A Novel Integrated Biorefinery for the Valorization of Residual Cardoon Biomass: Overview of Technologies and Process Simulation

Abstract

Lignocellulosic biomass is currently widely used in many biorefining processes. The full exploitation of biomass from uncultivated or even marginal lands for the production of biobased chemicals has deserved huge attention in the last few years. Among the sustainable biomass-based value chains, cardoon crops could be a feedstock for biorefineries as they can grow on marginal lands and be used as raw material for multipurpose exploitation, including seeds, roots, and epigeous lignocellulosic solid residue. This work focused on the technical analysis of a novel integrated flowsheet for the exploitation of the lignocellulosic fraction through the assessment of thermochemical, biochemical, and extractive technologies and processes. In particular, high-yield thermochemical processes (gasification), innovative biotechnological processes (syngas fermentation to ethanol), and extractive/catalyzed processes for the valorization of cardoon roots to FDCA and residual solid biomass were modeled and simulated. Inulin conversion to 2,5-Furandicarboxylic acid was the main conversion route taken into consideration. Finally, the novel process flowsheet, treating 130,000 t/y of residual biomass and integrating all proposed technologies, was modeled and assessed using process simulation tools to achieve overall mass and energy balances for comparison with alternative options. The results indicated that cardoon biorefining through the proposed flowsheet can produce, per 1000 tons of input dry biomass, 211 kg of 2,5-Furandicarboxylic acid and 140 kg of ethanol through biomass gasification followed by syngas fermentation. Furthermore, a pre-feasibility analysis was conducted, revealing significant and potentially disruptive results in terms of environmental impact (with 40 kt_CO2eq saved) and economic feasibility (with an annual gross profit of EUR 30 M/y).

1. Introduction

In the last decade, the need to increase sustainability in all of the productive sectors has led to the development of novel industrial models making use of bioresources. The biobased industry is characterized by an initial prevalent growth of biofuel-driven processes and, more recently, by the development of novel biorefineries for the full exploitation of biomass through cascade valorization [1]. Lignocellulosic biomass includes many vegetable matrices that can be exploited for both the production of energy and value-added products [2]. The use of dryland farming to grow biomass crops is considered sustainable as they can be produced on lands that are not in competition with the food chain. Among these, cardoon (Cynara cardunculus L.) is a perennial biomass native to the Mediterranean area that has different species, some of which have high adaptability to marginal soils and low water and nutrient requirements [3]. Cardoon is typically cultivated to produce seed oils destined for many industrial sectors [4]. Its lignocellulosic components can be divided into two main parts: the epigeal biomass, including the head and the stems, and the hypogeal biomass made up of the underground roots [5]. Bioplastics can be produced by extracting oil from cardoon seeds, while lignocellulosic residue coming from stems and roots can be valorized to produce energy and/or valuable products. The overall lignocellulosic quantity is about 80–90% of the total weight collected, and it can be valorized through various conversion processes, including combustion, producing thermal and electrical energy, conversion to ethanol or other chemicals, e.g., 1,4-Butanediol (BDO), through a sugars platform [6], and gasification [7].

In particular, the gasification process uses sub-stoichiometric conditions to convert biomass into a mixture containing hydrogen (H₂) and carbon monoxide (CO), along with additional gas components in different proportions depending on the technology and the gasifying agent. Syngas can be used in many applications, including modified internal combustion engines and fuel cells for the production of energy, as well as catalytic reactors for the production of biofuels and chemicals [8,9]. These chemical processes are the more traditional conversion route of syngas. However, fermentation processes that convert syngas into alcohols or other chemical building blocks have emerged more recently. Significant investments have been made by Lanzatech to produce ethanol from industrial off-gases (e.g., from steel mills) [10]. In principle, the advantage of syngas fermentation concerning the sugars platform lies in the complete conversion of the biomass biopolymers, including lignin. On the other hand, as the fermentation of syngas is a heterogeneous process carried out by acetogenic bacteria under strictly anaerobic conditions that requires suitable bioreactor geometry, its overall theoretical efficiency could be lower than that of chemical processes [11]. Besides the annual yield of epigeal biomass in the range of 12–14 t/ha, cardoon crops also provide a yield of roots in the range of 3–4 t/ha, corresponding to the management of pluriannual crops [12]. Cardoon roots contain up to 36% inulin water-soluble sugars, over 80% of which are inulin [13]. The latter is a polysaccharide made of fructose units, with a degree of polymerization ranging from 2 to 60. It can be easily extracted and hydrolyzed to achieve oligomers and fructose for both nutraceutical and chemical applications [14]. In the chemical industry, fructose could be the precursor of the 5-Hydroxymethylfurfural (HMF) and its derivative product, 2,5-Furandicarboxylic acid (FDCA), an important chemical intermediate in novel polyester synthesis. One of the newly investigated polyesters is PolyEthylene 2,5-Furanoate (PEF), which has emerged as a standout candidate due to its excellent barrier properties coupled with its enhanced thermal and mechanical characteristics compared to PET. These attributes position Avantium to significantly disrupt the plastic packaging material industry, extending its influence across the value chain from plant-based feedstock to multiple end applications, such as packaging (e.g., bottles, trays, and pouches), textiles, films, and various everyday items. The potential market for these applications is substantial, estimated to be worth over USD 200 billion annually [15]. Furthermore, PEF’s structural similarity to terephthalic acid (TPA) makes it a viable substitute for fossil-based TPA, a key compound in polymer and resin production, with a potential market size of several hundred million euros per year [16]. In addition to PEF, ethanol can be utilized to produce a range of derived biochemicals with greater added value and a lower market size, as well as for energy-driven applications (fuels and additives), which have lower added value but a very large market size. Several promising biochemicals have emerged in recent years and are expected to see high demand in the future. Plant cell wall chemistry-based products, such as ethylene (Braskem Inc., São Paulo, Brazil), isobutanol (Gevo Inc., Englewood, CO, USA), farnesene (Amyris Inc., Barra Bonita, Brazil), epichlorohydrin (PTT, Map Ta Phut, Thailand), p-xylene (Virent, Madison, WI, USA), acrylic acid and adipic acid (ADM and BASF, Ludwigshafen, Germany), 5-HMF (AVA Biochem, Zug, Switzerland), and others, could play a pivotal role in the bioeconomy. Many of these chemicals are considered base or platform chemicals, which serve as intermediates for developing a wide array of household chemical commodities [17]. Brazil is a leading country in the bioethanol industry, with public policies that promote the further use of ethanol, potentially making it more advantageous and widely used than gasoline. The ongoing tax reform discussions in Brazil could also incorporate these considerations to improve energy taxation on different fuels. Enhancing the use of ethanol through carbon pricing strategies could include imposing a carbon tax on carbon-intensive fuels. Other policies might seek to implement a mandated ethanol blend [18].

