Bacterial Nanocellulose Derived from Banana Leaf Extract: Yield and Variation Factors

Abstract

Bananas are one of the most important crops worldwide. However, a large amount of residual lignocellulosic biomass is generated during its production and is currently undervalued. These residues have the potential to be used as feedstock in bio-based processes with a biorefinery approach. This work is based on the valorization of banana leaf and has the following objectives (i) to determine the effect of certain physical and environmental factors on the concentration of glucose present in banana leaf extract (BLE), using a statistical regression model; (ii) to obtain Bacterial Nanocellulose (BNC), using BLE (70% v/v) and kombucha tea as fermentation medium. In addition, the physicochemical properties of BNC were evaluated by X-ray diffraction (XRD), Fourier transform infrared (FTIR), and thermogravimetric analysis (TGA). The results indicate that storage time, location, leaf color, and petiole type are factors related to BLE concentration, which is reduced by approximately 28.82% and 64.32% during storage times of five days. Regarding BNC biosynthesis, the results indicate that the highest yield, 0.031 g/g, was obtained at 21 days. Furthermore, it was determined that the highest production rate was 0.11 gL⁻¹h⁻¹ at 11 days of fermentation. By FTIR, it was determined that the purification step with NaOH (3M) should be carried out for approximately two hours. This research supports the development of a circular bioeconomy around the banana value chain, as it presents a way of bioprocessing residual biomass that can be used to produce bioproducts.

1. Introduction

Banana is a perennial tropical crop belonging to the Musaceae family and is one of the most important crops in the world in terms of metric tons harvested [1,2]. It is also of economic and food importance for many developing countries [3]. The banana production system generates different by-products as (i) starchy material, rejected fruits that do not reach the commercial standard [4,5], this by-product is used for the production of flour, cattle feed, or snacks; and (ii) lignocellulosic biomass: rachis, leaf, and pseudostem [2]. However, these lignocellulosic residues are usually not valorized [6,7,8]. Banana leaves can be used as packaging for certain foods [9], but conventionally they are left on the plantation ground. This practice is thought to benefit the crop; however, it has been determined that it can cause a nutritional imbalance in the plant [10], in addition to generating environmental and health problems [11,12].

On the other hand, the rachis with the banana bunches arrives at the collection center, where the banana is finally packaged. The rachis is piled up to be discarded [13,14]. It is estimated that in Ecuador, the main banana exporter worldwide, the waste/product ratio is 3.79, and the annual waste production is 2.65 Mt of biomass on a dry basis [15]. The valorization of residual biomass based on the circular bioeconomy model would be a sustainable strategy that could generate new sources of employment, important for food security and in line with some of the Sustainable Development Goals (SDGs) [16].

For the residual biomass to be valorized, it must first pass through a pretreatment stage. The operations used in the pretreatment stage can be physical, chemical, biological, or a mixture of these [17,18,19,20]. Physical pretreatment involves using different techniques such as steam, crushing, grinding, ultrasound, microwaves, or drying. On the other hand, chemical pretreatment makes use of acidic, basic, or oxidizing reagents. In contrast, biological pretreatment is based on fungi [21] and enzymes [22,23,24,25]. The pretreatment step usually consists of solubilizing the hemicellulose structure and reducing the lignin composition of the biomass [23,26], which facilitates enzyme access to the polymers in the enzymatic hydrolysis stage of cellulose [27,28]. Enzymes are used to reduce the complex sugars present in the biomass, thereby increasing the concentration of simple sugars, such as glucose, galactose, arabinose, and xylose. These sugars serve as a carbon source in the fermentation stage [29,30,31].

Different studies demonstrate the use of this biomass to obtain (i) bioenergy, such as bioethanol [32,33,34,35,36], and biogas [37,38], and (ii) compounds of interest, such as biofertilizers [39,40], lactic acid [41], activated carbon [42], biopolymers [43], bacterial nanocellulose [44], among other compounds of industrial interest. In this sense, banana leaves can be used as a raw material for obtaining different bioproducts. It is mainly composed of cellulose (21.90–32.56%), hemicellulose (25.80–12.00%), and lignin (39.10–17.00%), expressed as a percentage of dry weight [35,45,46]. Likewise, it has been determined that the total phenolic content is 2731.49 ± 14.41 mg eq. of gallic acid/100 g of fresh matter, being a potential source of polyphenols, among which epigallocatechin gallate is included [47].

