The growing ecological and environmental unrest across the world demands greater utilization of green materials (Thomas et al., 2018). Among the many green materials in recent research, an emerging interest has been observed in cellulose, particularly, nanocrystalline cellulose. Nanocrystalline cellulose is a nano-size counterpart of cellulose having one dimension within the nanometer range and still retains cellulose’s chemical structure. It boasts high specific surface area, crystallinity, mechanical strength, and active functionalization groups which are linked to its size. Due to these properties, various industries (e.g., healthcare, electronics, construction, food packaging, etc.) find significance for this green material (Yu et al., 2021). As such, nanocrystalline cellulose and other nanotechnologies are recognized as a crucial part of a new industrial revolution (Pradhan et al., 2022).

A sustainable source of nanocrystalline cellulose is biomass. It is non-fossilized, organic matter that can participate in the carbon cycle which is sourced from plants, animals, and other organisms (Yu et al., 2021). However, in promoting a circular economy, first-generation feedstock biomass is not recommended as a source of green materials. Instead, recent research interest has been focused on the efficient valorization of biomass waste.

In the Philippines, cassava processing has led to significant waste generation. About 2.6 million metric tons of cassava are processed yearly for food production (Philippine Statistics Authority, 2022). Cassava is used for its starch which leaves behind waste pulp (roughly 10–30% of the original) (Akaracharanya et al., 2010). This contains 15–17% cellulose and boasts a low lignin content (1–4%) (Yu et al., 2021). For these reasons, cassava waste pulp shows great potential for waste valorization. Through it, nanocrystalline cellulose could be produced—a high-value product perfect for improving the economy of the cassava-producing region.

The extraction/synthesis of nanocrystalline cellulose from lignocellulosic biomass involves a two-step procedure. First, pretreatment of the raw material is performed to remove most of the hemicellulose, lignin, and other impurities. Many methods can be used to perform pretreatment, but a common route is to treat the lignocellulosic biomass with an alkaline solution and bleaching (Liu et al., 2016; Bacha and Demsash, 2021; Dien and Anh, 2021; Widiarto et al., 2017). On the other hand, the second step is the extraction procedure using acid hydrolysis.

Acid hydrolysis of lignocellulosic biomass degrades the amorphous domains of the cellulose while leaving the crystalline regions (nanocrystals) intact (Jonoobi et al., 2015). Typically, strong inorganic acids are utilized for extracting nanocrystalline cellulose, particularly sulfuric acid. It is known to be an active hydrolysis agent for shortening the cellulose chain (Dien and Anh, 2021). Furthermore, sulfate-ester groups are formed during sulfuric acid hydrolysis which results in a dispersed stable colloidal system, unlike other acids which result in agglomeration (Morais et al., 2013).

However, the use of strong, inorganic acids as hydrolysis agents does not support green and sustainable chemistry (Ji et al., 2019). Strong acids are highly corrosive and difficult to recover. Its shortcoming is the generation of wastewater during the washing process of pretreated nanocellulose (Gupta and Shukla, 2020).

In this regard, several studies considered the application of organic acids to hydrolyze cellulosic material to produce nanocrystalline cellulose. Recently, formic acid was used to substitute strong acids in nanocrystalline cellulose since it is less corrosive to equipment and easily recovered by distillation (Du et al., 2016). To give the generated nanocrystalline cellulose surface functionality, functional groups (such as ester groups and carboxylic groups) can also be simultaneously added during the hydrolysis process. As a result, organic acid hydrolysis has been viewed as a viable strategy for the environmentally friendly manufacturing of high-performance nanocrystalline cellulose (Wang et al., 2021). However, a disadvantage of using formic acid is comparatively low hydrolysis efficiency even at high concentrations and temperatures (Du et al., 2016). To address this concern, the use of strong acids as a catalyst for nanocrystalline extraction may be adopted (Wang et al., 2020; Wang et al., 2021; Xie et al., 2019).

