Formulation and Characterization of a Plasma Sterilized, Pharmaceutical Grade Chitosan Powder

Abstract

Chitosan has great potential as a pharmaceutical excipient. In this study, chitosan flake was micronized using cryo-ball and cryo-jet milling and subsequently sterilized with nitrogen plasma. Micronized chitosan was characterized by laser diffraction, scanning electron microscopy (SEM), conductometric titration, viscometry, loss on drying, FTIR, and limulus amebocyte lysate (LAL) assays. Cryo-jet milling produced mean particle size of 16.05 μm, 44% smaller than cryo-ball milling. Cryomilled chitosan demonstrated increased hygroscopicity, but reduced molecular weight and degree of deacetylation (DD). SEM imaging showed highly irregular shapes. FTIR showed changes consistent with reduced DD and an unexplained shift at 1100 cm⁻¹. Plasma treated chitosan was sterile with <2.5 EU/g after low-pressure plasma and <1.3 EU/g after atmospheric pressure plasma treatment. Plasma treatment decreased the reduced viscosity of chitosan flake and powder, with a greater effect on powder. In conclusion, pharmaceutical grade, sterile chitosan powder was produced with cryo-jet milling and plasma sterilization.

1. INTRODUCTION

Chitosan (CS) in the presence of moisture is a soft, ductile biopolymer with great potential as a biomedical material, especially as an excipient and immune adjuvant. CS is obtained commercially by deacetylating chitin extracted from either crustacean shells via chemical extraction or fungi via enzymatic extraction and consists of β-1,4-linked 2-amino-2-deoxy-D-gluocopyranose and N-acetamido-2-deoxy-D-glucopyranose moieties that are randomly distributed throughout the polymer chain. Since chitin is the second most abundant biopolymer on Earth and crustacean shell waste generated by the seafood industry is an environmental problem in coastal areas, CS is a widely available, green, and economical biomaterial. Chitin becomes CS once at least 50% of the D-glucosamine moieties in the polymer chain are deacetylated (Pillai, Paul & Sharma, 2009). However, there is disagreement over the naming of chitin and CS based on degree of deacetylation, which has resulted in a proposal for naming chitin and CS based on solubility in acetic acid (Badawy & Rabea, 2011; Kumar, 2000). CS is soluble in aqueous acetic acid, whereas chitin is not.

As an excipient, CS enhances drug penetration through tissues and epithelial barriers by loosening gap junctions, maintains drug in the area of interest through bioadhesion between cationic amino groups of CS and anionic tissues, and controls drug release over time by keeping drugs bound until physical degradation (Artursson, Lindmark, Davis & Illum, 1994; Dodane, Amin Khan & Merwin, 1999; Jameela, Misra & Jayakrishnan, 1994; Kristl, Šmid-Korbar, Štruc, Schara & Rupprecht, 1993; Luessen, de Leeuw, Langemeyer, de Boer, Verhoef & Junginger, 1996; Nordtveit, Vårum & Smidsrød, 1994; Schipper, Vårum & Artursson, 1996; Takeuchi, Yamamoto, Niwa, Hino & Kawashima, 1996; van der Lubben, Verhoef, van Aelst, Borchard & Junginger, 2001). The biological functions of CS are dependent on its physical properties, including molecular weight (MW), degree of deacetylation (DD), salt form, and pH at which it is used (Dodane & Vilivalam, 1998; Kumirska, Weinhold, Thöming & Stepnowski, 2011). This physical-functional relationship necessitates careful characterization of CS formulations. Since powder and hydrogel forms of CS – especially those composed of particles in the low micron range – have been identified as the ideal form of CS for pharmaceutical applications and micronization of materials is a common process in pharmaceutical manufacturing, the goal of the present study was to formulate and characterize a pharmaceutical grade CS powder that can also serve as a precursor to a hydrogel when dissolved in dilute acids (van der Lubben, Verhoef, van Aelst, Borchard & Junginger, 2001).