Banana leaves have been traditionally used as packaging for certain foods [48,49]. However, new applications for this residual biomass are being studied. Tarrés et al. [6], obtained lignocellulosic micro/nanofibers (LCMNF) where the results determined that this biomass has the potential to be used in paper manufacturing with lower production costs and higher yields than the cellulose nanofiber (CNF) production method. Regarding bioenergy production, Suhag et al. [27] reported a maximum bioethanol yield of 0.38 g/g sugar, using dried banana leaf as a carbon source. The use of banana leaf extract has also been investigated. Chai et al. [50] determined that pressed banana leaf juice has a high glucose content (16.6 gl⁻¹), and from this result, they produced lipases. Tan et al. [7] used the juice extracted from banana frond (JEBF), which contains a total sugar of 14% with the amount of glucose (18.9 gl⁻¹), sucrose (13.29 gl⁻¹), and fructose (15.63 gl⁻¹) with a total volume of 0.33 l JEBF/kg banana leaf with a theoretical yield of 65% for obtaining bioethanol. These results demonstrate the potential of banana leaf extract as a carbon source in the fermentation process of different microorganisms.

The valorization of waste biomass employing bio-based processes is a current trend [51,52,53,54]. However, there are different barriers to these developments. High production costs compared to products obtained from petroleum [53]. In addition, the variability of the quality, physical, structural, and chemical composition attributes of biomass [55], can technically and economically affect the operation of a biorefinery [56,57,58]. These attributes possibly vary due to certain factors such as environment, crop management, and location; however, this is not clear. Moreover, it is not known how this would affect the yield of fermentable sugars in the juice extracted from the banana leaves.

This research analyzes how the quality attributes of banana leaves influence the concentration of its reducing sugars and their use as a carbon source for the production of bacterial nanocellulose (BNC) using a symbiotic culture of bacteria and yeast (SCOBY) [59,60]. SCOBY is a microbial consortium including yeast, acetic acid bacteria (AAB), and lactic acid bacteria (LAB). In addition, SCOBY performs well in different media such as fruit juices, corn liquor, and media containing polyphenols at relatively low costs. It is used as a starter culture, together with black or green tea, to prepare Kombucha tea (KT) [60,61]. Green tea mainly provides four polyphenolic derivatives: (-)-epicatechin (EC), (-)-epicatechin-3-gallate (ECG), (-)-epigallocatechin (EGC) and (-)-epigallocatechin-3-gallate (EGCG). Likewise, black tea mainly provides two polyphenolic derivatives: theaflavin and thearubigins [61]. In addition, they are used as a source of nitrogen for fermentation [62].

KT has remarkable nutritional properties [62,63,64,65] and its consumption has increased in recent years [66,67,68,69]. For the preparation of this beverage, it is usually left to ferment for 7 to 14 days in static, aerobic conditions and the absence of light [59,66]. As a side stream of the fermentation, a biofilm containing BNC is obtained, known as tea fungus (TF), which is formed at the gas-liquid interface of the container [70,71,72,73]. BNC has applications in different sectors such as medicine, food, and cosmetology [44,70,74], can be produced sustainably [69], and exhibits unique physical and biochemical properties [73]. However, it must be purified because, during fermentation, melanoidins are produced that are embedded in TF. Therefore, the biofilm must undergo physical or chemical treatment for its purification and subsequent characterization. Analyzes performed for BNC characterization include (i) thermogravimetric analysis (TGA), (ii) Fourier transform infrared (FTIR), and (iii) X-ray diffraction (XRD).

This work aims to: (i) determine the relationship between the morphological characteristics of banana leaves and the content of reducing sugars in the extracts obtained from them and, (ii) characterize the bacterial nanocellulose formed from banana leaf extract (BLE) as the sole carbon source for fermentation according to its physicochemical properties.

2. Materials and Methods

In the first part, the methods for analyzing the influence of certain factors of the banana leaf on the concentration of reducing sugars are indicated. In the second part, the procedure for obtaining BNC from BLE and kombucha tea as fermentation medium is described, and the methodology for the determination of the physicochemical properties of BNC by FTIR, TGA, and XRD.