To the best of available knowledge, no reports have been made on formic acid hydrolysis of waste cassava pulp. Hence, in this paper, nanocrystalline cellulose (NCC) was synthesized using a mixed formic/H₂SO₄ acid hydrolysis of waste cassava pulp with sulfuric acid (5wt%) as the catalyst. Formic acid concentrations of 50-88wt%, hydrolysis temperature of 70–90°C, and reaction times of 1–6 hours were investigated. Finally, optimization of the reaction parameters was performed based on the product yield. To the best of available knowledge, no reports have been made on formic acid hydrolysis of waste cassava pulp. Hence, in this paper, nanocrystalline cellulose (NCC) was synthesized using a mixed formic/H₂SO₄ acid hydrolysis of waste cassava pulp with sulfuric acid (5wt%) as the catalyst. Formic acid concentrations of 50-88wt%, hydrolysis temperature of 70–90°C, and reaction times of 1–6 hours were investigated. Finally, optimization of the reaction parameters was performed based on the product yield.

NCC was synthesized using mixed acid (H₂SO₄/HCOOH) hydrolysis with varying acid concentration, temperature, and time, while keeping the solids-to-acid ratio constant. Most samples were white in aqueous suspension, but some were dark brown. This indicates the presence of other soluble hydrolysis products that have either been readsorbed into the NCC or bound to the immobile water layer on the NCC surface (Wang et al., 2021). Table 1 shows the yield results of the various parameters set by the Box-Behnken Design. In numerical optimization, the yield was maximized.

The theoretical maximum yield is approximately 67.373% NCC by weight, achieved by hydrolyzing cellulose with 71 wt% formic acid at 77°C for 3.6 hours. This was verified through three replicated runs, yielding 65.272 ± 2.295 with a 3.12% error from the theoretical value. These results align with a similar study using the same acids, which found the best yield with 75 wt% formic acid at 80°C for 3 hours (Wang et al., 2021). The optimized parameters were used for the run S5F71, which was freeze-dried for tests such as FTIR, SEM, and DLS.

Effects of acid concentration, temperature, and time to yield. Based on the results of the Box-Behnken design, acid concentration significantly affects NCC yield. Low concentrations (e.g., 50%) do not fully hydrolyze the amorphous cellulose regions, hindering efficient NCC extraction from crystalline regions (Wang et al., 2021). Higher concentrations, particularly 69%, increase yields at all temperatures and hydrolysis times, but very high concentrations (e.g., 88%) drastically reduce yields to 30–40% due to hydrolysis of both amorphous and crystalline regions, which is reflected in a dark brown color in some samples. Temperature and hydrolysis time also impact yield. Lower times produce lower yields as the amorphous regions remain unhydrolyzed, while excessive times cause over-degradation, forming by-products like glucose, xylose, and furfural instead of nanocrystals (Wang et al., 2021). Temperature, on its own, has a minor effect, possibly due to the short interval or all tested temperatures being optimal for formic acid hydrolysis. Determining optimal parameters for NCC extraction is challenging due to varying methodologies. Strong inorganic acids like sulfuric acid achieve efficient hydrolysis at lower temperatures and shorter times, at 45°C for 60 min (Liu et al., 2016), whereas weaker organic acids like formic acid require longer times and higher temperatures, as seen in studies (Du et al., 2016; Aprilia et al., 2018). Some studies explore high-temperature, low-time extraction, such as 95°C for 30 mins with corncob residue (Liu et al., 2016) and birch (Du et al., 2016). Some studies use only formic acid, requiring longer times and higher temperatures (Aprilia et al., 2018), while others use catalysts like FeCl₃ (Du et al., 2016) or H₂SO₄ (Wang et al., 2021), reducing the temperature, time, and acid concentration needed for optimal yield.

Figure 1. FTIR Analysis of Waste Cassava Pulp, Purified Cellulose, And NCC (S5F71)