A powder form of CS was also desired to enhance the effectiveness of a novel sterilization method for CS based on non-thermal nitrogen gas plasma (NtNP) (de Oliveira Cardoso Macedo, de Macedo, Gomes, de Freitas Daudt, Rocha & Alves, 2013). Plasma is considered the fourth state of matter and consists of an ionized gas that has emergent properties the gas alone does not, such as magnetism and conductivity. Use of NtNP for sterilization of CS is necessitated by the fact that conventional sterilization methods like dry/wet heat, radiation, and chemical sterilants cause caramelization of the polysaccharide, chain scissions, and/or may leave residual toxic residues in the material (de Oliveira Cardoso Macedo, de Macedo, Gomes, de Freitas Daudt, Rocha & Alves, 2013; Franca et al., 2013; Lim, Khor & Koo, 1998; Lim, Khor & Ling, 1999; Marreco, da Luz Moreira, Genari & Moraes, 2004; Norzita, Norhashidah, Maznah, Nurul Aizam Idayu Mat, Nor Akma & Norafifah Ahmad, 2013; Rao & Sharma, 1995; Rosiak, Ulański, Kucharska, Dutkiewicz & Judkiewicz, 1992; San Juan et al., 2012). These physicochemical changes result in changes in the biological properties of CS making these methods non-optimal for sterilizing CS. Since NtNP is a surface treatment, it benefits from a large surface to volume ratio, which is accomplished when a material is micronized.

Knowledge on micronizing hard and crystalline materials is extensive, but lacking on softer materials. It is vital to characterize particle size and size distribution of powders since these properties influence flowability, dissolution, release kinetics, and more (Koennings, Sapin, Blunk, Menei & Goepferich, 2007; Miranda, Millán & Caraballo, 2007; Mullarney & Leyva, 2009). Reports in the literature of techniques for generating micronized CS powders are especially sparse. Techniques for micronizing CS that have been reported include dense gas anti-solvent precipitation, supercritical-assisted atomization, microsphere precipitation, high speed planetary ball mill, and ultrafine milling (Chien, Li, Lee & Chen, 2013; Gimeno, Ventosa, Boumghar, Fournier, Boucher & Veciana, 2006; Reverchon & Antonacci, 2006; Yao, Peng, Yin, Xu & Goosen, 1995; Zhang, Zhang, Jiang & Xia, 2012; Zhang, Zhang & Xia, 2014). We identified cryomilling as an optimal method for generating a CS powder based on previous studies that show it preserves functional properties of proteins and starches by reducing the energy input needed to fractionate particles into smaller sizes (Dhital, Shrestha, Flanagan, Hasjim & Gidley, 2011; Ehmer, 2010; Tran, Shelat, Tang, Li, Gilbert & Hasjim, 2011). Additionally, cryomilling overcomes the soft ductile nature of CS, which reduces the effectiveness of traditional milling techniques in micronizing soft materials like CS (Garmise et al., 2006; Saleem & Smyth, 2010).

In the present study, two cryomilling techniques were tested, cryo-jet and cryo-ball milling. Ball and jet mills were chosen for this study since they are the only milling machines commonly used to reduce particles to 5 μm or less in dry conditions. (Vatsaraj et al 2003) Both ball and jet mills are thought to reduce particle sizes using the same mechanism(s), which is by breaking particles along cracks or fractures that already exist at the micro- or nanoscale. Although jet milling is the most commonly used technique for producing particles in the lower micrometer range and is the gold standard for manufacture of inhalable particles of small molecular drugs, this study is the first reported use of cryo-jet milling for micronizing CS (Ehmer, 2010). Once the optimal CS powder formulation was identified, defined as the micronized powder with the smallest mean particle size, it was sterilized with NtNP to form a sterile, pharmaceutical grade CS powder. Physicochemical properties of the CS were characterized before and after both milling and NtNP sterilization.

2. MATERIALS AND METHODS

2.1 Reagents

All chemical reagents including glacial acetic acid, lactic acid, sodium acetate, hydrochloric acid, sodium hydroxide, and tryptic soy agar were obtained from Sigma Aldrich or Fisher Scientific and were of analytical grade.

2.2 Cryomilling

CS derived from crab shells in the form of 1–10 mm flakes was obtained from Scion Biomedical, Inc. (Miami, FL). CS flakes were filtered with fine mesh to remove flakes > 8 mm in diameter and approximately 30 g of filtered CS flake was placed in a 25 mL zirconium oxide jar. Six milling balls made of either zirconium oxide (10 mm diameter) or stainless steel (5 mm) were added to the jar and the CS flakes were subsequently milled under cooling with liquid nitrogen at 30 Hz for up to 30 minutes in the Retsch CryoMill system (Verder Scientific, Inc., Newtown, PA). For cryo-jet milling, CS flakes were filtered, as described above, to remove flakes > 8 mm, 470 g of filtered CS was placed in the Micron-Master jet pulverization system (The Jet Pulverization Co., Inc., Moorestown, NJ), cooled with liquid nitrogen, and milled with a jet stream of liquid nitrogen for 30 minutes.