2.1. Determination of Factors Affecting the Concentration of Reducing Sugars in Banana Leaf Extract

2.1.1. Collection and Pretreatment of the Banana Leaf

The samples were obtained from two banana exporting farms of the Cavendish subgroup (Musa acuminata) located in the Tres Postes (−1.983744822954974, −79.6114196251947) and Mariscal Sucre (−2.1013910, −79.4960170), according to the Global Positioning System (GPS), a precinct in the province of Guayas-Ecuador. Thirty leaves per farm were collected from 30 plants in each farm, trying to choose green leaves without symptoms of necrosis, chlorosis, or insect perforations, and fully expanded from the third layer of leaves from the top in plants at the harvest stage. The samples were coded and transported in plastic bags to the laboratory. The leaves were washed with abundant running water and sponges to remove dust, organic matter, or residues and then left to dry in the sun for two hours for further processing. The banana leaves were stored at ambient conditions (28 °C and 1 atm).

2.1.2. Banana Leaf Treatment and Reducing Sugars Estimation

Before obtaining the BLE, some physical attributes such as fresh weight (g), length from the base of the leaf blade to the apex (cm), and width at the widest part of the blade (cm) were determined. In addition, indicators of leaf blade color and petiole type were determined visually on both the upper and lower sides. The color palette of the banana descriptor guide of the International Network for the Improvement of Banana and Plantain served as the basis for these descriptions [74]. Subsequently, the extract of each sample was obtained by passing the leaves three times through a mill called “trapiche”, this equipment is made up of three rollers that press the banana leaf, like the one used to obtain sugar cane juice. Between each subsample, the “trapiche” was washed with abundant running water to minimize the influence of the extracts of other samples. The process of obtaining the extract was based on the methodology described in [7]. The juice from each subsample was collected in 2000 mL beakers and stored at low temperature (4 °C). The extract from each subsample was collected in a 2000 mL Erlenmeyer flask and stored at a low temperature (4 °C). The extract samples were centrifuged at 3500 RPM for 35 min at −4 °C; a centrifuge was used (Thermo Scientific Sorvall ST 16R, Dreieich, Germany). Reducing sugars were analyzed by the 3,5-dinitrosalicylic acid method [75,76]. The calculation of the concentration of reducing sugars in the BLE was performed using the D-glucose standard curve.

2.1.3. Data Analysis

The relationship between the response variable: reducing sugar concentration and the factors: farm location, leaf weight, leaf length, leaf width, and volume of extract obtained, were analyzed using the scatter plot matrix; in addition, the Analysis of Variance and Linear Regression were used to determine the statistical model that best describes the process. All statistical tests were analyzed at 5% significance. RStudio software (version 4.0.3) was used.

2.2. Juice Extracted from the Banana Leaf as a Means of Obtaining BNC

2.2.1. Culture Medium, Collection, and Purification of the Membrane

To make the infusion, two liters of distilled water were heated for 15 min at 60 °C, followed by the addition of 20 gL⁻¹ of green tea (Sangay brand) for 15 min. Then the temperature of the infusion was expected to drop to 25 °C to be inoculated with 10% Kombucha mother tea (pH 4, 5° Brix, and Specific Gravity 1.025). The SCOBY could include yeasts of the species Saccharomyces spp., Zygosaccharomyces spp., and Brettanomyces bruxellensis [59,77], acetic acid bacteria of the species Komagataeibacter spp, and lactic acid bacteria [77,78]. Next, BLEs were pasteurized for 15 min at 65 °C, separately, and standardized at 4° Brix. The fermentation process was carried out in 500 mL glass jars, previously sterilized and dried. The total volume of the ferment was 150 mL, whose composition is 5 g of SCOBY, BLE (70% v/v), and it was completed with the new Kombucha tea. The mouths of the vials were covered with surgical gauze and a plastic band. The ferments were stored at room temperature for 21 days, without shaking, following the methodology described in [78,79].

Figure 1 shows the process diagram for obtaining BNC from the mixture of BLE, green tea infusion, vinegar, and SCOBY. The process is like that of the KT production, with BLE being used as the carbon source. During fermentation, a membrane is produced at the liquid-air interface of the ferment, which is harvested and washed with sodium hydroxide (3M) for purification, thus eliminating the presence of melanoidins and microorganisms. Finally, the membrane is washed with distilled water and repeatedly drained until neutralized (pH 7) [71,80]. The residual NaOH solution is used to treat the residual bagasse from the leaf, and finally, this effluent is treated for final disposal.

2.2.2. Acidity Level (pH) of Kombucha Tea Based on BLE

The acidity level of the ferment was monitored every 24 h with a digital pH meter (APERA instruments, LLC-PC60, EE.UU., Columbus, OH, USA).