FTIR Analysis. FTIR spectroscopy was used to analyze the functional groups of untreated waste cassava pulp, cellulose, and nanocrystalline cellulose (S5F71). Comparing the FTIR spectra (Fig. 1) confirms the removal of lignin and hemicellulose during cellulose isolation. The structure of cellulose, a polysaccharide with β-D-glucose units linked by β(1–4) glycosidic bonds, shows most bands associated with carbon, hydrogen, and oxygen bonds. In the FTIR spectrum, a band at around 1720 cm⁻¹, present in untreated waste cassava pulp but absent in cellulose and nanocrystalline cellulose, is attributed to the C = O stretching vibration of hemicellulose and lignin carboxylic groups. A band at 1244 cm⁻¹, found only in waste pulp, is linked to ether linkages (C–O–C) of lignin (Widiarto et al., 2017), indicating lignin and hemicellulose removal during isolation. The band between 3500 to 3200 cm⁻¹, attributed to O-H stretching vibration, appears higher in nanocrystalline cellulose, indicating higher cellulose content (Travalini et al., 2017). The bands at 560–620 cm⁻¹ are due to out-of-plane C-H bending or ring deformation of glucose units. The band at around 2900 cm⁻¹ corresponds to C-H stretching vibration, and in-plane C-H bending is observed at 1365 cm⁻¹. The band at 1420–1450 cm⁻¹ is associated with CH2 bending, while the peak at around 1030 cm⁻¹ corresponds to C-O-C stretching in the pyranose ring of nanocrystalline cellulose (An, 2020). A band at around 2270 cm⁻¹ in the cellulose sample, not associated with cellulose, lignin, or hemicellulose structures, suggests possible sample contamination, explaining its absence in the other FTIR spectra.

Morphological Analysis. Scanning electron microscopy (SEM) was used to examine the morphology of purified cellulose from waste cassava pulp and nanocrystalline cellulose (S5F71). The pretreated cassava pulp was oven-dried, and the nanocrystalline cellulose suspension was freeze-dried before imaging. The samples were coated with gold and observed at an accelerating voltage of 3.0 to 20.0 kV. As shown in Fig. 2, the pretreated cassava pulp formed a tightly packed network resembling a mountainous 3-D surface.

In contrast, the NCC sample appeared as long fibers (see Fig. 3), indicating the effect of mixed acid hydrolysis. The extraction was effective in breaking the amorphous regions of the cassava pulp, yielding nanocrystals. However, NCC samples should typically appear as short rods instead of long fibers. These lengthy fibers are due to the self-aggregation of NCC during drying (Liu et al., 2016). During freeze-drying, NCCs interact with each other as ice crystals form in the distilled water suspension. Nanocellulose gathers around the edges, forming hydrogen bonds on adjacent surfaces, leading to longer fiber organization (Liu et al., 2016). Despite this, most NCC samples measured less than 100 nm when measured manually. Additionally, SEM analysis may have skewed the nano-size of the material. While SEM is commonly used for imaging nanocellulose, the required gold coating can enlarge the structure, preventing accurate observation of the true NCC size (Liu et al., 2016). However, the nano-size is confirmed by other analyses.

Particle Size Distribution and Zeta Potential. The particle size distribution (Fig. 4) of the optimized NCC sample (S5F71) was measured using dynamic light scattering (DLS). While DLS is ideal for spherical particles, it provides quick quantitative estimates for NCC (which is rod-like in shape) and can approximate particle length (Du et al., 2016). The NCC suspension was diluted with distilled water before measurement. Freeze fried samples were not used to avoid the effects of self-aggregation during drying (Liu et al., 2016).

The hydrodynamic diameter ranged from 15 nm to 687 nm, with 80% passing at 58.10 nm and 95% at 93.60 nm. This is comparable to similar studies, whose average values were 452 nm, and 375 nm to 1631 nm (Du et al., 2016). The smaller size in this study is attributed to sulfuric acid, which improved hydrolysis efficiency as a catalyst in small concentrations (Wang et al., 2021). Sulfuric acid also contributed to an average zeta potential of -14.3 mV, a higher absolute value than previously reported values (~-6 mV), indicating the presence of surface sulfate groups (Du et al., 2016). This increased zeta potential reduces particle aggregation, enhancing application in water phases (Du et al., 2016).

Equation 1 is used to determine the % yield of nanocrystalline cellulose. This is the ratio of the NCC obtained and the bleached cassava pulp’s mass.

