelling for Supercritical Fluid Extraction of Phytochemicals from Terminalia Chebula Pulp and Optimization Using RSM Coupled with Desirability Function (DF) and Genetic Algorithm (GA) and ANN with GA. Ind. Crops Prod. 2021, 170, 113769. [Google Scholar] [CrossRef]
  34. Boisbunon, A.; Canu, S.; Fourdrinier, D.; Strawderman, W.; Wells, M.T. Akaike’s Information Criterion, Cp and Estimators of Loss for Elliptically Symmetric Distributions. Int. Stat. Rev. 2014, 82, 422–439. [Google Scholar] [CrossRef]
  35. Karanicola, P.; Patsalou, M.; Stergiou, P.Y.; Kavallieratou, A.; Evripidou, N.; Christou, P.; Panagiotou, G.; Damianou, C.; Papamichael, E.M.; Koutinas, M. Ultrasound-Assisted Dilute Acid Hydrolysis for Production of Essential Oils, Pectin and Bacterial Cellulose via a Citrus Processing Waste Biorefinery. Bioresour. Technol. 2021, 342, 126010. [Google Scholar] [CrossRef]
  36. Mollaei, S.; Sedighi, F.; Habibi, B.; Hazrati, S.; Asgharian, P. Extraction of Essential Oils of Ferulago Angulata with Microwave-Assisted Hydrodistillation. Ind. Crops Prod. 2019, 137, 43–51. [Google Scholar] [CrossRef]
  37. Chen, F.; Jia, J.; Zhang, Q.; Yang, L.; Gu, H. Isolation of Essential Oil from the Leaves of Polygonum Viscosum Buch-Ham. Using Microwave-Assisted Enzyme Pretreatment Followed by Microwave Hydrodistillation Concatenated with Liquid–Liquid Extraction. Ind. Crops Prod. 2018, 112, 327–341. [Google Scholar] [CrossRef]
  38. Clark, B.C.; Chamblee, T.S. Acid-Catalyzed Reactions of Citrus Oils and Other Terpene-Containing Flavors. In Developments in Food Science; Charalambous, G., Ed.; Off-Flavors in Foods and Beverages; Elsevier: Amsterdam, The Netherlands, 1992; Volume 28, pp. 229–285. [Google Scholar]
  39. Duwee, Y.S.; Kiew, P.L.; Yeoh, W.M. Multi-Objective Optimization of Pectin Extraction from Orange Peel via Response Surface Methodology: Yield and Degree of Esterification. Food Meas. 2022, 16, 1710–1724. [Google Scholar] [CrossRef]
  40. Ye, R.; Tian, K.; Hu, H.; Li, P.; Tian, X. Extraction Process Optimization of Essential Oil from Mellissa officinalis L. Using a New Ultrasound-Microwave Hybrid-Assisted Clevenger Hydrodistillation. Ind. Crops Prod. 2023, 203, 117165. [Google Scholar] [CrossRef]
  41. Fernández-Delgado, M.; del Amo-Mateos, E.; Coca, M.; López-Linares, J.C.; García-Cubero, M.T.; Lucas, S. Enhancement of Industrial Pectin Production from Sugar Beet Pulp by the Integration of Surfactants in Ultrasound-Assisted Extraction Followed by Diafiltration/Ultrafiltration. Ind. Crops Prod. 2023, 194, 116304. [Google Scholar] [CrossRef]
  42. Gandolfi, S.; Ottolina, G.; Riva, S.; Fantoni, G.P.; Patel, I. Complete Chemical Analysis of Carmagnola Hemp Hurds and Structural Features of Its Components. BioResources 2013, 8, 2641–2656. [Google Scholar] [CrossRef]
  43. Ortiz-Sanchez, M.; Solarte-Toro, J.C.; Orrego-Alzate, C.E.; Acosta-Medina, C.D.; Cardona-Alzate, C.A. Integral Use of Orange Peel Waste through the Biorefinery Concept: An Experimental, Technical, Energy, and Economic Assessment. Biomass Conv. Bioref. 2021, 11, 645–659. [Google Scholar] [CrossRef]
  44. Jimenez-Gonzalez, C.; Ponder, C.S.; Broxterman, Q.B.; Manley, J.B. Using the Right Green Yardstick: Why Process Mass Intensity Is Used in the Pharmaceutical Industry to Drive More Sustainable Processes. Org. Process Res. Dev. 2011, 15, 912–917. [Google Scholar] [CrossRef]
  45. Midolo, G.; Cutuli, G.; Porto, S.M.C.; Ottolina, G.; Paini, J.; Valenti, F. LCA Analysis for Assessing Environmenstal Sustainability of New Biobased Chemicals by Valorising Citrus Waste. Sci. Rep. 2024, 14, 21418. [Google Scholar] [CrossRef]
Sustainability 17 00293 g001
Figure 1. Simplified scheme of the sequential and combined processes for the integrated recovery of limonene and pectin from OPW.
Figure 1. Simplified scheme of the sequential and combined processes for the integrated recovery of limonene and pectin from OPW.
Sustainability 17 00293 g001
Sustainability 17 00293 g002
Figure 2. Response surface plots of selected responses from the models described in Table 2 at the optimal RT of 130 min: (A) Essential oil mass (EO) against the effect of LSR and pH, 3D surface, and contour plot; (B) limonene purity (LP) against the effect of LSR and pH, 3D surface, and contour plot; (C) pectin mass (PM) against the effect of LSR and pH, 3D surface, and contour plot; (D) electric consumption (EC) against the effect of RT and the combined effect of LSR and pH; (E) desirability plot with the resulting optimal point at fixed residence time.