An important step consists of exploiting the processes and technologies that valorize residual cardoon biomass to produce these compounds. Our research group is actively developing several experimental activities for all three main processes studied in this work. However, while conducting small-scale experimental validation is ideal, resource constraints currently limit our capacity for such studies. Furthermore, small-scale experimental data may not be representative of an industrial scale, making it challenging to base a techno-economic–environmental analysis on such data. Also, for this reason, in the present paper, a novel integrated biorefinery layout was investigated for the conversion of lignocellulosic biomass from cardoon to biobased products. The process layout included the extraction of inulin from roots followed by its conversion to FDCA via HMF. The lignocellulosic residues of roots after extraction were then mixed with the epigeal biomass and fed to the gasification unit. The syngas mixture containing CO and H₂ in a ratio of 1:1 was converted to ethanol through syngas fermentation. Finally, the unconverted syngas was used to produce heat and electricity. In the scope of building up an overall mass and energy balance, the latest technological developments of the selected processes were considered and coupled with process simulations.

2. Inulin Extraction and FDCA Production

2.1. Cardoon Root Processing

Cardoon roots can be considered a natural source of inulin that is simple to extract and available in large quantities (more than 30%wt). Due to the limited favorable period for harvesting the roots, storage plays a key role in the supply chain. The complete drying of cardoon roots before storage has to be proven to prevent the degradation of inulin during up to 6 months of storage. Because drying is a costly operation, energy saving and the effect on the percentage of inulin content in cardoon roots dried at different moisture contents before storage have to be assessed [19].

2.2. Inulin Exctraction

Raccuia et al. [20] studied inulin extraction from different C. Cardunculus L. genotypes. They examined the yields of roots and the total biomass per hectare (t_DM/ha) for each genotype. Starting from these results, an average value of the weight percentage of roots with respect to the total biomass was calculated and used to determine the amount of annual roots processed in the simulation. Furthermore, it was taken into account that the roots are harvested every three years, unlike the epigeal biomass, which is instead taken from the field annually. The composition of Cynara cardunculus roots may vary according to several factors, including variety, region, harvest season, climate conditions, and time of storage [12]. Raccuia et al. [20] also reported the total sugar content of the roots and its characterization on average for all of the genotypes.

Inulin is also present in many plants of the Asteraceae family, of which chicory (Cichorium intybus L.) and Jerusalem artichoke (Helianthus tuberosus L.) are the most interesting for its production. Moreover, extraction from the various natural substrates has been the subject of numerous studies and applications [12,21].

Inulin is a fructose-based oligosaccharide with a degree of polymerization ranging from 2 to 60 [22] Some methods, such as those based on hot water extraction, are traditionally used on an industrial level, while others involve more innovative techniques and are still on a lab scale. The hot water extraction of inulin consists of two steps: extraction and purification. The first step is carried out at 70–80 °C and lasts 1 to 2 h. The juice generated from the initial operation contains impurities and must therefore undergo several purification steps. These steps include liming and carbonation for protein degradation and the elimination of peptides and colloids, the use of cationic and anionic ion exchange resins for demineralization, and active carbon treatment for decolorization. Then, after filtration, evaporation, and, finally, spray-drying, the inulin is obtained from the juice [21].

2.3. Inulin Hydrolysis, HMF, and FDCA Production

From inulin hydrolysis, fructose can be produced. This sugar is currently used in the food industry as an ingredient in functional foods due to its nutritional properties. However, within the context of a biorefinery, it could serve as an interesting intermediate, acting as a precursor to HMF and FDCA, which are potential substitutes for petroleum-based building blocks [11].

The production of HMF from inulin can generally take place either through direct conversion of inulin to HMF or through the production of fructose through inulin hydrolysis and the subsequent dehydration of this sugar to HMF. Inulin hydrolysis involves two approaches: chemical hydrolysis and enzymatic hydrolysis. In chemical hydrolysis, diluted or concentrated acids, like HCl, H₂SO₄, etc., are used. This simple method has important advantages, such as a low-cost and easily available acid catalyst and a short hydrolysis time, but it also has the drawback of causing corrosion and environmental impacts. On the other hand, enzymatic hydrolysis involves an environmentally friendly process and low corrosion problems. Still, the high enzyme cost, long processing time, and feedback inhibition represent its major disadvantage [23]. To date, many studies have been carried out on the dehydration of fructose to HMF. Catalytic reaction systems, such as solid acid catalysis, magnetic carbon-based catalysis, and photocatalysis, have been investigated, and the yield obtained ranged between 94 and 70%. The problems of all of these approaches were related to low product selectivity, difficulty in separating the product from the catalytic system, and many by-products [24]. On the other hand, limited information on the direct conversion of inulin into HMF is available. Among these, Li et al. [25] studied the effects of different factors on the dehydration to HMF of fructose and inulin, such as a heterogeneous catalyst, substrate loadings, the reaction temperature, and time. They used CM-SO₃H, an SO₃H-functionalized carbonaceous material, as the catalyst, and the ionic liquid 1-butyl-3-methylimidazolium chloride ([BMIM][Cl]) as the solvent. Under the optimum conditions, they obtained a yield of 59% for the direct dehydration of inulin compared to 83.5% obtained with fructose as the substrate. Concerning the synthesis of FDCA from HMF, the oxidation route has received increased attention in recent years. Heterogeneous metal catalysts based on noble metals have been investigated, showing excellent performance in FDCA production. However, these catalysts have significant drawbacks, including high costs and environmental impact. Electrocatalysts and biocatalysts were also considered in many studies to test green and low-cost alternative catalysts. Both types of biocatalysis methods, enzyme and whole cell catalysis, attracted more attention due to their simple processes and high selectivity, with the latter also having the advantage of avoiding the purification of the enzyme [24]. In the present paper, FDCA was produced in a GVL/H₂O system. The advantage of this system is the higher solubility of FDC compared to commonly used solvents. Moreover, because FDCA is an effective catalyst for fructose dehydration, this process eliminated the use of a corrosive (acid) one.