2.3 Particle Sizing

Particle size analyses were performed on powders by suspending CS in water and measuring particle size with a Horiba LA-930 laser diffraction analyzer (HORIBA Instruments, Inc., Irvine, CA).

2.4 Degree of Deacetylation

Degree of deacetylation (DD) of both CS flake and powder was determined by conductometric titration. Conductometric titration was performed by dissolving dried CS sample of known mass (about 0.100 g) into 10 mL of 0.1 N hydrochloric acid (HCl) then 90 mL of distilled water. The CS solution was then titrated with a standard 0.1 N sodium hydroxide (NaOH) solution using a 10 mL buret while the solution conductivity was monitored as a function of the volume of NaOH added with an Orion Benchtop Conductivity Meter (Model 162) equipped with an Orion Conductivity Cell (Model 013030) (Raymond, Morin & Marchessault, 1993). During the titration, the temperature of the solution was kept constant (30 °C) by using a water bath since the conductivity is a function of temperature. In a typical conductometric titration curve, there are two deflection points. The first deflection point corresponds to the neutralization of excess H⁺ ions of the strong acid, HCl. After all excess H⁺ ions are neutralized, then the neutralization of the weak acid, the ammonium salt in CS, starts. After the ammonium is completely neutralized, the conductivity again goes up with a higher value of slope due to the excess of OH⁻ ions of NaOH added, which is the second deflection point. Thus, the range between the first and the second deflection points corresponds to the neutralization of the protonated amine groups of CS. As a result, the number of moles of NaOH used between the first and second deflection points equals the number of moles of amine groups of the CS sample. The percent DD was calculated by the following equation:

$$\%\text{DD}=\frac{(v2-v1)\times M_{\mathit{NaOH}}\left(\frac{\mathit{mol}}{L}\right)\times \frac{161.16g}{\mathit{mol}}}{\mathit{Mass}of\mathit{chitosan}\mathit{sample}(g)}\times 100$$ (1)

where (v₂-v₁) is the difference in volume in liters between the two deflection points, M_NaOH is the molarity (mol/L) of standard NaOH solution, and 161.16 is the molar mass of chitosan.

2.5 Morphologic Characterization

CS particles were imaged via scanning electron microscopy (SEM) using a Phenom G1 (Model 800-03103-02) (FEI Co., Netherlands) microscope. Prior to imaging, a small amount of CS powder or flake was fixed on conductive carbon tape and mounted on the support. The sample was then sputtered with an approximately 6 nm layer of gold/palladium (Au/Pd) using a Quorum Technologies SC7620 Mini Sputter Coater (Laughton, East Sussex, UK), which deposits 10 nm coating/45 sec.

2.6 Hygroscopicity

Hygroscopicity of CS flake and powder was measured by loss on drying using a 120VAC Moisture Balance IL-50.001 (Summit Measurement Inc., South Deerfield, MA), an automatic moisture balance with a capacity of 50 g and resolution of 0.001 g. CS samples were placed on the balance and the masses before and after drying via vacuum were recorded. Loss on drying was calculated as the percent difference between the initial and final mass.

2.7 Molecular Weight

Viscosity of CS samples was measured using a 2–10 centistoke glass capillary ASTM 1 Ubbelohde viscometer (Fisher Scientific, Pittsburgh, PA). Viscosity measurements were performed by dissolving chitosan in a solvent containing 0.3 M acetic acid and 0.2 M sodium acetate (0.3 M AcOH/0.2 M NaOAc) to a final CS concentration of 1–2 mg/mL (0.1%–0.2%). The AcOH protonates the amino group which solubilizes CS and the NaOAc salt screens electrostatic interactions between different CS chains. The 0.3 M AcOH/0.2 M NaOAc solvent was chosen since previous studies have shown it prevents aggregate formation, unlike other solvents like 0.1 M AcOH/0.2 M NaCl (Rinaudo, Milas & Dung, 1993; Roberts & Domszy, 1982). Efflux times were measured for each of 3 concentrations (1 mg/mL, 1.5 mg/mL, and 2 mg/mL) and the following viscosities were determined using the given equations:

$$\text{Relative}\text{viscosity}{\mathrm{n}}_{\text{rel}}=\frac{t_{CS}}{t_{\mathit{sol}}}$$ (2)

$$\text{Specific}\text{Viscosity}{\mathrm{n}}_{\text{sp}}={\mathrm{n}}_{\text{rel}}-1$$ (3)

$$\text{Reduced}\text{viscosity}{\mathrm{n}}_{\text{red}}=\frac{n_{sp}}{c}$$ (4)

where t_CS represents efflux time for chitosan, t_sol represents efflux time for the solvent, and c represents chitosan concentration in g/mL. Intrinsic viscosity ([n]) was calculated using the Huggins plot in which the reduced viscosities for the 3 chitosan concentrations were extrapolated to infinite dilution (i.e. zero concentration). Intrinsic viscosity is related to molecular weight by the Mark-Houwink equation:

$$[\mathrm{n}]={\text{KM}}^{\mathrm{\alpha }}$$

where K and a are constants for the specific polymer, solvent, and temperature combination. The K and a values for CS with a %DD of approximately 80% and the solvent 0.3 M AcOH/0.2 M NaOAc at 25 °C are 0.074 and 0.80, respectively (Brugnerotto, Desbrières, Roberts & Rinaudo, 2001).

2.8 FTIR

Molecular composition of CS flake and powder was assessed via Fourier Transform Infrared Spectroscopy (FTIR). To generate FTIR spectra, films of CS salts were cast by dissolving the CS flake or powder in 1% acetic acid. The films were then converted to the free amine form of CS by washing films in alcoholic base prior to imaging with a Nicolet iS50 FT-IR instrument (Thermo Fisher Scientific, Madison, WI) using a diamond attenuated total reflectance (ATR) attachment and 32 scans at 4 cm⁻¹ resolution.

2.9 Plasma Sterilization

CS samples were sterilized via NtNP treatment with either atmospheric pressure plasma (ApP) or low-pressure plasma (LpP) systems. The ApP instrument is a custom built instrument housed at North Carolina State University that has been described in detail previously and was operated at 4.8 kW power and 5 kHz frequency (Cornelius, 2007). Helium was used as the carrier gas at a flow rate of 10 L/min. The LpP instrument is the IoN40 plasma system (PVA-TePla America, Corona, CA) and the gas used was medical grade nitrogen, power was 500 Watts, pressure was maintained at 300 mTorr, and gas flow was 0.5 L/min.

Prior to NtNP treatment, CS samples were weighed and placed into paper-backed sterilization pouches. The pouches used in the ApP instrument were sealed and subsequently flushed with 100% nitrogen gas whereas the pouches used in the LpP instrument were left unsealed. The CS-containing pouches were placed into the ApP instrument or directly on the metal tray of the IoN40. For the LpP, the chamber was evacuated down to <100 mTorr and flushed with 100% nitrogen gas 3 times prior to beginning the NtNP treatment. Evacuating and flushing the chamber served to ensure the CS-containing pouches were filled completely with 100% N₂ gas instead of air.

Previous studies showed that sterility doses, defined as a 6-log reduction in bacterial spores, were achieved at 15 mins with both the ApP and LpP instruments when using nitrogen gas. As is commonly done in medical device and pharmaceutical manufacturing, an overkill dose of 2X the sterility dose, 30 minutes in this case, was used in these studies. Sterility of each NtNP-treated CS sample was confirmed by diluting each sample 40X with pyrogen-free water to overcome the anti-microbial properties of CS and subsequently cultured in standard polystyrene petri dishes containing tryptic soy agar at 37°C for 72 hours. Colony-forming units were enumerated by manually counting each culture plate.

2.10 Endotoxins

Endotoxins are measured in endotoxin units (EU), a unit of biological activity based on the United States Pharmacopoeia Reference Endotoxin Standard, since different types of endotoxins elicit varying degrees of immune responses. Endotoxins were extracted from CS by incubating CS samples in a 40X volume of pyrogen-free water (w/v) for 2 hours at 37°C. The extracts were then syringe filtered through a 0.22 μm mixed cellulose esters filter (EMD Millipore, Inc., Darmstadt, Germany) to remove any residual CS that can cause false positives with Limulus amebocyte lysate (LAL) assays. The endotoxin concentration in the filtrate was subsequently quantitated using a handheld LAL device (Endosafe PTS device, Charles River Laboratories, Charleston, South Carolina) and FDA-certified 0.5–0.005 EU/mL PTS cartridges. The LAL assay is the current gold standard method for quantitating endotoxins in drugs and medical devices.

3. RESULTS AND DISCUSSION

3.1 Particle Size Analysis

Cryomilling raw CS flake with either a cryo-ball or a cryo-jet mill resulted in a fine, white powder, as shown in Figure 1, consisting of mostly <100 μm particles. Average particle size and size distribution, including median and the diameter larger than 95% of the particles (D95), for each milling machine and parameter are presented in Table 1. Cryo-ball milling CS flakes for 15 minutes with 10 mm zirconium oxide balls generated CS powder with the largest mean particle size. Doubling the grind time to 30 minutes reduced the mean particle size by 21% and D95 by 15%. Using 50% smaller stainless steel balls reduced the mean particle size and D95 by an additional 57% and 55%, respectively. The smallest mean particle size generated by cryomilling (28.5 μm) was 24% smaller than the mean particle size generated by a standard ball mill after 8 hours (37.53 μm) in a previous study (Zhang, Zhang & Xia, 2014).

Figure 1.

Macroscopic images of CS flake (left) and cryo-jet milled powder (right).

TABLE 1.

Mean, Median and D95 values for cryomilled chitosan.

	Mean	Median	D95
Cryo-ball Mill (10 mm balls; 15 mins)	84.5 μm	71.0 μm	209.7 μm
Cryo-ball Mill (10 mm balls; 30 mins)	66.8 μm	52.2 μm	179.0 μm
Cryo-ball Mill (5 mm balls; 30 mins)	28.5 μm	20.4 μm	80.3 μm
Cryo-jet Mill (30 mins)	16.05 μm	15.62 μm	24.24 μm

Cryo-jet milling CS for 30 minutes produced the smallest particles, with a mean particle size (16.05 μm) that was 44% smaller than the smallest mean particle size produced by cryo-ball milling (28.5 μm). Although the cryo-jet mill proved to be the superior method for milling CS into powder in the present study, a previous study showed that CS powder with a particle size distribution similar to cryo-jet milling can be produced with a cryo-ball mill when coupled with thermocatalytic destruction of CS (Laka & Chernyavskaya, 2006). Furthermore, another study found that CS flakes can be reduced to submicron sizes using an ultrafine milling technique (Zhang, Zhang, Jiang & Xia, 2012).

3.2 Viscosity and Molecular Weight

Cryomilling reduced the intrinsic viscosity and molecular weight (MW) of CS as shown in Figure 2. This reduction is a result of scissions of the chitosan polymer chain. A detailed discussion of this mechanism is provided in section 3.5. Since MW has a dramatic influence on the biological properties of CS, additional studies are needed to understand the influence of powders with different particle size distributions on biologic properties (Kumirska, Weinhold, Thöming & Stepnowski, 2011).

Figure 2.

Effect of cryo-jet milling on intrinsic viscosity (A) and molecular weight (B) of CS. Flake is raw CS and powder is milled CS.

3.3 Degree of Deacetylation

Changes in the DD are shown in Figure 3. The 9.6% reduction in DD after cryo-jet milling appears to be another consequence of the scissions in the CS polymer chain that occur during cryomilling. Specifically, we postulate that chain breaks lead to changes in functional groups in CS that may lead to chemical reactions. However, other mechanisms are possible. For example, high temperatures are known to occur, albeit for very short times (< 10⁻⁶ seconds), at the fracture sites during cryo-jet milling and this might cause slight caramelization of the CS (Rumpf, 1973; Weichert & Schönert, 1976). Thus, additional studies are needed to better understand the mechanism(s) responsible for this change in %DD with cryomilling.

Figure 3.

Effect of cryo-jet milling on degree of deacetylation of CS. Flake is raw CS and powder is milled CS.

3.4 Hygroscopicity

The hygroscopicity of CS was increased by 25% after milling with loss on drying of the cryo-jet milled CS at 11.12% compared to 8.34% for the CS flake, reflecting the increase in surface area of the cryomilled material. The increase in hygroscopicity of CS after cryomilling is similar to the % loss on drying for the thermocatalytic ball milled CS powder reported by Laka and Chernyavskaya. The high hygroscopicity of CS powder must be taken into account when making CS solutions, which are commonly used for biomedical, and especially pharmaceutical, applications (Ahmadi, Oveisi, Samani & Amoozgar, 2015; Heffernan, Zaharoff, Fallon, Schlom & Greiner, 2011).

3.5 Morphologic Characterization

Morphologic characterization of the micronized CS via SEM imaging showed highly irregular particle shapes after cryo-jet milling, as shown in Figure 3. These irregular particle shapes are thought to be a result of the fact that most polymers at room temperature exhibit considerable tensile toughness or work to rupture, compared to glassy, brittle materials such as inorganic glasses. The appearance of the milled CS suggests the CS flakes are torn apart in a random fashion, leaving highly diverse particle shapes and sizes. The majority of particles appeared either sphere-like or spindle shaped. The size of these shapes is small enough that individual polymer chains are likely broken in the milling process, which leads to the reduced intrinsic viscosity and MW shown in Figure 2. For example, if we take 5.15 x 10⁻¹⁰ m as the diameter of the glucopyranose sugar ring, and assume that there is a degree of polymerization of 1000, then the CS chain is 0.5 μm in length and the particles range in diameter down to 16 μm and smaller.

3.6 FTIR

FTIR spectra for CS flake and powder are shown in Figure 4 and reveal changes in the molecular composition of CS after cryomilling. Specifically, the amine peak at 3400 cm-1 is lower in the CS powder compared to the CS flake, which correlates with the conductometric titration data and indicates loss of amine groups (is this the best way to state what the peak reduction indicates). There is no change in the amide I peak at 1640 cm-1, but the CS powder shows a rightward shift at 1100 cm-1 compared to the CS flake. The significance of this peak shift is not currently understood.

Figure 4.

SEM images of CS flake before cryo-jet milling at 500X magnification (A) and resultant CS powder after cryo-jet milling at 1000X (B) and 5000X (C) magnification.

3.7 Plasma Decontamination

Plasma treatment resulted in a sterile CS with endotoxin levels of 1.26 EU/g (ApP) and 2.46 EU/g (LpP). This is a very low level of endotoxins. Most medical or pharmaceutical grades of CS contain > 100 EU/g, although most are certified to be < 100 EU/g. Plasma sterilization also resulted in a significant reduction in viscosity of both flake and powder forms of CS, as shown in Figure 6.

Figure 6.

Changes in reduced viscosity of CS flake and cryo-milled powder after treatment with sterility and overkill doses of NtNP using LpP and ApP systems.

3.8 Solubility

In the process of making CS hydrogels for the tests described in previous sections, it was noted that cryomilled CS powder solubilized significantly faster and better in dilute acids than CS flakes. Thus, milling increased the solubility of CS. This is again a result of the increased surface area of the milled CS compared to the raw CS flake.

3.9 Biological Properties

Our lab has previously shown in a rat model of arterial and venous hemorrhage that the hemostatic properties of CS are preserved after NtNp-sterilization (Crofton, Chrisler, Hudson, Inceoglu, Petersen, & Kirsch, 2016). We have also found that micronized, NtNP-sterilized chitosan retains its drug delivery efficacy in vivo (unpublished data). Specifically, we treated mice bearing orthotopic bladder tumors with micronized, NtNP-sterilized CS and the anti-tumor cytokine IL-12 (CS+IL-12). We found there was no statistical difference in the anti-tumor efficacy of CS+IL-12 when the CS is micronized and NtNP-sterilized as described in the present study compared to its native form.

4. CONCLUSION

Cryo-ball and cryo-jet milling are effective methods for producing micron-sized CS particles and increasing the surface to volume ratio of CS. Cryo-milling also increases the hygroscopicity and decreases the %DD, intrinsic viscosity, and therefore the molecular weight, of CS. Cryo-jet milling produced a 44% smaller mean particle size than cryo-ball milling after 30 minutes of milling. However, in the case of cryo-ball milling, smaller grinding balls may produce much smaller mean particle sizes, potentially within the range of cryo-jet milling, but it is unclear whether there would be any advantage to cryo-ball milling over cryo-jet milling. Plasma-treated CS was sterile with <2.5 EU/g, but NtNP treatment caused a significant reduction in CS MW. This effect was greater with micronized CS powder, which is predicted to be due to the increased surface area of the powder. Future studies must be aimed at elucidating how different particle sizes influence biological properties of CS.

Figure 5.

FTIR spectra of CS flake and powder showing changes in molecular composition caused by cryo-jet milling. Blue represents chitosan flake, red represents chitosan powder. Both flakes and powder were analyzed in the free amine form.

Highlights.

• Chitosan was micronized by cryo-ball and cryo-jet milling.

• Mean particle size after cryo-jet mill (16.05 μm) was smaller than cryo-ball mill.

• Cryomilling reduced molecular weight by 26% and degree of deacetylation by 9.6%.

• Nitrogen plasma treatment produced sterile chitosan with endotoxins at <2.5 EU/g.

• Nitrogen plasma treatment also reduced molecular weight of chitosan.

Abbreviations

CS

chitosan

NtNP

non-thermal nitrogen plasma

ApP

atmospheric pressure plasma

LpP

low-pressure plasma

DD

degree of deacetylation