2.2.3. Determination of Fresh Weight and Thickness of BNC

The polymeric membrane was harvested from the fermentation media every 24 h until day 21; then, the membrane was weighed on an analytical balance. Next, the thickness of the BNC formed in the fermentation media was measured with a vernier caliper at ten different points, and their values were averaged as described in [79,80,81].

2.2.4. BNC Production

The membrane formed at the liquid-gas interface of the ferment was harvested, dried on a glass plate at 105 °C for 1 h, and the dry weight was calculated. The volume of the culture medium was 150 mL. Equations (1) and (2) were used to calculate the yield and production rate, respectively [82,83].

$$\mathrm{BNC}\mathrm{yield}\left(\mathrm{g}/\mathrm{g}\right)={\mathrm{W}}_{\mathrm{S}}·{\left({\mathrm{S}}_{\mathrm{o}}-{\mathrm{S}}_{\mathrm{e}}\right)}^{-1},$$

(1)

$$\mathrm{BNC}\mathrm{production}\mathrm{rate}\left({\mathrm{gL}}^{-1}{\mathrm{h}}^{-1}\right)=\mathrm{W}·{\left(\mathrm{V}·\mathrm{t}\right)}^{-1},$$

(2)

where: Ws dry weight of BNC (g), S_o and S_e mass of the substrate (g) at the beginning and at the end of fermentation, respectively, W is the amount of BNC produced (g), V is the volume of culture medium (L), and t, is the time of culture fermentation (h).

2.2.5. Fourier Transform Infrared (FTIR)

The membrane obtained on day 21 was purified with NaOH (3M) to eliminate the presence of microorganisms and biochemical compounds present in the sample. Different treatment times were experimented with, 0, 0.5, 1.0, 1.5 and 2.0 h. The treated samples were analyzed by FTIR. The spectra were recorded by the attenuated total reflectance (ATR) technique, in the range of 4000 to 600 cm⁻¹, accumulating 32 spectra with a resolution of 4 cm⁻¹ in a Spectrum GX spectrometer (Perkin Elmer, Waltham, MA, USA).

2.2.6. Thermogravimetric Analysis (TGA)

TGA was performed on the purified membrane. Standard TGA mode used nitrogen (99.99%) as an equilibrium purge gas, flow rate 10 mL/min, nitrogen (99.99%) 40 mL/min as sample purge gas [78].

2.2.7. X-ray Diffraction (XRD)

Measurements were made using Cu Kα1 radiation (wavelength 1.54059 Å), in parallel beam configuration, using the following system: the incident parallel slit was 5°, the incident slit was 0.2 mm, the length-limiting slit was 10 mm, the receiver parallel slit analyzer was 0.5° and is operated at 45 kV and 200 mA [44,63].

3. Results and Discussion

3.1. Determination of Factors Affecting the Concentration of Reducing Sugars in Banana Leaf Extract

Over the course of a month, the process of selection and classification of banana leaves was carried out, based on the following parameters (input variables): location of the farm, the time elapsed from harvesting to obtaining the extract, fresh weight of the leaf including petiole, length of the leaf including petiole, the width of the leaf, the color of the upper surface of the leaf and petiole canal. Additionally, the density of BLE was calculated, giving a value of 1020 g/mL at 25 °C and 1 atm. Subsequently, following the steps described in Figure 1, the leaves are subjected to the extraction process, which is characterized by (output variables): extract volume, BLE yield per unit leaf mass, and reducing sugars concentration.

Table 1. Banana leaf sample indicators.

Location	Time (day) 1	Length (cm)	Width (cm)	Weight (g) 2	BLE (mL)	Banana Leaf Color	Canal Petiole	Yield BLE (% w/w)	Reducing Sugars (gL−1)
Mariscal Sucre	1	283.00	102.00	1752.30	215.00	5e7804	Wide	12.52	18.00
		291.00	109.00	2204.00	270.00	5e7804	Wide	12.50	18.40
		299.00	100.00	1831.00	226.00	486904	Wide	12.59	17.60
	2	244.00	100.00	656.56	83.00	719604	Open	12.89	13.50
		237.00	72.00	786.63	130.00	486904	Wide	16.86	17.60
		215.00	81.00	789.79	63.00	618604	Wide	8.14	16.40
	3	263.00	100.00	1392.03	70.00	6e8904	Wide	5.12	13.50
		256.00	100.00	1592.74	165.00	b0dc04	Wide	10.57	13.50
		297.00	104.00	2131.50	375.00	486904	Open	17.95	18.00
	4	290.00	106.00	1802.00	205.00	486904	Straight	11.60	13.50
		242.00	60.00	473.81	60.00	5e7804	Open	12.92	5.30
		273.00	82.00	954.00	75.00	b0dc04	Overlapping	8.02	11.50
	5	203.00	68.00	535.61	41.00	618604	Wide	7.81	9.80
		254.00	94.00	948.48	37.00	bfd404	Wide	3.98	13.10
		309.00	88.00	1254.01	20.00	bfd404	Wide	1.63	9.40
Tres postes	1	256.00	108.00	1608.81	220.00	719604	Open	13.95	17.60
		268.00	105.00	1862.38	210.00	334c04	Open	11.50	17.60
		254.00	105.00	1790.47	245.00	3b5d04	Wide	13.96	16.00
	2	286.00	92.00	1211.55	136.00	b0dc04	Open	11.45	15.60
		269.00	101.00	1334.91	205.00	9bba04	Open	15.66	13.50
		266.00	99.00	1094.56	140.00	dfeb04	Wide	13.05	17.60
	3	258.00	95.00	1220.18	125.00	719604	Wide	10.45	17.20
		250.00	91.00	1384.55	155.00	9bba04	Wide	11.42	17.60
		205.00	78.00	971.91	120.00	bfd404	Open	12.59	13.90
	4	268.00	89.00	1232.00	145.00	3b5d04	Straight	12.01	13.90
		263.00	103.00	1763.00	250.00	5e7804	Open	14.46	17.60
		316.00	101.00	1774.00	130.00	486904	Straight	7.47	13.90
	5	262.00	97.00	1340.25	150.00	6e8904	Open	11.42	13.50
		260.00	106.00	970.60	90.00	dfeb04	Open	9.46	13.50
		200.00	90.00	1350.37	95.00	dfeb04	Open	7.18	13.50

¹ Values obtained by averaging three replicates (leaves) per day. ² Values of the sum of the weight of the leaf blade, midrib, and petiole.

shows the mean values of the quantitative variables concerning farm location. During the experiment, it was observed that as the storage time of banana leaves increased, physical characteristics such as weight and color of the upper surface varied. Regarding leaf length and width, these values coincide with those reported by Suada [84], who obtained the following results: leaf length 257.70 ± 27.66 cm, and leaf width 70.23 ± 5.83 cm. On the other hand, Tan et al. [7] determined that the percentage of juice extracted and filtered was 33% (v/p), and Chai et al. [50] estimated a percentage of 36% (v/p); however, in this work, a maximum average value of 13% (v/p) was calculated. In addition, it was determined that the Mariscal Sucre and Tres Postes samples reduced their BLE yield by 64.32% and 28.82% on the fifth day, respectively. There is no report so far on BLE performance at different times and environmental storage conditions. This study shows that storage at room temperature would produce large losses in juice yield and is certainly not a practice that should be carried out. In addition, the information obtained could serve as a baseline for future research that seeks to optimize the storage stage in biorefinery processes based on the use of banana leaves as raw material.

Concerning the concentration of reducing sugars, Oliveira et al. [85], determined that the water-soluble extractives of the banana leaf variety ‘Dwarf Cavendish’ are mainly composed of reducing sugars (16.00%); this is due to the presence of starch in this morphological region of the plant. Chai et al. [50] determined that the content of fermentable sugars in banana leaf juice was 29.09gL⁻¹, of which 55% is glucose. Tan et al. [7] found that banana leaf juice contained 14% of total sugars. In this study, the highest concentration of reducing sugars was found to be 18 gL⁻¹. The use of banana leaf extract eliminates the pretreatment stage such as acid/alkali hydrolysis and enzymatic hydrolysis that have been traditionally used [7,86,87,88,89], and replaces the costly glucose in the fermentation stage [50], making the process of obtaining reducing sugars sustainable.

In Figure 2, the box diagrams of each of the quantitative variables concerning the location are presented. Furthermore, the level of correlation of quantitative variables with respect to reducing sugars (RS) concentration can be graphically established. This is determined by observing the width of the ellipse; the greater the width, the less correlation will exist between the variables. In the case of the treatment days, thin ellipses are observed concerning the RS, indicating a strong correlation between the variables. However, the correlations of RS with the other variables are low.