3. Syngas Fermentation and Ethanol Production

3.1. Cardoon Lignocellulosic Residues’ Gasification

A number of papers have investigated the use of Cynara cardunculus residual biomass for energy production via gasification. Due to its high volatile matter content at more than 75%, Cynara C. is indeed particularly suitable for gasification and subsequent conversion in syngas. First, Encinar and Gonzalez [26] studied the steam gasification of Cynara C. in a fixed bed gasifier, with an energy yield of the gasification process between 0.5 and 0.8. Temperature and partial water pressure exerted positive effects on the main parameters of the process, increasing the reaction rate, the gas yield and production, the conversion, and the energy generated per kilogram of initial residue. The main gases generated were H₂, CH₄, CO, and CO₂, with a higher heating value (HHV) between 10 and 11 MJ/Nm³. The gas in the greatest proportion was H₂, and the gas in the smallest proportion was methane. The H₂/CO ratio in the exit gas varied between approximately 3 and 8. The experimental results show that the water–gas shift reaction is the main determinant of the composition of the gases. For this reaction, an increase in temperature leads to a greater formation of CO, and an increase in partial water pressure leads to a greater formation of CO₂. Because steam gasification produces a gas with low nitrogen content and high hydrogen content, it can be utilized in fuel cell units for electricity production. If the CO content is high, the gas can also serve as feedstock for Fischer–Tropsch synthesis [27]. Zabaniotou et al. [28] also investigated syngas production from Cynara c. in a pilot-scale fixed-bed gasifier (FBG) at different equivalent ratios and temperatures in the range of 700–900 °C at atmospheric pressure. The syngas produced was rich in H₂ (11–28% volume fraction, N₂ free) and CO, with an LHV of 9.84 MJ/m³. The experimental results of Cynara C. gasification showed that the yield (v/v%) of syngas was optimized at a higher temperature and a lower λ, while for H₂ the maximum yield in v/v% was obtained at a lower temperature. Syngas (H₂ + CO) reached 72.9 v/v% with a lower heating value (LHV) of 9.84 MJ/Nm³ (based on N₂ free gas) at a temperature of 900 °C and λ = 0.2. H₂ achieved a syngas yield of 22 v/v% at 700 °C and λ = 0.2. According to the results, producer gas is recommended for H₂ production rather than for CHP [29]. Serrano et al. [30] investigated the use of magnesite and olivine as bed materials in a bubbling fluidized bed reactor (BFB) and obtained a gas rich in hydrogen content (26–30% volume fraction, N₂-free). Both studies used air as a gasifying agent. Christodoulou et al. [7] used an O₂ steam and magnesite as the bed material and achieved a gas with 34.5% and 55.4% (volume fraction, N₂-free), respectively, H₂ and CO₂ and a value of 5.3 MJ/Nm³ for LHV. Bed agglomeration is one of the problems when using cardoon biomass. The melting point of cardoon ash is low (460 °C), and because the ash is rich in alkaline elements (K, Ca) both promote agglomeration problems. In the present paper, experimental data of syngas derived from Cynara cardunculus gasification performed in a bubbling fluidized bed using air as a gasifying agent and olivine as the bed material were considered to feed the syngas fermentation section. This choice offers high gas yield and lower nitrogen concentration; furthermore, the main compounds are balanced with the same volumetric percentage [30].

3.2. Syngas Fermentation

Syngas fermentation is a more recent biorefinery approach for the production of biobased products [31]. The main advantage of this biorefinery scheme is that gasification can transform up to 90% of the biomass into a fermentable platform (syngas) [32]. Depending on the type of micro-organism used in fermentation, several bioproducts, like ethanol, acetic acid, PHA, and BDO, can be obtained [33]. The limits of this process are the need for high-quality syngas and the bioreactor geometry. In particular, in these bioreactors, the syngas compounds (CO, CO₂, H₂) with very low solubility in water need to be transferred to the liquid phase, where the fermentation reaction takes place through micro-organisms [34]. In recent years, several studies have focused on developing models or process simulations for syngas fermentation to produce ethanol or other chemicals. Almeida Benalcázar et al. constructed a thermodynamics-based black box model of the primary microbial reactions integrated with a mass-transfer-based model of a bubble column bioreactor [35], and they analyzed the gas production processes as the main process parameter of syngas fermentation [11]. More innovative bioreactor configurations were studied by Jang et al. [36], who considered a hollow fiber membrane bioreactor (HFMBR) for microbial CO conversion to ethanol. Inhibition of Clostridium autoethanogenum cell growth was observed to be proportional to the electrolyte concentration. However, kinetic simulations predicted that overall reactor performance using the electrolytes increased when compared to results obtained in an acid-buffered basal medium. This suggests the applicability of this approach to biofuel production. Phillips et al. [37] developed the Wood–Ljungdahl biochemical pathway model used by chemoautotrophs. Important concepts discussed include gas solubility, mass transfer, thermodynamics of enzyme-catalyzed reactions, electrochemistry, and cellular electron carriers, and fermentation kinetics. Potential applications of these concepts include acid and alcohol production, hydrogen generation, and conversion of methane to liquids or hydrogen. A review of the techno-economic analysis of gasification–syngas fermentation revealed a competitive advantage of hybrid gasification–syngas fermentation technology in producing biofuels. This technology outperforms gasification-mixed alcohol catalytic conversion and enzymatic hydrolysis fermentation processes. Safarian et al. [38] developed a simulation model based on the non-stoichiometric equilibrium method to analyze the gasification performance of 20 herbaceous and agricultural biomasses. This model is linked to syngas fermentation and product purification units for ethanol production. To study and understand the synthesis gas fermentation process, it was necessary, first of all, to identify a mathematical model capable of adequately describing the kinetics of fermentation reactions by microorganisms and the transport phenomena of matter from the gas phase to the liquid phase. Among the different models described in the literature, the one provided by de Medeiros et al. [39] was the most exhaustive, as it included a set of chemical reactions and detailed kinetic expressions:

$$4\mathrm{C}\mathrm{O}+2{\mathrm{H}}_2\mathrm{O}\to {\mathrm{C}}_2{\mathrm{H}}_4{\mathrm{O}}_2+2{\mathrm{C}\mathrm{O}}_2$$