Figure 2. Response surface plots of selected responses from the models described in Table 2 at the optimal RT of 130 min: (A) Essential oil mass (EO) against the effect of LSR and pH, 3D surface, and contour plot; (B) limonene purity (LP) against the effect of LSR and pH, 3D surface, and contour plot; (C) pectin mass (PM) against the effect of LSR and pH, 3D surface, and contour plot; (D) electric consumption (EC) against the effect of RT and the combined effect of LSR and pH; (E) desirability plot with the resulting optimal point at fixed residence time.
Sustainability 17 00293 g002
Sustainability 17 00293 g003
Figure 3. Mass balance flowchart of the combined hydrodistillation process performed at the optimal conditions.
Figure 3. Mass balance flowchart of the combined hydrodistillation process performed at the optimal conditions.
Sustainability 17 00293 g003
Table 1. Experimental conditions from central composite design (CCD). Liquid-to-solid ratio, LSR natural and coded units; pH, natural and coded units, and residence time (min), RT natural and coded units. Block 1 is the factorial points (standard order 1–8) and central points (standard order 9–11); Block 2 is the axial points (α = 1.67332).
Table 1. Experimental conditions from central composite design (CCD). Liquid-to-solid ratio, LSR natural and coded units; pH, natural and coded units, and residence time (min), RT natural and coded units. Block 1 is the factorial points (standard order 1–8) and central points (standard order 9–11); Block 2 is the axial points (α = 1.67332).
Std OrderRun OrderLSRpHRTBlock
152.5−1.002−1.0090−1.001
2104.11.002−1.0090−1.001
382.5−1.0041.0090−1.001
414.11.0041.0090−1.001
522.5−1.002−1.001501.001
674.11.002−1.001501.001
762.5−1.0041.001501.001
844.11.0041.001501.001
933.30.0030.001200.001
1093.30.0030.001200.001
11113.30.0030.001200.001
12151.96−1.6730.001200.002
13124.641.6730.001200.002
14173.30.001.33−1.671200.002
15183.30.004.671.671200.002
16143.30.0030.0069.8−1.672
17163.30.0030.00170.21.672
18133.30.0030.001200.002
Table 2. Regression equations and analysis of the selected responses related to the selected independent coded variables (LSR; pH and RT). Essential oil mass (EO [mg]), limonene purity (EP [%]), pectin mass (PM [g]), and electric consumption (EC [kW]); coefficient of determination (R2); adjusted coefficient of determination (R2Adj); root mean square error (RMSE); mean absolute error (MAE); Mallows’ Cp (Cp).
Table 2. Regression equations and analysis of the selected responses related to the selected independent coded variables (LSR; pH and RT). Essential oil mass (EO [mg]), limonene purity (EP [%]), pectin mass (PM [g]), and electric consumption (EC [kW]); coefficient of determination (R2); adjusted coefficient of determination (R2Adj); root mean square error (RMSE); mean absolute error (MAE); Mallows’ Cp (Cp).
ResponseModelR2R2AdjRMSEMaECp
EO38.3 − 23.8 × LSR − 0.7 × pH + 16.2 × LSR2 + 24.8 × pH20.7540.67819.04614.3311.378
LP73.9 − 8.7 × LSR + 27.4 × pH − 8.7 × pH20.8310.79511.78810.2742.699
PM1.6 − 0.4 × LSR − 0.4 × pH + 0.2 × LSR20.8110.7700.2500.2291.339
EC2.7 + 1.0 × RT + 0.3 × LSR × pH0.8900.8760.3770.2861.284
Table 3. Predicted and experimental values at the optimal conditions (i.e., LSR of 2.5, pH value of 3.7, and RT of 130 min.) along with the confidence interval and standard deviation for each response selected in this study.
Table 3. Predicted and experimental values at the optimal conditions (i.e., LSR of 2.5, pH value of 3.7, and RT of 130 min.) along with the confidence interval and standard deviation for each response selected in this study.
Sequential ProcessCombined Process
ExperimentalPredictedExperimental
EOg/100 gDW0.93 ± 0.10.56 ± 0.10.52 ± 0.1
LP%95.8 ± 1.1100.0 ± 5.192.4 ± 0.3
PMg/100 gDW17.3 ± 2.511.0 ± 0.910.6 ± 1.2
ECkWh11.2 ± 1.02.7 ± 0.92.6 ± 0.4
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Paini, J.; Midolo, G.; Valenti, F.; Ottolina, G. One-Pot Combined Hydrodistillation of Industrial Orange Peel Waste for Essential Oils and Pectin Recovery: A Multi-Objective Optimization Study. Sustainability 2025, 17, 293. https://doi.org/10.3390/su17010293

AMA Style

Paini J, Midolo G, Valenti F, Ottolina G. One-Pot Combined Hydrodistillation of Industrial Orange Peel Waste for Essential Oils and Pectin Recovery: A Multi-Objective Optimization Study. Sustainability. 2025; 17(1):293. https://doi.org/10.3390/su17010293

Chicago/Turabian Style

Paini, Jacopo, Giusi Midolo, Francesca Valenti, and Gianluca Ottolina. 2025. "One-Pot Combined Hydrodistillation of Industrial Orange Peel Waste for Essential Oils and Pectin Recovery: A Multi-Objective Optimization Study" Sustainability 17, no. 1: 293. https://doi.org/10.3390/su17010293

APA Style

Paini, J., Midolo, G., Valenti, F., & Ottolina, G. (2025). One-Pot Combined Hydrodistillation of Industrial Orange Peel Waste for Essential Oils and Pectin Recovery: A Multi-Objective Optimization Study. Sustainability, 17(1), 293. https://doi.org/10.3390/su17010293

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop