Can Chitosan Be Depolymerized by Thermal Shock?

Abstract

Chitosan is a biopolymer with a wide range of applications. It typically requires depolymerization to achieve a desired molecular weight for specific uses. This study investigated the potential for depolymerizing chitosan by thermal shock and grinding to produce nanochitosan. A series of thermal shock cycles combined with grinding were performed to assess the influence of drying temperature, residence time, and number of thermal cycles on the molecular weight, particle size, and crystallinity of chitosan. The thermal shock reduced the molecular weight and particle size of chitosan within the first hour of treatment, with optimal conditions achieved at a drying temperature of 90 °C and residence time inside the oven of 5 min. These conditions resulted in a molecular weight of 15.0 kDa with an average diameter of 136 nm. Thermal shock can be considered an effective method for chitosan depolymerization with grinding serving to standardize the particle size. This optimized process offers promising applications where low-molecular-weight chitosan is required, including biomedical, agricultural, and food industries, as well as the potential for reducing time and energy consumption.

1. Introduction

Chitosan is a non-toxic biopolymer derived from chitin, the second most abundant natural polymer after cellulose. Chitin is commonly found in the exoskeletons of crustaceans like shrimp, crabs, and lobsters, as well as in the cuticles of insects and cell walls of certain yeasts and fungi. Chitosan is composed of D-glucosamine (2-amino-2-deoxy-D-glucose) and N-acetyl-D-glucosamine (2-acetamide-2-deoxy-D-glucose) copolymers linked by β-glycosidic bonds (1 → 4) [1].

Physical and chemical properties of chitosan are dependent on its production process. Particle diameter and molecular weight are the variables that most influence the final characteristics of chitosan, especially in biomedicine or drug delivery systems [2]. A key advantage of chitosan lies in its versatility, allowing it to be physically modified into various forms including powders, nanoparticles, flakes, microspheres, foams, membranes, and fibers [3].

Chitosan nanoparticles, or nanochitosan, have extensive applications in nanomedicine, pharmaceutical, nutraceutical, food, and environmental sectors. According to Vert et al. [4], nanoparticles are defined as particles of any shape with dimensions ranging between 1 and 100 nm. The 100 nm threshold is crucial because the unique physical and chemical properties that distinguish nanoparticles from bulk materials typically emerge below this critical size. The study and application of nanoparticles is a growing field due to the distinct characteristics between nanomaterials and macroscopic materials [2]. Thus, the development of appropriate techniques for producing nanochitosan has become imperative to ensuring desirable properties in nanotechnology and nanomaterials.

The literature predominantly describes chemical and enzymatic processes for chitosan production [5,6,7,8]. However, the role of physical processes, such as drying, grinding, and sieving, has been less explored. Prior research from our group has demonstrated that these physical processes can influence the final characteristics of chitosan. For example, the drying methods investigated by Arantes et al. [1] and the grinding processes studied by Alves et al. [9] both contributed to reducing the molecular weight and particle diameter of chitosan. Literature studies for chitosan depolymerization suggest methods such as the use of enzymes [10] or extremely acidic/alkaline solutions [11,12]; oxidative degradation by catalysts with active sites [13]; ultraviolet-irradiated oxygen treatment [14]; plasma solution applied to a chitosan–metal complex [15]; and the use of gamma radiation [16]. Although these studies show promising results in terms of chitosan depolymerization, the physical method of thermal shock offers several advantages over conventional chemical and enzymatic approaches, including faster processing times, lower costs, and the elimination of additional separation and purification steps.

Therefore, this study evaluated the effect of thermal shock and grinding on the depolymerization of chitosan. A central composite design with two independent variables (drying temperature and residence time) investigated the effects of different drying conditions on molecular weight, particle size, and crystallinity of chitosan. Thermal shock, which occurs during the drying stage, aids in fragmenting chitosan macromolecules into smaller molecules. This technique is essential for producing nanochitosan with enhanced applicability due to its lower molecular weight and smaller particle size.

2. Materials and Methods

2.1. Experimental Design

A thermal cycle is defined as a repeated alternation between oven temperature and temperature outside the oven during chitosan drying for a predetermined time [1]. This process induces thermal shock in the material and cyclically modifies its equilibrium state, leading to chitosan depolymerization. The experimental conditions for thermal shock were based on preliminary studies from our group [9,17] to identify the optimal range of temperature and residence time inside the oven. Commercial chitosan from shrimp shells (Polymar, Fortaleza, Brazil) was used in the experimental design. Initial tests aimed to establish the residence time of samples outside the oven at 30 °C during the thermal shock. These tests set the residence time outside the oven at five minutes. Thermographic images were captured using a thermal camera (model PTi120, Fluke, São Paulo, Brazil) with the oven set at 110 °C. Figure 1a,b display the thermographic images taken when the empty tray was immediately removed from the oven and after five minutes, respectively. These images indicated that five minutes was insufficient time for the tray to cool to approximately 30 °C, as it remained at an average temperature of approximately 53 °C.

In order to assess the temperature variation under the experimental conditions, this procedure was repeated with the tray containing chitosan samples. Figure 2a,b illustrate the thermographic images taken after the sample tray had been outside the oven for three and six minutes, respectively.

This test was conducted in duplicate. In the first test, the sample reached a temperature of about 34 °C after 9 min, while in the second test it reached the same temperature after 11 min. Consequently, it was determined that the tray required approximately 10 min to cool to approximately 30 °C, and the residence time of samples outside the oven was fixed at 10 min for the experimental design.

The effects of thermal shock and grinding on chitosan depolymerization were evaluated using a central composite design (CCD) with two independent variables: drying temperature and residence time inside the oven at five levels (

Table 1. Drying temperature and residence time inside the oven for the CCD.

Run	Drying Temperature (°C)	Residence Time inside the Oven (min)
1	90	5
2	90	15
3	130	5
4	130	15
5 (C)	110	10
6	82	10
7	138	10
8	110	3
9	110	17
10 (C)	110	10

).

The response variables analyzed by the CCD included moisture content (%), average diameter (nm), molecular weight (kDa), crystallinity (%), and crystallite size (nm). These variables were evaluated using analysis of variance (ANOVA) with a 90% confidence interval (p < 0.1) in Statistica^® software (trial version 14.0.0, TIBCO Software) to validate the model and identify the independent variables with significant effects. Considering the complexity of the system, a 90% confidence interval was chosen due to imposed temperature limitations to avoid chitosan degradation during drying which can reduce the ability to detect the main effects.

For the CCD assays, commercial chitosan was dissolved in a 0.3 mol L⁻¹ acetic acid solution at a concentration of 1.3% w/v. The mixture was mechanically stirred for 24 h to ensure complete solubilization. Then, chitosan was precipitated by adding the solubilized chitosan into a 50% w/v sodium hydroxide (NaOH) solution with continuous stirring to form a precipitate. The sample was repeatedly washed and filtered until a neutral pH was achieved (pH 7). The washed and filtered samples were placed in stainless-steel trays and subjected to thermal shock for 24 h for each experimental run. The dried material was ground for four hours in a ball mill in a ratio of 4 kg of porcelain balls to 86 g of chitosan [9]. A sieve with 150 mesh (106 µm) openings was used to homogenize samples. The sieved samples were stored in plastic bags until characterization analyses.

2.2. Drying Kinetics

The drying kinetics was based on the CCD conditions that resulted in the lowest molecular weight (sample 1: drying temperature of 90 °C and residence time inside the oven of 5 min). This assay investigated the influence of residence time and grinding on the molecular weight and particle size of chitosan as a function of time. The residence time of samples outside the oven at approximately 30 °C was fixed at 10 min, as determined in the CCD experiments. Samples were collected at regular time intervals over a total period of 24 h. Thermographic images were captured using a thermal camera (model PTi120, Fluke, São Paulo, Brazil) throughout the drying process.

2.3. Characterization

2.3.1. Moisture Content

Moisture content of chitosan samples after thermal shock was performed in a moisture analyzer (MOC63u, Shimadzu, Tokyo, Japan) with an automatic drying mode. Approximately 2 g of each sample was used for analysis. Moisture content was calculated by the equipment when the moisture change rate per 30 s was below 0.05% at 120 °C. Results of CCD samples are expressed as average values ± standard deviation of two measurements. Results of drying kinetics are expressed for each point collected as a function of time.

2.3.2. Hydrodynamic Particle Size

Hydrodynamic particle size was measured by dynamic light scattering (DLS) with a particle size analyzer (Nanoptic 90, Bettersize, Dandong, China). Approximately 100 mg of chitosan was solubilized in 75 mL of a 2% v/v acetic acid solution. The mixture was agitated in a shaker incubator (SL-222, Solab, Piracicaba, Brazil) at 35 °C and 150 rpm for 24 h. The solution was sonicated with a probe sonicator (Vibra Cell, Sonics, Newtown, CT, USA) at a pulse of 2 s/2 s for five minutes with an amplitude of 70%. The sonicated sample was placed in a quartz cuvette and diluted with distilled water before measurements. The results were reported as the average hydrodynamic diameter (D50) and polydispersity index (PDI) from the volumetric particle size distribution.

2.3.3. Viscosity Average Molecular Weight

A chitosan solution was prepared at a concentration of 2.06 mg·mL⁻¹ in a 1:1 buffer solution of 0.3 mol·L⁻¹ acetic acid and 0.2 mol·L⁻¹ sodium acetate (pH ~ 4.5). The mixture was stirred at 150 rpm and 35 °C for 24 h, following the method of Signini and Campana-Filho [18]. After filtration, a 20 mL aliquot was added in a calibrated Ubbelohde capillary viscometer (constant K = 0.03) maintained at 25 ± 0.01 °C in a thermostatic bath and connected to a viscosity measuring system with an automatic dilution mode (model PVS 1, Lauda Scientific, Lauda-Königshofen, Germany) for viscosity measurements. The solution was serially diluted in the capillary by adding the 1:1 acetic acid/sodium acetate buffer solution. The flow time values used for determining the intrinsic viscosity (η) corresponded to the averages of three independent measurements, with a variation of less than 0.1%. The concentration of chitosan solution (g·mL⁻¹) and the reduced viscosity (mL·g⁻¹) were plotted and linearized for each sample. The linear regression coefficient was correlated with the molecular weight by the Mark–Houwink–Sakurada equation (Equation (1)):

$$M_v=\sqrt[\alpha ]{{\displaystyle \frac{\eta }{k}}}$$

(1)

where M_v is the viscosity average molecular weight (Da), α is a constant characteristic of the polymer molecule (0.76), η is the intrinsic viscosity of the solution (linear coefficient), and k is a constant dependent on the polymer, temperature, and solvent (0.076) [19].

2.3.4. X-Ray Diffraction

The semi-crystalline structure of chitosan was analyzed by X-ray diffraction (XRD, D2 PHASER, Bruker, Karlsruhe, Germany). The diffractograms were obtained using CuKα radiation at 30 kV, 10 mA, continuous scanning at 1.5°·min⁻¹, and readings taken in the 2θ interval between 5 and 80°. The crystallinity was obtained using the Profex 5.5.2 software. The crystallite size was determined by the Scherrer equation (Equation (2)):

$$D={\displaystyle \frac{K\lambda }{\beta \cos \left(\theta \right)}}$$

(2)

where D is the average crystallite diameter (nm), λ is the radiation wavelength (0.1542 nm), β is the full width at half maximum of the highest intensity peak (in radians), K is the Scherrer constant (assumed to be 0.94 for spherical crystallites), and θ is half the Bragg angle of the most intense signal (in radians) [20].

2.3.5. Degree of Deacetylation

The degree of deacetylation (DD) of chitosan was determined by conductometric titration [21]. Approximately 200 mg of chitosan was dissolved in 40 mL of 0.05 mol·L⁻¹ hydrochloric acid (HCl) and agitated in a shaker incubator (SL-222, Solab, Piracicaba, Brazil) at 35 °C and 125 rpm for 24 h. The solution was titrated with 0.17 mol·L⁻¹ NaOH until a decline, stability, and increase in conductivity values were observed. Conductance was measured using a portable conductivity meter (model NT-CVM, Nova Técnica, Piracicaba, Brazil). The DD was calculated by Equation (3):

$$DD\left(\%\right)={\displaystyle \frac{16.1\left[base\right](V_2-V_1)}{m}}$$

(3)

where [base] is the concentration of the NaOH solution (mol·L⁻¹), V₂ and V₁ are the volumes of NaOH solution at the second and first equivalence points, respectively, and m is the mass of chitosan (g).

2.3.6. Attenuated Total Reflection—Fourier-Transform Infrared Spectroscopy

The interactions of chemical groups in commercial chitosan and chitosan after drying kinetics were analyzed by attenuated total reflection–Fourier-transform infrared spectroscopy (ATR-FTIR) in a spectrophotometer (Spectrum 65 model, Perkin Elmer, Shelton, CT, USA). The infrared spectra were obtained with 32 scans, resolution of 4 cm⁻¹, and wavenumber between 600 and 4000 cm⁻¹.

3. Results and Discussion

3.1. Characterization of Samples

The yield of all experiments was approximately 80% relative to the initial mass of chitosan (approximately 100 g). The moisture content of commercial (unprocessed) and chitosan samples after thermal shock is presented in

Table 2. Moisture content, average diameter (D50), polydispersity index (PDI), molecular weight (M_v), crystallinity, and crystallite size of chitosan samples before and after thermal shock.

Sample	Temperature (°C)	Residence Time (min)	Moisture (%)	D50 (nm)	PDI	Mv (kDa)	Crystallinity (%)	Crystallite Size (nm)
Commercial	-	-	-	428	0.18	126.4	43.5	4.3
A1	90	5	38.0 ± 3.1	136	0.34	15.0	26.7	2.8
A2	90	15	14.3 ± 1.1	151	0.32	23.8	27.6	2.5
A3	130	5	10.4 ± 1.3	77	0.46	21.7	29.4	2.8
A4	130	15	9.1 ± 0.8	131	0.30	20.8	29.1	2.6
A5	110	10	9.1 ± 0.4	53	0.51	47.2	18.8	1.2
A6	82	10	34.2 ± 4.2	153	0.35	36.2	26.8	2.4
A7	138	10	10.3 ± 0.9	86	0.42	40.1	23.2	2.0
A8	110	3	53.5 ± 1.3	105	0.30	27.3	25.7	2.2
A9	110	17	6.8 ± 2.1	287	0.28	31.7	24.6	2.2
A10	110	10	11.0 ± 0.5	123	0.34	30.3	26.7	3.0

. The highest moisture content of 53.5% was observed for sample A8, which had the shortest residence time inside the oven (3 min) during thermal shock at 110 °C. Conversely, the lowest moisture content of 6.8% was found for sample A9 that had the longest residence time inside the oven (17 min) at 110 °C. In general, higher moisture content was observed in samples with shorter residence time inside the oven, leading to non-homogeneous and incomplete drying.

The relationship between moisture content and the two independent variables—drying temperature (T, °C) and residence time inside the oven (t, minutes)—is described by Equation (4). The ANOVA results for moisture content are presented in

Table 3. Analysis of variance (ANOVA) of moisture content for CCD.

Source of Variation	SS	DF	MS	Fcalculated
Regression	1995.220	5	399.044	4.427
Residual	360.552	4	90.138
Total	2355.772	9

SS: sum of squares; DF: degrees of freedom; MS: mean squares; F: Fisher test.

. The calculated F-value (F_calculated = 4.427) exceeded the critical F-value obtained from the Fisher distribution, considering the degrees of freedom in the regression and residuals for a 90% confidence interval (F_{critical (5, 4, 10%)} = 4.05) [22]. Since F_calculated > F_critical, the proposed regression model is considered valid (R² = 0.92) and provides a good fit for the experimental data.

$$Moisture\left(\%\right)=292.165-3.151T+0.010T^2-14.858t+0.320t^2+0.056Tt$$

(4)

From these data, the response surface (Figure 3) and Pareto chart (Figure 4) were constructed. A negative correlation between drying temperature and residence time inside the oven with the moisture content was observed during the thermal shock treatment. Among these factors, residence time demonstrated a more pronounced influence on moisture content, as previously discussed.

The average diameters, molecular weights, crystallinity, and crystallite size of chitosan before and after thermal shock are listed in Table 2. It is important to note that all samples from the CCD were ground before these analyses, as described in Section 2.1. Although all experimental conditions resulted in reductions in these parameters compared with the unprocessed sample, the differences between them were minimal and no significant effect (p > 0.1) could be identified among the experimental region studied.

When the drying was conducted at 90 °C with a residence time of 5 min (sample A1), the lowest molecular weight of 15.0 kDa was achieved (Table 2), representing an 8-fold reduction relative to the initial molecular weight of commercial chitosan of 126.4 kDa.

DLS results (Figure 5) indicated that the commercial chitosan exhibited the largest average diameter of approximately 428 nm. After the thermal shock, chitosan samples showed an average diameter ranging from 53 to 153 nm, representing a reduction in particle size by a factor of 3 to 8 compared with the initial size, except for sample A9 that had an average diameter of approximately 287 nm. Sample A9, treated at 110 °C with a residence time of 17 min, underwent fewer thermal shock cycles, highlighting the influence of the number of cycles on particle diameter. This result suggests that the number of thermal cycles can have a more prominent effect on particle size probably due to the repeated exposure to abrupt temperature changes which induces stress within the polymer chains, leading to their breakdown into smaller molecules. Considering the PDI results, a PDI of about 0.3 is considered acceptable for drug delivery applications with a homogeneous distribution [23]. In this study, commercial chitosan presented the lowest PDI value of 0.18 while processed samples presented slightly higher values ranging from 0.28 to 0.51.

The XRD results presented in Table 2 indicate that all chitosan samples experienced a reduction in both crystallite size and crystallinity by a factor of 2 to 4 compared to the commercial sample. The XRD patterns of samples depicted in Figure 6 are characteristic of semi-crystalline materials with two primary diffraction peaks observed at 2θ angles of about 10 and 20°. The peak at approximately 10° was associated with reflections from the (010) planes, corresponding to the α-chitin structure with orthorhombic crystals, while the peak at 20° was attributed to the (110) and (020) planes [1]. After the thermal treatment, chitosan samples exhibited an amorphous region instead of the original peak at 10° in the commercial chitosan, as best viewed in the detail of Figure 6. This transformation in the diffractogram can be attributed to the selective destruction of the (010) planes during the thermal shock, which facilitates depolymerization, as suggested by Alves et al. [24].

The DD of the commercial chitosan and the central point sample (A5) was determined by conductometric titration. Both the commercial and chitosan samples had a DD of 80% (Appendix A, Figure A1). This finding suggests that the heat treatment did not affect the DD of chitosan, i.e., the heat treatment primarily influenced the depolymerization of chitosan by breaking down the macromolecules into smaller and simpler molecules without altering the DD.

3.2. Drying Kinetics

The drying kinetics monitored the number of thermal cycles over 24 h, culminating in a total of 87 cycles. The experimental conditions for the drying kinetics were based on previous CCD results with the lowest molecular weight (sample 1: 90 °C with a 5 min residence time inside the oven). The moisture content and M_v of chitosan samples decreased progressively with increasing numbers of thermal cycles, as shown in Figure 7. A notable reduction in molecular weight was observed during the first hour of the kinetics, corresponding to the initial four thermal cycles. Specifically, the molecular weight decreased from 126.4 to 44.4 kDa, representing a reduction of approximately 65% from the initial value. Following this initial decline, the molecular weight stabilized, and values ranged between 28.2 and 47.4 kDa for the entire duration of the experiment. These results suggest that the duration of heat treatment required to produce nanochitosan could be reduced from the previously established 24 h (as reported in Alves et al. [9,24]) to just one hour, thereby optimizing energy consumption, reducing costs, and shortening production time.

After 24 h of drying kinetics, the final sample reached a molecular weight of 40.1 kDa. Upon grinding, the molecular weight further decreased to 28.9 kDa, representing a reduction of approximately 11 kDa. This result suggests that grinding has a limited impact on further reducing the molecular weight, as the chitosan samples had already achieved molecular weight values between 28.2 and 47.4 kDa during the final stages of drying, as previously discussed. To further validate this observation, the molecular weight of the central point sample of CCD (sample A5) was measured before and after grinding. The pre-grinding molecular weight was 52.7 kDa, while post-grinding, it decreased slightly to 47.2 kDa (Table 2), again demonstrating the minimal influence of grinding on molecular weight reduction after the heat treatment.

The temperature monitoring of samples throughout the drying kinetics, captured by thermal imaging, is illustrated in Figure 8. The thermographic images reveal a gradual increase in sample temperature immediately after each thermal cycle. Initially, the sample temperature was at approximately 27 °C. After four thermal shocks (equivalent to one hour of kinetics), the temperature rose to 41 °C, and by the final thermal shock (after approximately 24 h and 87 thermal cycles), the temperature reached about 65 °C. These observations support the earlier discussion that the initial thermal cycles exert the most remarkable influence on molecular weight reduction, as the most intense temperature fluctuations occur during the early stages of the drying process.

The FTIR spectra of commercial chitosan and chitosan samples after the first hour of kinetics are shown in Figure 9. Typical absorption bands of functional groups were detected at around 3346 cm⁻¹ (stretching vibration of –OH and –NH), 2925 cm⁻¹ (stretching vibration of CH₂), 1621 cm⁻¹ (stretching of –C=O in amide I), 1539 cm⁻¹ (N–H bending vibration and C–N stretching in amide II), and 1048 cm⁻¹ (vibration of ether C–O–C functional group) [1,25]. These spectra, which also corroborate the DD results, indicate that the physical process of chitosan depolymerization by thermal shock results in the cleavage of β-glycosidic bonds, culminating with the reduction in molecular weight without chemical modification of functional groups of chitosan. Similar results were reported in the literature for chitosan depolymerization by chemical and enzymatic processes [10,12].

4. Conclusions

This study demonstrates the efficacy of thermal shock and grinding in depolymerizing chitosan, with a focus on optimizing process efficiency and reducing energy consumption. The experiments revealed that the number of thermal cycles influences the reduction in molecular weight and particle size of chitosan to nanoscale levels, particularly within the first hour of treatment. The thermal shock cycles induce stress within the polymer chains, leading to their breakdown into smaller molecules. Optimal conditions were identified as a drying temperature of 90 °C and a residence time inside the oven of 5 min, resulting in the lowest molecular weight of 15.0 kDa and an average particle size of 136 nm. The initial cycles had the greatest impact on molecular weight reduction due to temperature gradients, facilitating heat and mass transfer, particularly in samples with high moisture content. Grinding was found to have a relatively minor effect on further molecular weight reduction, suggesting that its primary utility lies in standardizing particle size. These findings can contribute to large-scale industrial applications where low-molecular-weight chitosan is desirable. Potential applications include the production of nanochitosan for biomedical uses, such as in drug delivery, tissue engineering, and wound healing, as well as in agriculture, for bio-based pesticides, and in the food industry. The process’s reduced time and energy requirements also meet the sustainable development goals of chitosan production. Future research should focus on assessing the functional properties of nanochitosan in these specific applications.