(1)

$$4{\mathrm{H}}_2+2{\mathrm{C}\mathrm{O}}_2\to {\mathrm{C}}_2{\mathrm{H}}_4{\mathrm{O}}_2+2{\mathrm{H}}_2\mathrm{O}$$

(2)

$${\mathrm{C}}_2{\mathrm{H}}_4{\mathrm{O}}_2+2\mathrm{C}\mathrm{O}+{\mathrm{H}}_2\mathrm{O}\to {\mathrm{C}}_2{\mathrm{H}}_4\mathrm{O}+2{\mathrm{C}\mathrm{O}}_2$$

(3)

$${\mathrm{C}}_2{\mathrm{H}}_4{\mathrm{O}}_2+2{\mathrm{H}}_2\to {\mathrm{C}}_2{\mathrm{H}}_4\mathrm{O}+2{\mathrm{H}}_2\mathrm{O}$$

(4)

In particular, because reactions (1) and (3) produce CO₂, while reaction (2) consumes it, reaction (2) has to be enhanced to reduce global CO₂ production and increase the target product yields of ethanol and/or acetic acid [40].

In the present paper, syngas fermentation kinetics were modeled in order to optimize the fermentation conditions in terms of substrate and microorganism concentrations, thus maximizing the ethanol yield [39]. The performance of the bacterium Clostridium ljungdahlii can be considered best when a low partial pressure of H₂ and CO (due to the presence of nitrogen in the syngas) is available.

4. Novel Integrated Biorefinery Flowsheet for the Exploitation of Lignocellulosic Cardoon Biomass

4.1. Methodology and Process Description

An integrated novel biorefinery layout (Figure 1) was assessed for treating 130,000 t/y of residual biomass. The plant size implies a cultivated area of about 10,000–20,000 hectares, with overall management and transport costs reflecting the guidelines for the sustainable production of energy crops [41].

The proposed novel flowsheet introduces several innovative process pathways and integrations (mass and energy):

-

Residual roots are valorized first through high-added-value compound extraction processes, and then the solid residues after extraction are used for a thermochemical process, following a cascade valorization approach.

-

Typical second-generation sugar platforms cannot convert a very high percentage of the raw material, as they primarily utilize the sugar content (mainly cellulose), while lignin and other solid residues must be valorized through thermochemical processes. In contrast, this flowsheet directly converts all solids to syngas with very high yields (about 200% in mass, if gasification is performed with air).

-

The syngas is converted using novel biotechnological processes (syngas fermentation), and the off gases after fermentation are used as fuel for gas turbines and in a heat recovery and steam generation section, resulting in very high energy efficiencies compared to solid-to-energy systems.

-

Mass integration is achieved from the residual solids after extraction to the gasification section (cascade approach).

-

Energy integration is possible using the cogeneration section, which produces electricity for all of the plant’s power needs. Additionally, it supports heat integration due to the high heat consumption of ethanol distillation columns and the medium temperature requirements of the inulin extraction and conversion sections.

Optimized technologies and relevant process conditions were retrieved through a detailed literature analysis, along with process simulation and, where required, process scaling. The proposed biorefinery scheme involved the use of epigeal biomass (stem, leaves, and inflorescence) for syngas production, followed by fermentation to ethanol. Cardoon roots were preliminary processed to extract the inulin [42], and the exhaust solid residue was sent for gasification. The typical cardoon cultivation practice provides that over short cycles (about 3 years), the plants are collected and uprooted. Seeds, stems, and roots are then separated. Inulin was hydrolyzed to fructose and further purified [43] before dehydration to HMF [44] and further oxidation to FDCA [45]. The amount of total sugar in the roots was 370 g/kg_DM based on Raccuia et al. [20]. Besides inulin, cardoon roots also contain additional components; the hypogeal part of the plant contains large amounts of sugars, of which the most abundant is sucrose.

Table 1. Root composition used in the process simulation.

Compound	g/kgROOTS
inulin	333.37
sucrose	26.27
glucose	0.962
fructose	9.62
other solid	629.78

reports the average root composition [15].

Before being fed to the leaching step, roots were delivered to conveyor belt washing equipment to remove biomass residues, stones, earth, etc. Then, they were transferred to a drum dryer, where the moisture content was reduced to 20%. Next, a size reduction step was carried out in a shredder to produce chips with a thickness of around 1 cm. At this point, the material was transferred into a tank where it was diluted with 50 t/h of water to meet a solid load of 64% while the temperature was maintained at 70 °C for 1 h. Under these conditions (

Table 2. Summary of process conditions and yield for the inulin valorization sections’ operation.

Operation	T (°C)	P (bar)	Catalytic System	Yield w/w
Inulin leaching	70	1.013	---	34.7% (inulin and sugars/rootsDM)
Inulin hydrolysis	40.2	2.013	Endoinulinase on aminoethyl cellulose	90% (fructose/inulin)
HMF production	179.9	20	FDCA	51% HMF/fructose
FDCA production	110	40	Pt/C	91% FDCA/HMF

), inulin and the other sugars were extracted from the roots. The slurry containing the dissolved sugars was passed through a basket centrifugation. The solid stream from the basket underwent another extraction step, followed by further filtration in a filter press. The reactors were modeled as continuously stirred tank reactors with operational conditions extrapolated from Bastioli et al. [22].

According to [22], in each of the two extraction stages, it was assumed to extract 94% of inulin and 99% of the other sugars, while the total yield of inulin and sugars with respect to the root’s dry matter was 34.7%. After filtration into a plate and a frame filter, two streams were obtained: a solid residue and an aqueous solution containing inulin. The first part was sent to the gasification step, where it was mixed with the epigeal residual biomass and converted into syngas. The second part, a solution containing inulin, glucose, fructose, and sucrose, was subjected to evaporation to achieve an inulin concentration of 7%, which is suitable for subsequent enzymatic hydrolysis. After concentration, the solution containing inulin was fed to a packed bed hydrolysis reactor where the endoinulinase enzyme was immobilized (at a concentration of 39.3 U inulinase activity/g dried) in an aminoethyl cellulose packed bed. The system reacted continuously for 3.8 h at around 40 °C and 2.013 bar. The operational conditions, the reaction system, and the yield were extrapolated from the experimental work of Kim et al. [23]. After being separated from the other sugars, the solution was concentrated again to reach 23.08% w/w of inulin and sent to HMF and FDCA production steps; modeling was carried out based on work of Motagamwala et al. [45]. According to [45], fructose is converted to HMF, Levulinic acid (LA), and furfural in a thermostatic continuous stirred tank reactor at 179.9 °C and 20 bar. The reaction system was completed by introducing into the reactor a solution containing gamma valerolactone (GVL), which acts as a solvent together with water, and FDCA, which acts as a catalyst and is the recycled stream from the subsequent FDCA manufacturing process. It is assumed that glucose, unextracted inulin, sucrose, and unreacted fructose are degraded to humin. The latter was then separated from the products through adsorption on activated carbon, which was also previously introduced into the reactor. At this point, FDCA was obtained by oxidating HMF with oxygen into a continuous fixed bed reactor filled with a Pt/C catalyst. The temperature of the system was 110 °C, and the pressure was 40 bar. FDCA can be separated from the GVL/H₂O solvent system through crystallization, so the mixture coming from the FDCA synthesis process was sent to a crystallization unit, where it was cooled to 25 °C. This decrease in temperature caused FDCA to precipitate and, after basket centrifugation, >99% pure FDCA was obtained. Epigeal biomass was gasified in a fixed/fluidized bed gasifier using air to reduce investment and operating costs associated with air separation into oxygen and nitrogen (Air Separation Unit). Despite this, the partial pressures of the main components (CO and H₂) and the higher heating value (HHV) of the syngas were lower than those expected when using oxygen. The operating parameters, including gas cleaning and the final syngas composition, as shown in

Table 3. Syngas composition [30] and process conditions.

Gasifier technology	Bubbling fluidized bed
Gasifier process conditions	Air at 320 °C, gasifier at 800 °C, ER 0.2, bed material: olivine
Cleaning treatment	Cyclones, hot filter, tar trap
Compound	%volDRY (yi)
CO	17.0
H2	16.7
CH4	5.0
CO2	16.9
N2	44.4
Process condition
Pressure (bar)	1.025
Temperature (°C)	37

, were obtained from Serrano et al. [30].

The gaseous mixture from gasification was sent to a fermentation Trickle Bed Reactor (TBR). The type of reactor is justified by the need to maximize the contact between the liquid and the gaseous phase (high k_L*a) [46]. Furthermore, it allowed for maintaining low ratios between the liquid and the gas phases (which is impossible in a CSTR), providing various degrees of freedom in the process, such as gas and liquid recycling. This approach also resulted in low operating costs due to the absence of moving parts [33].

Downstream of the fermentation process, the unconverted syngas (not absorbed by the liquid phase in the bioreactor) was sent to a cogeneration section, where it was used in a gas turbine for electricity production. The amount of hot combustion fumes was able to generate enough steam to support the low-pressure steam demands of the distillation columns where ethanol purification took place. Figure 2 includes the flowsheet of the two processes necessary, as proposed by Safarian et al. [38]: recovering the ethanol present in the gas phase leaving the bioreactor and obtaining a 99.9% pure ethanol stream.

4.2. Process Simulation

A software-assisted flowsheet was developed for inulin extraction and conversion to FDCA based on a plant processing capacity of 92,918 Mt/y (15% moisture) of hypogeal biomass. The biorefinery section aimed at the valorization of inulin was sized considering annual root production of 78,980 t_DM/year.

The flowsheet consisted of four sections: inulin extraction, inulin hydrolysis, HMF production, and FDCA production. The leaching reactors were modeled as continuous stirred tank reactors, while the hydrolysis reactor was modeled as a plug flow packed bed reactor. The operational conditions for these reactors were extrapolated from the experimental work of Bastioli et al. and Kim et al. [23]. The enzyme inulinase immobilized in the reactor bed converted inulin into fructose and sucrose in fructose and glucose, allowing for conversion of both oligosaccharides at 90%. Fructose was dehydrated to HMF with a 70% yield system, and, subsequently, after having concentrated the solution, HMF was oxidized to FDCA over a Pt/C catalyst with 91% yield [38]. Levulinic acid (LA) and furfural were coproduced with HMF in 8.7% and 3.4% molar yields, respectively. The excess water contained in the filtrate and generated during fructose dehydration was removed through distillation and fed to a catalytic fixed bed reactor containing RuSn4/C for hydrogenation. Levulinic acid was converted to GVL, and the quantity compensated for that lost in the upstream operations, so the solution containing GVL and a certain amount of FDCA could be concentrated and recycled to the HMF production stage.

The epigeal biomass of cardoon and the solid residue left after inulin extraction were sent to a bubbling fluidized bed gasifier. Considering the average biomass humidity reported in the literature, which is 15%, the epigeal biomass entering the gasification reactor is approximately 51,000 t/y. Meanwhile, the solid residue from inulin extraction amounts to 49,757.4 t/y (dry matter). The mass yield is higher than 100%, as the post-gasification solid residue is minimal (7%), while the inert nitrogen fraction is relatively high due to the use of air as a gasifying agent. Syngas fermentation modeling was performed using two distinct equation sets, one for the liquid phase, where the biological reactions happened, and another for the syngas phase. The substrate for the bacteria’s growth is represented by CO, H₂, and CO₂ (also a reaction product). A simplified reaction network is shown in reactions (1)–(4) [39]. According to the literature [33,47], the best bioreactor configuration consists of a Trickle Bed Reactor in which higher k_L*a values can be achieved. This setup allows for a simple reactor design without moving parts, facilitating the management of fermentation time and achieving flat profiles of CO and H₂. The biochemical reactions in a TBR favor the transportation of gases to the liquid phase. Concerning the streams’ configuration, a concurrent input of gases and liquids was considered because it can increase the final concentration of ethanol and avoid flooding problems [48]. Based on the literature, CO and H₂ concentration levels ensuring high yields and selectivity for most microorganisms were in the range of 0.01–0.30 mmol/L. An analysis of the material exchange coefficients k_L*a was implemented to identify the best reactor configuration.

The overall simulation assessed the best combination of CO and H₂ levels and inoculum concentrations to maximize the ethanol yield. By elaborating on the key assumptions utilized in the proposed process simulations, we conducted a rigorous analysis of microorganism kinetics for syngas fermentation, supported by detailed simulation and modeling. Additionally, an extensive sensitivity analysis of the initial concentration of microorganisms, as well as the concentrations of CO and H₂, was performed.

To recover the ethanol present in the vapor stream, a low-temperature (0 °C) partial condensation process was implemented. The non-condensable gases remained in the gas phase and were sent to the next cogeneration section. The condensed ethanol was mixed with the beer stream. Then, this mixture was split into two streams, one to be sent to the bioreactor and the other to the column for rectification to send as a recycled stream to the bioreactor, and a stream that proceeded to the distillation columns. The recycled current was recompressed at 1.5 bar through a pump and sent back to the bioreactor. The main stream was compressed to 2 bar and sent to a first distillation column, where compounds more volatile than ethanol were recovered at the top by a partial condenser, together with residual ethanol. This stream was then remixed with the gas stream exiting the bioreactor upstream of the condenser. The bottom current consisted of water and microbial biomass, acetic acid coproduced with ethanol, etc. The liquid distillate stream was composed of at least 40% wt of ethanol and sent to a subsequent rectification column, where the distillate took on a composition similar to the azeotropic one (at least 93% wt of ethanol). Downstream of this second column, the process stream with azeotropic composition was sent to a separator with molecular sieve technology (an ideal separator in the process simulator) to reach a final composition of ethanol equal to 99.9%. The object of the analysis was the performance of the distillation columns (the recovery and composition of the overhead streams) and the total amount of thermal energy needed by the two reboilers. The temperatures at which the bottom currents of the columns vaporize should not exceed 150 °C. To optimize the conditions of the distillation columns, a sensitivity analysis was carried out in which the two reflux ratios and the distillate flow rates were varied together with the number of plates of the distillation columns. In particular, the total recovery of ethanol must be maximized with the constraint of obtaining a concentration of ethanol in the headstream of the second column equal to at least 93% wt. If the maximum possible value of ethanol recovery corresponds to a quantity of thermal energy of the reboilers lower than that obtained from the cogeneration section, the latter can be minimized. The gas stream not converted in the bioreactor was sent to the cogeneration section (Figure 3). This stream has a calorific value directly proportional to the quantity of CO and H₂ not absorbed and converted into the liquid phase. A good part of the combustible composites was represented by the quantity of methane present (about 56 kmol/h). The gas turbine was simulated as in Giuliano et al. [9]; a compressor and a turbine were energetically coupled, and, between the two equipment, there was an equilibrium reactor that minimized the Gibbs Free Energy to simulate total gas combustion. A flow rate of air with oxygen over at least 40% concerning the stoichiometric value was varied to maximize the amount of electricity supplied by the gas turbine. A heat exchanger downstream of the turbine was used to estimate the amount of heat (and steam) obtainable from the cooling of the flue gases up to a temperature of 150 °C.

4.3. Environmental and Economic Feasibility Analysis

In addition to the technical aspects of the proposed novel flowsheet for valorizing cardoon residual biomass, a pre-feasibility analysis was conducted to estimate the potential advantages derived from it in both environmental and economic terms.

Table 4. Environmental and economic parameters used to assess the pre-feasibility analysis.

Impact Parameter	CO2eq Emissions (−If Savings)	Cost (−If Gaining)
CO2 captured from air [49]	−1.493 kgCO2/kg	-
Cultivation and harvesting [6,50]	1.94 kgCO2eq/kg	EUR 40/t
Residual biomass transport [6,51]	0.07 kgCO2eq/kg	EUR 15/t
Enzymes [52]	-	EUR 5/kg
Fresh water [49]	6.52 kgCO2eq/t	EUR 0.4/t
Wastewater treatment [49]	500 kgCO2eq/t	EUR 10/t
Green electricity [49]	−668 kgCO2eq/MWhe	−EUR 50/MWhe
Ethanol [49]	0.348 kgCO2eq/kg	−EUR 0.75/kg
FDCA [16,53]	−2 kgCO2eq/kg	−EUR 2/kg

shows the CO₂ equivalent emission parameters used in this work. For each process section and each input/output mass or energy flow, an environmental impact was identified along with the potential savings derived from production processes replacing fossil-based alternatives. Specifically, this part of the work aimed to evaluate the environmental impact of FDCA and ethanol production in a cardoon-based biorefinery in terms of equivalent CO₂ emissions/savings. The phases considered were the following.

-

CO₂ savings related to carbon dioxide binding from cardoon growth;

-

CO₂ emissions during the cultivation phase, which includes cultivation and harvesting of residual thistle biomass in the field;

-

Transportation, involving the movement of biomass collected from the field to the biorefinery plant;

-

The biorefinery phase or the transformation phase of lignocellulosic residue into FDCA and ethanol, also considering the impact of the solvent (fresh water, while GVL was considered recycled), reactant (enzymes) consumption, and wastewater treatment;

-

Direct CO₂ emissions from flue gases;

-

The impact/savings of bioproduct end use in relation to fossil-based alternative carbon emissions.

For the economic evaluation, the same cost flow was used to calculate a potential annual profit, without considering the capital costs and related costs (maintenance, labor, etc.).

4.4. Results

Table 5. Mass and energy balances for 78,980 t/y of wild cardoon roots (dry matter) processed.

Raw Materials	In (t/y)	Out (t/y)
Wild cardoon roots (dry matter)	78,980	-
Inulin	26,330	1667
Sucrose	2075	105.2
Glucose	76	3.8
Fructose	760	38.5
Other solid	49,740	49,740
5-HMF	0	0.01240
FDCA	0	16,635
Utilities	MW
Electricity demand	1.23
Heating demand	50

provides a summary of mass and energy balances, process conditions, and yield for each operation. Air gasification of 100,000 t/y of lignocellulosic cardoon produced 201,000 Nm³/h. Microbial conversion of syngas could require optimal conditions in terms of the CO/H₂ molar ratio. Simulation of syngas fermentation provided inputs on the specific syngas upgrade needed to achieve the optimal gas composition to maximize the process yield. Figure 3 displays the results of the simulation of syngas fermentation.

The data indicated that for CO concentrations lower than 0.015 mmol/L and for initial microorganism concentrations higher than 2 g/L, the death kinetics are faster than the growth kinetics, and the obtained results are in agreement with the literature’s results [40]. Starting from X_IN = 2 g/L, raising the inoculum concentration led to an increase in the final concentration of EtOH, which, for CO = 0.025 mmol/L, also reached concentrations close to 40 g/L. As before, the microorganism was able to grow on a CO concentration of 0.01 mmol/L or higher. Using the CO molar flowrate (192 kmol/h), an estimated flowrate of fresh liquid necessary to obtain a final concentration of 40 g/L was calculated (

Table 6. Mass balances for the syngas fermentation section.

t/h	Clean Syngas	Gas Output	To DIST	Unconverted Syngas	Flue Gas	Pure Ethanol
H2O	2.2	1.1	32.1	-	4.1	-
Ethanol	-	0.8	1.8	0.1	-	1.8
Acetic Acid	-	-	0.4	-	-	-
O2	-	-	-	-	6.0	-
N2	14.0	14.0	-	14.0	52.6	-
CO	5.4	0.5	-	0.5	-	-
CO2	8.4	12.3	-	12.3	16.5	-
CH4	0.9	0.9	-	0.9	-	-
H2	0.4	0.2	-	0.2	-	-
Total	29.1	29.8	34.3	28	79.2	1.8

). INTX packaging (Intalox saddles) was implemented with a nominal size of 6 mm, as it has the highest value of surface area compared to the volume of the same packed bed. An optimal value of 32 t/h and a recycling flow rate of 2100 t/h were obtained from a sensitivity analysis study of the liquid flow rates and the recycling flow rates.

The value of the overall yield to ethanol was 14,020 t/y of ethanol. Overall recovery of 96.0% was therefore obtained. As shown in

Table 7. Results of process optimization of the distillation columns.

	Beer Column	Rectifying Column
Reflux ratio	3.2	6.0
Distillate/feed ratio	0.06	0.55
Reboiler heat (MWt)	7.4	3.5
Condenser heat (MWt)	4.0	3.0

, the optimization of the distillation columns led to the optimal parameters being set in the simulation to maximize the recovery of ethanol, thus obtaining an energy load to the columns of less than 11 MWt.

The results concerning the cogeneration section consist of the production of green electricity through the gas turbine and the production of low-pressure steam through thermal recovery from the combustion gases coming from the turbine. By changing the pressure level and the airflow to be sent to the gas turbine, it was possible to maximize electricity production, obtaining a value of 6.0 MWe. The pressure at which the fuel gas + air mixture is compressed is equal to 50.2 bar, while the air flow rate is equal to 51.4 t/h. The electricity requirements of the process are estimated to be 1.23 MW for FDCA production and 1.15 MW for the compressors and pumps for the syngas fermentation and ethanol purification sections, resulting in green electricity net production of 3.62 MWe. The energy use of the whole process was optimized by partially matching the heating of 69.2 MW and the cooling requirement of 19.3 MW. After heat integration, significant energy recovery of about 19.2 MW was achieved.

Regarding the pre-feasibility analysis performed, the environmental impact was calculated using CO_2eq emissions and savings. The presented novel biorefinery resulted in potential CO₂ savings following the proposed methodology (Section 4.3):

-

CO₂ savings by cardoon growth: −192.6 kt_CO2eq;

-

CO₂ emissions for cultivation, harvesting, and transportation: 34.6 kt_CO2eq;

-

Freshwater and wastewater treatment emissions: 40.7 kt_CO2eq;

-

Direct CO₂ emissions from flue gases: 123.8 kt_CO2;

-

Green electricity production savings: −18.1 kt_CO2eq;

-

Ethanol emissions (including end use emissions by combustion): 4.9 kt_CO2eq;

-

Savings from the FDCA replacing fossil-based alternatives: −33.3 kt_CO2eq.

From this, the global biorefinery savings were found to be approximately −40.2 kt_CO2eq. The main contribution includes the CO₂ captured from the atmosphere during cardoon growth, balancing both direct CO₂ emissions from the combustion of unreacted gases in the cogeneration section and indirect biorefinery emissions (e.g., from freshwater usage). Bioethanol’s environmental benefits (−31.6 kt_CO2eq) were considered lost due to end use emissions, as it is a fuel. A biorefinery plant achieving overall CO₂ savings can represent a significant opportunity to decrease GHG emissions without radically transforming the industry sector’s production and utilization processes by producing biorefining compounds that replace fossil-based alternatives.

From an economic point of view, the pre-feasibility analysis concluded positively regarding the potential profitability of the proposed configuration, with an annual profit (excluding capital costs, auxiliaries, and taxes) of approximately EUR 30.4 M/y. The main cost was attributed to raw material collection and transportation, with an annual cost of about EUR 7.1 M/y. Enzymes and chemicals represented the second highest cost, with a total annual expenditure of about EUR 6.6 M/y, due to their high cost and the significant percentage of enzyme consumption for the inulin hydrolysis step. The highest revenues were from FDCA sales (approximately EUR 33.3 M/y), followed by ethanol at about EUR 10 M/y. This resulted in a gross profit, significantly contributing to the potential economic feasibility of the proposed biorefinery flowsheet. This, along with the potential environmental savings, make this study particularly interesting and promising for the future.

Discussion

The valorization of cardoon lignocellulosic residues can be a particularly interesting challenge for replacing fossil-based chemicals and fuels because cardoon can grow in marginal lands without conflicts with the agricultural sector. The methodology proposed in this paper aims to establish the mass and energy yields of a novel biorefinery layout designed to produce FDCA and biofuel ethanol. This is achieved through inulin extraction and biomass conversion to syngas, followed by syngas fermentation. The plant size was equal to 79,000 t/y of roots and 50,000 t/y of epigeal residues. The results present final mass yields of FDCA (based on total roots) and ethanol (based on residual biomass of the gasifier) equal to 21 and 14%, respectively.

Table 8. Mass yield comparison for cardoon residual lignocellulosic biomass valorization.

Work	Final Product	Mass Yield (%)	Energy Efficiency (%) 3
Present	Ethanol + FDCA	14.0 1 + 21.1 2	22.7 4
Temporim et al. [4]	Electricity	-	18.0
Pesce et al. [56]	Methane	12.8	40.1
De Bari et al. [6]	BDO	13.3	-
Espada et al. [54]	Ethanol	12.9	21.0
Fernandes et al. [55]	Ethanol	10.1	16.4
Castellini et al. [50]	Biodiesel	4.1	9.0

¹ Based on lignocellulosic biomass of the gasifier. ² Based on roots. ³ Considering an LHV of cardoon biomass equal to 16 MJ/kg. ⁴ Only for ethanol.

shows the main literature outputs in terms of mass yields using cardoon biomass. In particular, Espada et al. [54] and Fernandes et al. [55] obtained ethanol yields of 12.9 and 10.1%, respectively. In contrast to the present work, these studies considered a second-generation sugars platform (pretreatment, enzymatic hydrolysis, sugar fermentation, ethanol purification), implying lower recovery of carbon to ethanol. On the other hand, the present work assessed a novel process of integration (biomass gasification and syngas fermentation) for potentially converting whole biomass components into the final products. To make a comparison with the performance of other studies of biofuel production through cardoon biomass valorization, energy efficiency in terms of LHV was considered. In particular, Temporim et al. [4] assessed cardoon valorization through polygeneration (electricity, heat, and cold) by analyzing its LCA. Pesce et al. [56] studied the anaerobic digestion of cardoon silage experimentally. In real scale conditions, 128 kg/t_DM of biomethane was generated. Cardoon residual biomass valorization was assessed by Castellini et al. [50], who considered a biorefinery converting cardoon biomass into biodiesel using oleaginous microorganisms. In contrast to this work (which is focused on extraction and thermochemical-based processes), that one concentrated on steam explosion pretreatment, enzymatic hydrolysis, lipid production, lipid extraction, and alkali-catalyzed transesterification. Numerical results showed a yield of biodiesel of 4.1%wt., which is 10% lower than the ethanol mass yield obtained in the present work. Only biomethane production through anaerobic digestion shows an energy efficiency (about 40%) higher than syngas fermentation in terms of ethanol (23%). The second-generation sugars platform used to produce ethanol or biodiesel is less competitive, with 9 to 21% energy efficiency. On the other hand, in this process, lignin can be converted into energy, and this raises the overall process efficiency. In general, the potential of the gasification route in terms of energy recovery in the final products is of great interest, as it is capable of transforming lignin. Compared to traditional processes of ethanol production from fossil sources, the fermentation of synthesis gas offers several advantages. These include the elimination of the need for expensive metal catalysts, greater specificity of microbial catalysts under fermentation conditions, the operation of bioreactors under environmental conditions, and the elimination of noble metal poisoning problems. Furthermore, Kennes et al. [57] carried out a comparison between bioethanol production via a sugars platform and syngas fermentation. In particular, they assessed a comparison between the economic analysis of both processes, finding similar production costs, essentially, even if the two processes have many differences.

On the other hand, the cardoon residues’ chemicals have only been assessed by De Bari et al. [6]. Additionally, in this investigation, the conversion of the epigean parts of the plants was considered through the second-generation sugars platform. An integrated biorefinery processing 60 kton/year of residual biomass to produce 8.0 kton/year of 1,4-butanediol (bio-BDO) was assessed. This mass yield of 13% is lower than the FDCA yield obtainable from inulin extraction from cardoon roots in the present work.

5. Conclusions

In this paper, technologies and processes converting cardoon residual biomass through the syngas platform and inulin conversion were considered within the scope of analyzing a novel integrated biorefinery flowsheet. In particular, two main material streams were achieved: cardoon roots containing high concentrations of inulin for FDCA and residual lignocellulosic material (epigean and residues of roots) of ethanol through syngas fermentation. This novel proposed flowsheet presents an opportunity to better valorize high-added-value compounds of cardoon roots using a cascade approach, which involves initial extraction processes followed by energy-driven/thermochemical processes. By employing high-yield conversion processes (gasification) and coupling them with innovative biotechnologies (syngas fermentation), the flowsheet also integrates mass and energy flows. Consequently, the final mass yield of products reached 35% w/w. This process scheme could be particularly beneficial for processing metal-contaminated biomass, as the gasification process typically converts them into inert material. Furthermore, a pre-feasibility analysis based on a carbon footprint and operating costs assessment highlighted significant potential environmental advantages, resulting in negative CO_2eq emissions (savings) of approximately −40.2 kt_CO2eq. The analysis also revealed a very high annual gross profit (revenues minus raw material annual costs) amounting to approximately EUR 30 M/y. This substantial profit indicates a significant opportunity to make the proposed biorefinery scheme economically sustainable while also considering the energy self-sufficiency of the system.

6. Future Directions

Following the described advantages and opportunities, future directions of the proposed research can be summarized into several points for deeper exploration:

-

Investment cost analysis for inulin’s conversion to FDCA, considering the costs and performances of catalysts/biocatalysts.

-

Identification and optimization of critical steps in the process, such as effective tar removal and optimization of the CO/H₂ mixture to maximize microbial growth.

-

Assessment of the realistic environmental impacts of each process stream and the target products considered.

In general, the overall process’s profitability and environmental advantages require a more detailed and precise analysis.