Materials. AR-grade sodium hydroxide pellets were purchased from Sigma-Aldrich (Massachusetts, United States). The sodium hydroxide pellets have a molar weight of 40.00, an assay of ≥ 98.5%, and is soluble in water for up to ~ 1260 g/L at 20°C. Sodium hypochlorite solution of industrial grade was purchased from Mabuhay Vinyl Corporation (Iligan City, Lanao del Norte, Philippines). The solution has a weight percent of 7.77 wt% (analyzed using ASM Method D2022). Reagent-grade formic acid was sourced from ExpertQ®, ACS, Scharlau AC10852500 (Barcelona, Spain). The acid has a concentration of ≥ 88%, an alkalimetric assay of ≥ 88%, a molar weight of 46.03, and a pH of 2.2. Sulfuric acid and hydrochloric acid were purchased at Scharlau (Barcelona, Spain). Waste cassava pulp was provided by Ricor Mills Corporation (Brgy. Panadtalan, Maramag, Bukidnon).

Pretreatment. Wet cassava pulp was sun-dried for one week and sieved with Mesh No. 40. Further drying was performed in a convective oven. Dried cassava waste pulp (20g) was refluxed with 400 mL NaOH (4%) at 90°C for 2 hours with constant stirring. Two hundred (200) mL of 4% (w/v) NaOCl was then added maintaining the temperature 80°C for 1 hour with constant stirring. Vacuum filtration, neutralization with HCl, and washing with distilled water followed until a neutral pH was reached. The obtained cellulosic material was dried in a convective oven at 80°C for 4 hours. It was then milled in a blender and sieved through Mesh No. 50.

Acid Hydrolysis. Pretreated pulp was hydrolyzed with a mixed acid system with a solids-to-acid ratio of 1:30 (w/w). Sulfuric acid was employed as a catalyst in formic acid hydrolysis (0.05g H₂SO₄: 1g total acid solution). Formic acid concentrations of 50-88wt%, hydrolysis temperature of 70–90°C, and reaction times of 1–6 hours were investigated. The hydrolysis was done in a water bath under constant magnetic stirring. At completion, the solution was diluted with five (5) times of distilled water to quench the reaction. The resulting solution was then centrifuged at 4500 rpm for 30 minutes to remove the liquid from the solids. The solution was then neutralized using 0.5 N NaOH. The solids were washed again with distilled water and centrifuged at 4500 rpm for 1 hour as many times until the solution became neutral. Finally, the NCC was separated from the unhydrolyzed cellulose by centrifugation at 4500 rpm for 2 mins. The supernatant, which is rich in NCC, was collected for further analysis.

Characterization of NCC. The yield of the extraction process was determined using gravimetry. It was calculated as the percentage ratio of the NCC obtained and the bleached cassava pulp’s mass. For each obtained NCC suspension, the volume was determined. A sample was then taken and dried until constant weight. The mass after drying was recorded and considered the concentration of the NCC suspension (g/mL). Eq. 1 was then used to determine the yield.

Pretreatment of the waste pulp was performed using 4 wt% NaOH and 4% (w/v) NaOCl. The process resulted in a purified cellulose yield of 16.33%. Meanwhile, mixed H2SO4/formic acid hydrolysis was utilized for NCC extraction. Formic has shown potential for NCC synthesis due to its ease of recoverability and reuse. Still, it is a weak organic acid, and the addition of 5 wt% of sulfuric acid boosted the hydrolysis efficiency. A concentration range of 50–88 wt% for formic acid, temperature of 70–90°C, and hydrolysis time of 1–6 hours were investigated. As expected, increasing all three factors resulted in higher NCC yields. However, too much increase also produced negative results.

The FTIR analysis of the waste cassava pulp and the purified cellulose, and NCC samples revealed effective pretreatment. Lignin and hemicellulose were removed resulting in higher peaks associated with cellulose—indicating more cellulose content. Morphological and particle size analysis of the optimized NCC sample and purified cellulose revealed the effect of acid hydrolysis on the cellulose. As observed through scanning electron microscopy, the amorphous regions were removed revealing fibers of crystalline regions in the NCC sample. Meanwhile, dynamic light scattering reported hydrodynamic diameters within 15 nm to 687 nm. The presence of sulfuric acid as a catalyst reduced NCC size. Furthermore, it contributed to a higher zeta potential compared to previously reported formic acid hydrolysis studies. This increased value indicates less tendency for particle agglomeration, thereby improving the material's application in water phases.