In Table 1, the factors of the completely randomized Experimental Design model are analyzed. The model is shown in Equation (3).

$$\mathrm{RS}\left({\mathrm{gL}}^{-1}\right)=\mathsf{\mu }+\mathrm{Time}+\mathrm{Width}+\mathrm{Weight}+\mathrm{Length}+\mathrm{Location}+{\mathrm{Color}}_{\mathrm{Leaf}}+{\mathrm{Canal}}_{\mathrm{petiole}}+\mathrm{Error}$$

(3)

The time and location factors have a highly significant effect, the petiole carcass type factor has a significant effect at 5%, and finally, the leaf color factor is significant at 10% (

Table 2. Anova of the linear regression model.

	Sum Sq	Df	F Value	Pr(>F)
Time	81.421	1	38.0784	0.0000007555	***
Banana leaf Length	0.663	1	0.3099	0.5817536
Location	29.662	1	13.8719	0.0007806	***
Weight of Banana leaf	2.365	1	1.1062	0.3010395
Banana leaf width	2.050	1	0.9586	0.3351280
Cod_color banana leaf	39.652	10	1.8544	0.0917887	.
Canal_petiole	23.151	3	3.6090	0.0240843	*
Residue	66.286	31

Signif. codes: ‘***’ 0.001; ‘*’ 0.05; ‘.’ 0.1; ‘ ‘ 1.

).

When analyzing these factors as a mixed regression model, the results obtained in

Table 3. Linear regression model coefficients.

	Estimate	Std. Error	T Value	Pr(>|t|)
Intercepto	22.705961	2.686999	8.450	0.00000000152	***
Time	−1.134597	0.183866	−6.171	0.00000075549	***
Banana leaf length	−0.004871	0.008750	−0.557	0.581754
Location (Mariscal Sucre)	−2.077652	0.557834	−3.724	0.000781	***
Weight of banana leaf	−0.001057	0.001005	−1.052	0.301039
Banana leaf width	0.032122	0.032809	0.979	0.335128
Cod_color[T.3b5d04]	−4.107568	1.623970	−2.529	0.016729	*
Cod_color[T.486904]	−2.459556	1.830761	−1.343	0.188875
Cod_color[T.5e7804]	−2.711858	1.657470	−1.636	0.111924
Cod_color[T.618604]	−4.695788	1.965974	−2.389	0.023190	*
Cod_color[T.6e8904]	−6.070761	1.946733	−3.118	0.003908	**
Cod_color[T.719604]	−4.512197	1.706995	−2.643	0.012760	*
Cod_color[T.9bba04]	−3.166541	1.755115	−1.804	0.080928	.
Cod_color[T.b0dc04]	−1.468817	2.420680	−0.607	0.548416
Cod_color[T.bfd404]	−4.646796	1.903255	−2.441	0.020531	*
Cod_color[T.dfeb04]	−4.025819	1.792450	−2.246	0.031973	*
Canal_petiole[Wide range]	1.178651	0.589376	2.000	0.054347	.
Canal_petiole[Overlapping]	−1.495045	1.319143	−1.133	0.265759
Canal_petiole[Straight]	−0.963140	1.001189	−0.962	0.343496

Signif. codes: ‘***’ 0.001; ‘**’ 0.01; ‘*’ 0.05; ‘.’ 0.1; ‘ ‘ 1.

indicate that storage time and the location from which the sample is obtained are very significant factors affecting RS recovery. As the time elapsed from banana leaf collection to juice extraction increases, the RS concentration decreases by 1.134 units for each day. This indicates that prolonged storage under ambient conditions reduces the RS concentration, possibly due to the presence of microorganisms, such as fungi and bacteria, thus causing leaf biodeterioration. Similar results have been observed in sugarcane [90,91]. In this sense, Solomon [88] determined that most sugar factories in India have an average delay of 3 to 5 days between harvesting and milling, losing 1.0–1.3 sucrose units from the cane. He further indicates that this is due to factors such as humidity, ambient temperature, cane variety, invertase activities in the cane, and maturity stage. According to Wyse [90], in sugar beetroots, the reduction of sucrose content during postharvest is due to the continuous metabolic activity of living cells and the presence of endogenous enzymes capable of degrading sugars. It is important to note that no fungal growth was observed on banana leaves during this research. Likewise, it is observed that the location is a highly significant factor, i.e., when the leaves are collected from the Mariscal Sucre parish, the RS concentration is reduced by 2.07 units. This is due to the fact that the physicochemical characteristics of the biomass depend on the environment in which the crop is grown [15,45].

Mariscal Sucre and Tres Postes are located at an altitude of 10 and 9 m above sea level, respectively. Tres Postes has a fine, clayey soil texture and an ideal climate for banana and other food crops [91]. On the other hand, the Mariscal Sucre precinct has a fine soil type in 49% of the territory, and the rest is medium soil type. It also has a humid tropical Mediterranean climate with an average annual atmospheric temperature between 25 and 26 °C [92]. Several factors can affect banana plant composition, including acidic pH soils, soil type, and excessive application of chemical fertilizers. These factors could be analyzed in future research.

On the other hand, it has been determined that leaf color varies with respect to color #334c04 (Hex Color Codex), observing significant relationships concerning the decrease in RS concentration, which could be due to leaf senescence. Likewise, it was determined that the wide-ranging petiole type with erect margins positively influenced RS concentration by 1.18 units. These relationships between the different factors studied in this work have not been analyzed by other researchers.

To determine which variables are related to glucose concentration, the generalized linear model was used. The following statistical values were estimated; Residual standard error: 1.462 on 31 degrees of freedom, Multiple R²: 0.8423, Adjusted R²: 0.7507; F-statistic: 9.196 on 18 and 31 DF, p-value: 0.00000006194. In this sense. The estimated parameters that have some degree of significance are expressed in Equation (4) with their respective coefficient.

$$\begin{gathered} \mathrm{RS}\left({\mathrm{gL}}^{-1}\right)=22.706-1.135·\mathrm{Time}-2.078·{\mathrm{Location}}_{\mathrm{MS}}-4.108·{\mathrm{Color}}_{\mathrm{T}.3\mathrm{b}5\mathrm{d}04}-4.696·{\mathrm{Color}}_{\mathrm{T}.618604}-6.071 \\ ·{\mathrm{Color}}_{\mathrm{T}.618604}-4.512·{\mathrm{Color}}_{\mathrm{T}.719604}-4.647·{\mathrm{Color}}_{\mathrm{T}.\text{bfd}404}-3.167·{\mathrm{Color}}_{\mathrm{T}.9\text{bba}04}-4.06 \\ ·{\mathrm{Color}}_{\mathrm{T}.\text{dfeb}04}+1.179·{\mathrm{Canal}}_{\mathrm{petiole}-\mathrm{width}} \end{gathered}$$

(4)

3.2. Banana Leaf Extract as a Means to Obtain BNC

3.2.1. Culture Medium, Collection, and Purification of the Membrane

For the fermentation stage, BLE (70% v/v), mother vinegar (10% v/v) were used as sugar sources. Green tea infusion provided nutrients containing nitrogen, vitamins, and minerals [65]. The culture medium resulted in a BNC polymer with a surface area equal to the dimensions of the container with the presence of melanoidins. The membrane is then purified with NaOH. In addition, in this process, the leaf bagasse remains as a residue that undergoes a washing treatment with the residual solution generated in the BNC purification stage.

Different agricultural and food wastes have been used as a carbon source for the production of BNC [93,94,95,96]. With respect to the use of residual biomass from the banana value chain, rotten banana [83], as well as banana peel [44,81], have been used as a carbon source in the fermentation stage to obtain this biopolymer. In this work, BLE and Kombucha tea have been used for the first time as a medium for BNC biosynthesis.

In Figure 3, the products that are obtained from the banana leaf recovery process are shown. In Figure 3a the juice of the extracted banana leaf is observed. Figure 3b shows the membrane formed at the liquid-gas interface, whose dark color is due to the formation of melanoidins.

3.2.2. Acidity Level (pH) of Kombucha Tea Based on BLE

The pH parameter of the fermentation process is closely related to microbial growth [59]. These assimilate the glucose present in the medium to carry out their various metabolic processes [97,98]. Therefore, due to these biochemical and metabolic processes, acetic and gluconic acid are generated, which lowers the pH of the fermentation medium [99,100,101,102]. In this sense, the lowest acceptable pH value should not be less than 3 [99]. Figure 4 shows the increase in pH at different fermentation times. The BLE-based Kombucha tea starts with a pH of 6.39 and reaches a pH of 3.5 after 21 days of fermentation.

3.2.3. Determination of Fresh Weight and Thickness of BNC

The main source of carbon during fermentation was the reducing sugars present in the banana leaf extract. The BNC is collected by simply removing it from the fermentation medium since this biopolymer forms a membrane that floats on the fermenting liquid-gas interface [100]. This membrane serves as a physical barrier to protect the microbial consortium against external agents [63,81,103], and is thought to help reduce the loss of oxygen from the medium caused by increased acidity [102]. As shown in Figure 5a, the results indicate that the weight of the CNB increases during the first 14 days, after which time the mass of CNB does not increase significantly. If the BNC weight increases significantly, it could precipitate to the bottom [79].

In Figure 5b, it is observed that the thickness of the membrane increases with time. These results are in agreement with those obtained in different studies [81,98,104,105]. Ramirez [104] determined that the optimum time for BNC recovery is 21 days.

3.2.4. BNC Production

BNC production and yield were estimated for the different fermentation times. Figure 6a shows that the yield increases until approximately day 15, after which time it remains constant. This agrees with that indicated by some authors who observed better results in an average of 15 days [81,96]. Likewise, Figure 6b shows that the production rate is highest at 11 days and decreases after. A limitation of this study is that the fermentation kinetics for the production of CNB has not been determined.

3.2.5. Fourier Transform Infrared (FTIR)

The BNC samples obtained from the fermentation were purified with NaOH (3M), and the effect of treatment time was observed by FTIR. Figure 7 shows the FTIR profiles of BNC using BLE (70%) at different treatment times with NaOH. The spectra obtained showed typical BNC bands, which are the broad peak located in the region of 3200–3400 cm⁻¹ [44,106,107], which corresponds to stretching vibrations of cellulose OH groups, while the peak around 2900 cm⁻¹ is related to C-H stretching [71,79]. The peak at 1634 cm⁻¹ is attributed to the OH bending of adsorbed water [44,71,106,107]. The peaks that appeared between 1055–1049 cm⁻¹ correspond to C-O stretching at C3; C-C stretching; and C-O stretching at C6 [71,108]. The band at 896 cm⁻¹ is attributed to C-O-C stretching at β (1,4) glycosidic. Finally, the peak at 658 cm⁻¹ is observed from C-OH bonding out of plane vibrations [44,71,79,107]. The spectra with treatment times of 1.5 and 2 h show defined peaks corresponding to the cellulose I profile.

3.2.6. Thermogravimetric Analysis (TGA)

TGA was performed on the BNC sample treated with NaOH (3M) for two hours. In Figure 8, it is observed in the TGA spectrum that the BNC film loses mass around 100 °C. This could be due to the evaporation of the moisture present in the sample [71,106,107]. The pyrolysis of biosynthesized BNC in a medium with BLE (70% v/v) presents the highest peak at 343.78 °C; this result is within the reported by different studies [71,107,109]. It is estimated that BNC starts at a higher temperature range (340–360 °C), associated with the complete degradation of BNCs, including depolymerization, dehydration, and decomposition of glucose [108].

3.2.7. X-ray Diffraction (XRD)

The diffractogram of BNC obtained from BLE (70% v/v) and purified with NaOH (3M) for 2 h is presented in Figure 9. In this sense, Santos et al. [109] indicate that the diffractogram of BNC has two dominant diffraction peaks, one between 14° and 15°, and another between 22° and 24°. Each of the peaks presents the two crystalline phases in cellulose, Iα, and Iβ.

In Figure 9, it is observed that the XRD curve presents some characteristic peaks of crystalline cellulose I at 2 θ: 14.5°, 16.5°, 22.5°, and 34.5°. These results are in agreement with various studies [107,108,109].

4. Conclusions

Some feedstock factors can affect the yield of a bio-based production process. Testing these yields through predictive modeling can help in the development of sustainable biorefineries. In this sense, banana leaves are potential sources of reducing sugars that can be used in bio-based processes with a biorefinery approach. However, to implement this, it is necessary to know how storage time, feedstock source location, and physical characteristics relate to the process yield. It has been determined that banana leaf yield decreases by approximately 28.82% to 64.32% during storage times of three to five days. The linear model relating these factors to reducing sugar yield has an adjusted R² of 0.7507. Previous research on banana leaf valorization had not considered these aspects of importance for industrial scale-up.

Furthermore, it has been shown that BNC films can be successfully obtained using banana extract as a carbon source and SCOBY as a starter culture. The pH change of the medium, the weight, and the thickness of BNC were evaluated. Yield (0.031 g BNC/g fermentation medium) and production rate (0.11 g L⁻¹h⁻¹) were also calculated. In addition, the physicochemical properties of BNC were analyzed by FTIR, XRD, and TGA, demonstrating the presence of Nanocellulose.

This approach based on the use of leaf extract to obtain bio-based compounds could contribute to the development of more sustainable processes and boost the creation of new value chains based on the concept of circular bioeconomy.